Application Research Of Freeway Control And Management Based On Multi-agent

With the rapid development of the national economy, the accelerated process ofurbanization and motorization, people’s daily long, medium and short-distance travel demandare increasing. Highway is the most important public traffic infrastructure and the mobilecarrier of the national economy for the national collection and distribution. It plays animportant role of organic links and effective communication in all aspects of socialproduction, distribution, exchange and consumption. Therefore, it reflects a country and theregion’s level of the transportation development, and even the overall level of economicdevelopment.Multi-Agent comes from people’s understanding of the artificial intelligence. It’s goal isto achieve the "proxy" that is intelligent and can replace human beings to deal with the affairs.Multi-agent has some characteristics, such as autonomy, collaboration, communication andcross. MAS get the overall system efficiency far more the ability of single agent through theactions coordination of the multiple agents. It provides new ideas for solving random andcomplex problems. This paper applies the Multi-agent technology to the control andmanagement of traffic flow for the distributed, nonlinear and time-varying random features ofthe highway transport system. It has good feasibility and superiority.The main work of this paper includes the following aspects:(1) Reviewed the highway traffic control and management, multi-agent technology,as well as domestic and foreign-based traffic control and management of multi-agenttechnology application status;(2) Summarized the freeway traffic flow theory, including the continuity equation, thebasic parameters of the model, the macroscopic fluid model as well as other traffic flowmodel; analysis of the principle of control of the highway system and intelligent controlmethods;(3) Multi-agent technology has combed the agent works, characteristics, classification,Multi-Agent architecture to apply to the problem domain, and finally used in highway trafficcontrol and management of the feasibility and advantages of MAS analyzed;(4) On the basis of the theoretical method and the multi-agent technology, built theframework of the Coordinated Control System, modeling the Agent of the system, expressedthe Agent with knowledge, described the Agent’s internal database, function modules andStatus Interface, study the communication and collaboration mechanisms of the system. (5) In order to verify the theory, Selected Guangzhou an expressway in topical sectionsas a simulation object, a model of total trip spend time (TTS) and sections of density havebeen established. Using genetic algorithm optimization solution, and the results wereanalyzed.