With the rapid development of society and economy, as well as the fast urbanization, the number of city motor and traffic flow increased dramatically. As a result, the city becomes crowded and congested, lagging the economic development and urbanization process. As a complex giant system, the city traffic signal control is a random problem. So city traffic control research has important theoretical significance and practical value. At the same time, adaptive dynamic programming (ADP) is a class of relatively new optimization method. ADP is received more and more emphasis in theory and practical applications for its ability in studing and optimization. In this paper, we study the traffic signal intelligent control problems based on ADP under the background of city traffic singal control problem. Firstly, we pay attention to urban traffic control systems and their control problems. We also present the research background and significance of this paper. The research aims, as well as this paper’s main work and missions are introduced. Secondly, we present the development of adaptive dynamic programming. Several types of adaptive dynamic programming are illuminated. We present their structure and calculation. Thirdly, the basic model of junction traffic flow is introduced. A traffic signal intelligent controller is designed based on adaptive dynamic programming algorithm. And some simulations are carried out to compare the performance of the proposed controller with the normal pertimed signal control, traffic-actuated signal control, and so on. Fourthly, for dealing with the training problem in the normal adaptive dynamic programming, this paper adapts the artificial bee colony theory for training adaptive dynamic programming quickly. As a result, the adaptive dynamic programming can achieve an optimal control policy more quickly. It is important for applying the adaptive dynamic programming in real time traffic signal control. Fifthly, the normal adaptive dynamic programming chooses initial parameters randomly. But the random initial parameters lead to uncertain performance of this algorithm. For dealing with this disadvantage, this paper adapts the people’s knowledge in the adaptive dynamic programming. Thus, the training of the adaptive dynamic programming can begin with a better direction. Sixthly, this paper designs a new cooperative multi-intersections control algorithm based on the adaptive dynamic programming to deal with area traffic cooperative ...
修改评论