The Optimal Control Strategy Of Intelligent Traffic Light Timing In The Single Intersection

With the rapid development of economy in our country in recent years, the transportation has become an indispensable part in People's Daily life. This directly led to the vehicle ownership has increased dramatically in our country, the problem such as urban traffic congestion is becoming more and more serious. Traffic light is the most widely used way of traffic management, the commonly used control method of traffic lights has two kinds: one is solid time control, namely the allocated time of every direction of traffic lights is constant.the another is period of time control, namely the allocated time of traffic light changes over time. Both of the two method has good effect in traffic management, but, it also lead to the vehicles blocking in the direction with insufficient green time and no car pass through the intersection in the green time of some direction with less vehicle.This paper analyzed the shortages of recent control method and proposed two kinds of control strategies. It takes the MATLAB to simulate the data respectively and compared the optimization results of this paper with the traditional intelligent algorithm optimization results. The results have proved the feasibility of the two kinds of control strategies. This paper mainly completed the following work:A traffic light heuristic control strategy based on real-time traffic is proposed and the mathematical model of green time distribution based on real-time traffic conditions is established. As a result, the adaptive allocation of intersection traffic signal under different traffic conditions is realized.To make up the deficiency of the traditional intelligent algorithm, this paper proposed an evolutionary algorithm based on State-space model(SEA). Global convergence and search efficiency of SEA was analyzed by homogeneous finite Markov chain.For the signal timing optimization problems of traffic intersections, a mathematical model of the average waiting time of intersection vehicles is developed. Then the model combined with SEA to optimize the signal timing problems, which aimed to minimize vehicle average waiting time.