Intelligent Traffic Controller Annie M. Santiago MSECE, 10795413
March 28, 2008
Table of Contents
I. Table of Contents
…………..
2
II. Chapter 1: Introduction
…………..
3
1.1 Background of the Study
…………..
3
1.2 Objectives
…………..
5
1.3 Statement of the Problem
…………..
5
…………..
6
2.1 Review of Related Studies
…………..
6
2.2 Review of Related Literature
…………..
7
…………..
11
…………..
11
3.1.1 Sensors
…………..
11
3.1.2 Signal Assembly
…………..
12
3.1.3 Controller
…………..
12
3.2 The Fuzzy Design
…………..
12
3.3. Implementation Schedule
…………..
14
…………..
15
III. Chapter 2: Review of Related Studies and Literature
IV. Chapter 3: Methodology 3.1 The Instruments
V. Chapter 4: References
2
CHAPTER 1: Introduction
1.1 Background of the Study
Living in a developing country, Filipinos are adversely inherited with various challenges in politics, economy, education system, and even transportation system. Mo Twister in his radio segment, Good Times with Mo, argues that We are dealing with problems like corruption, education system, and inflation everyday and nobody is complaining about it. But when it comes to traffic, everybody is bothered; everybody is complaining (Magic 89.9). Gumatay even argues that there is a greater possibility for a political candidate that has a platform of improving the traffic system to win compared to a candidate that has anticorruption as platform. Overlooking on the exaggeration of his statement and extracting the intellect that lies between the lines will reveal the factual setting of the traffic system in the Metro. Indeed, Gumatay’s thought on the Philippine traffic situation actually represents the sixteen millions of unhappy people roaming the streets of the Metro because of the unpleasant scenario every single day. Another robust allegation was dropped by Erik Van Mechelen (2007) in his article in The People’s Media Company entitled “Traffic in Manila, Philippines - Simply Devastating”. Mechelen said, “Manila, if wanted, could issue a single traffic report on a single day for all roads for the entire year. And this report would only require three words - very very busy. Plain and simple.”
3
The statement written in one of the review section in the blog site explains that the daily encounter of the devastating traffic jam cannot be blamed to the country’s infrastructure. It is no surprised that the situation in the streets of the Metro will be as such because majority of the population, either students or workers commute to each destinations, which in effect makes the roads busier. Traffic jam problem is also true to other countries. In fact, the same inconvenience was experienced by William L. Potts in 1920, a police officer in Detroit. Though during that time, the traffic consisted only of buggies, pedestrian and few automobiles, Potts figured out a way to adapt railroad signal for street use. This led to the invention of the first 4-way, 3 color traffic light (http://www.google.com/patents). Certainly, traffic light contributes a significant role to the management of vehicles and pedestrians in the road, specifically, in Metro Manila. Majority of the intersections in the Metro use time controlled traffic light. Having this system, each color in the traffic light is controlled by a constant cycle time. Every single state, which is associated with corresponding color, will alter after certain amount of time, say 30 se conds. Clearly, using time controlled traffic light is not an optimal solution for the increasing number of road users. It would be more feasible to pass more cars at the green interval if there are fewer cars waiting behind the red lights. Having many cars in the southbound lane wait for the traffic light to change to green light when no vehicle is crossing the westbound direction is a waste of time an d energy. At the present, several limitations of the existing traffic light can be listed. Being the above mentioned as the first and second, the existing controller has no display of remaining seconds before the next state. In the pedestrian’s point of view, showing the
4
remaining time before switching to the next color is necessary. This way, the pedestrian will know if they still have sufficient time to cross the road. This will prevent the crossers to be stuck in the middle of the road when a change of state occurs.
1.2. Objectives
The study aims to provide a traffic controller which reveals the remaining time until a change is to occur in its state. Yet important to note that the remaining time should match the color of the traffic light. For example, green timer for the remaining “GO” state and red timer for the “STOP” state. It is also an object of the study to locally create a system utilizing fuzzy logic to control the traffic system being the density of cars present in the green and red interval and the cycle time as the input parameters. The change from one state to another state of the traffic light is dependent on how many cars present in the red and green interval. The more number of cars in an interval, the longer the time before the next state will occur prior to that interval. Briefly, a locally made traffic controller device constructed in accordance with fuzzy logic system and has a color-oriented numeric display that shows the remaining time, in seconds, of each state before it switches to another.
1.3. Statement of the Problem
This study aims to surpass the limitations mentioned above. It is another objective of the study to further improve the “Traffic Light” patented by Dimon Simon, CA in July 31, 2001 by employing fuzzy logic system. Dimon’s invention particularly relates to
5
traffic light that shows the remaining time before the change of state through a count down timer. Specifically, it aims to answer the following; 1. What is the effect of the fuzzy traffic light in terms of traffic time? 2. What is the effect of the fuzzy traffic light in commuter’s fare savings? 3. What are the effects of the color coded timer in pedestrian’s safety? Furthermore, while in the process it is important to satisfy the following issues; 4. What are the advantages of using the fuzzy logic system compared to other technologies? 5. Will there be an increase in terms of efficiency, reliability and accuracy in the traffic system using fuzzy logic system?
CHAPTER 2: Review of Related Studies and Literature
This section presents the writings and past inventions related to the study and as well as the related information gathered by the researcher.
2.1 Review of Related Study
. A telephone interview was done by the researcher with a computer programmer in MMDA, Traffic Engineering Center, Mario Melendrez, said Actually, buong metro Manila gumagamit ng SCATS, kayalang, 1 Region lang yung gumagana. Sira kasi yung server nung tatlo pang Regions. Bandang
6
Ortigas to Makati lang yung gumagana, mahal din kasi magpagawa (Actually, the whole Metro Manila is using SCATS, however, only one Region is operating. The servers of the other three regions are not working. Only the one around Ortigas to Makati is the one working, it’s expensive to have a repair )( Melendrez, M. telephone interview, February 28, 2008). He even reveals that the whole system that the country is currently using is acquired all the way from Australia, which added to the cost of deployment. Apart from that, the maintenance cost added to the inconvenience. It would be a big savings if the whole system, the hardware and the software, is locally made. The country, when it comes to technology expertise is good enough to perform or invent a better solution to merging problems such as this. Steven Kaehler in his paper entitled “Fuzzy Logic – An Introduction” discusses some practical applications and implementation of the Fuzzy Logic system. The journal also discusses why fuzzy logic is a very useful tool in solving control system problem ranging from simple, small, embedded microcontrollers to large networked, multichannel PC and control systems.
2.2 Review of Related Literature
A study conducted by TransCore (2002), a solution provider company that implements SCATS, proved this hypothesis. SCATS is a proven adaptive system, includes software, hardware and all the materials that goes in between, that is currently controlling more than 20,000 intersections on six continents. SCATS’s ability to improve traffic signal operations and provide benefits to the traveling public has been proven in
7
deployments across the United States. In TansCore study, 31.8% reduction in waiting time was gathered. In Chula Vista, California, waiting time was reduced to 28 % and 24% were reduced at Menlo Park, California. The result of Transcore’s study is interesting but the system was tested only in the United States where the traffic condition is man y times better than the Philippines. United State’s roads are wider compared to the Philippines. It could have been better if the study were extended to developing countries such as the Philippines. Same system as Transcore was released by its competitor, T yco Integrated System. Tyco also conducted study that tests improvements caused by the new system. They conducted the experiment in New South Wales, in Australia. They observed the travel time, stop time and fuel savings. In New South Wales, 20 % savings in travel time, 40 % reduction in stop time and 12% less fuel usa ge. Tyco’s study sounded more appealing for the study was conducted at the most populous state in Australia, which is somehow similar to the Philippines. Both Tyco’s and Transcore’s papers only focused on the traffic flow improvement. Vehicles are not the only users of the road, there are pedestrians whose safety should be thought about also. It would have been better if the time for the stop state is visible for the pedestrians. In this way, they would know if it is still safe to cross the street. Crossing the street when you don’t know how much time is left before the green light signals and suddenly being caught right at the middle of the busy road is not a joke. One good paper was gathered by the researcher from a patent paper database, www.google/patents.com, Jerome Lemelson, Texas, patented a computer controlled traffic light and named it as “Intelligent Traffic light”. His patent paper, Intelligent
8
Traffic Light (2001), discusses the objective, theories and technical specification of his invention. This is a system and method for controlling traffic and traffic light and selectively distributing warning messages to motorist. The researcher uses fuzzy logic to dynamically derive traffic light phase-splits based on traffic flow patterns and other factors such as weather conditions, predicted increases in traffic for rush hours or special events. For example, the time split between red and green for a given traffic light cycle). Jerome said that the primary objective of this invention is to provide an improved traffic controller and method to improve safety and congestion on roads. Having these objectives, it is therefore clear to expect that there are improvements on waiting time which is one of the objectives of the researcher. But then, just like Tyco and Transcore, only improvements in vehicle traffic was noted. Similar to the discussed paper, Case Study: Fuzzy Traffic Light Controller discusses
how
a
fuzzy
system
was
applied
to
the
traffic
controller
(http://www.doc.ic.ac.uk/~nd/surprise_96/journal/vol4/sbaa/report.html). This study had, cycle time, cars behind red light and cars behind green light as the input parameters and the light display as the output. A paper entitled “Traffic Light” patented by D. Simon (2001) discusses the principles and technical specifications of a time controlled traffic light with a count down timer http://www.google.com/patents/traffic light). One of the objectives of the invention is to improve pedestrian’s safety by employing a count down timer visible to the pedestrians. The timer display is same color as the state it is currently displaying. Example red time display for stop “state”, green time display for “go state” and so on.
9
I find the study interesting for its focus is in pedestrian’s safety. But the problem is that the technology that it used is obsolete. Though the invention originated in California, the system that it used is based on constant time control. Another study compares traffic situation in the Philippines and Thailand. Hokao et al. concluded in this study that Traffic Impact Assessment System is ideal to be implemented in the Philippines than in Thailand as the country already has an established set of laws imposed in transportation system. However the biggest concern rose was the budget limitation. The country has stingy norms when it comes to road development. Credit must be given to this study for considering many points that may affect in analyzing
the
TIA
implementation
in
the
country.
Examples
of
such
are
institutionalization and legislation, unavailability of established and applicable standard, poor accessibility management, budget limitation, lack of knowledge in transportation system and pedestrian’s safety. However it has been discussed in the study that limiting urbanization might improve the traffic system. It has been believed that road problem rose from construction areas like road construction and commercial sites construction. It is a fact that construction affect traffic situation in the country, but limiting urbanization to improve traffic system is an illogical idea. It’s not healthy to prevent the country from improving because it has an adverse effect. Every improvement has its own drawbacks, but in order to improve more, another solution to overcome the drawbacks should be thought about. A dreamer can only succeed b y improving continually.
10
CHAPTER 3: Methodology
3.1 Instruments
This section presents the instruments used to come up with a working system and prototype. This includes the sensors used, the assembly which comprises of several electrical and mechanical parts and the controller, both computer and the microcontroller. 3.1.1 Sensors In order to identify the number of cars in road (both in front and back of traffic light) a metallic sensor will be buried in all lanes. One sensor per lane is enough. These are positioned as shown in the figure below. Figure 1. Sensor Positioning in Road Lanes
The sensor right below each traffic light is a counter that counts the number cars coming to the intersection and the second sensor counts the cars passing the traffic lights. The amount of cars between the traffic lights is determined by taking the difference of the
11
reading between the two sensors. For example, the number of cars behind traffic light North is s7-s8. The distance D, chosen to be 50 meters, was used to determine the maximum density of cars allowed to wait in a very crowded situation. This is done by adding the number of cars between to paths and dividing it by the total distance. For instance, the number of cars between the East and West street is (s1-s2)+(s5-s6)/400. 3.1.2 Signal Assembly The signal assembly included red stop indicator, green passage light indicator and amber caution light indicator. It also had a timer that was adapted to display remaining time in color same as the current state it represented. 3.1.3 Controller A microprocessor, a computer which includes a Central Processing Unit (CPU) was needed in order to control the parameters to drive the desired output. Software interface such as Microsoft C++ was required also.
3.2 The Fuzzy Design
Figure 2. Fuzzy Design Process Flow
Fuzzification
Rule Evaluation
Defuzzification
The whole system starts with fuzzification, a process in which the inp uts were derived to be a member of a membership function to graphically describe a situation. As
12
before, firstly the inputs and outputs of the design has to be determined. Assuming red light is shown to both North and South streets and distance D is constant, the inputs of the model consist of : 1) Cycle Time 2) Cars behind red light 3) Cars behind green light
The cars behind the light are the maximum number of cars in the two directions. The corresponding output parameter is the change of the current cycle time. Once this is done, the input and output parameters are divided into overlapping member functions, each function corresponding to different levels. For inputs one and two the levels and their corresponding ranges are zero, low, medium, high, and chaos. For input 3, the levels are very short, short, medium, long, very long, limit. The levels of output, change of cycle time are no, probably no, maybe, probably yes, and yes.
Next, the rule should be identified. The set of rules as before are formulated using a series of if-then statements, combined with AND/OR operators. Ex: if cycle time is medium AND Cars behind Red is low AND Cars behind Green is medium, then change is Probably Not. With three inputs, each having approximately 5,5,and 6 membership functions, there are a combination of not more than 200 rules. However using the minimum or maximum criterion some rules are combined that makes the number of rules lesser.
13
Deffuzification came in after the rule has been identified. This process, also mentioned above converts the fuzzy set output to real crisp value. The method used for this system is center of gravity:
Figure 3. The System Block Diagram Showing both Input and Output Parameters INPUTS
OUTPUTS
Cycle Time
No. of cars behind red light No. of cars behind green light
Fuzzy Logic
Fuzzy Cycle Time Fuzzy Time Display
Figure 3 shows the whole system used. As shown, the invention used Fuzzy logic processing 3 input parameters; cycle time, number of cars behind red light and number of cars behind green light driving two output parameters; Fuzzy cycle time and fuzzy display time.
3.3 Implementation Schedule
14
CHAPTER V: References Aziz, S.A & Parthiban, J. (2006). Case Study: Fuzzy Traffic Light Controller , p.5. Abstract retrieved February 9, 2008 from http://www.doc.ic.ac.uk . Gumatay, M. (Announcer).(2008). Good Times with Mo [FM Radio Broadcast]. Manila, Philippines: DWMM. Lemelson et al (2001). US Patent No. 6317058. Incline Village, Nevada. Retrieved March 9, 2008 from http://www.google.com/patents. Mechelen E. (2007). Traffic in Manila, Philippines-Simply Devastating. The People’s Media Company, p.2. Simon, D. (2001). US Patent No. 6268805. Richmond, California. Retrieved January 10, 2008, from http://www.google.com/patents. Transcore, San Francisco. (2007) SCATS Adaptive Traffic Control System. Retrieved March 12, 2008, from the Transcore website: http://www.transcore.com. Tyco Integrated System, Australia. (2007) SCATS Advantage. Retrieved March 12, 2008 from the Tyco website: http://www.tycois.com
15
