Research On The Optimal Control Strategy Of Urban Regional Traffic Signal Timing Coordination

In recent years,China’s economic development has continued to rise steadily,and the number of private car ownership has increased accordingly.Although private cars provide convenience to people’s lives,they also bring about traffic congestion problems and unnecessary troubles for people’s lives and work.Traffic congestion makes vehicles travel slowly,reduces driver satisfaction,and leads to waste of urban road resources.At the same time,vehicles will emit a large amount of CO,NO2 and other harmful substances,causing environmental pollution and causing a certain degree of damage to human health and social development.Influence.Therefore,without changing the original road structure,an efficient and reasonable traffic signal control scheme is urgently needed to improve the efficiency of urban road traffic and avoid road congestion.In this paper,the research on the urban area traffic signal control method is carried out.The main contents are as follows:(1)Due to the huge and complex structure of urban traffic road network,when carrying out regional traffic coordination control,the road network is simplified into a directed graph that is easy to calculate,and the traffic sub-areas are dynamically divided according to the principle of step accessibility.The rationality of the sub-area division results is evaluated by comparing the variance of the road segment space occupancy in the sub-area with the variance threshold.In the future,appropriate adjustments can be made on the basis of the original sub-area division according to the actual traffic operation status of the regional road network and changes in traffic flow.(2)The cross port is the most basic control unit of signal control.It is a prerequisite for regional traffic coordination control.For the traffic optimization control of a single intersection,this paper adopts the asynchronous advantage reinforcement learning algorithm.Through multi-threaded parallel technology,the real-time data obtained from the traffic environment is analyzed by feature analysis,so as to accurately perceive the traffic operation status of the intersection.The reinforcement learning algorithm is used to make multiple agents learn and train in the traffic environment at the same time,and the internal parameters are continuously optimized in the shortest time to obtain the optimal traffic signal timing scheme,and compared with the timing signal control and Q-learning algorithm,Prove the rationality and efficiency of the algorithm.(3)On the basis of single-intersection traffic signal control,further study the coordinated control of regional traffic signals.The coordinated control of regional traffic signals is not only to achieve the optimal operation of a single intersection,but also to comprehensively consider the actual situation of the entire regional road network,so that the overall traffic operation effect is optimal..After dividing the regional road network into multiple traffic sub-zones,a game is introduced,and it is combined with the asynchronous advantage reinforcement learning algorithm to improve the operation effect of regional traffic signal control.The actual traffic environment is simulated by VISSIM simulation software,and the experimental comparison with the traditional signal control method proves that the method can improve the traffic efficiency of road traffic.