The Cooperative Control And Optimization Of Urban Transportation Signal Control System Based On Multi-agent

With the fast development of the society,process of urbanization and urban population has increased rapidly,the urban problem is becoming more and more serious,and seriously affects the normal operation of urban.Transportation is the life line of urban,scientific management and control is one of the important ways to improve the operation efficiency of urban transportation,meanwhile,it is an important symbol of modernization,informationalization,intellectualization of urban.Currently,since the city land resource is scare,advanced traditional traffic control methods can improve the capacity of the existing transportation network and communication state,which ensures traffic safety and unimpeded.Firstly , the development of cooperative control methods of urban transportation system was introduced.The theory and technique of artificial intelligence and its application to the urban transportation system were described.Especially,the possibility and superiority of multi-agent technology used in the cooperative control of urban transportation system were pointed out.Secondly,according to the characteristics of multi-agent and transportation system, analysis on the deficiency of existing traffic modes,distributed control structures and traffic control systems based on multi-agent were carried out.The system gives every intersection agent the establishment of control strategy,the traffic control was conducted based on the intersection agent's own knowledge and its coordination control.It would improve the ability of intelligent decision.The coordinated function of control system was improved to meet the complicated variable features of transportation system flexibly.In the paper,the intersection agent optimization model based on the distributed multi-agent traffic coordination control system structure was studied by adopting genetic algorithms as its solving method.Then,the intersection agent optimization model on the basis of genetic algorithms of real-code with the average vehicle delay of the intersection as objective function was established under the model of traffic cooperative control system structure based on multi-agent.The result of the condition on utilizing the model to optimize the signal phasing and timing indicated that the control effect and calculation efficiency were enhanced.Finally,game theory was introduced,the cooperative mechanism of the simple network composed by three intersection agents was established and the game process of intersections was explored.The control strategy of intersection was optimized according to the actual intersection situation,next applying Matlab 7.0 to realize optimal operation.For this method,performance index were significantly improved.It makes it possible to realize real time intelligent control and provide theoretical basis for the cooperative control of multiple intersections.