ACC Adaptive Cruise Control_2003.pdf

2003

The Program

Order Number

ISBN

Automotive Electrics/Automotive Electronics Batteries Alternators Starting Systems Lighting Technology Electrical Symbols and Circuit Diagrams Automotive Sensors Automotive Microelectronics

1 987 722 153 1 987 722 156 1 987 722 170 1 987 722 176 1 987 722 169 1 987 722 131 1 987 722 122

3-934584-21-7 3-934584-22-5 3-934584-23-3 3-934584-24-1 3-934584-20-9 3-934584-50-0 3-934584-49-7

1 987 722 138 1 987 722 135

3-934584-62-4 3-934584-47-0

1 987 722 175

3-934584-40-3

1 987 722 179

3-934584-41-1

1 987 722 174 1 987 722 164 1 987 722 162

3-934584-39-X 3-934584-38-1 3-934584-36-5

1 987 722 102 1 987 722 159 1 987 722 101 1 987 722 160 1 987 722 105 1 987 722 155 1 987 722 154 1 987 722 161 1 987 722 178

3-934584-26-8 3-934584-27-6 3-934584-28-4 3-934584-29-2 3-934584-30-6 3-934584-32-2 3-934584-31-4 3-934584-33-0 3-934584-34-9

1 987 722 136

3-934584-48-9

1 987 722 103 1 987 722 177 1 987 722 134

3-934584-60-8 3-934584-44-6 3-934584-64-0

1 987 722 165

3-934584-45-4

1 987 722 166 1 987 722 150 1 987 722 132

3-934584-46-2 3-934584-25-X 3-934584-53-5

Diesel-Engine Management Diesel-Engine Management: an Overview Electronic Diesel Control EDC Diesel Accumulator Fuel-Injection System Common Rail CR Diesel Fuel-Injection Systems Unit Injector System/Unit Pump System Radial-Piston Distributor Fuel-Injection Pumps Type VR Diesel Distributor-Type Fuel-Injection Pumps VE Diesel In-Line Fuel-Injection Pumps Gasoline-Engine Management Emissions-Control Technology for Gasoline Engines Gasoline Fuel-Injection System K-Jetronic Gasoline Fuel-Injection System KE-Jetronic Gasoline Fuel-Injection System L-Jetronic Gasoline Fuel-Injection System Mono-Jetronic Spark Plugs Ignition M-Motronic Engine Management ME-Motronic Engine Management Gasoline-Engine Management: Basics and Components Safety, Comfort and Convenience Systems Conventional and Electronic Braking Systems ESP Electronic Stability Program ACC Adaptive Cruise Control Compressed-Air Systems for Commercial Vehicles (1): Systems and Schematic Diagrams Compressed-Air Systems for Commercial Vehicles (2): Equipment Safety, Comfort and Convenience Systems Audio, Navigation and Telematics in the Vehicle

The up-to-date program is available on the Internet at: www.bosch.de/aa/de/fachliteratur/index.htm

The Bosch Yellow Jackets Edition 2003

The Bosch Yellow Jackets

AA/PDT-04.03-En

Expert Know-How an Automotive Technology

Safety, Comfort and Convenience Systems

ACC Adaptive Cruise Control

ACC

Order Number 1 987 722 134

Expert Know-How an Automotive Technology

ISBN-3-934584-64-0

Automotive Technology

• • • •

Radar ranging, radar modules ACC electronic module, system network Operation, object detection and selection, control Data processing and transmission

Robert Bosch GmbH

Imprint

Published by: © Robert Bosch GmbH, 2003 Postfach 1129, 73201 Plochingen. Automotive Aftermarket Business Sector, Department AA/PDT5. Product Marketing, Diagnostics & Test Equipment. Editor-in-Chief: Dipl.-Ing. (FH) Horst Bauer. Editorial team: Dipl.-Ing. Karl-Heinz Dietsche, Dipl.-Ing. (FH) Thomas Jäger, Authors: Prof. Dr. rer. nat. H. Winner, Dr.-Ing. K. Winter, Dipl.-Ing. (FH) B. Lucas, Dipl.-Ing. (FH) H. Mayer, Dr.-Ing. A. Irion, Dipl.-Phys. H.-P. Schneider, Dr.-Ing. J. Lüder, Dr.-Ing. E. Zabler, Dr. rer. nat. V. Denner, Dr.-Ing. M. Walther and the editorial team in co-operation with the responsible technical departments of Robert Bosch GmbH.

Unless otherwise stated, the above are all employees of Robert Bosch GmbH, Stuttgart. Reproduction, duplication and translation of this publication, either in whole or in part, is permissible only with our prior written consent and provided the source is quoted. Illustrations, descriptions, schematic diagrams and the like are for explanatory purposes and illustration of the text only. They cannot be used as the basis for the design, installation, or specification of products. We accept no liability for the accuracy of the content of this document in respect of applicable statutory regulations. Robert Bosch GmbH is exempt from liability, Subject to alteration and amendment. Printed in Germany. Imprimé en Allemagne. 1st edition, April 2003. English translation of the 1st German edition dated: April 2002 (1.0)

Robert Bosch GmbH

ACC Adaptive Cruise Control Robert Bosch GmbH

Robert Bosch GmbH

Contents

4 4 4 6

System overview Benefits and applications Operation Components

45 Future developments 45 Sensor technology 45 Function 48 Frequently asked questions

7 Radar ranging 7 Physical principles of measurement 10 Radar modules

52 Glossary of ACC terms

15 15 17 19

ACC sensor & control unit Physical structure Adjustment Electronic hardware

54 54 55 58 60 64

22 22 23 23 24 24

System network System architecture Drivetrain control Brake control Corner sensing systems Safety concept

66 66 66 66 69 70

28 Controls and displays 28 Function 28 Design and method of operation 32 32 33 34

Object detection and selection Radar-signal processing Object selection Course prediction

38 38 39 42

ACC control sequence Control-unit structure Control-unit functions Limits of function

Sensors Automotive applications Yaw-rate sensors Steering-wheel-angle sensors Acceleration sensors Wheel-speed sensors Data processing in the vehicle Requirements Microcomputer ECU Complete system Severe demands on electronic systems

70 History of development 72 Data transfer between automotive electronic systems 72 System overview 72 Serial data transfer (CAN) 77 Prospects 78 Index of technical terms 78 Technical terms 80 Abbreviations

Robert Bosch GmbH

Mentally, driving is a highly demanding activity – a driver must maintain a high level of concentration for long periods and be ready to react within a split second to changing situations. In particular, drivers must constantly assess the distance and relative speed of vehicles in front and adjust their own speed accordingly. Those tasks can now be performed by Bosch’s electronic Adaptive Cruise Control (ACC) system, which is an extension of the conventional cruise control system. Like a conventional cruise control system, ACC keeps the vehicle at a set constant speed. The significant difference, however, is that if a car with ACC is confronted with a slower moving vehicle ahead, it is automatically slowed down and then follows the slower vehicle at a set distance. Once the road ahead is clear again, the ACC accelerates the car back to the previously set cruising speed. In that way, ACC integrates a vehicle harmoniously into the traffic flow. Of course, the driver can always override the automatic control system at any time. When the accelerator is pressed, the vehicle responds in the normal way. And when the accelerator is released, the ACC adjusts the vehicle’s speed back to the set cruising speed. A brief press of the brake pedal is enough to deactivate the ACC. This publication in the “Bosch Technical Instruction” series provides comprehensive information about the system structure, its components and method of operation, and the different approaches to this interesting topic.

Robert Bosch GmbH 4


System overview

ACC Adaptive Cruise Control ACC (Adaptive Cruise Control) simplifies the task of driving a car because it relieves the driver of the mentally demanding task of keeping a check on the car’s speed, thus enabling relaxed and safe driving behind slower vehicles.

System overview Benefits and applications It is that “following slower vehicles” function in particular that is perceived by the driver as a major gain in convenience and as a substantial mental relief. The side effects of that function also include improved road safety due to greater distances between vehicles and greater relaxation on the part of the driver. The main area of application for ACC (Figure 1) is on motorways and multilane trunk roads with light to relatively high traffic densities. Although use of the system in traffic jams and urban conditions may be desirable, at present this remains an objective for future systems as the technical difficulties associated with such a function demand considerable further development of sensor capabilities (see also the chapter “Future Developments”).

6 2 C

6

N

A

6

4 5

6

6

7 1 6

AN C

Fig. 1 1 ACC sensor & control unit 2 Engine-management ECU 3 Active intervention in braking via ESP 4 Controls and display 5 Intervention at the engine via EM system with ETC (Electronic Throttle Control or EGAS) (gasoline engines) or EDC (diesel engines) 6 Sensors 7 Transmission intervention (optional)

Bosch ACC Adaptive Cruise Control

3

æ UAE0732-1Y

1

Operation Every Adaptive Cruise Control (ACC) system incorporates the standard, basic cruise control function whereby the vehicle is held at a constant speed selected by the driver. That function, also referred to in this manual as “set speed control”, is employed primarily when there is no vehicle in front forcing the driver to adopt a slower speed than the set speed. It also comes into effect when the vehicle in front of the car with ACC is travelling faster. The essential difference in function between ACC and standard cruise control is that a car with ACC will safely follow a vehicle that is travelling at a slower speed than the cruising speed to which the driver has set the ACC (Figure 2). If the vehicle in front is travelling at a constant speed, a car fitted with ACC will follow it at the same speed and a virtually constant distance. That is because the distance between the two vehicles is – within a broad speed range at least – virtually proportional to their speed. That “constant time gap” is equal to (regardless of speed) the time required for the most forward point on the car fitted with ACC to reach the momentary position of the rearmost point on the vehicle in front. Changeover between the two main functions of the system is performed automatically

Robert Bosch GmbH ACC Adaptive Cruise Control

5

260

240

220

220

180 200

260

220

240

æ UFS0011Y

40

60

140 160

180 200

120 100 80

140 160

120 100 80

140 160

180 200

40

60

100 80

120

140 160

180 200

ACC function. The main application for ACC is in light to relatively heavy traffic conditions

120 100 80

2

System overview

Fig. 2 a Car with ACC approaching a slower vehicle when travelling at a constant speed (set cruising speed) b Car with ACC reduces speed to match speed of slower vehicle in front c After vehicle in front turns off, car with ACC accelerates back to set speed originally selected


System overview

without intervention by the driver. If the situation should change, e.g. as a result of another vehicle pulling into the gap between the two vehicles from another lane and thereby itself becoming the vehicle in front of the ACC car, the necessary readjustment is also carried out automatically without the need for driver input. In order to adjust the vehicle’s speed, the ACC system electronically opens the throttle within defined limits by means of the enginemanagement system in order accelerate or electronically applies the brakes in order to decelerate. Components In order to detect vehicles in front and to measure the distance and the speed of such vehicles, the ACC system requires a ranging sensor. In Europe, this takes the form of a microwave radar transmitter/receiver. It is incorporated within the same unit that also performs the control functions and is therefore referred to as the ACC Sensor & Control Unit (ACC SCU).

3

Adjustment and control of vehicle speed is effected by means of existing subsystems which, however, are modified for the purposes of ACC (Figure 3): Engine-management system with torque control, e.g. Motronic with ETC (Electronic Throttle Control or EGAS) (gasoline engines) or EDC (diesel engines) Electronic brake modulation system with active brake-pressure-increase capability (generally based on ESP Electronic Stability Program). In order to ensure reliable functioning of the ACC (including when cornering), the ESP provides other important sensor signals relating to dynamic handling parameters in addition to the deceleration facility. For ultimate driving convenience, the combination of ACC with an automatic gearbox is also desirable. The system incorporates special switches and displays to enable the driver to activate functions and set the desired speed and time gap. The instrument cluster then indicates the current settings and other ACC-related information.

Basic structure and components of ACC system

Radar-sensor check unit Sensors for yaw rate, lateral acceleration, wheel speed and steering angle

ECUs

Vehicle movement

Motronic Gearbox Object selection

Distance controller

ESP

Object detection Radar sensor

Engine Gearbox Brake

æ UAE0733-1E

6


7

echo. If the signal is reflected back directly, that time interval, τ, is represented by formula below, where d is the distance to the reflector and c is the speed of light.

Radar ranging Physical principles of measurement

Reflection RADAR (Radio Detecting and Ranging) emits an electromagnetic beam from an antenna. Objects in the path of the beam reflect it back to the radar antenna. RADAR beams can be reflected by any electrically conductive material, and therefore particularly by metal-bodied motor vehicles. For that reason, RADAR is ideally suited to use as a distance measuring device for cars. In addition, RADAR is superior to optical systems in poor weather conditions (e.g. fog and rain) because of its longer wavelength. Other conceivable ranging devices (e.g. optical rangers) require objects to have surfaces with good light-reflecting properties. Objects whose reflective surfaces are not clearly visible or are dirty cannot be reliably detected. Echo timing Measurement of distance by all RADAR systems is based on direct or indirect timing of the interval between transmission of the RADAR signal and reception of the signal

τ = 2d/c At a distance, d, of 150 m and assuming c ≈ 300,000 km/s, the time interval, τ ≈ 1.0 µs. Doppler effect If the object detected is moving relative to the RADAR transmitter/receiver at a relative speed, υrel, the signal echo undergoes a frequency shift, fD, relative to the transmitted signal. At the speed differences encountered in this scenario (Figure 1), that frequency shift is given by fD = –2fC · υrel/c where fC is the signal carrier frequency At the RADAR frequency normally used for ACC of fC = 76.5 GHz, a frequency shift of fD ≈ –510 · υrel/m results, i.e. 510 Hz at a relative speed of –1 m/s (distance to detected object is closing).

Use of radar ranging system (Doppler effect)

υ2

fD

d Licht !

fC

υ1 υrel = υ2- υ 1

æ UFS0001Y

1

Radar ranging

Fig. 1 d Object distance fC Carrier frequency fD Differential frequency υ1 Road speed of vehicle 1 υ2 Road speed of vehicle 2 υrel Relative speed

Robert Bosch GmbH 8


2

Radar ranging

Frequency modulation Direct measurement of the time interval is complicated. For that reason, it is normally measured indirectly. One of the methods used is known as FMCW (Frequency Modulated Continuous Wave). Instead of timing the interval between transmission of the signal and reception of the echo, an FMCW RADAR system compares the frequencies of the transmitted signal and the echo. The basic prerequisite for a meaningful measurement is a transmission frequency that alters over time. This is normally achieved by using a VCO (Voltage-Controlled Oscillator) to modulate the transmission frequency at a linear rate of change, m = df/dt (Figure 2a). By the time the echo signal is received after an interval of τ = 2d/c, the transmission frequency has altered by the differential frequency fD = τ · m. Thus the time interval, and therefore the distance of the object, can be established indirectly by determining the frequency difference between the transmitted and received signals. For its part, the differential frequency can be obtained by means of a mixer and a low-pass filter. For the purposes of determining the frequency, the signal is digitized and converted into a frequency spectrum with the aid of an FFT (Fast Fourier Transform). A peak in the spectrum at fD (Figure 2b) corresponds to an object distance of

Modulation of transmission frequency (with positive frequency gradient)

a

∆B B

Frequency B

C

B

B

D

ges1

τ Time J b

Power 2

∆B

Fig. 2 a Positive addition of Doppler shift to differential frequency b Effect of Doppler shift c Distance versus relative speed Carrier frequency (modulated transmission frequency) fD Differential frequency ∆f Doppler shift fges1 = fD – ∆f Overall frequency shift (positive frequency gradient) fp Positive frequency shift due to Doppler effect τ Time interval

B

p

B

Differential frequency

D

d = fD · c/2m

B

D

c

@

Distance

æ UFS0012E

Relative speed υrel

fC

The differential frequency information contains, however, not only the interval component but also the Doppler-shift component, which combine to give the overall frequency difference described by the formula fges 1 = fD – ∆f. This circumstance means that in the first instance there is an ambiguity with regard to interpretation. That is because in addition to an individual differential frequency, a linear combination of object distance and relative speed has to be taken into account, which is represented on the “Distance versus relative speed” graph by a straight line (Figure 2c).


a

Frequency B

∆B

B

C

B

ges2

Time J b ∆B

B

Echo-angle detection In order to determine the angle at which the RADAR locates an object, multiple RADAR “lobes” are transmitted and analysed.

Fig. 3 a Negative addition of Doppler shift to differential frequency b Effect of Doppler shift c Distance versus relative speed

B

D

n

Differential frequency

B

D

Positive frequency gradient

@

υrel

Negative frequency gradient

Distance

Carrier frequency (modulated transmission frequency) fD Differential frequency ∆f Doppler shift fges2 = fD + ∆f Overall frequency shift (negative frequency gradient) fp Positive and fp Negative frequency shift due to Doppler effect d ~ fp + fn Object distance υrel ~ fp – fn Relative speed fC

c

Relative speed υrel

Each RADAR beam has a characteristic “antenna pattern”. For a defined target, the amplitude of the signal echo has a characteristic dependency on the angle at which the signal is received by the RADAR (Figure 4). On the other hand, the reflective properties of a located target are unknown. Thus no definite conclusion as to the angle of incidence of the signal can be drawn from the information provided by a single RADAR beam.

9

Modulation of transmission frequency (with negative frequency gradient)

æ UFS0013E

However, this method must also be applicable when there is more than one target object present. To make that possible, the procedure has to be extended by the addition of more modulation cycles so that unambiguous allocation of target frequencies to objects is possible.

3

Power 2

That ambiguity can be resolved by the use of multiple FMCW modulation cycles with differing rates of frequency change. When the transmission frequency is modulated with a different rate of change, although there is still an ambiguity between distance and relative speed, that ambiguity is expressed by a different linear relationship on the “Distance versus relative speed” graph. If for example, the second frequency gradient used is the inverse of the first frequency gradient, the relationship illustrated by Figure 3a results. A negative frequency gradient accordingly results in addition of the Doppler shift, ∆f, to the differential frequency, fD, that results from the time interval (Figure 3b). The linear progression of distance plotted against relative speed corresponding to the negative gradient now intersects the straight line for the positive frequency modulation gradient. The point of intersection of the two lines then provides the correct values for object distance and relative speed (Figure 3c).

Radar ranging

Robert Bosch GmbH 10


4

Radar ranging

Radar modules

Antenna patterns of various RADAR beam lobes

Function The actual ranging unit of the ACC SCU is the RADAR transmitter/receiver unit or RTC (RADAR Transceiver). Its functions are as follows:

a 2

3

Generation of high-frequency radar signals in the range 76...77 GHz Separation and emission of three simultaneous RADAR beam lobes Subsequent reception of the echoes of those beams reflected from target objects, and Conditioning of these echo signals for subsequent digital electronic signal processing.

Signal amplitude

1

0 Angle b

10

The transceiver also contains an electronic circuit for high-precision stabilisation of the transmission frequency and linear frequency modulation.

1

A Bosch radar unit has the following technical specifications (Table 1):

1 2 3 2/1

Left lobe Center lobe Right lobe Signal amplitude ratio between pairs of beam lobes

0.1

2 /1 3/2 3 /1

0.01 - 8°

- 4°

0 Angle

4°

æ UFS0014E

Fig. 4 a Overlap of antenna patterns b Scanned angular range of RADAR beam lobes

Signal amplitude ratio

100

8°

Adequate conditions for making that judgement are provided, however, by the use of multiple RADAR lobes. As their antenna patterns are sensitive within a different angular range in each case, a comparison of the amplitudes for a signal echo at the various RADAR lobes can provide the basis on which to determine the angle of incidence of the beam. The quality of the amplitude comparison depends on the overlap of the individual antenna patterns (Figure 4a). In order to be able to scan an angular range of 8°, three parallel RADAR lobes with a lobe width of +/– 2° are used (Figure 4b). Table 1

1

Technical specifications of a Bosch radar unit

Range

2...120 m

Detectable relative speed

–50...+50 m/s

Angular range

±4°

Resolution

0.85 m; 1.7 m/s

Scanning rate

10 Hz

Frequency range

76...77 GHz

Mean transmission power

approx. 1 mW

Bandwidth

approx. 200 MHz


Design and method of operation The basic components of the RTC (RADAR transceiver, Figures 5 and 6) are the following: A high-frequency oscillator (Gunn oscillator) for generating the radar signal RADAR transceiver

6

RADAR transceiver (schematic diagram)

A divider circuit for the antenna feed and return-signal mixing An electronic frequency modulator and reference oscillator, and A signal preamplifier.

æ UFS0015Y

5

Radar ranging

+ 8V supply Switch on signal for Gunn-effect oscillator FLL-ASIC frequency control circuit Harmonic mixer

Input for saw-tooth control voltage Frequency monitoring

DRO

Mic.

+ 5V supply

12.65GHz 2 3-channel pre-amplifier ASIC

2 Radar signal, center 2

Lens (Fresnel)

Directional coupler Antenna Mixer feed point

Radar signal, left

Radar signal, right

Ground

æ UAE0705-1E

VCO Voltagecontrolled oscillator

11



7

Radar ranging

whereby certain semiconductors (gallium arsenide) produce microwave oscillations when subjected to strong electrical fields. The Gunn diode is enclosed in a ceramic casing and fitted in an aluminum oscillator block.

Gunn oscillator (arrangement of components)

1 2

3

4 Fig. 7 11 Fixing screw 12 Supplementary circuit board 13 Insulator 14 Coil spring 15 Ferrite sleeve 16 Bias choke 17 Oscillator body 18 Locating pin 19 Gunn semiconductor element 10 Frequency-tuning pin 11 Power-tuning pin

5 6

7

æ UFS0016Y

8

10

8

11

9

Divider circuit (in situ)

1

2

3 4

5

æ UFS0017Y

Fig. 8 1 Micro printedconductor circuit on quartz glass substrate 2 Wilkinson power divider with two surface resistors 3 Directionally selective signal separator for receive and transmission signals 4 3 antenna patches 5 7 mixer diodes

8

The contact pin for the power supply incorporates filter structures and a resonator disc that, in combination with the rectangular internal cross-section of the block, forms a cavity resonator. The oscillation characteristics of the diode are essentially determined by the geometry of the components surrounding it. Consequently, extremely exacting standards of precision are required in the manufacture of those components. The frequency can be varied within the useful range of 76...77 GHz by altering the applied voltage, which explains why this component is referred to as a voltage-controlled oscillator (VCO). The high-frequency energy generated is passed to a divider circuit via a square waveguide integrated in the oscillator.

Gunn oscillator At the centre of the oscillator (Figure 7) is an electronic semiconductor component made of n-type gallium arsenide that produces electromagnetic oscillations of a very high frequency when a DC voltage is applied to it. It is called a Gunn diode after the American physicist I. G. Gunn. In 1963, he discovered the effect now known as the Gunn effect

Divider circuit and antenna feed The divider circuit (Figure 8) is an electrical printed-conductor circuit with gold conductor tracks on a quartz glass substrate only 0.17 mm thick. First of all, a small proportion of the transmitted power is tapped off and fed into the frequency modulator described further on. The energy is then divided between three separate but identical transmitter/receiver branches. Each of those branches contains a double ring junction and is terminated by a rectangular element known as an antenna patch, the function of which is to transmit and receive a radar beam lobe. On top of the antenna patches there are components made of dielectric material. These pre-focus the emitted energy and beam it onto an antenna lens that, in similar fashion to an optical lens, focusses the radar beam into three overlapping beam lobes, each with an angular width of approximately 4°. A greater degree of focussing would only be


9

Radar ranging

13

Signal path in divider-circuit mixer

1

2

Fig. 9 A Transmission signal B Receive signal

5

1

5

3 2

3

A

possible with a larger antenna lens. As each beam lobe overlaps its neighbour by about half its width, the overall beam scanning width is 8° extending outwards from the radar module on the front of the vehicle. The Bosch system is a monostatic radar system, that means it uses the same antenna arrangement in reverse to receive the radar echoes. This type of system requires less space than a bistatic system with separate transmitter and receiver set-ups. It is therefore better suited to use in a motor vehicle. Mixer The first ring junction in each divider circuit branch splits the power fed to it so that the antenna patch (Figure 9) receives only approximately half the power. The other half is fed to another ring junction. Simultaneously arriving at that point is the radar echo energy received by the antenna patch.

B

æ UFS0002Y

4

This second ring junction in each branch of the divider circuit combines with the antiparallel diodes connected at that point to form a mixer in which an electrical signal is generated from the transmission and received energy. The frequency of that signal is equal to the difference between the transmitted and received frequencies. It is this electrical signal that is actually the useful radar signal. Its frequency, which is in the range of 20...200 kHz, contains the information about the distance ahead and the relative speed of the detected objects. The differences in amplitude between the three branches of the circuit are analysed to determine the echo angle. The electronic circuitry of the radar transceiver receives the useful signal via two signal lines in each case.

4 5

Micro printedconductor circuit on quartz glass substrate Wilkinson power divider with two surface resistors Directionally selective signal separators for receive and transmission signals 3 antenna patches 7 mixer diodes



Radar ranging

Pre-amplifier The overall transmission power of the ACC radar module is only approximately 1 mW. Consequently, the electrical voltages of the useful signals are so small that, before they can be processed, they have to be amplified several million times in a specially designed three-channel integrated amplifier circuit. The amplifier circuit’s frequency-dependent amplification characteristic ensures that echo signals even from far distant objects are reliably processed. Echoes from distant objects produce higher mixed frequencies and lower voltage amplitudes because the more distant the reflective object is, the weaker is the received radar signal. For that reason, they have to be amplified even more. Frequency-control electronics As all the essential information is contained within the frequency of the useful signal, fluctuations in the transmission frequency or inconsistencies in the linear progression of the transmitted-frequency gradient would lead to misinterpretation of the receive signals. The Bosch ACC radar module is therefore equipped with high-speed electronic frequency control which compares the transmitted frequency with the required setting and adjusts it as necessary about one million times a second. At the same time, it ensures that the frequency consistently remains within the range of 76...77 GHz, which are the statutory limits imposed by telecommunications authorities for longer-range motor-vehicle radar systems. In order to perform those functions, the frequency control contains not only the main oscillator but also a reference oscillator in the form of a DRO (Dielectric Resonator Oscillator) with a nominal frequency of 12.65 GHz. This is an electronic resonator circuit consisting of a power transistor and a

dielectric resonator element for stabilising the frequency (like the Quartz circuit in a clock or watch, this oscillator is extremely stable over a prolonged period of time and a wide range of temperatures). The energy from the DRO is fed into a “harmonic mixer” in the divider circuit. This mixes the base frequency multiplied by a factor of six (6 · 12.65 = 75.9 GHz) with a small proportion of the output from the main oscillator, with the result that mixed frequencies of 100...1100 MHz are obtained. That signal is the input for the electronic frequency control. Following further division (the frequencies are still too high for processing by the standard electronics) the signal is fed into a “discriminator” and converted into a voltage that is proportional to the frequency. That is then compared with the required frequency setting which is also provided in the form of a voltage. If there is a difference, the supply voltage of the oscillator is adjusted until the required frequency is obtained. The required frequency itself is, of course, a variable quantity. The signal processing unit specifies its value so as to obtain the transmission frequency alteration rate of 200 MHz per millisecond that is required for interpretation of the return signals. In addition, there are permanently programmed maximum and minimum levels that ensure that even if the frequency control fails, the frequency cannot go beyond the permissible frequency band.


Physical structure

Requirements The purpose of the ACC radar module dictates that it has to be fitted at the front of the vehicle. As a consequence of that, it must satisfy the following requirements: Temperature resistant within the range –40 °C...≥ +80 °C, Proof against splashwater and pressurised steam, Immune to vehicle vibrations due to poor road surfaces, Resistant to stone impact, As small as possible. In addition there must be a facility for adjusting the radar sensor, as very precise alignment of the centre of the radar beam with the centerline of the car is essential to the response of the vehicle when the ACC is active.

Lens Generally, two fundamental physical quantities determine the size of a radar unit (Figure 2): the external dimensions of the antenna system, and its focal length, i.e. the distance of the beam source from the back of the lens. At a given frequency, the diameter of a lens is determined by the desired beam concentration. The greatest possible utilization of the lens area is achieved by positioning the 2

ACC sensor & control unit (ACC SCU)

ACC sensor & control unit (sectional view)

1

æ UFS0004Y

1

15

Components The Bosch ACC electronic module (Figure 1) incorporates not only the actual radar sensor but also the entire electronic circuits for vehicle-speed control. That is why it is referred to as the ACC SCU (ACC Sensor & Control Unit). This module does not require additional fitting space or extra wiring, as all communication between the ACC system and the engine management and brake control systems and the display instruments takes place via the vehicle’s existing data bus (CAN).

2

8

3

7

4

6

5

æ UFS0005Y

ACC sensor & control unit


Fig. 2 1 Circuit board 1 2 Oscillator block 3 Beam sources (rod emitters) 4 Lens 5 Lens heater contact 6 Circuit board 3 7 Circuit board 2 8 Radar transceiver




beam source at the optimum distance from the lens. The lens of the Bosch radar module is made of a special temperature and stone-impact resistant plastic. It is part of the plastic modulecasing cover and also seals the casing against outside conditions. If required, the lens can also incorporate an electric heater element in order to prevent it becoming covered in ice or snow which might interfere with the beam (wet snow in particular has a marked attenuating effect on radar beams). Electronic components The basic electronic circuitry of the ACC SCU is made up of three printed-circuit boards and the radar transceiver unit (Figures 2 and 3).

Radar transceiver unit The radar transceiver unit is mounted directly on circuit board 1, thus keeping the connections as short and as insusceptible to interference as possible. Circuit board 1 Circuit board 1 contains all the components necessary for digital signal processing (calculation of object positions and speeds from raw radar data). Its central component is a digital signal processor. Circuit board 2 On circuit board 2 there is another processor (16-bit microcontroller) which performs all calculations for vehicle speed control, as well as a voltage regulator and other switching and monitoring components. Circuit board 3 Circuit board 3 contains the connector and the driver modules for connection to the vehicle’s electrical system and for communica-

ACC sensor & control unit (ACC SCU), electrical components

76,5 GHz Radar transceiver

SignalProcessing Unit (2 W)

(2 W) Fig. 3 11 Dielectric resonator oscillator (DRO) 12 Rod emitters 13 SRAM 14 Flash 15 16-bit microcontroller 16 Terminal, 5 V (digital) 17 Switch, 3 A 18 MQS connector and CAN transceiver 19 Gunn oscillator 10 CC610 ASIC 11 Switching controller, 4.1 V 12 DSP 56002 13 Regulator, 8 V 14 Terminal, 5 V (analog) 15 K-line interface

1

2

9 Radar transceiver unit

3

CAN Diagnosis

(1 W + 8 W)

4

10

Plus (Term. 87) Plus (Term. 15)

Regulation Processing Unit + voltage regulator

11 12

Circuit board 1

5

13

6

7

Mass 8

14 15

Circuit board 2

Circuit board 3 with connector

æ UFS0003E

3


Outer casing and adjusters The Bosch radar sensor and control unit is housed in a pressure die-cast aluminum casing. The electronic printed-circuit boards are fitted inside it in such a way that the heat generated by the circuit components is dissipated as efficiently as possible. On the outside of the casing are three locating eyes with plastic ball-joint sockets which form the fixings for the module. A screw with specially shaped collar passes through each of these eyes and screws into a plastic thread in the mounting bracket.

ACC sensor & control unit (ACC SCU), physical structure

1

3 4 5 6

8

9

y

1

3

2

x 4

Fig. 5 1 Adjusting screw 1 for vertical alignment 2 Screw 2 (fixed anchor point) 3 Lens 4 Adjusting screw 3 for horizontal alignment x Axis for vertical adjustment y Axis for horizontal adjustment

The positioning of the screws at the corners of the casing allows the module to be tilted in two planes (Figure 5). Adjusting screw 1 tilts the casing in the vertical plane (x axis is the axis of rotation), while screw 3 adjusts the casing in the horizontal plane (y axis is the axis of rotation). Screw 2 acts in both cases as the fixed anchor point and is not adjusted.

Adjustment Adjustment involves two stages: determining the vehicle’s longitudinal axis (centre line), and aligning the radar axis parallel to the vehicle’s longitudinal axis.

7

17

Adjustment of ACC module in horizontal and vertical planes (front view)

In order to correct any inaccuracies in its fitting position on the vehicle, the unit can be aligned with this adjustment mechanism in a similar fashion to a headlamp. The mounting bracket is adapted to the particular vehicle, with some versions having pinioned adjusters in cases where the mounted position does not allow access to the adjusting screws from the front.

2

æ SFS0018Y

4

5

æ SFS0019Y

tion with the CAN bus, plus interferencesuppression chokes and capacitors. Integrated connections effectively combine all three circuit boards into a single electrical module. As those connections are flexible, they allow the individual circuit boards to be folded over one another so as to save space inside the ACC SCU casing (Figure 4).


The vehicle centre line can be determined using conventional wheel-alignment methods.

Fig. 4 1 Lens 2 Pressure equalizer 3 Casing front cover 4 Radar transceiver 5 Rigid folding printed-circuit board 6 12-pin connector for MQS contacts 7 Casing base 8 Adjusting screws (to suit particular vehicle) 9 Sensor bracket




The adjustment accuracy of the ACC sensor, i.e. the accuracy with which the sensor is aligned with the vehicle centre line is very important for the correct functioning of the ACC. Misalignment of the sensor in the horizontal plane can impair the unit’s ability to accurately locate target objects because it results in misinterpretation of the angular position of vehicles in front. As a consequence, approach responses may be impaired and vehicles in other lanes may be taken as target objects. Misalignment of the sensor in the vertical 6

plane can reduce its range and produce errors in echo-angle detection. The degree of adjustment accuracy demanded is determined by the lane prediction and echo-angle evaluation functions and by plausibility algorithms (sequences of computing operations). Misalignment of the sensor has an effect on those functional components that is similar to an offset error. Upwards of approximately 0.3° of horizontal misalignment, the impairment of function starts to be discernible to the driver. The degree of adjustment accuracy demanded should therefore be well inside that limit.

ACC sensor & control unit (ACC SCU), cutaway view

2

3

Fig. 6 1 Lens 2 Radar transceiver 3 Beam sources (rod emitters)

æ UFS0020Y

1


Design and method of operation The digital electronic circuitry (Figure 7) can be subdivided into the RPU (Regulation Processing Unit) and the SPU (Signal Processing Unit). The central component of the SPU is a DSP (Digital Signal Processor). This is a highly integrated electronic component that was originally used in car audio applications. It is ideally suited to performing large numbers of mathematical operations (e.g. multiplication and division) very quickly. It is thus the ideal component for performing the necessary calculations for the object detection, distance, speed, angle and tracking functions. For those computing operations, the signals have to be available in digital form. The task of converting them into the required form is performed by a highly complex CC610 circuit. It was developed by Bosch specifically for ACC signal processing.

Electronic hardware

Digital electronic circuitry Functions The digital electronic circuitry must do the following: Digitise radar signals Perform FFTs (Fast Fourier Transforms) Calculate target-object distance, relative speed and angular position Control vehicle speed (maintain a safe distance from the vehicle in front), predict vehicle course and perform self-diagnosis Exchange data with the ESP (Electronic Stability Program), Motronic, transmission control system and dashboard display via CAN Enable diagnosis via a connector Activate the lens heater under certain conditions Monitor voltages and signals.

ACC radar sensor (processing unit), schematic circuit diagram

V+

+8V power supply

8V regulator

+5V power supply

V+

5V regulator

Gunn-oscillator activation signal 5V regulator

Frequency monitoring

Lens heater EEPROM Flash

SRAM

ADR

16

Left lobe signal 2

Central lobe signal 2

Right lobe signal 2

SDADC 12 bit SDADC 12 bit SDADC 12 bit

16 Bit µC

DSP

A0,1

2

CANL

K line

CTRL

DATA 24

CONTROL

CANH

2

4

Sc 2

3

XTAL

æ UFS0006E

D/A 10 bit

CAN

CC 610

Input saw-tooth control voltage

Diagnosis

7


Ground

Ground Q

SPU

RPU

Q

19




An additional integral DAC (Digital-Analog Converter) generates an incremental voltage gradient. This provides the specified levels for FLL modulation (Frequency Locked Loop, see also the section “Frequency control”) and results in linear modulation of the transmission frequency. While the voltage rises along that gradient, the three mixed radar signals are amplified in the pre-amplifier, digitised with a resolution of 12 bits, filtered and passed through an FFT (Fast Fourier Transform). The FFT enables extremely fast conversion of the timing signals into frequency signals (Figure 8). The DSP controls the timing of the modulation process and obtains the results from the CC610 circuit via the parallel interface. The data is temporarily stored in an SRAM with fast read/write access. Once the two double gradients for frequency modulation have been completed, the mathematical operations described above are performed. The time required for completion of a single 8

cycle is 80...100 ms. The program required for controlling the procedure is stored in a separate SPU Flash EEPROM. The target-object data comprising details of the object attributes “distance”,“speed”,“angle”, etc. is transmitted via the serial interface to the RPU. The function of the RPU is described in a separate section of this manual. The single-chip controller of the RPU incorporates all circuit components such as processor, RAM, CAN controller, ADC (analog-digital converter), counters, and digital interfaces with the EEPROM (erasable, programmable, non-volatile, read-only memory), with the SPU, with the diagnostic module and with the oscillator for timing-pulse generation. The program, which is stored in an integral Flash EEPROM, can also be modified by the vehicle manufacturer after installation in the vehicle, assuming the appropriate interfaces are provided.

Conversion of timing signals to frequency signals using FFT

Voltage

1.5 V

0

-1.5 0

0.0002

0.0006

0.001

s

Time

60

40

20

0 Fig. 8 a Timing signals b Frequency signals

0

10000

30000

20000 Frequency

40000

æ UFS0021E

Voltage level (absolute)

mV

Hz


The analog/digital converters monitor the voltages. If, for example, the monitored power supply voltage drops below a specified level, the ACC function is disabled. The stabilized generated power-supply voltages are also monitored for compliance with specific tolerances. In the event of a fault, the ACC function would be disabled, a “System inactive” signal sent to the display and a fault code stored in the EEPROM. The CAN interface module enables reliable digital data communication with the partner control units on the vehicle. In recent years, the CAN bus has become established as the standard for serial data transmission in vehicle systems. Common transmission rates are 250...500 kBit/s. At such fast transmission speeds, special precautions have to be adopted. The methods employed include the use of suitable filters. They prevent harmonic interference that could adversely affect such things as radio signal reception in the vehicle. In order to enable fault diagnosis in the course of vehicle servicing, any faults that occur have to be stored. The ACC SCU has an integral EEPROM for that purpose. A certain area of that memory is set aside for storing faults. The information stored in that memory area can be read and interpreted using a special diagnostic tester unit connected via a diagnosis interface. In addition, the vehicle manufacturer can program the EEPROM with supplementary, vehicle-specific data. The diagnosis module forms a bidirectional interface with the diagnostic tester. If the diagnostic tester sends the command “Read fault memory”, that instruction is interpreted by the RPU controller. The controller then reads the data from the EEPROM and converts it into a protocol which is readable by the diagnosis tester. The diagnosis module also performs a protective function by shielding the sensitive controller from direct exposure to other vehicle systems.


The control function for the lens heater switches on the heater filament in the lens in cold weather conditions. As the surface of the lens heats up, it prevents the build-up of ice or snow on the lens. Both snow and ice can attenuate the radar beam to a certain degree and thus limit its intended range. Although a monitoring circuit could ensure that under such circumstances the ACC function were disabled, such an arrangement would restrict the availability of the system in poor weather conditions. By employing pulse-width modulation, the lens heater control circuit is a versatile setup that is responsive to temperature and power supply voltage. Voltage regulators In order that the digital and analog components can operate correctly, they require a power supply with a constant voltage. The task of keeping that voltage constant is performed by a number of voltage regulators. The power supply provided by the battery or alternator would destroy the sensitive electronic components. The voltage peaks of ±100 V that occur and a superimposed AC voltage of ±2 V have to be filtered out because of their potential for causing malfunctions. In addition, the ACC SCU has to be protected from the effect of inadvertent battery pole reversal or the vehicle being started with a 24 V car battery. In order to dissipate the power loss it is necessary to share this task between two voltage regulators. These provide the power supply for the analog and digital components of the RPU, SPU and RTC. The Gunn oscillator is supplied by an 8 V voltage regulator.

21


System network

System network System architecture Function of system architecture As ACC performs a function that involves interaction with a number of other systems, the system architecture plays a key role. Only with a suitable system architecture can the various contributory functions be interlinked in such a way that an effective and reliable overall function is achieved. A particular challenge to the design of the system architecture is the fact that the contributory systems are often developed by different, and in many cases, competing suppliers and sometimes vary even within the same vehicle model.

Structure of the system network A general view of the basic structure of the ACC system and how it is incorporated in the vehicle’s overall system network has already been given in Figures 1 and 3 in the chapter “System overview”. The ACC sensor & control unit detects vehicles ahead and calculates the acceleration1) required in order to maintain a safe distance. 1

That required acceleration is subsequently converted into appropriate instruction signals for the engine management and braking control systems. The ACC system is therefore not an autonomous system but depends on other partner systems within a network in order to be able to perform its function. Method of operation of the system network Figure 1 provides an overview of the contributory partner systems that are required for the overall function of the ACC. Implementation of the ACC speed-control instructions is carried out by the engine management and braking control systems. Conversely, the ACC requires information from those partner systems about the current status of the vehicle, e.g. its speed, acceleration, whether it is cornering, the current engine torque, etc.

1) Acceleration in this case refers to both positive and negative acceleration (i.e. deceleration).

ACC within the network of associated partner systems

ACC sensor & control unit Radar transceiver

Signal processing

ACC control

Drivetrain control Engine ACC/CC management input signals

Relevant to ACC

Transmission control Engine torque control

Power/torque ratio

Drivetrain control

CAN ESP Electronic Stability Program Electronic Sensor Stability signal preProgram (ESP) processing

Sensors

Deceleration control

Displays and controls Set "speed"

Selected "Target object "time gap" detected" indication

æ SFS0022E

22


The displays and controls are also accessible via the CAN. The ACC requires details of the driver’s instructions (set speed, selected time gap) and also passes information back to the driver (e.g. whether a target object has been detected). The controls are also required by the conventional cruise control system and are therefore generally interpreted by the engine-management system. The transmission-control system is not used as an actuator system by the ACC. Nevertheless, it needs information from the transmission control system on the current effective power/torque ratio of the transmission. The data transmission medium is the controlunit bus, or CAN (Controller Area Network). This links the individual systems with one another. Often, there are also other devices connected or available via gateway functions. Not only does the network use a defined manner of data transfer, there is also a specified convention for the data content of network communications. This arises from the allocation of tasks. Depending on the installed partner systems, therefore, the content of the individual interface signals can vary. Drivetrain control The ACC system needs to be able to intervene in the engine-management system in order to effect a required rate of acceleration or level of engine torque. Most modern engine-management systems afford this possibility (e.g. ETC (Electronic Throttle Control) systems, Motronic ME7, EDC electronic diesel control). They make use of the existing internal interface with the conventional cruise control system. Consequently, power transmission can be controlled on the basis of engine torque without the knowledge of engine data maps.

System network

Brake control If the deceleration due to the engine and transmission drag is insufficient to slow the vehicle down at the rate required by the ACC, active application of the brakes is necessary. There are two ways in which this can be achieved.

Active brake servo unit A brake servo unit with electronically controlled actuation of the pedal linkage offers the facility for automatic operation of the brakes rather than by “manual” operation on the part of the driver. The basic requirement for such an arrangement is a modified diaphragm design and an additional pneumatic proportional control valve. The brake light continues to be controlled by the brake-light switch on the brake pedal. An extra “release switch” indicates when the driver is braking “manually”. Provided there is a pressure sensor at the master-cylinder outlet port (as is usual with the ESP vehicle-dynamics control system), active braking by the ACC is also frequently controlled via a pressure or torque interface. Hydraulic brake actuator Electronically controlled brake actuators are already widely used by vehicle safety systems such as TCS and ESP – though only in potentially hazardous vehicle-handling situations and not under normal driving conditions. With improved control techniques, the hydraulic actuators (generally motor-driven pumps and valves) can be triggered so as to facilitate brake control. So that the brake lights are operated when the brakes are automatically applied, as well as when they are applied by the driver, a second switch signal is generated in addition to that triggered by the action of the brake pedal. The SBC (Sensotronic Brake Control) electrohydraulic brake system can perform the required functions without any additional hardware whatsoever. As a “brake-by-wire” system, SBC is ideal for combination with

23



System network

ACC. For this system, the ACC instructions are merely an additional control input. Bend-sensing systems Present-day ACC systems use ESP sensor signals to ascertain the way in which the vehicle is moving. Typically, the CAN transfers the measured data from the ESP control unit to the ACC control unit. In this way the cost of a separate set of sensors for the ACC can be avoided.

The ESP sensors described below are available to the ACC (the chapter “Sensors” contains detailed descriptions of the design, usage and method of operation of those sensors). Yaw-rate sensor The yaw-rate sensor detects the rate of rotation of the vehicle around its vertical axis. The physical principle on which operation is based is measurement of the Coriolis force. Under the effect of a rotational movement, the pattern of oscillation of an oscillating mass is made to change. Steering-wheel-angle sensor The physical principle on which operation is based is measurement of the angle of rotation of the steering column. Depending on the type of application, these sensors may detect movement by means of sliding contacts or by proximity sensing. Acceleration sensor The physical principle on which operation is based is measurement of the deflection of a flexibly mounted mass under the effect of inertial forces acting along or across the vehicle axis. Wheel-speed sensor The signals from the wheel-speed sensors are used by the control unit concerned to determine the speed of rotation of the wheels. The following types of sensor are used:

Passive (inductive) speed sensors with a reluctor ring attached to the wheel hub. Active speed sensors with a multipolar ring consisting of a series of adjacent magnets of alternating polarity which is attached to the wheel hub. The sensing probe thus detects the change in magnetic flux. Safety concept Purpose of the safety concept The aim of the ACC safety concept is to prevent faults in the ACC system causing critical driving situations or vehicle handling scenarios. At the same time, however, the object is to minimise the limitations on system availability resulting from the safety measures. The safety concept must ensure that the ACC control unit has fail-safe characteristics, and must facilitate selectively targeted fault diagnosis by providing for

shut-down of the radar transceiver, deactivation of ACC function, and recording a fault in the fault memory. This demands that all possible fault scenarios are reliably distinguished and detected and a response appropriate to the nature of the fault is initiated. Structure of safety concept The generally accepted methods of monitoring safety-related systems are the principles of diversity and redundancy. With diverse systems, all computing operations are performed simultaneously on different types of computer with different software. With redundant systems, identical hardware and software is simply duplicated on multiple systems. With the increasing complexity of today’s automotive-control-unit functions, bit-forbit identity of computing results obtained from diverse systems is not achievable. Instead of a straightforward check for identity, a complex plausibility algorithm which tolerates divergence of results within defined limits has to be developed. However, ab-


solutely comprehensive fault detection is then no longer possible. In addition, in the development of control units it can be said that as objectives, both diversity and redundancy are in conflict with the aims of cost and size reduction. For those reasons, a monitoring concept was developed for the ACC control unit which is based on the specific processor structure of the ACC module and takes equal account of the complexity of the tasks and the specific safety requirements of the system. As a consequence, the ACC control unit with its twin-processor structure and the associated internal communication facilities satisfies the safety requirements in respect of redundant hardware structures and monitoring units. The monitoring concept of the ACC control unit is subdivided into three logical levels which are located within the two controller units and the external partner control units. “Component monitoring” level The component monitoring level consists of two independent sections of the two controllers. Its functions are restricted to detection of faults within the controller’s peripherals. It is not linked in any way with monitoring of the computation logic. Examples of component monitoring include

monitoring of the radar transceiver detection of sensor misalignment detection of sensor “blindness” power-supply monitoring monitoring of the CAN data bus lens-heater monitoring

“Function monitoring” level The function monitoring level is similarly implemented independently on both controllers. Each controller performs internal tests on its own computation logic. Outside of the ACC control unit, there are also localized tests carried out by the partner control units. These involve those control

System network

units checking the plausibility and consistency of ACC messages. In that way, ACC malfunctions that result in implausible CAN signals or irregular CAN transmission cycles can be detected. Examples of function monitoring include

internal processor hardware tests internal processor checksum tests checking of CAN checksums checking of CAN message counters CAN “time-out” monitoring

“Reciprocal monitoring” level The reciprocal monitoring level involves the interaction of the two controllers within a common monitoring structure. The essential difference from the function monitoring level is that the monitoring and the monitored functions are not running on the same hardware; instead, each of the two controllers monitors the other. Examples of reciprocal monitoring include checksum verification of internal communications monitoring of internal communication timing calculation and reciprocal checking of test functions.

25



System network

Method of operation of the safety concept The error messages from the individual monitoring procedures are centrally analysed by the control unit. There is a differentiated response to errors depending on their severity and the current dynamic status of the vehicle as well as the traffic situation. The possible responses are as follows: unrestricted continuation of ACC function, no fault indication, recording of fault for subsequent diagnosis, completion of ACC deceleration sequence followed by indication of fault and recording of fault for subsequent diagnosis, immediate deactivation of ACC function combined with fault indication and recording of fault for subsequent diagnosis. Furthermore, a distinction is made between reversible and irreversible faults as follows: reversible faults disable the ACC function only while the fault is present, irreversible faults disable the ACC function for the duration of the current journey. Thus in all fault scenarios, the ACC will be functional again if the fault is no longer present after the ignition has been switched off and on again. The only exception to this is sensor misalignment; in that case, the ACC function must be re-enabled by a service centre. The majority of faults recorded for subsequent diagnosis can be allocated to one of the following categories: Control unit fault (requires replacement of control unit) Power supply voltage too high/low Temperature too high Sensor misalignment CAN bus hardware fault Fault in communication with partner control units Fault signal received from a partner control unit

If the ACC shuts down due to a fault, the vehicle can still be driven without any other functions being restricted in any way. The car does not have to be taken to a service centre immediately. Only a few component failures within the ACC control unit can only be detected by a single monitoring process. In most cases, more than one fault detection function will be triggered, depending on the nature of the fault. The example set out below illustrates how the different monitoring levels complement each other. Let us assume the power supply voltage for the controller units is incorrect. The component monitoring level has a voltage check for detecting such faults whereby the supply is passed through a monitoring circuit and compared with tolerance limits. However, this demands that the monitoring controller continues to operate correctly despite the assumed voltage discrepancy. If, however, the assumed fault leads to a malfunction on one of the two controller units, this can be detected by the reciprocal monitoring of internal communication. The most likely course of events in this example, though, is the total failure of both controllers, which would be detected as a CAN “time-out” error by the function monitoring level on one of the partner control units.


Very severe demands are made on the ECU

Basically, the ECU in the vehicle functions the same as a conventional PC. Data is entered from which output signals are calculated. The heart of the ECU is the printed-circuit board (pcb) with microcontroller using high-precision microelectronic techniques. The automotive ECU though must fulfill a number of other requirements. Real-time compatibility Systems for the engine and for road/traffic safety demand very rapid response of the control, and the ECU must therefore be "real-time compatible". This means that the control's reaction must keep pace with the actual physical process being controlled. lt must be certain that a real-time system responds within a fixed period of time to the demands made upon it. This necessitates appropriate computer architecture and very high computer power. Integrated design and construction The equipment’s weight and the installation space it requires inside the vehicle are becoming increasingly decisive. The following technologies, and others, are used to make the ECU as small and light as possible: Multilayer: The printed-circuit conductors are between 0.035 and 0.07 mm thick and are “stacked” on top of each other in layers. SMD components are very small and flat and have no wire connections through holes in the pcb. They are soldered or glued to the pcb or hybrid substrate, hence SMD (Surface Mounted Devices). ASIC: Specifically designed integrated component (Application-Specific Integrated Circuit) which can combine a large number of different functions. Operational reliability Very high levels of resistance to failure are provided by integrated diagnosis and redundant mathematical processes (additional processes, usually running in parallel on other program paths).

Environmental influences Notwithstanding the wide range of environmental influences to which it is subjected, the ECU must always operate reliably. Temperature: Depending upon the area of application, the ECUs installed in vehicles must perform faultlessly during continual operation at temperatures between –40°C and + 60...125°C. In fact, due to the heat radiated from the components, the temperature at some areas of the substrate is considerably higher. The temperature change involved in starting at cold temperatures and then running up to hot operating temperatures is particularly severe. EMC: The vehicle's electronics have to go through severe electromagnetic compatibility testing. That is, the ECU must remain completely unaffected by electromagnetic disturbances emanating from such sources as the ignition, or radiated by radio transmitters and mobile telephones. Conversely, the ECU itself must not negatively affect other electronic equipment. Resistance to vibration: ECUs which are mounted on the engine must be able to withstand vibrations of up to 30 g (that is, 30 times the acceleration due to gravity). Sealing and resistance to operating mediums: Depending upon installation position, the ECU must withstand damp, chemicals (e.g. oils), and salt fog. The above factors and other requirements mean that the Bosch development engineers are continually faced by new challenges.

Hybrid substrate of an ECU

æ UAE0744Y

Demands placed on a control unit

27


Controls and displays

Controls and displays Function Controls and displays constitute the direct interface between the ACC system and the driver. Operation and interpretation should be as straightforward, unambiguous and intuitive as possible (in other words, a control operation or the meaning of a display indication should be immediately obvious). Design and method of operation Specifically with regard to the controls and displays on the driver’s instrument panel, there is a large degree of scope for variation in design which vehicle manufacturers use to differing extents (Example: Figure 1). For that reason, this description will restrict itself to the typical controls and displays and their functions without consideration of individual design variations. As control operations are frequently acknowledged by a display indication, the particular control and its associated display indication are dealt with together in each case.

Activation Although ACC is used frequently, it still has Fig. 1 1 Speedometer with LEDs for indicating the desired speed setting (“ACC active”) 2 Relevant target object detected (“ACC active”) 3 Indication of “Selected distance” by means of car pictograms (displayed for 6 seconds after ACC activation and when setting is changed) or error message “ACC inactive” or instruction “Clean sensor” 4 Stand-by

1

to be switched on by the driver. On some vehicle models, it first has to be enabled by a master switch. On other models, it is automatically set to stand-by when the ignition is switched on. The conditions for activation include the following: The vehicle speed must be higher than the minimum possible cruising-speed setting. The brake pedal must not be depressed. The handbrake must be off. No fault must have been detected on the ACC SCU or the ACC system as a whole. Assuming the conditions for activation are satisfied, the ACC commences speed control as soon as the driver operates a control provided for activation of the system. An important precondition for commencement of speed control is, of course, availability of the initial settings for “desired speed” and “desired time gap” so that the driver can immediately be informed of the active settings and can adjust them if required. For that reason, the display of those settings is absolutely essential, at least at the time at which the system is activated.

Driver’s instrument panel incorporating ACC displays (example)

1

2

3

4

æ UFS0007Y

28


In order to ensure unambiguous distinction from other functions, the ISO (International Organization for Standardization) has defined a symbol (Figure 2). That symbol may be used to indicate that the system is on stand-by or that it is active.

2

ISO display symbols for ACC activation

Fig. 2 a ACC active/on stand-by b ACC malfunction

æ UFS0008Y

b

3

ACC controls on the steering wheel (example)

1

There are four functions for setting the speed: 2

1. Adoption of the actual speed as the desired speed (Set). 3

2. Adoption of the next increment above the actual speed as the desired speed (Set +).

4

æ UFS0009Y

3. Adoption of the next increment below the actual speed as the desired speed (Set –). 4. Adoption of the stored speed setting as the desired speed (R, Resume).

29

a

Setting and display of “desired speed” All currently known control concepts combine the operations of activating the system and setting the desired speed – i.e. as soon as the driver first operates the control for setting the desired speed while the system is on stand-by, the ACC is simultaneously activated (Figure 3). Although ACC frequently uses the same controls as conventional cruise control, the method of setting the desired speed is significantly different. In particular, operation in practice has shown that with ACC, larger speed-setting increments are more helpful to the driver. For example, instead of the speed increments of around 1 km/h used on conventional cruise control systems, steps of between 5 and 10 km/h have proved to be more effective with ACC. With those larger increments, it is easier to make larger adjustments to the desired speed, e.g. when exiting a stretch of road where there is a low speed limit because of roadworks and entering a section of “clear” motorway, or vice versa.


Fig. 3 1 Resume: recalls the last selected speed setting (“ACC passive”) Displays and selects the set distance from three possible settings (“ACC active”) 2 “+” button: adopts the speed currently indicated by the speedometer as the set speed (“ACC passive”) Increases the set speed in increments of 10 km/h (“ACC active”) 3 “–” button: same by analogy as “+” button, i.e. decreases the set speed in decrements of 10 km/h 4 “I/O” button: Switches the ACC system on/off when it is “off”/”active”, and to “ACC passive”



4


Activating ACC and setting “desired speed” using four buttons and a display integrated in the speedometer scale

υset

–

υset

The Set and Step functions are combined, though often in different ways by the various vehicle manufacturers (Figure 4). The following are typical combinations:

+

Step + combined with Set or Set + 70

90 110 130

50

150 170

30

190

10

The setting is indicated either on the speedometer scale (Figure 4) or on a separate digital display.

210

Resume

R

SET+

+

+ STEP+

ACC active

SET–

–

Off

0

STEP–

–

æ SFS0023E

ACC passive

5

Step – combined with Set, Set – or Resume

Controls for setting “desired distance”/ “desired time gap”

a 1

1 2 3

1 Fig. 5 a Adjuster wheel b Incremental switch c Button for cycling through sequence of options 1 2 3

“Green” zone, large distance “Amber” zone, medium distance “Red” zone, short distance

2

3

æ SFS0024Y

3

c

Even with that range of settings there are a number of different control concepts, including infinitely variable adjustment using an adjuster wheel (Figure 5a) incremental switch (Figure 5b), and button for cycling through a sequence of options, e.g. long, medium, short, long, medium, ... etc. (Figure 5c).

b

2

Setting“desired distance”or“desired time gap” The desired distance or time gap from the vehicle in front depends not only on the driver’s personal preference, but also on traffic and weather conditions. In order to be able to accommodate those variations, all manufacturers offer at least three different settings within the range of 1.0 ... 2.0 s (time gap).

Some or all of those options may be offered depending on the control concept. Once the desired speed has been set, it can be increased in the above increments by repeated and/or continuous pressing of the button (Step +)/(Step –).

When the time gap is altered, the driver is informed of the setting selected. There are two methods of displaying that information as shown in Figure 6. Indication of “object detected” In addition to the absolutely essential display of “desired speed” and “desired distance”, an indication that an object has been detected has become an established feature. It informs the driver whether or not the ACC sensor has detected a relevant object (i.e. a vehicle ahead).


6

Figure 7 shows some examples of the possible display symbols.

a

3

2

yellow red

b Fig. 6 a Symbolic representation of straight-ahead view b Side-on view

1

2

1

3

æ SFS0025E

As well as “genuine” faults on the various control units within the network utilised by the ACC, transient faults can also cause the system to shut down.

1

green

Other display functions An unwanted but essential display item is the error message that appears when the system is shut down or fails to activate due to a fault. In addition to plain-language messages, there is also an ISO symbol that can be displayed for this purpose.

In particular, if the sensor’s “vision” is impaired, e.g. by a thick layer of wet snow, the system informs the driver of the problem and shuts down. 7

Deactivation Deactivation is effected in similar fashion to a conventional cruise control system by operating an OFF switch or the brake pedal. Other conditions for deactivation include incompatible vehicle operating statuses and vehicle speeds below the minimum possible speed setting.

2 3

“Green” zone, large distance “Amber” zone, medium distance “Red” zone, short distance

Display options for indicating “object detected”

a

B

A

Partial deactivation is provided for in the event of active intervention by the TCS or ESP slip-control systems. In such cases, the ACC may still operate the brakes but the acceleration facility is disabled. This allows a deceleration sequence already in progress to be completed. In order to restore full ACC function, the driver must manually reactivate the system.

31

Display symbols for indicating “desired distance”/ ”desired time gap” setting

b A

B

45

æ SFS0026Y

If the detected object is found to be travelling at a slower speed than the currently set desired speed, it is classified as a ranging object.


Fig. 7 a Symbolic representation of straight-ahead view b Side-on view A B

No relevant object Relevant object detected



Object detection and selection

Object detection and selection Radar-signal processing Fourier Transform All simultaneously located objects (i.e. different vehicles) produce characteristic signal attributes – the frequency of each individual signal is determined by the distance and relative speed of the object, while the amplitude is dependent on the reflective properties of the object. All echo signals superimposed make up the return signal.

The return signal is first passed through an analog-digital converter and then subjected to spectral analysis in order to determine the distance and relative speed of the objects. A powerful algorithm (calculation procedure) known as an FFT (Fast Fourier Transform) is applied for this purpose. It converts a sequence of scanned equidistant timing-signal levels into a sequence of spectral (differentiated) power-density values with equidistant frequency intervals. In the classical FFT algorithm, the quantity must be a power of 2 (e.g. 512, 1024, 2048). In the case of the frequencies, the calculated spectrum reveals particularly high power densities which are assigned to the radar echoes. In addition, the spectrum also contains noise signals that are generated within the sensor and superimposed on the useful signals originating from the target objects. The spectral resolution is determined by the number of scanned levels and the scanning rate. Detection Detection involves the isolation of the characteristic frequency signals that are reflected by target objects. Because of the widely varying strengths of signal from different objects, and even from the same object at different times, a special type of detector is used. That detector must, firstly, find all the signal peaks originating from real objects. Secondly, however, it must be insensitive to signal components that are produced by noise or

interference signals. The noise signal generated within the radar unit itself, for example, does not occupy a fixed position within the spectrum but rather is frequency and timedependent. Every spectrum is first of all subjected to noise analysis. Based on the spectral distribution of the noise component, a threshold curve is then defined. Only signal peaks that are above that threshold are then interpreted as target frequencies. Object identification Although the echo signals in each modulation cycle contain information about the distance and relative speed of the detected objects, they cannot be definitely assigned to individual objects. Only by comparing the detection results between modulation cycles can the distance and speed information be assigned to specific objects. A detected target frequency is made up of a distance-dependent component and a relative-speed-dependent component. Thus, in order that the distance and relative speed of a particular object can be determined, target frequencies from multiple modulation cycles must match up with one another. For a physically present radar target, the “multi-cycle FMCW method” requires the identification in every modulation cycle of a target frequency that results from the distance and relative speed of the object (cf chapter “Radar ranging sensor”). Allocation of signal to object becomes difficult if the spectra contain large numbers of target frequencies. The angular position of a radar target relative to the radar axis is determined by comparing the amplitudes of a signal from the same object in the three adjacent antenna lobes.


An object that was identified in the last detection cycle at a distance, d, and travelling at a relative speed, υr, will have moved in the period, ∆t, between the last detection cycle and the one that follows it so that it would then be expected to be detected at the distance de = d + υr · ∆t If we also take into account the fact that the detected object may accelerate or decelerate during that period, then there is an area of uncertainty around the distance de within which the object’s new position may fall.

Object selection The first step in the selection of relevant objects is the determination of the lateral position, dyc (course offset) of an object relative to the predicted course of the ACC system’s own vehicle. As shown in Figure 1, it is determined firstly by the lateral offset, dyv, relative to the vehicle’s longitudinal axis. In order to arrive at that figure, the information determined by the sensor for lateral offset of the object relative to the sensor axis, xS, is transformed on the basis of the sensor offset, dySensor, into a figure representing the object offset relative to the vehicle’s centre line, xF.

1

If the following detection cycle does actually find an object within the expected zone based on distance and relative speed, it can be assumed that it is the same vehicle. Because, therefore, the previously detected object has been identified again in the current detection cycle, the object data is filtered on the basis of the “historical” data. If, however, a previously detected object is not identified in the current detection cycle (e.g. because it is outside the radar beam’s field or because its echo signal is too weak), the predicted object data continue to be used.

33

Echo signals are also analysed for indications of sensor “blindness” and radar-component malfunctions.

Determination of lateral object offset, dyc, relative to course (course offset)

d yvCourse 1

d yv

4

d yc

xF Fig. 1 1 Object 2 Sensor 3 ACC vehicle 4 Course

d ySensor xS

d

Additional object-tracking procedures are necessary if an object produces multiple echo signalsfromdifferentdistances.Thisistypically the case with large trucks. Such vehicles must be combined into a single object.

α

2

A C C

3

æ UFS0027Y

Tracking The tracking function compares the signal data from the detected objects in the current detection cycle with the signal data from the last detection cycle.


dyv Lateral offset dyc Course offset dyvCourse = ky · d2/2 predicted course where d is calculated distance to object ky current course curvature dySensor Sensor offset xF Vehicle centre line xS Sensor axis α Angular offset of object from sensor axis



Secondly, however, by plotting a predicted course, dyvCourse = ky · d2/2 (e.g. using a parabolic projection as an approximation of the arc) the offset relative to the predicted course can be represented by dyc = dyv – dyvCourse Thus the determination of dyc depends on the nature of the description of the ACC vehicle’s course – for which there are various procedures, some of which are examined a little further on. The second step involves the calculation of the object’s lane probability, “spw”, in each detection cycle. This figure indicates the degree of probability that the detected radar object is in the same lane as the ACC vehicle. The ACC vehicle’s lane is described by means of geometrical projections which take account not only of the “lane width” but also other factors such as the “uncertainty of course prediction”. The lane probability, “spw”, is the input variable for the integral object attribute “plausibility”, “plaus”. The latter quantity is used as an index of the relevance of the object when combined with the frequency and reliability of detection. It also takes account of characteristics of the sensors such as “accuracy of angle detection” and “detection capability”. If there is a positive probability that the object is in the same lane as the ACC vehicle, the attribute “plaus” (plausibility) can be increased. If, on the other hand, the object is not in the same lane in the current detection cycle, or if it is not detected at all, “plaus” is reduced.

Fig. 2 1 Object 1 2 Object 2 3 ACC vehicle A Course A B Course B dyc Course offset

The object is only selected as a target object if there is more than a certain minimum degree of probability that it is in the same lane as the ACC vehicle. Similarly, the known ACC systems take account only of moving objects travelling in the same direction as the ACC vehicle. Because of the risk of misdetection and the present impossibility of

object classification (i.e. determining whether the detected object is a metal can or a stationary vehicle) the ACC system ignores stationary objects. Course prediction Performance quality Course prediction plays a decisive role in determining which of the vehicles detected ahead of the ACC vehicle are in its path, and therefore has a substantial influence on the quality of ACC performance.

In the example illustrated in Figure 2, the response of an ACC vehicle travelling in the left-hand lane and following a constant curved path, Course A, is guided by the vehicle that is ahead of it in the same lane, i.e. Object 1. This produces the desired behaviour and the ACC vehicle maintains a constant distance from Object 1. The straight-ahead course, Course B, would incorrectly take account of a slower vehicle, Object 2 , in the right-hand lane in a 2

Course prediction and object selection

B 1 2 A dyc

3

A C C

æ UFS0028Y

34


situation such as that illustrated. This would cause the ACC vehicle to slow down in a manner that would be unexpected and disconcerting for the driver. A reliable course prediction capability is therefore of great benefit in reducing the risk of incorrect object selection as illustrated in the above example. The basic variable for determining the course of the vehicle is initially the “trajectory curvature”. This defines the change of direction of the ACC vehicle as a function of the distance travelled. Supplementary to that information, the current and past positions of moving or stationary objects may also be used to determine the projected course of the vehicle. Future ACC systems will make use not only of navigation systems but also of video systems with image-analysis capabilities to determine the curvature of the vehicle’s path. Trajectory-curvature calculation The trajectory curvature, k, defines the change of a vehicle’s direction relative to distance travelled. It is given by the formula R = 1/k A number of vehicle sensors are used to determine the curvature of the vehicle’s trajectory, whereby it is assumed that all calculations are only applied inside the vehicle’s stable-handling limits. In other words, they do not apply to situations in which the vehicle is skidding or there is a high level of wheel spin. The ACC systems known at present require a yaw rate corrected for offset in order to be able to determine vehicle course. This is obtained either directly by the ESP system using the signals from the steering-wheel-angle sensor, lateral-acceleration sensor, wheelspeed sensors and yaw-rate sensor, or by the ACC system itself using offset correction.


The yaw rate, dψ/dt, being the rotation of the vehicle around its vertical axis, defines the current curvature, ky, of the vehicle trajectory at the speed of travel, υx, in the formula ky = (dψ/dt) / υx The trajectory curvature is generally averaged, e.g. using a simple low-pass filter. ESP sensor data for calculating trajectory curvature Apart from the yaw-rate sensor, known ESP systems also make use of three other sensors which allow the curvature to be calculated by the methods set out below. In order to calculate the course curvature, ks, from the steering-wheel-angle, δ, two other vehicle parameters in the form of the steering-gear ratio, isg, and the wheelbase, dax, are required. The following formula then gives a very good approximation of ks under the conditions that are typical for ACC operation: ks = δ /(isg · dax) Calculation of the trajectory curvature, ka, from the lateral acceleration, ay, also requires the vehicle speed, υx, thus: ka = ay/υx2 Calculating the curvature, kv, from the wheel speeds requires the relative wheel-speed difference, ∆υ/υx, and the track measurement, day. In order to minimise power-transmission effects, the speed difference ∆υ = (υl – υr) and the linear speed at the non-driven wheels are also calculated. kv = ∆υ/(υx · day) Although all the methods described can be used to determine the trajectory curvature, they have varying degrees of suitability in different conditions. They differ primarily under conditions where there is a crosswind, where the road is banked, where there are differences in wheel radius, and with regard to their accuracy of measurement at different speeds.

35



As Table 1 shows, the curvature ky calculated from the yaw rate has the best overall suitability of all the methods discussed. Nevertheless, a further improvement in signal quality can be obtained if several or all of the methods are used in combination and their various results compared. In particular, this will be possible if the ACC vehicle is also equipped with a vehicle dynamics control system such as the Electronic Stability Program (ESP). In that case, all the sensors referred to above will be present. Bend prediction On roads with pronounced changes of direction (e.g. on winding motorways), there is a potential for incorrect selection of target object when using the ESP-assisted coursecurvature calculation, which defines the current trajectory of the vehicle. Prediction of course curvature at a distance is theoretically possible using the approaches described below. Prediction based on radar data Two different methods are conceivable using radar data. 1. Analysis of the lateral movement of vehicles in front as the basis for predicting a bend In this case, the collective lateral movement of a number of vehicles ahead of the ACC vehicle is taken as an indication of an approaching bend. Associated misinterpretations such as may be caused by vehicles changing lanes must be prevented. 2. Analysis of stationary objects at the edge of the road as a means of predicting the course of the road In this case, approximation procedures can be used, though objects more distant from the roadside must be reliably ignored.

Navigation systems Predictive curvature information at defined intervals can theoretically be calculated on the basis of digital road maps supported by datum points along the route at distances of no more than 100 meters. The curvature could then be detected, for instance, 50 m before the bend using interpolation procedures assisted by reference to the available datum points. Inherent problems arise from factors such as digital road maps that are insufficiently accurate or which do not reflect recent changes to the road layout. Additional information (e.g. the number of lanes or the type of road) will in future enable wider application. Video-image analysis A very effective but expensive method is lane identification with the aid of a video camera and image analysis. This technology was adopted on the first ACC system for the Japanese market. Since then, however, there have been no other known ACC systems that use video as a means of acquiring data.

3

Production of ACC radar sensor at Bosch factory in Backnang

æ UFS0029Y

36


Method

Curvature calculation based on Steering-wheel-angle Yaw rate ks

ky

Lateral acceleration ka

Immunity to cross wind

––

+

+

+

Immunity to road camber

––

+

––

+

Immunity to wheel-radius differences

o

+

+

–

Accuracy at low speeds

++

o

––

–

Accuracy at high speeds

–

o

++

–

Key to suitability rating

+ + high suitability, + good suitability o acceptable suitability, – poor suitability, – – unsuitable

4

37

Comparison of alternative methods for curvature calculation

Wheel speeds kv

Table 1

Test set-up at Bosch research facility: detection of moving objects using radar sensors

æ UFS0030Y

1



ACC control sequence

ACC control sequence Control-unit structure The basic structure of the ACC control unit is illustrated in schematic form in Figure 1. Levels 1 to 3 have already been described in detail in the chapter “Object detection and selection”. The present chapter focusses primarily on Levels 4 to 6 (Figure 1).

Level 1 At this first level of signal processing (functional level) the available physical quantities are measured (e.g. signal frequencies, echo timings, signal amplitudes, etc.). The ACC detects some of that data itself (e.g. radar signal data), while other information is obtained from the sensors of external systems (e.g. wheel-speed sensor data from the ESP electronic stability program which controls vehicle dynamics).

1

Level 2 At this level (plausibility checking level) the physical data acquired at the previous level is processed with reference to its relevance for the specific vehicle concerned. The result is an initial group of radar target objects that are potentially of relevance to the ACC control sequence, each with a set of attributes comprising “distance”, “relative speed” and “lateral position”. In addition, the information from the ESP sensors is also analyzed at this level in order to determine the curvature of the vehicle’s course. Level 3 This next stage of processing (verification level) involves selection of the specific target object which is to be the basis for control from the group of potentially relevant objects. The selected vehicle then becomes the “target vehicle”. In virtually every case, that target vehicle will be the vehicle that is im-

Basic structure of ACC control unit

Level 1

Level 2

Radar data

Wheel-speed sensor

Yaw-rate sensor

Radar object detection

Determination of course curvature

Level 3

Object selection course prediction, tracking

Level 4

ACC control

Level 5

Linear-speed control

Level 6

Other sensors

Engine management Drivetrain

Active brake intervention

æ SFS0031E

38


mediately in front of the ACC vehicle and in the same lane. Exceptions to that rule will occur primarily when vehicles ahead of the ACC vehicle change lanes, or the ACC vehicle does so itself. In such cases, a group of several possible target vehicles is offered. The decision is then deferred to the next level. For correct selection of the target vehicle or vehicles, effective course prediction and tracking are indispensable requirements (see also the chapter “Object detection and selection”). Level 4 Following selection of the target vehicle, this level is where the actual modulation sequence takes place. The result is the calculation of a “required acceleration”. If, after object classification at Level 3, there should be more than one target vehicle selected, the control algorithms can be calculated for several potential target objects and subsequently assessed. The conventional cruise-control function and the bend-detection function may also take place at this stage. Level 5 At Level 5, it is now a case of implementing the output variable “required acceleration” from the fourth processing stage through linear-speed control. To begin with, this involves selection of the actuation system that can effect the desired acceleration. In the case of positive and small negative rates of acceleration, that will be the drivetrain. If, however, the engine drag effect is insufficient to achieve the required negative acceleration, the alternative actuation system is selected, which actively intervenes in the braking system to produce a retardation effect. Both actuation systems must compensate for interference effects that arise from changes in the resistance to motion (particularly variations in the gradient of the road).


Level 6 This last level of the overall processing sequence involves the generation and control of forces acting on the vehicle’s wheels by two alternative actuation systems. The first of those actuation systems is the drivetrain, which is controlled primarily by the engine management system, although a degree of modulation is also achievable by means of the transmission. The second actuation system consists of a pneumatic or hydraulic braking system which is the chief means of bringing about deceleration. It is controlled by active intervention (i.e. without driver input) on the part of a control system in order to produce a braking effect. Control-unit functions The ACC control unit (Figure 2 overleaf) incorporates the following functions that are described in more detail below:

cruise control, constant gap, bend detection, and acceleration control.

Output of the command signals to the actuation systems takes place via a torque or acceleration interface. Cruise-control function The driver uses the controls to set a desired constant speed. The control unit then first calculates the required adjustment for changing the current vehicle speed to match the desired vehicle speed. It should be noted in this connection that the speed indicated on the display instrument is in advance of the actual speed. The display and control concept of ACC systems currently fitted on vehicles has inherent characteristics that dictate that in the two situations outlined below, there may be a substantial difference (unintended by the driver) between the current vehicle speed and the set speed.

39



Resumption of the cruise-control function using the “Resume” button (Resume function) adopts the last desired speed setting as the current desired speed. It is possible that the driver may have accelerated the vehicle to a much higher speed than the set desired speed before pressing the “Resume” button to reactivate the ACC function. The driver may cancel active ACC function by pressing the accelerator. Thus the vehicle may be travelling at a much higher speed than the set desired speed. In either situation, the driver may not be fully aware of how large the difference is between the actual and the desired speed. The ACC cruise-control function assists the driver in such situations by gradual adjustment of vehicle speed. Follow-on control This second function requires selection of 2

the target vehicle from which a constant gap is to be maintained. To that end, the target object data is compared with the geometry of the ACC vehicle’s predicted path. If there is more than one vehicle in the predicted path, then normally, the nearest of the vehicles ahead of the ACC vehicle is selected as the target vehicle. Ideally, the vehicle selected will be the one whichproducesthelowestrequiredacceleration as specified by the control unit. However, this necessitates feedback of the control unit output to the target-vehicle selection procedure. Once the target vehicle has been selected, a required acceleration is calculated on the basis of the object distance and relative speed. The required distance is calculated from the desired/required time gap, τSet, specified by the driver thus: dSet = τSet · υF

ACC control loop showing control-unit functions

Vehicle with ESP sensors

Actuation systems (engine management braking system transmission control)

Wheel speed Yaw rate Steering-wheel angle Acceleration

Linear-speed control

Acceleration signal

ACC controller Bend detection

Cruise control

Desired speed

Calculated required distance

Driver Desired time gap Radar system

Control-mode selection

Follow-on control

Object selection Object detection

Target object selection

æ SFS0032E

40


The required time gap is generally in the range of 1 to 2 seconds with a tendency towards longer times at slower speeds. That range can usefully be divided into three levels so that the driver is then offered three time-gap programs, i.e. “Short”, “Medium” and “Long” as illustrated in Figure 3.


41

The limited “field of view” of the sensor in tight bends also leads to situations where the target vehicle selected for the constantgap function may no longer be visible. In such situations, the bend-detection function prevents the ACC from immediately accelerating the vehicle.

3

The choice of control parameters represents a compromise between two opposing optimization objectives.

ACC control unit time-gap programs (stationary)

s

Time gap

3

2

1 2 3

1

æ UFS0033E

The first optimization is the adjustment of the actual speed and object distance as quickly as possible to match the required status represented by a relative speed of υrel = 0 and the specified object distance. The second optimisation objective requires that, for the sake of comfort, the system should react as gradually as possible to small changes in the distance and the speed of the vehicle ahead. A non-linear control characteristic resolves this optimization problem, though changes in relative speed provoke more sensitive reactions than changes in object distance. As a rule of thumb, it can be assumed that a relative speed of 1 m/s triggers roughly the same required acceleration as a divergence of 5 ... 10 m from the specified object distance.

0 50

100

150

km/h

Vehicle speed

4

Fig. 3 Time-gap programs: 1 “Long” 2 “Medium” 3 “Short”

Field of view of radar sensor on a straight road and in a bend

2α Range κ

d Range

A C C

1

æ UFS0034Y

Bend-detection function Although the ACC system is primarily designed for use on motorways (with relatively long bend radii), it can also be used on roads with sharper bends. A number of specific considerations have to be taken into account, however. As a convenience system, the ACC should not surprise the driver by abrupt changes in linear acceleration in bends. Because of the limited width of the radar beam (Figure 4), the ACC system modifies the allowable acceleration to suit the reduced “visibility” when negotiating tight bends.

Fig. 4 1 ACC vehicle k Curvature of bend 2αRange Radar field of view dRange 2αRange/k




Selection of control mode At the same time as calculating the required settings for the constant-gap function, the control unit does the same for the desiredspeed function and the bend-detection function. A subsequent control-mode selection procedure analyses the various specified settings and ensures that the ACC never follows the vehicle ahead at a speed greater than the desired speed set by the driver. Linear-speed control The ACC control unit calculates a required setting on the basis of acceleration rates. A separate acceleration-control sequence converts that required setting into actual vehicle acceleration. That involves first of all selecting the appropriate actuation system – either the drivetrain or the braking system – as dictated by the required setting and the current status. Subsequently, the actuation system concerned calculates the control operations necessary to implement the required acceleration. The main requirements that must be satisfied by linear-speed control are smooth transitions when changing from drivetrain to braking system (and vice versa), and compensation for interference effects such as uphill or downhill gradients. Interfaces with actuation systems It would be easiest if the acceleration setting required by the ACC control unit could be passed directly to the actuation systems. However, that would require an underlying acceleration-control system in the actuation systems which does not always exist. In practice there are primarily two interfaces as outlined below.

Torque interface Most engine-management systems operate on the basis of engine torque. They handle ACC instructions in the same way as conventional cruise-control instructions. For that reason, only minor adaptations to the engine-management system are required in order to implement a specified output torque demanded by the ACC. Within the ACC unit, however, the required acceleration has to be converted into a required engine torque. This requires calculation of the forces in the drivetrain and estimation of the positive or negative gradient of the road, or else performance of the torque control within the ACC control unit. The torque interface is only of any use if the actuation system can reliably implement the specified torque. Acceleration interface Most commonly used braking systems (Smart Booster, ESP or SBC employing hydraulic brake intervention) support the acceleration interface. Limits of function Speed range ACC systems are primarily intended for use on motorways and fast trunk roads. The sensors currently used have a detection range for vehicles in the same lane on a straight road which starts at a distance of about 40 m. This means that they have difficulty detecting vehicles on roads with tight bends and in urban traffic. For that reason, the minimum speed limit for operation of ACC systems is between 30 and 50 km/h (depending on system design, see also Figure 4 above).

Because of the comfort considerations of common ACC systems, the upper speed limit is in the range of 160...200 km/h.


Linear dynamics As absolutely correct control-unit response cannot be guaranteed, the effects of controlunit function (i.e. acceleration and deceleration) must be limited. Such limitation may relate both to absolute acceleration and to its variation over time. Whereas the upper acceleration limits represent levels that are also normal for conventional cruise-control systems (approx. 0.6...1.0 m/s2), the deceleration limit for ACC systems with active braking is typically 2.5 m/s2. That figure is generally sufficient for speed regulation. But although such a rate of deceleration is clearly discernible by the driver, it is still only a quarter of the maximum possible deceleration on a dry road. As a result of the limitations on deceleration combined with the similarly limited range of the radar sensor, there is also a maximum differential speed that the ACC system is capable of responding to without intervention by the driver (Table 1). The tempting conclusion to call for a greater sensor range in order to obtain an earlier response is not achievable for the following reasons: The reliability with which correct lane allocation can be achieved diminishes rapidly with increasing target distance. The probability of an overtaking manoeuvre increases as the differential speed rises and can only be definitely confirmed in the immediate proximity of the target object. This produces a conflict of aims: with a high differential speed, on the one hand an early reaction is necessary, while on the other hand it is precisely in such a situation that the probability of overtaking is very high so that premature deceleration is undesirable.


The effect of a change of lane by vehicles ahead, or by the ACC vehicle itself, is that the point of response is determined not simply by object distance but also by the point at which the object is recognised as being in the ACC vehicle’s lane. For that reason, the driver must expect that in the event of lane changes, the ACC may not be able to cope with the speed difference. In bends with a radius of less than 1000 m, the ACC sensor’s view can be restricted by roadside structures and vehicles in the adjacent lane, so that although constant gap maintenance remains possible through the corner, the forward scanning range is insufficient for early reaction if confronted with a new slower vehicle. Although in a few cases higher detection quality is achievable, this is often at the cost of transparency and can therefore make assessment by the driver more difficult. Stationary objects ACC has the fundamental capability to distinguish stationary objects from moving ones. The radar system measures the relative speed, υrel, j of an object and by comparing it with the ACC vehicle’s own speed, υF, is able to obtain the absolute speed, υj, of the object thus: υj = υF + υrel, j However, stationary objects are normally ignored by the ACC constant-gap function. There are two main reasons for this: ACC is a convenience system. Its deceleration capability is therefore not designed to be able to brake the vehicle in time to avoid collision with a stationary object.

43




It is currently not technically possible for the system to decide with sufficient certainty whether or not an object is in the same lane as the ACC vehicle. With the large number of stationary objects at the roadside, it would therefore be highly likely that the ACC would react to one of them by mistake. For those reasons, the Bosch ACC system operates according to the following strategy:

1

Stationary objects are taken account of and analysed by the sensor system only at low speeds. Only moving or stopping objects are taken account of for the purposes of constant gap maintenance. This virtually excludes the possibility of mistaken deceleration in response to a stationary object at the roadside. ACC prevents acceleration of the vehicle if it detects stationary objects in its own lane.

Maximum approach speed

At a distance, d, an even rate of deceleration, a, will allow adjustment to a relative speed of

υrel = – 2d · a

Target vehicle

ACC vehicle =

υ1 > υ2

υ2

æ UFS0035E

υ rel = υ 2 - υ 1 @

For various pairings of d and a, the table below details the corresponding maximum approach speed that can be adjusted to. The average deceleration has to be based on a smaller figure than the maximum deceleration since deceleration generally takes time to build up.

d m

Table 1

a m/s2

–υrel m/s

–υrel km/h

50

–1

10

36

100

–1

14

51

150

–1

17

62

50

–2

14

51

100

–2

20

72

150

–2

24

88


Future developments Sensor technology At the forefront of continuing development of the ACC sensor system are the preparations for wider distribution of ACC systems. In the very near future ACC will also be available on mid-range and small cars.

The efforts of the developers are thus essentially directed towards three key areas: Components that can be produced economically in large numbers, particularly for the high-frequency circuits of the radar sensor: Up to now, such components have only been available for aircraft, military or radio-relay applications in small numbers and at relatively high prices. Development of these items is particularly important in order to be able to make ACC affordable to large numbers of motorists. Further reduction of the external dimensions of the sensor & control unit. The overall size of the unit needs to be reduced to less than half its present volume as even the Bosch ACC system – which is the smallest currently available – is difficult to accommodate at the front of many vehicles. Advances in the area of electronic componentry will enable more compact and more highly integrated modules. Expansion of the detection range. In order to make ACC usable on roads with tighter bends, the sensor beam width needs to be increased to about double its current span (to around 16°). This can be achieved by making the lens antenna smaller and/or increasing the number of radar beam lobes from the present three to four or five. Broadening the field of vision in that way would mean spreading the radar energy over a wider area. Without other modifications, this would produce an undesirable reduction in beam range. Development activities are therefore focussed on improving the circuitry in order to increase

Future developments

the signal quality so that the range can at least be maintained at the same distance as the present system. In addition to the activities mentioned above in the area of ACC sensor technology, Bosch is also working on the development of new, supplementary near-vicinity sensors. These will be used in future driver-assistance systems such as close-range radar (Figure 1 overleaf). Function Perfecting the present range of functions Perfecting the present range of functions in the course of product improvement will give future systems a greater degree of reliability in the selection of target objects and even better adaptive performance. The latter will be evident primarily in more balanced response in lane-change situations and in bends.

Use of ACC in traffic jams To enable use of ACC in traffic jams in the future, three function levels are to be added to the present capability: Braking to a standstill Two preconditions must be satisfied in order to enable braking to a standstill behind a target vehicle: The target vehicle must not be “lost” at close range, i.e. the ranging sensor must cover a sufficiently wide area (upwards of 15° beam width if fitted centrally). Once the vehicle has been brought to a standstill, the system must ensure that it is not inadvertently allowed to roll forwards. One solution might be an electronically controllable parking brake operated by the ACC.

45



1

Ambient sensors

1

b

Test vehicle with video-sensor technology that recognises road signs Following at a constant gap on motorways or in urban “stop-and-go” traffic

2

3

C ranlose ge

Fig. 1 Driver-assistance systems with multiple sensors for motor vehicles a

Future developments

e g n

g n o

ra

L

b

3

m id u e ge M n ra

æ UFS0036E

2

Long range: the 77-GHz radar tracks the vehicle ahead in the same lane; it detects the distance and relative speed of the target vehicle as the basic functions for the constant-gap function (range 120 m, width of field ±8°) Close range: one or more 24-GHz radar sensors scan a broad area immediately in front of the vehicle; vehicles cutting in sharply in front can be reliably detected (range 14 m, width of field ±50°) Medium range: a stereo camera scans the lane ahead of the vehicle for the constant-gap function; it also detects road signs and measures object dimensions (range 50 m, width of field ± 20°)

a

1


Future developments

Follow-on function at low speeds When this function is active, the ACC vehicle “locks onto” the vehicle ahead and follows it at a constant time gap. The preconditions and consequences associated with making such a function suitable for use in slow-moving traffic on a motorway (Figure 1) are the following:

With currently available technology, this is a challenge that can only be met by means of an image-producing sensor (e.g. stereo video sensor). Thus a “stop-and-go” system with “normal” ACC function would have a triple sensor system (Figure 1):

Sensor technology that ensures total coverage of the area in front of the vehicle from a distance of 1 m upwards across the full width of the vehicle so that vehicles cutting in very sharply can also be detected. An additional sensor system for close range (e.g. with radar technology optimised for close-range detection). A function that prevents the vehicle pulling away automatically after it has been stationary longer than a certain period.

A radar ranging sensor similar to the present technology for the longer distances (10...120 m) An image-producing sensor (e.g. stereo video) for medium-range detection (5...50 m) A specially optimized radar system for close-range operation (0.5...10 m).

This mode of operation can and should only be used for typical traffic-jam situations. As a result there are restrictions with regard to the maximum applicable speeds (approx. 20...30 km/h) and deactivation of the function if there is no target vehicle within close range. Stop-and-go in urban traffic The most difficult function to implement is the “stop-and-go” function for heavy urban traffic. At speeds of 0...50 km/h or 0...60 km/h, the system must be capable of automatically distinguishing between relevant and irrelevant objects. As, in contrast with ACC function at high speeds, stationary objects also have to be taken into account, absolutely reliable distinction is very difficult. Depending on the traffic conditions, a car parked at the kerbside and partially blocking the carriageway may represent an obstruction that you have to stop behind or which you can manoeuvre around.

The overlaps between the ranges of the three sensor systems will allow the reliability of detection to be increased. But in spite of the highly involved technology, a dilemma still remains. Response to stationary objects is necessary, and it is achievable since the decision to react to them can be left until they are at a distance of < 50 m. At speeds of ≥ 70 km/h though, this distance is no longer sufficient. At greater distances, however, the frequency of incorrect decisions is so high that stationary objects have to be ignored. As a consequence, “stop-and-go” and normal ACC function have to be implemented as two separate modes of operation which offer the driver different functionalities. The driver will then have to adjust to the differing circumstances. For that reason, driver input will probably be required in order to switch from “stop-and-go” to the lower-level “normal ACC” function.

47



Frequently asked questions

Frequently asked questions How easy is ACC to use? Using ACC is comparable with operating a conventional cruise-control system and is thus very quickly learned. A short introductory phase for learning the basic controls is followed by a familiarization phase lasting several weeks during which the driver internalizes the limits of the system and establishes a personal usage profile. Comprehensive trials1) using test subjects (including long-term tests) revealed no indications of problems with the operation and function of ACC. Instead, virtually all users underlined the reduction of mental effort and the greater convenience. Does ACC still function in fog and rain? In general, yes. Nevertheless, atmospheric propagation conditions do affect the radar beam’s electromagnetic wave to a certain degree. Raindrops and the water droplets in fog disperse a proportion of the wave. Thus there is a degree of attenuation dependent on the droplet size and frequency. In extremely heavy rain, attenuation levels of approx. 30 dB/km have been measured. In view of the ACC functional range of approx. 150 m, however, such a level of attenuation is not drastic. The analyzable return signal is weaker and, therefore, the maximum range of the radar is reduced.

In such weather conditions, the range of the sensor for less distinct targets (e.g. motorcycles) may thus be reduced. But at the same time, sensible drivers will moderate their speeds in such weather conditions so that the ACC will not require such a long range in any case. Are radar beams dangerous? The ACC SCU produces a radar beam with a frequency in the range of 76...77 GHz. That is equivalent to a wavelength of approx. 4 mm. Because of the high frequency of the beam, its effect on humans is comparable with that of heat radiation (infrared range). The average emitted power is only approx. 1 mW and is thus at least 500 times lower than the radiation from a mobile telephone. Research has shown that this level of radiation is absolutely uncritical in terms of its effect on sensitive parts of the human organism. Even pointing such a beam directly at the human eye has no known negative effects. Do radar sensors interfere with one another? Radar sensors only interfere with one another if they are operating simultaneously within the same frequency band. A number of properties on the part of the radar sensors ensure that if this occurs at all, it only does so very sporadically. Firstly, the radar beam is only activated for the precise period that it is actually required. This alone reduces the likelihood of mutual interference to less than 10 %. Secondly, FMCW modulation ensures that for each time unit, only an effective bandwidth of less than 500 kHz within the frequency band of 76...77 GHz is used. This makes mutual interference between radar sensors extremely unlikely.

1) See: Markus Weinberger. Der Einfluss von Adaptive Cruise Control Systemen auf das Fahrverhalten Dissertation, Shaker Verlag, Aachen, 2001.

In addition, the filtering and plausibility checking of measured data ensures that interference signals do not cause the vehicle to react incorrectly.


In practical terms, therefore, mutual interference between radar sensors is virtually impossible. Do ACC radar sensors interfere with speed traps? Police speed-enforcement equipment is not affected by ACC systems. As far as radar equipment is concerned, it operates at much lower frequencies (< 35 GHz) or is based on entirely different methods of operation (photoelectric or laser beam). What happens if another vehicle cuts in very sharply in front of you? This question expresses the worry that a gap control system would respond by restoring the required gap without appropriately modifying the deceleration rate. In fact, however, the system response is designed to be similar to that of a real driver. If the gap reduces even further (i.e. if the relative speed is negative), then the speed difference is equalized relatively quickly. After that, however, the ACC vehicle “drops back” gradually in the same way as when the vehicle cutting in is travelling at the same speed. What happens if you are faced with an oncoming vehicle in your lane? Oncoming vehicles are totally ignored because it is not possible to define an appropriate reaction on the part of the ACC. The system is able to identify an oncoming vehicle by virtue of the fact that the relative speed measured by the radar sensor is greater than the ACC vehicle’s own speed (in the opposite direction) as measured by the ABS/ESP wheel-speed sensors. It is ultimately the responsibility of the drivers of the two converging vehicles to take the appropriate action in order to avoid a collision. When will cars be driven completely automatically? It is a dream that we have had almost since the invention of the automobile that we might


one day simply be able to tell the car where we want to go and it would do the rest. We would then lean back and relax and occupy ourselves in other ways such as reading the newspaper. Unfortunately, even in the information-technology age, that dream remains largely unrealisable. Even the most sophisticated sensor systems and powerful computers do not begin to approach the capabilities of a human being. And even if the technological capability existed, there would be serious legal issues to consider. The responsibility for driving the vehicle would shift from the driver to the manufacturer of the automatic vehicle. At present, it is difficult to imagine how this problem could be solved, because with manufacturer’s liability legislation in its present form, no company would be prepared to take on such responsibility. Nevertheless, there are one or two niches for automatically driven vehicles: 1. Special stretches of road (e.g. tunnels or bridges) with a clearly defined infrastructure. The chief motive for such an application would not be driver convenience but a higher rate of traffic flow, thus obviating the need for expensive road improvements. 2. Situations where traffic moves very slowly (no more than 10...20 km/h), and in which incorrect system response would only involve very slight risk for the ACC vehicle and for other traffic. Why are there not yet any collisionavoidance systems? Collisions are prevented either by braking or by an avoidance manoeuvre. The distance at which braking would have to begin in order to prevent a collision increases exponentially in relation to relative speed. The distance required for an avoidance manoeuvre, on the other hand, increases only at a linear rate (Figure 1).

49




Consequently, at closing speeds of > 50 km/h, an avoidance manoeuvre rather than braking is the appropriate reaction. Avoiding a collision by braking alone only makes sense at low relative speeds. Collision avoidance at higher relative speeds necessitates the driver steering around the obstacle or oncoming vehicle.

1

The real challenge for a collision-avoidance system, however, is the need to ensure an extremely low rate of false alarms. A false alarm, namely, would be very likely to create an extremely dangerous situation, particularly in the case of steering manoeuvres. Estimates indicate the necessity for less than one false alarm per 10 million kilometres, which represent an enormous challenge to the developers of such systems.

Avoiding a collision by braking or avoidance manoeuvre

100 m

Distance for start of manoeuvre

80

60

1 40

2 20

1 2

Braking Avoidance manoeuvre

0 0

20

40

60

80

100

120

km/h

Relative speed

æ UFS0037E

Fig. 1 Assumptions: Course deviation of 2.5 m Mean maximum deceleration = Mean maximum lateral acceleration of 8 m/s2


History of radar

Technology borrowed from the animal world RADAR (Radio Detection and Ranging) is a system that uses radio waves to locate distant objects and is traditionally employed primarily in aviation and shipping. It has also been widely used for military applications since the development of radar-assisted air defences in the Second World War. More recent areas of application include space exploration, weather forecasting and, now, motor vehicles where it is used to measure the distance between vehicles for the ACC (Adaptive Cruise Control) system. The idea for RADAR came from the sonar (Sound Navigation and Ranging) system which uses sound echoes to determine the distance and position of objects, and which itself was copied from the navigation techniques of certain animals. Bats, for example, make high-pitched sounds with frequencies in the ultrasonic range of 30...120 kHz. The echoes that bounce back off solid objects are picked up by the bat’s highly sensitive ears. That information then helps the bat to find its way around and to locate its prey.

æ UFS0038Y

History of radar

RADAR functions in a similar manner but by using radio signals instead of sound. Measurement of distance by RADAR is based on timing the interval between transmission of an electromagnetic wave signal and reception of the signal echo that is reflected back by an object in its path.

While radar systems used in aviation and shipping operate at frequencies between 500 MHz and 40 GHz, the frequency band approved for ACC is 76...77 GHz. Stages in the development of RADAR The development of electromagnetic detecting and ranging equipment with long-range capabilities was an enormous challenge to the designers. Only a minute part of the energy originally transmitted was reflected back by the target. For that reason a very high-energy signal that is concentrated in as narrow a beam as possible has to be produced. This demands highly sensitive transmitters and receivers using signals with a wavelength that is shorter than the dimensions of the target. The development of radar technology was marked by the following milestones and personalities: 1837 Morse: The telegraph becomes widely established. Here, electrical currents are used for the first time in communicating over longer distances. 1861/1876 Reis and Bell: Replacement of the telegraph by the telephone provides a much more direct and user-friendly method of telecommunication 1864 Maxwell, Hertz and Marconi: Existence of “radio waves” is theoretically and experimentally confirmed. Radio waves are reflected off metal objects in precisely the same way as light is reflected by a mirror. 1922 Marconi: The pioneer of radio provides the impetus for the continuation of earlier research into radio ranging 1925 Appleton and Barnett: The principle of radio-wave reflection is used to demonstrate the existence of conductive layers in the atmosphere Breit and Tuve: Development of pulse modulation which enables precise measurement of distances 1935 Watson-Watt: Invention of radar 1938 Ponte: Invention of the magnetron (velocity-modulated electron tube for generating high-frequency oscillations)

51


ACC Adaptive Cruise Control Glossary

Glossary Abbreviation Stands for

Explanation

ACC

Adaptive Cruise Control

Cruise-control system which automatically adapts speed if there is a vehicle ahead

ADC

Analog-Digital Converter

Converts analog signals to digital

Bistatic RADAR CAN

RADAR system with separate transmitter and receiver antennae Controller Area Network

Dielectric material

Data network for electronic control units Material with specific transmission properties for microwaves

DAC

Digital-Analog Converter

DRO

Dielectric Resonator Oscillator

DSP

Digital Signal Processor

ECD

Electronically Controlled Deceleration

EDC

Electronic Diesel Control

EEPROM

Electrically Erasable Programmable Read-Only Memory

EGAS

Electronic Gas (Pedal)

Electronic throttle-control system

ESP

Electronic Stability Program

Electronic vehicle-dynamics control system

Flash EEPROM

Flash Electrically Erasable Programmable ReadOnly Memory

FFT

Fast Fourier Transform

FMCWRADAR

Frequency-modulated Continuous Wave RADAR

Gallium

Converts digital signals to analog

Signal processor for performing complex calculations at high speed

Engine-management system for diesel engines

Mathematical procedure for fast conversion of timing signals to a frequency spectrum

Chemical compound with semiconductor properties

arsenide Yaw rate

Rate of vehicle rotation around its vertical axis

Robert Bosch GmbH ACC Adaptive Cruise Control Glossary

Abbreviation Stands for

Explanation

Gunn diode

Electronic component for generating highfrequency oscillations

MIC

Microwave Integrated Circuit

Mixer

Component or module for mixing signals generally for the purposes of demodulation (separating the information from the carrier)

Modulation

The impression of an information-bearing signal onto a carrier signal, e.g. by the multiplication of electrical signals relative to time

Monostatic RADAR

RADAR system which uses the same antenna for transmission and reception

Motronic

BOSCH trademark

Polyrods

Electronic engine-management system Dielectric material in conical shape for concentrating microwave beams

PU

Processing Unit

Control and signal-processing unit

RADAR

Radio Detection and Ranging

System for detecting and locating objects using microwave technology

RAM

Random-Access Memory

Memory whose data can be accessed randomly (usually also a volatile memory)

RPU

Regulation Processing Unit

ACC controller

RTC

Radar transceiver

RADAR transmitter and receiver unit

SBC

Sensotronic Brake Control

Electronically controlled power-braking system with hydraulic force transmission (electrohydraulic braking system)

SCU

Sensor and Control Unit

SPU

Signal Processing Unit

ACC signal processing unit

TCS

Traction Control System

System which improves vehicle handling by preventing wheel spin and increasing traction

VCO

Voltage Controlled Oscillator

53


Sensors

Automotive applications

Sensors Sensors register operating states (e.g. engine speed) and setpoint/desired values (e.g. accelerator-pedal position). They convert physical quantities (e.g. pressure) or chemical quantities (e.g. exhaust-gas concentration) into electric signals.

Automotive applications Sensors and actuators represent the interfaces between the ECUs, as the processing units, and the vehicle with its complex drive, braking, chassis, and bodywork functions (for instance, the Engine Management, the Electronic Stability Program ESP, and the air conditioner). As a rule, a matching circuit in the sensor converts the signals so that they can be processed by the ECU. The field of mechatronics, in which mechanical, electronic, and data-processing components are interlinked and cooperate closely with each other, is rapidly gaining in importance. These are integrated in modules (e.g. in the crankshaft CSWS (Composite Seal with Sensor) module complete with rpm sensor).

Lower levels of computing power are needed in the ECU, A uniform, flexible, and bus-compatible interface becomes possible for all sensors, Direct multiple use of a given sensor through the data bus, Registration of even smaller measured quantities, Simple sensor calibration.

Sensor integration levels

Sensors

Figure 1 SE Sensor(s) SA Analog signal conditioning A/D Analog-digital converter SG Digital ECU MC Microcomputer (evaluation electronics)

Depending upon the level of integration, signal conditioning, analog/digital conversion, and self-calibration functions can all be integrated in the sensor (Fig. 1), and in future a small microcomputer for further signal processing will be added. The advantages are as follows:

Transmission path

Conventional

SE

1st integration level

SE

SA

2nd integration level

SE

SA A

3rd integration level

SE

SA A

D

D

MC

ECU

Susceptible to interference (analog)

SA A

Multiple tap-off

Resistant to interference (analog)

A

Buscompatible

Immune to interference (digital)

SG

Buscompatible

Immune to interference (digital)

SG

D

D

SG

SG

æ UAE0037-1E

1

Since their output signals directly affect not only the engine’s power output, torque, and emissions, but also vehicle handling and safety, sensors, although they are becoming smaller and smaller, must also fulfill demands that they be faster and more precise. These stipulations can be complied with thanks to mechatronics.

Robert Bosch GmbH Speed and rpm sensors

Applications In vehicle’s with vehicle-dynamics control (ESP), the piezoelectric yaw-rate sensors (otherwise known as gyrometers) register the vehicle’s rotation about its vertical axis, for instance when cornering, but also when the vehicle swerves or goes into a skid. Design and construction The piezoelectric yaw-rate sensors are highprecision mechanical sensors. Two diametrically opposed piezoceramic elements (Fig. 1, 1 + 1) are used to cause sympathetic oscillations in a hollow metal cylinder. Another pair of piezoceramic elements (2 + 2) are used to control and maintain this oscillation at a constant amplitude which has four axially aligned oscillation nodes (offset by 45° to the direction of excitation). Refer to Figs. 1...3. When rotation takes place at a yaw rate Ω about the cylinder’s axis, the nodes are shifted slightly at the circumference due to the effects of Coriolis acceleration. The result is that in the nodes, which otherwise feature zero force, forces are now generated which are proportional to rotational speed and Piezoelectric yaw-rate sensor (measuring principle)

which are detected by a third pair of piezo elements (3 + 3). Using a fourth pair of piezo excitation elements (4 + 4) in a closed control loop, these forces are then controlled back to a reference value Uref = 0. The manipulated variable needed here is then carefully filtered and subjected to phase-synchronous rectification before being used as a highly accurate output signal. The selective, temporary change of the desired value to Uref = 0 permits an easy check of the overall sensor system (“built-in test”). This sensor’s temperature sensitivity necessitates a complex compensation circuit, and the materialbased aging of the piezoceramic elements necessitates painstaking preliminary aging. 2

Piezoelectric yaw-rate sensor

3

Piezoelectric yaw-rate sensor (design principle)

Ω

1

Uref

3

4’

5

2’

Ω

2

6

4

3’

Fig. 1 1....4 Piezo elements 5 Circuit 6 Bandpass filter (phase-locked) 7 Phase reference 8 Rectifier (phaseselective) UA Output voltage Ω Yaw rate Uref = 0 (normal operation) Uref 0 (“built-in” test)

1…4

6

5

6

7 UA

8

Ω 7

æ UAE0644-1Y

1’

æ UAE0662-1Y

1

55

UAE0789Y

Piezoelectric “oscillating drum” yaw-rate sensors

Piezoelectric yaw-rate sensors

Fig. 3 1....4 Piezo element pairs 5 Oscillatory cylinder 6 Baseplate 7 Connection pins Ω Yaw rate


Micromechanical yaw-rate sensors

Design and construction MM1 micromechanical yaw-rate sensor A mixed form of technology is applied in order to achieve the high accuracies needed for vehicle-dynamics systems. That is, two somewhat thicker oscillating elements (mass plates), which have been machined from a wafer using bulk micromechanics, oscillate in counter-phase to their resonant frequency which is defined by their mass and their coupling springs (>2 kHz). On each of these oscillating elements, there is a miniature, surface-type micromechanical capacitive acceleration sensor. When the sensor chip rotates about its vertical axis at yaw rate Ω, these register the Coriolis acceleration in the wafer plane vertical to the direction of oscillation (Figs. 1 and 2). These accelerations are proportional to the product of yaw rate and and the oscillatory velocity which is maintained electronically at a constant value. To drive the sensor, all that is required is a simple, current-carrying printed conductor on each oscillating element. In the permanent-magnet field B vertical to the chip surface, this oscillating element is subjected to an electrodynamic (Lorentz) force. Using a further, simple printed conductor (which saves on chip surface), the same magnetic field is used to directly measure the oscillation velocity by inductive means. The different physical construction of drive system

Micromechanical yaw-rate sensors Applications In vehicles with Electronic Stability Program (ESP), the rotation of the vehicle about its vertical axis is registered by micromechanical yaw-rate (or yaw-speed) sensors (also known as gyrometers) and applied for vehicle-dynamics control. This takes place during normal cornering, but also when the vehicle breaks away or goes into a skid. These sensors are reasonably priced as well as being very compact. They are in the process of forcing out the conventional high-precision mechanical sensors.

1

Structure of the MM1 yaw-rate sensor

200 m

3

Fig. 1 1 Retaining/guide spring 2 Part of the oscillating element 3 Coriolis acceleration sensor

2

2 Fig. 2 1 Frequency-determining coupling spring 2 Permanent magnet 3 Direction of oscillation 4 Oscillating element 5 Coriolis acceleration sensor 6 Direction of Coriolis acceleration 7 Retaining/guide spring Ω Yaw rate υ Oscillating velocity B Permanent-magnet field

UAE0790Y

1

MM1 micromechanical yaw-rate sensor Ω

3υ

N

2

S 1

B

7

6

5

4

0

2 mm

æ UAE0706-1Y

56


MM2 micromechanical yaw-rate sensor Whereas this silicon yaw-rate sensor is produced completely using surface-micromechanic techniques, and the magnetic drive and control system have been superseded by an electrostatic system, absolute decoupling of the power/drive system and measuring system is impossible. Comb-like structures (Figs. 3 and 4) electrostatically force a centrally mounted rotary oscillator to oscillate. The amplitude of these oscillations is held constant by means of a similar capacitive pick-off. Coriolis forces result at the same time in an out-of-plane tilting movement, the amplitude of which is proportional to the yaw rate Ω, and which is detected capacitively by the electrodes underneath the

4

MM2 yaw-rate sensor: Structure

50 m

2

1

Fig. 4 1 Comb-like structure 2 Rotary oscillator

MM2 surface-micromechanical yaw-rate sensor

CDrv1

CDrv2

CDrv Det1

1

CDrv Det2 0.5 mm

υ 2 CDet2

CDet1

Ω 3

-FC

+FC

æ UKI0044-1Y

3

57

oscillator. To avoid excessive damping of this movement, the sensor must be operated in a vacuum. Although the chip’s small size and the somewhat simpler production process result in considerable cost reductions, this miniaturization is at the expense of reductions in the measuring effect, which in any case is not very pronounced, and therefore of the achievable precision. It also places more severe demands on the electronics. The system’s high flexural stability, and mounting in the axis of gravity, serve to mechanically suppress the effects of unwanted acceleration from the side.

UAE0791Y

and sensor system serves to avoid undesirable coupling between the two sections. In order to suppress unwanted external acceleration effects, the opposing sensor signals are subtracted from each other. The external acceleration effects can be measured by applying summation. The high-precision micromechanical construction helps to suppress the effects of high oscillatory acceleration which is several factors of 10 higher than the low-level Coriolis acceleration (cross sensitivity far below 40 dB). Here, the drive and measurement systems are rigorously decoupled from each other.

Micromechanical yaw-rate sensors

Fig. 3 1 Comb-like structure 2 Rotary oscillator 3 Measuring axis CDrv Drive electrodes CDet Capactive pick-off FC Coriolis force υ Oscillatory velocity Ω = ∆CDet, measured yaw rate

Robert Bosch GmbH Position sensors

Steering-wheel-angle sensors

Steering-wheel-angle sensors Application The Electronic Stability Program (ESP) applies the brakes selectively to the individual wheels in order to keep the vehicle on the desired track selected by the driver. Here, the steering-wheel angle and the applied braking pressure are compared with the vehicle’s actual rotary motion (around its vertical axis) and its road speed. If necessary, the brakes are applied at individual wheels. These measures serve to keep the float angle (deviation between the vehicle axis and the actual vehicle movement) down to a minimum and, until the physical limits are reached, prevent the vehicle breaking away. Basically speaking, practically all types of angle-of-rotation sensors are suitable for registering the steering-wheel angle. Safety considerations, though, dictate that only those types are used which can be easily checked for plausibility, or which in the ideal case automatically check themselves. Potentiometer principles are used, as well as optical code-registration and magnetic principles. Whereas a passenger-car steering wheel turns through ±720° (a total of 4 complete turns), conventional angle-of-rotation sensors can only measure maximum 360°. This means that with the majority of the sensors actually used for this purpose it is necessary to continually register and store the data on the steering wheel’s actual setting. Fig. 1 1 Housing cover with nine equidistantly spaced permanent magnets 2 Code disc (magnetically soft material) 3 pcb with 9 Halleffect switches and microprocessor 4 Step-down gearing 5 Remaining 5 Hall-effect vane switches 6 Fastening sleeve for steering column

Design and operating concept There are two absolute-measuring (in contrast to incremental-measuring) magnetic angle-of-rotation sensors available which are matched to the Bosch ECUs. At any instant in time, these sensors can output the steering-wheel angle throughout the complete angular range.

Hall-effect steering-wheel-angle sensor (LWS1) The LWS1 uses 14 Hall-effect vane switches to register the angle and the rotations of the steering wheel. The Hall-effect vane switch is

similar in operation to a light barrier. A Hall-effect element measures the magnetic field of an adjacent magnet. A magnetic code disc rotates with the steering shaft and strongly reduces the magnet’s field or screens it off completely. In this manner, with nine Hall ICs it is possible to obtain the steering wheel’s angular position in digital form. The remaining five Hall-effect sensors register the particular steering-wheel revolution which is transformed to the final 360° range by 4:1 step-down gearing. The first item from the top in the exploded view of the LWS 1 steering-wheel-angle sensor (Fig. 1) shows the nine permanent magnets. These are screened individually by the magnetically-soft code disc beneath them when this rotates along with the steering shaft, and depending upon steering-wheel movement. The pcb immediately below the code disc contains Hall-effect switches (IC), and a micro1

Exploded view of the digital LWS1 steering-wheelangle sensor

1

2

3

4

5

6

æ UFL0029Y

58

Robert Bosch GmbH Position sensors

59

AMR steering-wheel-angle sensor LWS3 (principle)

1

2

5 6

3 4

æ UFL0030Y

7

3

AMR steering-wheel-angle sensor LWS3

4

AMR steering-wheel-angle sensor LWS4 for attachment to the end of the steering-column shaft

Fig. 2 1 Steering-column shaft 2 AMR sensor elements 3 Gearwheel with m teeth 4 Evaluation electronics 5 Magnets 6 Gearwheel with n > m teeth 7 Gearwheel with m + 1 teeth

æ UFL0031Y

Magnetoresistive steering-wheel-angle sensor LWS3 The LWS 3 also depends upon AMR (anisotropic magnetoresistive sensors) for its operation. The AMR’s electrical resistance changes according to the direction of an external magnetic field. In the LWS3, the information on angle across a range of four complete rotations is provided by measuring the angles of two gearwheels which are rotated by a third gearwheel on the steering-column shaft. The first two gearwheels differ by one tooth which means that a definite pair of angular variables is associated with every possible steering-wheel position. By applying a mathematical algorithm (a computing process which follows a defined step-by-step procedure) referred to here as a modified vernier principle, it is possible to use the above AMR method for calculating the steering-wheel angle in a microcomputer. Here, even the measuring inaccuracy of the two AMR sensors can be compensated for. In addition, a self-check can also be implemented so that a highly plausible measured value can be sent to the ECU. Fig. 2 shows the schematic representation of the LWS3 steering-wheel-angle sensor. The two gearwheels, with magnets inserted, can be seen. The sensors are located above them togther with the evaluation electronics. With this design too, price pressure forces the development engineers to look for innovative sensing concepts. In this respect, investigation is proceeding on whether, since it only measures up to 360°, a single AMR angle-of-rotation sensor (LWS4) on the end of the steering shaft would be accurate enough for ESP (Fig. 4).

2

1

2

3 4

æ UFL0032Y

processor in which plausibility tests are performed and information on angular position decoded and conditioned ready for the CANBus. The bottom half of the assembly contains the step-down gearing and the remaining five Hall-effect vane switches. The LWS1 was superseded by the LWS3 due to the large number of sensor elements required, together with the necessity for the magnets to be aligned with the Hall-IC.

Steering-wheel-angle sensors

Fig. 4 1 Steering column 2 Steering box 3 Steering-wheelangle sensor 4 Steering rack


Acceleration sensors and vibration sensors

Hall-effect acceleration sensors

Hall-effect acceleration sensors Applications Vehicles equipped with the Antilock Braking System ABS, the Traction Control System TCS, all-wheel drive, and/or Electronic Stability Program ESP, also have a Hall-effect acceleration sensor in addition to the wheelspeed sensors. This measures the vehicle’s longitudinal and transverse accelerations (depending upon installation position referred to the direction of travel).

Hall-effect acceleration sensor (opened)

1 3

a

Fig. 1 a Electronic circuitry b Spring-mass system 1 Hall-effect sensor 2 Permanent magnet 3 Spring

2

æ NAE0795Y

1

b

2

Design and construction A resiliently mounted spring-mass system is used in the Hall-effect acceleration sensors (Figs. 1 and 2). It comprises an edgewise-mounted strip spring (3) tightly clamped at one end. Attached to its other end is a permanent magnet (2) which acts as the seismic mass. The actual Hall-effect sensor (1) is located above the permanent magnet together with the evaluation electronics. There is a small copper damping plate (4) underneath the magnet. Operating concept When the sensor is subjected to acceleration which is lateral to the spring, the springmass system changes its neutral position accordingly. Its deflection is a measure for the acceleration. The magnetic flux F from the moving magnet generates a Hall voltage UH in the Hall-effect sensor. The output voltage UA from the evaluation circuit is derived from this Hall voltage and climbs linearly along with acceleration (Fig. 3, measuring range approx. 1 g). This sensor is designed for a narrow bandwidth of several Hz and is electrodynamically damped. 3

Hall-effect acceleration sensor

Hall-effect acceleration sensor (example of curve)

V

UH = const · a

4 U0

Output voltage UA

1

Φ

N S 3

2 a

2

1

4

0 -1g

0g Acceleration a

æ UAE0796E

IW

æ UAE0645Y

Fig. 2 1 Hall-effect sensor 2 Permanent magnet 3 Spring 4 Damping plate IW Eddy currents (damping) UH Hall voltage U0 Supply voltage Φ Magnetic flux a Applied (transverse) acceleration

3

1g

Robert Bosch GmbH Acceleration sensors and vibration sensors

Piezoelectric acceleration sensors

2

Piezoelectric acceleration sensors

61

Piezoelectric acceleration sensor (dual sensor for vertical mounting)

Application Piezoelectric bimorphous bending elements and two-layer piezoceramic elements are used as acceleration sensors in passenger-restraint systems for triggering the seat-belt tighteners, the airbags, and the roll-over bar.

1

Bending element from a piezoelectric acceleration sensor

a

1

Fig. 2 1 Bending element

For signal conditioning, the acceleration sensor is provided with a hybrid circuit comprised of an impedance converter, a filter, and an amplifier. This serves to define the sensitivity and useful frequency range. The filter suppresses the high-frequency signal components. When subjected to acceleration, the piezo bending elements deflect to such an extent due to their own mass that they generate a dynamic, easy-to-evaluate non-DC signal with a maximum frequency which is typically 10 Hz.

a =0

U A= 0

b

1

æ UAE0797Y

Design and operating concept A piezo bending element is at the heart of this acceleration sensor. It is a bonded structure comprising two piezoelectric layers of opposite polarities (“bimorphous bending element”). When subjected to acceleration, one half of this structure bends and the other compresses, so that a mechanical bending stress results (Fig. 1). The voltage resulting from the element bend is picked off at the electrodes attached to the sensor element’s outside metallized surfaces. The sensor element shares a hermeticallysealed housing with the initial signal-amplification stage, and is sometimes encased in gel for mechanical protection.

By “reversing” the actuator principle and applying voltage, the sensor’s correct operation can be checked within the framework of OBD “on-board diagnosis”. All that is required is an additional actuator electrode. 1

a=0

æ UAE0293-1Y

UA>0

Depending upon installation position and direction of acceleration, there are single or dual sensors available (Fig. 2). Sensors are also on the market which are designed specifically for vertical or horizontal mounting (Fig. 2).

Fig. 1 a Inoperative b Applied acceleration a 1 Piezo-ceramic bimporphous bending element UA Measuring-circuit voltage


Acceleration sensors and vibration sensors

Surface micromechanical acceleration sensors

Surface micromechanical acceleration sensors Application Surface micromechanical acceleration sensors are used in passenger-restraint systems to register the acceleration values of a frontal or side collision. They serve to trigger the seatbelt tightener, the airbag, and the rollover bar. Design and operating concept Although these sensors were initially intended for use with higher accelerations (50...100 g), they also operate with lower acceleration figures when used in passengerrestraint systems. They are much smaller than the bulk silicon sensors (typical edge length: approx. 100...500 µm), and are mounted together with their evaluation electronics (ASIC) in a waterproof casing (Fig. 1). An additive process is used to build up their spring-mass system on the surface of the silicon wafer.

1

The seismic mass with its comb-like electrodes (Figs. 2 and 3, pos. 1) is springmounted in the measuring cell. There are fixed comb-like electrodes (3, 6) on the chip on each side of these movable electrodes. This configuration comprising fixed and movable electrodes corresponds to a series circuit comprising two differential capacitors (capacity of the comb-like structure: approx. 1 pF). Opposed-phase AC voltages are applied across the terminals C1 and C2, and their superimpositions picked-off between the capacitors at CM (measurement capacity), in other words at the seismic mass. Since the seismic mass is spring-mounted (2), linear acceleration in the sensing direction results in a change of the spacing between the fixed and movable electrodes, and therefore also in a change in the capacity of C1 and C2 which in turn causes the electrical signal to change. In the evaluation electronics circuit, this change is amplified, and then filtered and digitalized ready for further signal processing in the airbag ECU. Due to the low capacity of approx. 1 pF, the evaluation electronics is situated at the sensor and is

Surface micromechanical acceleration sensors for airbag triggering (Example)

a

b

3

1

Fig. 1 a Side-airbag sensor b Front-airbag sensor 1 Casing 2 Sensor and evaluation chip 3 Cover

3 1

2

æ UAE0799Y

2

Robert Bosch GmbH Acceleration sensors and vibration sensors

either integrated with the sensor on the same chip, or is located very close to it. Closed-loop position controls with electrostatic return are also available. The evaluation circuit incorporates functions for sensor-deviation compensation and for self-diagnosis during the sensor start-up phase. During self-diagnosis, electrostatic forces are applied to deflect the comb-like structure and simulate the processes which take place during acceleration in the vehicle. 2

Comb-like structure of the sensor measuring element

Surface micromechanical acceleration sensors

63

Dual micromechanical sensors (4) are used for instance in the ESP Electronic Stability Program for vehicle dynamics control: Basically, these consist of two individual sensors, whereby a micromechanical yaw-rate sensor and a micromechanical acceleration sensor are combined to form a single unit. This reduces the number of individual components and signal lines, as well as requiring less room and less attachment hardware in the vehicle.

4

100 m

Lateral-acceleration sensor combined with yaw-rate sensor (dual sensor)

Ω

3

Fig. 2 1 Spring-mounted seismic mass with electrode 2 Spring 3 Fixed electrodes

2

Fig. 4 a Acceleration in sensing direction Ω Yaw rate

Surface micromechanical acceleration sensor with capacitive pick-off

1

2

3 C2

CM

C1

a

C1

C2 4

5

CM 6

7

æ UAE0801Y

3

a

æ UAE0678Y

UAE0800Y

1

Fig. 3 1 Spring-mounted seismic mass with electrodes 2 Spring 3 Fixed electrodes with capacity C1 4 Printed Al conductor 5 Bond pad 6 Fixed electrodes with capacity C2 7 Silicon oxide a Acceleration in sensing direction CM Measuring capacity


Wheel-speed sensors

Wheel-speed sensors Application It is from the wheel-speed sensor signals that the ABS, TCS, and ESP control units (ECUs) derive the wheel-rotation rates. These wheel speeds are applied in preventing the wheels blocking or spinning so that the vehicle’s stability and steerability are maintained. In vehicle navigation systems, the signals are used for calculating the distance travelled. Design and operating concept

Passive (inductive) wheel-speed sensors The inductive wheel-speed sensor's pole pin, surrounded by its coil winding, is installed directly above a trigger wheel (rotor) attached to the wheel hub. This soft-magnetic pole pin is connected to a permanent magnet which projects a magnetic field toward and into the trigger wheel. The continuously alternating sequence of teeth and gaps that accompanies the wheel's rotation induces corresponding fluctuations in the magnetic field through the pole pin and its coil winding. These fluctuations induce an alternating current in the coil suitable for monitoring at the ends of its winding.

Fig. 1 a Chisel pole pin: Radial installation, radial scan b Rhombus pole pin: Axial installation, radial scan c Round pole pin: Radial installation, axial scan 1 Sensor case with electrical connections 2 Permanent magnet 3 Soft-iron core (pole pin) 4 Winding 5 Trigger wheel

The frequency and amplitude of this alternating current are proportional to wheel speed, and with the wheel not rotating, the induced voltage is zero. Tooth shape, air gap, rate of voltage rise, and the ECU input sensitivity define the smallest still measurable rotation rate and thus, for ABS applications, the minimum switching speed. To ensure interference-free signal detection, the gap separating the wheel-speed sensor and the trigger wheel is only approx. 1 mm, and installation tolerances are narrow. The wheel-speed sensor is also installed on a stable mounting to prevent oscillation patterns in the vicinity of the brakes from distorting the sensor's signals. Various pole-pin configurations and installation options are available to adapt the system to the different

installation conditions encountered with various wheels. The most common variant is the chisel-type pole pin (also called a flat pole pin Fig. 1a) for radial installation at right angles to the pulse rotor. The rhombus-type (lozenge-shaped) pole pin (Figure 1b) designed for axial installation is located radially with respect to the trigger wheel. Both pole-pin designs necessitate precise alignment to the trigger wheel. Although precise alignment is not so important with the round pole pin (Figure 1c), the trigger wheel must have a large enough diameter, or less teeth.

1

Wheel-speed sensors: Pole-pin shapes and types of installation (DF6 as example)

a

1

2 3 4 5

b

c

1 2

1 2

æ UAE0476-1Y

64


A typical feature of the active wheel-speed sensor is the local amplifier circuit. Both components – measuring element and amplifier – are integrated in a single sensor casing. The active sensor requires a power supply of between 4.5 and 20 volts, and it is connected to the ECU by a two-conductor wire. The wheel-speed data is impressed on one of the two conductors (supply lines) as load-independent current. As with the inductive wheel-speed sensor, the current's

Data transmission identifying the wheel's direction of travel. This option is especially significant for the “hill-holding” feature, which relies on selective braking to prevent the vehicle from rolling backwards when starting off on a hill. Also used in vehicle navigation systems. Relay of information on sensor-signal quality, including a display indicating that the driver should have the vehicle serviced in order to check correct sensor functioning. 2

Active wheel-speed sensor showing a section of the multipole ring

1

2

3

æ UAE0688-1Y

The most important sensor components are either Hall or magnetoresistive elements, both of which generate a voltage that varies according to the magnetic flux through the measuring element. This voltage is then conditioned by the active wheel-speed sensor. One of the active sensor's advantages is the fact that in contrast to the inductive sensors, its output voltage is independent of the wheel speed. This fact permits monitoring to continue until the wheel is practically stationary.

65

frequency is proportional to wheel speed. This single-wire data-transmission strategy uses pre-conditioned digital signals. These are less sensitive to interference than the signals from the inductive sensor. The concept also features the following options:

3

Fig. 2 1 Multipole ring 2 Sensor element 3 Sensor case

Example of sensor installation in the wheel bearing

1

2

3

æ UAE0878Y

Active wheel-speed sensors The conventional inductive units are increasingly being replaced by active wheelspeed sensor types in which the function formerly performed by the trigger ring's teeth is taken over by peripheral magnets. These are installed around the periphery of a multipole ring so that their polarities alternate (Fig. 2). The sensor element of such an active wheel-speed sensor is located in the continuously changing fields generated by these magnets. Rotation of the multipole ring is thus accompanied by a continuous alternation in the magnetic flux through the sensor element. Compact dimensions combine with low weight to make the active wheel-speed sensor suitable for installation on and even within the vehicle’s wheel-bearing assemblies (Fig. 3). In the latter case, the bearing seal contains magnetic powder instead of fixed magnets. This means that a second function has been added and the bearing seal now becomes a multipole device.

Wheel-speed sensors

Fig. 3 1 Wheel bearing 2 Sensor 3 Multipole ring/ Bearing seal


Data processing in the vehicle

Requirements/Microcomputer/Electronic control unit (ECU)

Data processing in the vehicle Requirements

Microcomputer

Highly sophisticated state-of-the-art openloop and closed-loop control concepts are essential for meeting the demands for function, safety, environmental compatibility and comfort associated with the wide range of automotive subsystems installed in modernday vehicles. Sensors monitor the reference and controlled variables, which an electronic control unit (ECU) then converts into the signals required to adjust the final controlling elements/actuators. The input signals can be analog (e.g. voltage characteristic at pressure sensor), digital (e.g. switch position) or pulse-shaped (i.e. information content as a function of time; e.g. engine-speed signal). These signals are processed after being conditioned (filtering, amplification, pulse shaping) and converted (analog/digital); digital signal-processing methods are preferred. Thanks to modern semiconductor technology, powerful computer units, with their accompanying program and data memories, and special peripheral circuitry, designed specifically for real-time applications, can all be integrated on only a few chips. Modern vehicles are equipped with numerous digital control units (ECUs), e.g. for engine management, ABS, and transmissionshift control. Improved performance and additional functions are obtained by synchronizing the processes controlled by the individual control units, and by adapting (in real time) their respective parameters to each other. An example of this type of function is traction control (TCS) which reduces the driving torque when the drive wheels spin. Up to now, data between the control units (in the example cited above, ABS/TCS and engine management) has been exchanged mostly through separate lines. However, this type of point-to-point connection is only suitable for a limited number of signals. The data-transmission potential between the individual ECUs can be enhanced by using a simple network topology designed specifically for serial data transmission in automotive applications.

The microcomputer comprises both the central processing unit (CPU) for processing arithmetic operations and logical relationships, and special function modules to monitor external signals and to generate the control signals for external servo elements. These peripheral modules are largely capable of assuming complete control of real-time operations. The program-controlled CPU could only discharge these at the price of both additional complication and curtailment in the number of functions (e.g. determining the moment at which an event occurred). Computing power Apart from the architecture (e.g. accumulator, register machine) and the word length (4 ... 32 bits), the product of the internal clock frequency and the average number of clock pulses required per instruction determines the CPU’s power: Clock frequency: 1 ... 40 MHz (typical), Clock pulses per instruction: 1 ... 32 pulses (typical), depending on the CPU's architecture and the instruction (e.g. 6 pulses for addition, 32 pulses for multiplication).

Electronic control unit (ECU) Digital input signals For registering a switch position or digital sensor signals (e.g. rotational-speed pulses from a Hall-effect sensor). Voltage range: 0 V to battery voltage. Analog input signals Signals from analog sensors (lambda sensor, pressure sensor, potentiometer). Voltage range: Several mV up to 5 V. Pulse-shaped input signals Signals from inductive rpm sensors. After signal conditioning, they are further processed as digital signals. Voltage range: 0.5 V to 100 V.

… …

Counter 8 … 64 bit

…

(Timer, time processing unit, input capture, output-compare register)

I/O Signal acquisition and output with time reference

Resolution 50 ns Time range 50 ns … 1s

I/O Event counter

4-, 8-, 16-, 32-bit data circuit

…

Resolution 8 …10 bit 4 … 32 channels

I/O Analog/ Digital (A/D) converter

…

8 … 32 channels

I/O Digital inputs/ outputs (I/O)

Memory capacity 2 kbytes … 512 kbytes

Memory capacity 64 bytes … 32 kbytes

Bus

Program memory Read-only memory (ROM, EPROM, flash EPROM) For programs and permanent data records

Data memory (User memory) Volatile read/write memory (RAM) for variable data

…

Data rate 200 bit/s … 1 Mbit /s

(UART, SPI, CAN)

I/O Serial interface

…

Communication with external chips via address/ data bus

I/O Bus controller

Memory capacity 32 bytes … 1 kbyte

Data memory Non-volatile read/ write memory (EEPROM)


æ UAE0454-1E

Monitoring circuit (watchdog)

I/O Interrupt controller

Internal clock (oscillator)

Arithmetic and Logic Unit (ALU) 4-, 8-, 16-, 32-bit

Microprocessor Central Processing Unit (CPU)

1

Microcomputer

Robert Bosch GmbH Electronic control unit (ECU)

Microcomputer 67



Electronic control unit (ECU)

most unlimited number of logic operations can be established and data records stored and processed in the form of parameters, characteristic curves and multidimensional program maps.

Signal conditioning Protective circuits (passive: R and RC circuits; active: special surge-proof semiconductor elements) are used to limit the voltage of the input signals to acceptable levels (microcomputer operating voltage). Filters remove most of the superimposed noise from the useful signals, which are then amplified to the microprocessor's input voltage. Voltage range: 0 V to 5 V.

Output signals Power switches and power-gain circuits amplify the microprocessor's output signals (0 V .. 5 V, several mA) to the levels required by the various final-controlling elements/ actuators (battery voltage, several A).

Signal processing ECUs usually process signals in digital form. Rapid, periodic, real-time signals are processed in hardware modules specifically designed for the particular function. Results, e.g. a counter reading or the time of an event, are transmitted in registers to the CPU for further processing. This procedure substantially reduces the CPU’s interruptresponse requirements (µs range).

The amount of time available for calculations is determined by the open-loop or closed-loop control system (ms range). The software contains the actual control algorithms. Depending on the data, an al2

Signal processing in ECU

Signal conditioning

Microcomputer

Power controller 8

1

3 8 6 9

2

3

4

5

7

2 9

æ UAE0455-1E

Fig. 2 1 Digital input signals 2 Analog input signals 3 Protective circuit 4 Amplifier, filter 5 A/D converter 6 Digital signal processing 7 D/A converter 8 Circuit-breaker 9 Power amplifier

Robert Bosch GmbH Data processing in the vehicle

Complete system

69

between the individual control units. Procedures have been laid down for bus access, message structure, bit and data coding, error recognition and response, and the identification of faulty bus users (CAN).

Complete system Logistical concept (CARTRONIC) This concept divides the vehicle’s complete electrical system into conveniently dimensioned subsystems. Units with closelycoupled functions (that is, units with high rates of mutual data exchange) are combined in a sub-network. Although this logistical concept results in sub-networks with varying requirements on transmission capacity, demands on data exchange do not vary.

Transmission speed Multiplex bus: 10 kbit/s...125 kbit/s, Drivetrain bus: 125 kbit/s...1 Mbit/s, Telecommunications bus: 10 kbit/s...125 kbit/s. Latency time Latency time is defined as the time that elapses between the transmitter’s send request and the target station’s receipt of the error-free message. Multiplex bus: 5 ms...100 ms, Drivetrain bus: 0.5 ms...10 ms, Telecommunications bus: 5 ms...100 ms.

Topology At the logical level, all the known communications systems developed for automotive applications are based on a single serial connection of the ECUs. The physical layout employs one-wire or differential two-wire interfaces in bus form to interconnect the control units. Protocol The protocol consists of a number of a specific collection execution statements which are used to control data communications Bus-system interfacing

1

3

1 2

2

2

1 2

1 2

2

...

Drivetrain bus

1

1 2

2

2

...

Multiplex bus

1

2

1

Telecommunications bus

1 2

...

2

æ UAE0456-1E

3

Fig. 3 1 ECU 2 Bus controller 3 Gateway


Automotive microelectronics

Severe demands on electronic systems/History of development

Severe demands on electronic systems

History of development

Electronic systems in motor vehicles are exposed to extreme stresses (e.g. due to extreme temperature variations, unusual climatic conditions, poor road surfaces and the effects of corrosive substances). These are some of the requirements they must meet in order to be able to function reliably and without faults over long periods: Resistance to temperatures ranging from –40 °C ... 125 °C EMC (electromagnetic compatibility): immunity to external interference (e.g. mobile phone signals) and no emission of electromagnetic radiation likely to cause interference on other equipment Resistance to shocks and vibration Resistance to water and damp Resistance to corrosive fluids (e.g. oils and salt-water spray) Light weight Economical production costs and Secure and trouble-free mounting

The amount of electronic equipment in motor vehicles is continually increasing. Fig. 2 provides an overview of the growth of electronic equipment expressed as a proportion of vehicle cost. Because of their cost, electronic systems were initially reserved for vehicles at the luxury end of the market. This explains why in 1980 electronic equipment accounted for only half a percent of vehicle cost. From that time onwards, and particularly in the 1990s, that proportion grew rapidly as the price of electronic equipment continually dropped with the result that more and more electronic systems could be fitted to mid-range and even small cars. Gradually, more and more electronic systems were fitted to motor vehicles (Table 1). And the trend continues. The new science of mechatronics deals with the interaction between mechanical, electronic and data processing devices. 1

2

1958 1962 1965 1967

Proportion of cost per vehicle (overall average)

0.5 %

7%

1978 1979

1990

17 %

2000 Table 1

24 %

2010 Prognosis

æ UAE0855E

1980

1982 1982 1986 1986 1987 1989 1989 1989 1989 1991 1994 1997 1997 2000 2000

Historical development of electronic systems in motor vehicles (examples) DC generator with variode 3-phase alternator with variode Transistorized ignition D-Jetronic gasoline-injection system (pressure-controlled) Antilock braking system (ABS) Motronic (combined ignition and fuel-injection system) Electronic ignition system Knock control Electronic diesel control (EDC) Electronic throttle control (ETC) Traction control system (TCS) Electronic transmission-shift control (“stand-alone” system Tiptronic) Mono-Motronic (single-point injection system SPI) CAN (Controller Area Network) Vehicle navigation system (Travelpilot) Litronic ME-Motronic (integrated ETC) Electronic stability program (ESP) Radiophone (car radio + mobile telephone) MED-Motronic (gasoline direct injection) Adaptive Cruise Control (ACC)

Robert Bosch GmbH Automotive microelectronics

Miniaturization Transistorized ignition was first used on gasoline engines in 1965. It did away with the negative effect on ignition timing accuracy of erosion caused by electrical arcing between the contact-breaker points. The transistor had gained a foothold in the motor vehicle, heralding the start of the electronic age. But is wasn’t until electronic componentry was miniaturized that the decisive step was taken in making automotive electronic systems capable of the levels of performance that are taken for granted today. Enormous advances in miniaturization were made in the area of semiconductor components in particular, making it possible to integrate more and more functions within a component that occupied only a tiny amount of space. Every ECU contains microcontrollers that combine millions of transistor functions on semiconductor chips that take up only a few square millimetres. It has also been possible to substantially reduce the dimensions of power components such as output stages for controlling actuators. For example, multiple ignitionoutput stages are now combined in a single component. This means that an external ignition output stage is no longer required. It is now integrated in the engine-management

71

ECU. Consequently, the external ignitionoutput stage previously used can be dispensed with. The associated reduction in the number of components also improves the reliability of the system. The size of discrete components (resistors, capacitors) has similarly been significantly reduced. SMDs (Surface Mounted Devices) are soldered or bonded to the circuit board without wire connections. In spite of the continual growth in the number and complexity of the functions performed, the miniaturisation of electronic equipment has meant that the size of the ECUs continues to shrink (Fig. 3). Memory capacity Whereas in 4 kilobyte memory capacity was adequate for the modest requirements of a gasoline-engine management system in the late 1970s, 10 years later the figure had reached 30 kilobytes. The incorporation of more and more functions in the enginemanagement ECU led to an explosion in the demand for memory capacity. By the year 2000, the required capacity had reached 500 kilobytes. Other automotive electronic systems have followed a similar pattern of development. And there is no foreseeable end to this trend.

Gasoline-engine ECUs

a

b

æ UAE0857Y

3

History of development

Fig. 3 a 1979 Jetronic ECU with 290 components and a weight of 1.14 kg b 1996 Motronic hybrid ECU with 82 components and a weight of 0.25 kg

Robert Bosch GmbH Data transfer between electronic systems

System overview, serial data transfer (CAN)

Data transfer between automotive electronic systems Today’s vehicles are being equipped with a constantly increasing number of electronic systems. Along with their need for extensive exchange of data and information in order to operate efficiently, the data quantities and speeds concerned are also increasing continuously. For instance, in order to guarantee perfect driving stability, the Electronic Stability Program (ESP) must exchange data with the engine-management system and the transmission-shift control.

System overview Increasingly widespread application of electronic communications systems, and electronic open and closed-loop control systems, for automotive functions such as Electronic engine management (EDC and Motronic), Electronic transmission-shift control (GS), Antilock braking system (ABS), Traction control system (TCS), Electronic Stability Program (ESP), Adaptive Cruise Control (ACC), and Mobile multimedia systems together with their display instrumentation has made it vital to interconnect the individual ECUs by means of networks. The conventional point-to-point exchange of data through individual data lines has reached its practical limits (Fig. 1), and the complexity of current wiring harnesses and the sizes of the associated plugs are already very difficult to manage. The limited number of pins in the plug-in connectors has also slowed down ECU development work.

To underline this point: Apart from being about 1 mile long, the wiring harness of an average middle-class vehicle already includes about 300 plugs and sockets with a total of 2000 plug pins. The only solution to this predicament lies in the application of specific vehicle-compatible Bus systems. Here, CAN has established itself as the standard.

Serial data transfer (CAN) Although CAN (Controller Area Network) is a linear bus system (Fig. 2) specifically designed for automotive applications, it has already been introduced in other sectors (for instance, in building-installation engineering). Data is relayed in serial form, that is, one after another on a common bus line. All CAN stations have access to this bus, and via a CAN interface in the ECUs they can receive and transmit data through the CAN bus line. Since a considerable amount of data can be exchanged and repeatedly accessed on a single bus line, this networking results in far fewer lines being needed.

1

Conventional data transfer

Transmission-shift control Station 1

Engine management Station 2

ABS/TCS/ESP Station 3

Instrument cluster Station 4

æ UAE0745E

72


Applications in the vehicle For CAN in the vehicle there are four areas of application each of which has different requirements. These are as follows:

Multiplex applications Multiplex is suitable for use with applications controlling the open and closed-loop control of components in the sectors of body electronics, and comfort and convenience. These include climate control, central locking, and seat adjustment. Transfer rates are typically between 10 kbaud and 125 kbaud (1 kbaud = 1 kbit/s) (low-speed CAN). Mobile communications applications In the area of mobile communications, CAN networks such components as navigation system, telephone, and audio installations with the vehicle’s central display and operating units. Networking here is aimed at standardizing operational sequences as far as possible, and at concentrating status information at one point so that driver distraction is reduced to a minimum. With this application, large quantities of data are transmitted, and data transfer rates are in the 125 kbaud range. It is impossible to directly transmit audio or video data here.

Linear bus topology

Transmission-shift control Station 1

Engine management Station 2

CAN

ABS/TCS/ESP Station 3

Instrument cluster Station 4

æ UAE0283-2E

2

Serial data transfer (CAN)

Diagnosis applications The diagnosis applications using CAN are aimed at applying the already existing network for the diagnosis of the connected ECUs. The presently common form of diagnosis using the special K line (ISO 9141) then becomes invalid. Large quantities of data are also transferred in diagnostic applications, and data transfer rates of 250 kbaud and 500 kbaud are planned. Real-time applications Real-time applications serve for the open and closed-loop control of the vehicle's movements. Here, such electronic systems as engine management, transmission-shift control, and electronic stability program (ESP) are networked with each other. Commonly, data transfer rates of between 125 kbaud and 1 Mbaud (high-speed CAN) are needed to guarantee the required realtime response. Bus configuration Configuration is understood to be the layout and interaction between the components in a given system. The CAN bus has a linear bus topology (Fig. 2) which in comparison with other logical structures (ring bus and/or star bus) features a lower failure probability. If one of the stations fails, the bus still remains fully accessible to all the other stations. The stations connected to the bus can be either ECUs, display devices, sensors, or actuators. They operate using the Multi-Master principle, whereby the stations concerned all have equal priority regarding their access to the bus. It is not necessary to have a higher-order administration.

73



Content-based addressing The CAN bus system does not address each station individually according to its features, but rather according to its message contents. It allocates each “message” a fixed “identifier” (message name) which identifies the contents of the message in question (for instance, engine speed). This identifier has a length of 11 bits (standard format) or 29 bits (extended format). With content-based addressing each station must itself decide whether it is interested in the message or not (“message filtering” Fig. 3). This function can be performed by a special CAN module (Full-CAN), so that less load is placed on the ECU’s central microcontroller. Basic CAN modules “read” all messages. Using content-based addressing, instead of allocating station addresses, makes the complete system highly flexible so that equipment variants are easier to install and operate. If one of the ECUs requires new information which is already on the bus, all it needs to do is call it up from the bus. Similarly, provided they are receivers, new stations can be connected (implemented) without it being necessary to modify the already existing stations.

3

Figure 3 Station 2 transmits, Station 1 and 4 accept the data.

Addressing and message filtering (acceptance check)

CAN Station 1

CAN Station 2

CAN Station 3

Bus arbitration The identifier not only indicates the data content, but also defines the message’s priority rating. An identifier corresponding to a low binary number has high priority and vice versa. Message priorities are a function for instance of the speed at which their contents change, or their significance with respect to safety. There are never two (or more) messages of identical priority in the bus. Each station can begin message transmission as soon as the bus is unoccupied. Conflict regarding bus access is avoided by applying bit-by-bit identifier arbitration (Fig. 4), whereby the message with the highest priority is granted first access without delay and without loss of data bits (nondestructive protocol). The CAN protocol is based on the logical states “dominant” (logical 0) and “recessive” (logical 1). The “Wired And” arbitration principle permits the dominant bits transmitted by a given station to overwrite the recessive bits of the other stations. The station with the lowest identifier (that is, with the highest priority) is granted first access to the bus.

4

Bit-by-bit arbitration (allocation of bus access in case of several messages)

CAN Station 4 Bus line

Accept

Provision

Accept

Station 1 Selection

Selection

Station 2 Reception

Reception

Bus 0 1

Dominant level Recessive level

Selection

Reception

æ UAE0284-1E

Figure 4 Station 2 gains first access (Signal on the bus = signal from Station 2)

Send message

Station 3

1 0 1 0 1 0 1 0 Station 1 loses the arbitration

Station 3 loses the arbitration

æ UAE0742E

74


Message format CAN permits two different formats which only differ with respect to the length of their identifiers. The standard-format identifier is 11 bits long, and the extended-format identifier 29 bits. Both formats are compatible with each other and can be used together in a network. The data frame comprises seven consecutive fields (Fig. 5) and is a maximum of 130 bits long (standard format) or 150 bits (extended format).

The bus is recessive at idle. With its dominant bit, the “Start of frame” indicates the beginning of a message and synchronizes all stations. The “Arbitration field” consists of the message’s identifier (as described above) and an additional control bit. While this field is being transmitted, the transmitter accompanies the transmission of each bit with a check to ensure that it is still authorized to transmit or whether another station with a higher-priority message has accessed the Bus. The control bit following the identifier is designated the RTR-bit (Remote Transmission Request). It defines whether the message is a “Data frame” (message with data) for a receiver station, or a “Remote frame” (request for data) from a transmitter station.

The “Data field” contains the actual message information comprised of between 0 and 8 bytes. A message with data length = 0 is used to synchronize distributed processes. A number of signals can be transmitted in a single message (e.g. engine rpm and engine temperature). The “CRC Field” (Cyclic Redundancy Check) contains the frame check word for detecting possible transmission interference. The “ACK Field” contains the acknowledgement signals used by the receiver stations to confirm receipt of the message in noncorrupted form. This field comprises the ACK slot and the recessive ACK delimiter. The ACK slot is also transmitted recessively and overwritten “dominantly” by the receivers upon the message being correctly received. Here, it is irrelevant whether the message is of significance or not for the particular receiver in the sense of the message filtering or acceptance check. Only correct reception is confirmed.

5

CAN message format

Start of Frame Arbitration Field Control Field Data Field CRC Field ACK Field End of Frame Inter Frame Space 1 IDLE 1* 12* 6* 0...64* 16*

The “Control field” contains the IDE bit (Identifier Extension Bit) used to differentiate between standard format (IDE = 0) and extended format (IDE = 1), followed by a bit

75

reserved for future extensions. The remaining 4 bits in this field define the number of data bytes in the next data field. This enables the receiver to determine whether all data has been received.

0 Data frame Message frame

2* 7* 3* IDLE

æ UAE0285-1E

The transmitters with low-priority messages automatically become receivers, and repeat their transmission attempt as soon as the bus is vacant again. In order that all messages have a chance of entering the bus, the bus speed must be appropriate to the number of stations participating in the bus. A cycle time is defined for those signals which fluctuate permanently (e.g. engine speed).


Figure 5 0 Dominant level, 1 Recessive level. * Number of bits


Data transfer between electronic systems


The “End of frame” marks the end of the message and comprises 7 recessive bits. The “Inter-frame space” comprises three bits which serve to separate successive messages. This means that the bus remains in the recessive IDLE mode until a station starts a bus access. As a rule, a sending station initiates data transmission by sending a “data frame”. It is also possible for a receiving station to call in data from a sending station by transmitting a “remote frame”. Detecting errors A number of control mechanisms for detecting errors are integrated in the CAN protocol.

In the “CRC field”, the receiving station compares the received CRC sequence with the sequence calculated from the message. With the “Frame check”, frame errors are recognized by checking the frame structure. The CAN protocol contains a number of fixed-format bit fields which are checked by all stations. The “ACK check” is the receiving stations’ confirmation that a message frame has been received. Its absence signifies for instance that a transmission error has been detected. “Monitoring” indicates that the sender observes (monitors) the bus level and compares the differences between the bit that has been sent and the bit that has been checked. Compliance with “Bitstuffing” is checked by means of the “Code check”. The stuffing rule stipulates that in every “data frame” or “remote frame”, a maximum of 5 successive equal-priority bits may be sent between the “Start of frame” and the end of the “CRC field”. As soon as five identical bits have been transmitted in succession, the sender inserts an opposite-priority bit. The receiving sta-

tion erases these opposite-polarity bits after receiving the message. Line errors can be detected using the “bitstuffing” principle. If one of the stations detects an error, it interrupts the actual transmission by sending an “Error frame” comprising six successive dominant bits. Its effect is based on the intended violation of the stuffing rule, and the object is to prevent other stations accepting the faulty message. Defective stations could have a derogatory effect upon the bus system by sending an “error frame” and interrupting faultless messages. To prevent this, CAN is provided with a function which differentiates between sporadic errors and those which are permanent, and which is capable of identifying the faulty station. This takes place using statistical evaluation of the error situations. Standardization The International Organization for Standardization (ISO) and SAE (Society of Automotive Engineers) have issued CAN standards for data exchange in automotive applications:

For low-speed applications up to 125 kbit/s: ISO 11519-2, and For high-speed applications above 125 kbit/s: ISO 11898 and SAE J 22584 (passenger cars) and SAE J 1939 (trucks and buses). Furthermore, an ISO Standard on CAN Diagnosis (ISO 15765 – Draft) is being prepared.


Prospects Along with the increasing levels of systemcomponent performance and the rise in function integration, the demands made on the vehicle’s communication system are also on the increase. And new systems are continually being introduced, for instance in the consumer-electronics sector. All in all, it is to be expected that a number of bus systems will establish themselves in the vehicle, each of which will be characterized by its own particular area of application. In addition to electronic data transmission, optical transmission systems will also come into use in the multimedia area. These are very-high-speed bus systems and can transmit large quantities of data as needed for audio and video components. Individual functions will be combined by networking to form a system alliance covering the complete vehicle, in which information can be exchanged via data buses. The implementation of such overlapping functions necessitates binding agreements cover-

ing interfaces and functional contents. The CARTRONIC® from Bosch is the answer to these stipulations, and has been developed as a priority-override and definition concept for all the vehicle’s closed and open-loop control systems. The possible sub-division of the functions which are each controlled by a central coordinator can be seen in Fig. 1. The functions can be incorporated in various ECUs. The combination of components and systems can result in completely novel functions. For instance, the exchange of data between the transmission-shift control and the navigation equipment can ensure that a change down is made in good time before a gradient is reached. With the help of the navigation facility, the headlamps will be able to adapt their beam of light to make it optimal for varying driving situations and for the route taken by the road (for instance at road intersections). Car radios, soundcarrier drives, TV, telephone, E-mail, Internet, as well as the navigation and terminal equipment for traffic telematics will be networked to form a multimedia system.

CARTRONIC®: Design schematic

Functions

Vehicle coordination

Drive

Vehicle movement

Bodywork and interior

Mobile Multimedia On-board electrical system

Actuators Sensors Modules

æ UAE0674-3E

1

Prospects

77


Index of technical terms

Index of technical terms Technical Terms A ACC, 4-6 ACC adjustment, 17 ACC control sequence, 38-44 ACC electronic module, 15 ACC SCU, 10, 15 ACC sensor & control unit, 15-21 Acceleration-control interface, 42 Acceleration limits, 43 Acceleration sensor, 24 Acceleration sensors, Hall-type, 60 Acceleration sensors, piezoelectric, 61 Acceleration sensors, surface mechanical, 62-63 ACK field (CAN), 75 Activation (controls and displays), 28 Active brake servo unit, 24 Active speed sensor, 65 Adaptive Cruise Control, 4 Addressing (CAN), 74 Adjustment (radar module), 17-21 Ambient sensors, 45, 46 Amplifier circuit, 14 AMR steering-wheel-angle sensor, 59 Analog input signals (data processing), 66 Analog/digital conversion, 32 Angular position, determining, 9 Antenna feed, 12 Antenna patch, 12 Antenna pattern, 9, 10 Approach speed, 44 Approximation method, 36 Arbitration field (CAN), 75 Areas of application on motor vehicles (CAN), 73

B Basic structure (ACC control unit), 38 Bend-detection function, 41 Bend prediction, 36 Bend-sensing systems, 24 Brake actuation system, 24 Brake actuator, hydraulic, 24 Brake control, 24 Brake light operation, 23 Brake servo unit, active, 24 Braking system control, 24 Bus arbitration (CAN), 74 Bus configuration (CAN), 73 Bus topology (CAN), 73

C Calculating (trajectory curvature), 35 CAN, 72 CAN interface module, 21 CAN protocol, 74, 7 Carrier frequency, 7 CARTRONIC®, 77 Change of direction, 35 Close-range radar, 45 Collision, 50 Collision avoidance, 50 Components (ACC), 6 Constant-gap function, 40 Content-related addressing (CAN), 74 Control field (CAN), 75 Control intervention, 33 Control mode, 42 Control unit (data processing), 66 Control-unit bus, 23 Control-unit structure, 38 Control units in motor vehicles, 27 Controller functions, 39-44 Controller parameters, 41 Course offset, 33 Course prediction, 34-37 CRC field (CAN), 75 Cruise control, 4, 39

D Data field (CAN), 75 Data frame, 75 Data processing in motor vehicles, 66-69 Data transmission, 23 Data transmission, serial, 72-77 Deactivation, 31 Deceleration control, 39 Deceleration threshold, 43 Desired distance, 30 Desired speed, 28, 29 Desired time gap, 28, 30 Detection, 32-37 Determining course, 35 Determining trajectory curvature, 35 Development stages (RADAR), 51 Diagnosis module, 21 Diagnostic applications (CAN), 73 Dielectric resonator oscillator, 14 Differential speed, 43 Digital-analog converter, 20 Digital electronics, 19-21 Display, 28-31 Display functions (other), 31 Divider circuit, 12

Doppler effect, 7 Driver-assistance system, 45 Drivetrain control, 23

E Echo signals, 32 Echo timing, 7 Electrohydraulic braking system, 24 Electronic components (ACC sensor), 16 Electronic hardware, 19-21 Electronic systems (properties), 70-71 End Of Frame (CAN), 76 Engine-output control, 39 ESP sensor signals, 24 ESP sensors, 24

F False alarm, 50 Fast Fourier Transform, 20, 32 Fault detection (CAN), 76 Fault diagnosis, 26 Fault, irreversible, 26 Fault, reversible, 26 Field of vision (radar sensor), 41 Field of vision, 42 FMCW modulation cycles, 9 FMCW radar, 8 Fog, 48 Forces acting on wheels, 39 Fourier Transform, 32 Frequency modulation, 8 Frequency modulator, 14 Frequency range, 14 Frequency shift, 7 Frequently asked questions, 48-50 Function level, 38 Functional limits, 42-44 Future developments, 45

G Glossary, 52-53 Gunn diode, 12 Gunn oscillator, 12

H Hall-type acceleration sensors, 60 Hall-type steering-wheel-angle sensor, 58 History (electronic systems), 70-71 Hydraulic brake actuator, 24

Robert Bosch GmbH Index of technical terms

I Indication of “object detected”, 30 Input signals (data processing), 66 Interfaces with actuation systems, 42 Inter-frame space (CAN), 76 ISO symbols, 29

L Lane change, 44 Lane prediction, 18 Lane probability, 34 Latency time (data processing), 69 Lateral offset, 33 Lens (radar module), 15 Lens heater, 21 Linear dynamics, 43 Linear-speed control, 39, 42 Location (radar module), 15 Logistical concept (CARTRONIC) (data processing), 69

M Magnetoresistive steering-wheel-angle sensor, 59 Main oscillator, 14 Maximum approach speed, 44 Memory capacity (electronic systems), 71 Message format (CAN), 75 Microcomputers, 66 Micromechanical yaw-rate sensors, 56-57 Microwave radar, 6 Millimeter-wave radar, 6 Miniaturization (electronic systems), 71 Misalignment (radar module), 18 Mixer, 13 Modulation of transmission frequency, 8 Monitoring (components), 25 Monitoring (function), 26 Monitoring level, 38 Monitoring, reciprocal, 26 Multigradient FMCW method, 32 Multiplex application (CAN), 73

N Navigation systems, 36 Networking, 22 Noise analysis, 32 Noise signal, 32

O Object classification, 34, 44 Object identification, 32 Object selection, 34 Object tracking, 33 Objects, stationary, 44 Operation, 28-31 Output signals (data processing), 68 Overall system (data processing), 69

P Partial deactivation, 31 Passive wheel-speed sensor, 64 Performance quality, 34 Physical principles of measurement, 7 Piezoelectric acceleration sensors, 61 Piezoelectric yaw-rate sensors, 55 Plausibility checking level, 38 Pre-amplifier, 14 Principle of radar ranging, 7 Principles of measurement (physical), 7 Prospects (data transmission), 77 Protocol (data processing), 69 Pulse-type input signals (data processing), 66

Q Questions, frequently asked, 48-50

R Radar (module) adjusting mechanism, 17 Radar (module) housing, 17 RADAR (principle of ranging), 7 RADAR (Radio Detection and Ranging), 51 RADAR beam, 10 RADAR frequencies, 7 RADAR lobes, 9, 10 RADAR modules, 10-14 Radar ranging sensor, 7-14 Radar signal processing, 32 Radar transceiver unit, 16 RADAR transceiver, 10, 11 RADAR transmitter and receiver, 10 Rain, 48 Real-time applications (CAN), 73 Reference oscillator, 14 Reflection, 7 Required acceleration, 39 Requirements (data processing), 66 Requirements (radar module), 15 Response options, 26

S Safety concept, 25-27 Sensor and control unit, 15 Sensor technology (future developments), 45 Sensors, 54 Sensotronic Brake Control, 24 Serial data transmission (CAN), 72 Set, 29 Set Speed Control, 4 Shutdown due to fault, ACC, 27 Signal conditioning (data processing), 68 Signal echo, 7 Signal processing (data processing), 68 Spectral analysis, 32 Speed range, 42 Standardisation (CAN), 76 Start Of Frame (CAN), 75 Stationary objects, 44 Steering-angle sensors, 24, 58-59 Stop-and-go (future developments), 47 Surface micromechanical acceleration sensors, 62-63 Symbols (ISO), 29 System architecture, 22 System network, 22-27 System overview (ACC), 4-6 System overview (data transmission), 72

T Target frequency, 32 Target-object selection, 34 Time-gap programs, 41 Topology (data processing), 69 Torque-control interface, 42 Tracking, 33 Trajectory curvature, 35 Trajectory curvature changes, 36 Transceiver, 16 Transceiver unit, 16 Transfer rates, 21 Transmission frequency, 8 Transmission power, 14 Transmission speed (data processing), 69

79


Index of technical terms

Abbreviations

U Use on motor vehicles (sensors), 54

V Vehicle trajectory, 35 Video-image analysis, 36 Voltage regulators, 21

W Wheel-speed sensors, 24, 64 Wheel-speed sensors (active), 25 Wheel-speed sensors, passive (inductive), 25

Y Yaw rate, 35 Yaw-rate sensor, 24 Yaw-rate sensors, micromechanical, 56-57 Yaw-rate sensors, piezoelectric, 55

Abbreviations A ACC: Adaptive Cruise Control ACC SCU: ACC Sensor & Control Unit ADC: Analog-digital converter

C CAN: Controller Area Network D DAC: Digital-analog converter DRO: Dielectric Resonator Oscillator DSP: Digital Signal Processor

E ECD: Electronically Controlled Deceleration EDC: Electronic Diesel Control EEPROM: Electrically Erasable Programmable Read-Only Memory EMC: Electromagnetic compatibility ESP: Electronic Stability Program

F FFT: Fast Fourier Transform FMCW: Frequency Modulated Continuous Wave FMCW RADAR: Frequency Modulated Continuous Wave RADAR

I ISO: International Organization for Standardization

M MIC: Microwave Integrated Circuit

P PU: Processing Unit

R RADAR: Radio Detection and Ranging RAM: Random Access Memory RPU: Regulation Processing Unit RTC: Radar Transceiver

S SBC: Sensotronic Brake Control SCU: Signal and Control Unit SMD: Surface Mounted Device SPU: Signal Processing Unit T TCS: Traction Control System

V VCO: Voltage Controlled Oscillator

ACC Adaptive Cruise Control_2003.pdf

Recommend Documents