Optimization of Multi-Intersection Traffic Signal Timing Model Based on Improved Q-Learning

Abstract

With the economic development and the increase of car ownership, major cities in China are facing plenty of problems, such as traffic congestion and environmental pollution caused by traffic congestion. The effective solution to solve this kind of problem is to adopt intelligent control to the traffic signal timing scheme. Based on traditional Q-learning, this paper adds the influence of phase difference and congestion factors, and uses the traffic condition data of adjacent intersections and this intersection to predict the next state, so as to adjust the timing scheme of signal lights at each intersection in the road network. The VISSIM simulation experiments show that this method can effectively reduce the average vehicle delay time and the average number of parking times, alleviate traffic congestion, and improve the traffic efficiency in the overall road network.