Development of Multi-Agent System for Fire Accident Detection Using Gaia Methodology

(IJCSIS) International Journal of Computer Science and Information Security, Vol. 8, No. 1, April 2010

Development of Multi-Agent System for Fire Accident Detection Using Gaia Methodology 1

2

3

4

Gowri.R, Kailas.A, Jeyaprakash.R, Carani Anirudh 1

Asst. Professor, 2, 3, 4 Students Department of Information Technology, Sri Manakula Vinayagar Engineering College, Puducherry – 605 107. 1 [email protected], 2 [email protected], 3 [email protected], 4 [email protected]

Abstract—An agent is an encapsulated computer system that is capable of flexible and autonomous action in that environment in order to meet its design objectives. This paper presents the analysis model of MAS used to sensing the fire accident area in our surrounding, at the same time passing the information to the successive organization which is found nearby to the accidental area by the help of database server. By that it will decrease the travelling time taken by the organization to reach the destination. It collaborates with external components for performing tasks and reacts to situation appropriately. The MAS model is built using the GAIA methodology and the JADE agent framework.

•

A. Multi Agent System

When several agents interact they may form a multi-agent system. Characteristically such agents will not have all data or all methods available to achieve an objective and thus will have to collaborate with other agents. Also, there may be little or no global control. These agents are utilized when collaborative operations ought to be done in establishing communication system.

Keywords- Coordination, Crisis Response, GAIA, MAS, Simulation.

I.

INTRODUCTION

In the context of the Open Agent Architecture (OAA) [7, 8], agent is defined as any software process that meets the conventions of the OAA society. An agent satisfies this requirement by registering the services it can provide in an acceptable form, by being able to speak the Inter agent Communication Language (ICL), and by sharing functionality common to all OAA agents, such as the ability to install triggers, manage data in certain ways, etc. The term agent describes a software abstraction or a concept, similar to OOP term such as methods, functions and objects. The concept of an agent provides a convenient and powerful way to describe a complex software entity that is capable of acting with a certain degree of autonomy in order to accomplish tasks on behalf of its user. Agents are described in terms of its behavior. Some properties of agents are autonomy, interactivity, adaptivity, sociable, mobile, intelligent and trustworthy. Agents have different definitions that are being proposed based on their characteristics. They are •

•

•

Reactivity: Agent perceives the context in which they operate and react to it appropriately.

Persistence: Code is not executed on demand but runs continuously and decides for itself when it should perform some activity. Autonomy: Agents have capability of task selection, prioritization, goal directed behavior, decision making without human intervention. Social Ability: Agents are able to engage other components through some sort of communication and coordination, they may collaborate on task.

B. Gaia Methodology

Gaia [3] is intended to allow an analyst to go systematically from a statement of requirements to a design that is sufficiently detailed and it can be implemented directly in Jade platform. Analysis and design can be thought of as a process of developing increasingly detailed models of the system to be constructed.

Figure 1. GAIA Methodology (process and models)

Gaia provides an agent-specific set of concepts through which a software engineer can understand and model a complex system.

190

http://sites.google.com/site/ijcsis/ ISSN 1947-5500

Trusted by over 1 million members

Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.




The main Gaian concepts can be divided into two categories: abstract and concrete. concrete . Abstract entities are those used during analysis to conceptualize the system, but which do not necessarily have any direct realization within the system. Concrete entities, are used within the design process, and will typically have direct counterparts in the run-time system. C. Java Agent Development

JADE [6], is a software framework for multi-agent systems, implemented in Java that has been in development since at least 2001. The JADE platform allows the coordination of multiple FIPA-compliant agents. It provides a standard implementation of the FIPA-ACL [6] communication language in both SL and XML. Jade creates multiple containers for agents, each of which can be on the same computing system or different systems. Together, a set of containers forms a platform. Each platform must have a Main Container which holds two specialized agents called the AMS agent and the DF agent. The Disadvantages of JADE for simulation purposes, and the rigidity that may arise from using GAIA.

II.

ORGANIZATION BACKGROUND

complexity is high at the same time communication speed is not compatible to the crisis scenario. So safety and improved functionality absence happens to be the major drawbacks in present day system. But proposed system satisfies this entire drawback by using Agent Oriented Software Engineering concept. Each department has agent performing some role as already initialized to it. Agent communication speed is high compared with the real time system. III.

The existing system shows [1] how departments are working on the fire crisis area and how message alert signal had been sensed by the agent [2]. The only demerit is, it did not show the search operation of finding the nearby departments. The working status is, if any of the building had gotten fire accident the agent will be activated and calls the fire and police department for crisis response [5]. But the agent calls the generic number which was given by our government for the sake of peoples who needs helpline from fire station and police station, so it takes some more times to reach the destination. By this way the existing system was built. IV.

A. Police Station Policemen provide a string security to peoples, by putting cases against criminals. It includes several other departments for training the trainee level policemen in the organization. The Hierarchy of power in the police stations involves the following which is restricted for Indian Police. They are • •

Inspector Sub-Inspector and Constables

B. Fire Station

Firefighters, or firemen, are rescuers extensively trained primarily to put out hazardous fires that threaten civilian populations and property, to rescue people from car accident, collapsed and burning buildings and other such situations. The increasing complexity of modern industrialized life with an increase in the scale of hazards has stimulated both advances in firefighting technology and a broadening of the firefighter rescuer's remit. They sometimes provide services. Firefighting and firefighters have become ubiquitous around the world, wild land areas to urban areas, and on board ships. Firefighters are held in high regard around the world. This causes firefighting to be one of the five trusted public service professions. PROBLEM DEFINITION Time management is very important for all departments

EXISTING SYSTEMS

PROPOSED SYSTEM

Our proposed system provides many features that were not present in an existing system. The main feature is searching of nearby department which was not found in existing system. When an alarm signal is sensed by the agent, it starts to perform its work by searching the nearby departments and then passes the information to the corresponding departments [1]. According to the Gaia methodology, our project was developed and each phase in Gaia methodology has some features for implementation phase. The details of each phase is shown below, A. Requirement Phase

This phase shows the input and outputs of the project and also frame the analysis, design and implementation work. Before the GAIA based analysis, we went through an initial phase of requirement elicitation, based on identification of response processes for a Fire accident scenario. B. Analysis Phase

Analysis phase has led to four identification roles to our project they are, • • • •

Detect Fire service Police service Dsearch




layout of the project with some unwanted information but this information is deleted in the design ph ase. TABLE I.

GAIA Role Model

Role: Detect Description: Read the alarm sound then passes the message to the Fire station and Police station Protocol and Activities: Activities : sendsignal, fire message Permission: Permission : Read Responsibilities: 1. Liveness: sendsignal. Detect=fire message 2. Safety: True

Role: Fire service Description: Response according to the inputs collected from the block agent Protocol and Activities: Activities : receivemessage, fresponse Permission: Permission : Read, Change Responsibilities: 1. Liveness: fire service = receivemessage. fresponse 2. Safety: True

Role: Police service Description: Response according to the inputs collected from the block agent Protocol and Activities: Activities : receivemessage, presponse Permission: Permission : Read, Change Responsibilities: 1. Liveness: police service = receivemessage. presponse 2. Safety: True

Role: Dsearch Description: It connects to the database server by that it rece ives the fire station and police station phone no then passes the information to that stations Protocol and Activities: Activities : receivefmessage, sendpservice, sendfservice, receivepmessage, search Permission: Permission : Read, change Responsibilities: 1. Liveness: Dsearch = receivemessage.sservice sservice = sendpservice.sendfservice 2. Safety: True

The above roles are fixed to the agents for the behavioral action in an environment for our project each role is played by any type of agent so the AOSE helps the user to minimize the complexity of the agent creation. C. Design phase

During this phase the Agent model is achieved, along with the services and acquaintance models. The agent model defines the agent classes that will play specific roles and also creates agent types by aggregating roles. Each emerging agent type can be represented as a role that combines all the aggregated roles attributes (activities, protocols,

Figure 2. GAIA Agent Model

Gaia also provides service model in the design phase so that the user can easily design the service taken by an agent at the time of working against crisis. But our project mainly focussing on police and fire service to the crisis happening area so that the people are saved from the fire accident. TABLE II. Gaia Service Model Service

Calling Fire and Police service to crisis area

Inputs

Crisis area and Destination place

Outputs

Response to the crisis

Precondition

Collection of Fire station and Police station numbers in Database

Post condition

Going to crisis crisis area

The interactions model represents the interaction between the agents, connected through input or output. The Gaia interaction model denotes which action returns from a request along with the roles that can initiate a request and the corresponding responders. TABLE III.

Gaia Interaction Model




So the time taken by the organization to reach the destination has taken less time when compared with the real time system. D. Sequence Diagram for Fire Accident Detection Using MAS

A sequence diagram is a kind of interaction diagram that shows how processes operate with one another and in what order. In our project how agents communicates with each other and in what way it has to be communicated. The role plays the major work in an environment according to the situation the agent changes its characteristics then perform the work given by task manager (programmer). The sequence diagram shows timely basis of an agent communicating with other agent and the period of message transmission between each other is also determined using this diagram. In this project the below sequential diagram(fig :3) shows that how the agents going to be work and also sends the information between the agents .In the jade platform the agents communicate by the means of the ACL language, the ACL language is the common language used by the agent for the communication purpose. Here the sensor gets the input from the environment and that has to send that information to the different agent for the searching of the crisis located area and various types of the services to that particular place this has been done by the agents. The block agent percept the environment and it has to perform the action, when the block agent is activated then it will automatically initialize the police agent and the fire agent by the means of communication. The following diagram will gives the clear knowledge about the flow of information between the agents.

V. IMPLEMENTATION The steps to implement our project in JADE framework [4]: Define all the ACL messages by using interactions • models. • Define the needed data structures and software modules that are going to be used by the agents by using the Gaia roles and agent models. • Decide on the implementation of the safety conditions of each role. Define the JADE behaviours. Start by implementing • those of the lowest levels, using the various Behaviour class antecedents provided by JADE. The Gaia model that is useful in this phase is the roles model. Behaviours that are activated on the receipt of a specific message type must either add a message receiver behavior or receive a message at the start of their action. The first module is to detect the fire place in the building by the help of fire alarm [1,2] then the block agent collects the input from the fire alarm, and passes the fire message to the searching module, service module. Simply convert our project into jade framework to develop the multi agent system for fire accident detection in an area . During the implementation all the agent has been created in the jade platform using the containers and the communication between them is also be done by using it and then we have do compile the java file in the command prompt now start running the agent. The agent starts communicate with one another automatically and that also has been shown in below diagram (fig: 4 & 5) .




REFERENCES [1]

[2]

[3] [4] [5]

[6] [7]

[8]

Rafael A. Gonzalez,”Analysis and Design of a Multi-Agent System for Simulating a Crisis Response Organization” Proceedings of the 6th International ISCRAM Conference, 2009 . Frantisek Zboř il il jr., Frantisek Zboril,”Simulation for Wireless Sensor Networks with Intelligent Nodes” Tenth International Conference on Computer Modeling and Simulation, 2008. Michael wooldridge, Nicholas r. Jennings, David kinny, “The Gaia Methodology for Agent-Oriented Analysis and Design”, 2000. Pavlos Moraitis, Nikolaos Ι. Spanoudakis,”Combining Gaia and JADE for Multi-Agent Systems Development”, 2006. Khaled M. Khalil, M. Abdel-Aziz, Taymour T. Nazmy, Abdel-Badeeh M. Salem,”Bridging the Gap between Crisis Response Operations and Systems”, 2009. Pavlos Moraïtis, Eleftheria Petraki, Nikolaos I. Spanoudakis, “Engineering JADE Agents with the Gaia Methodology”, 2004. David L. Martin, Adam J. Cheyer, Douglas B. Moran, “The Open Agent Architecture: A Framework for Building Distributed Software Systems”, 1999. Peter Novak, “An open agent a rchitecture: Fundamentals”, 2008.

R.Gowri received her M.Tech. Degree in Computer Science & Engineering from Pondicherry Pondicherry University, Puducherry, India. She is currently working as Assistant Professor, Dept. of IT at SMVEC (Pondicherry University), University), Puducherry. Her research interests are in the areas of Database Management Systems, Algorithms and parallel Architecture, Agent Technology. Mrs. R.Gowri published in reputed National and International Conferences and Journals.

Figure 5. Command prompt Agent Agent Communication

VI. ADVANTAGES •

•

•

Decrease the time complexity by the department to reach the destination. Communication speed is very high when compared with the existing system. It reduces the human disaster and nature disaster very quickly. VII.

CONCLUSION

Thus the Multi-agent fire detection system can be utilized as an alternative to present day detection method. In the future of fire detection finds a competent technology that can handle much of accidents and performs parallel processing based on surrounding situation. This Multi-agent technology implementation in fire detection can be utilized generically for many other specialized fields where security of information is of high concern. Thus fire detection system using Multi-agent forms the pioneer leading to many new communication

various

A.Kailas is pursuing his B.Tech. Degree in Information Technology, Sri Manakula Vinayagar Engineering College, Pondicherry University, Puducherry, India. His research area includes Artificial Intelligence and Software Engineering. A.Kailas presented a paper in International conference on software engineering theory and practice 2010.

R.Jeyaprakash is pursuing his B.Tech Degree in Information Technology, Sri Manakula Vinayagar Engineering College, Pondicherry University, Puducherry, India. His research area includes Networks and Database Systems. R.Jeyaprakash published a paper in International conference on software engineering theory and practice 2010.

Carani Anirudh A is pursuing his B.Tech Degree in Information Technology, Sri Manakula Vinayagar Engineering College, Pondicherry University, Puducherry, India. His research area includes Distributed computing and Database systems.

Development of Multi-Agent System for Fire Accident Detection Using Gaia Methodology

Recommend Documents