关键词: 交通信号控制, 图卷积神经网络, 强化学习, 双延迟深度确定性策略梯度, 协同控制

Abstract:

To address the issues of uneven traffic flow at urban intersections, limited road capacity, and the poor coordination of existing traffic signal control algorithms, a traffic signal control algorithm based on graph convolutional reinforcement learning was proposed. By utilizing a multilayer perceptron, the dynamic features of vehicles and phase information at the controlled intersection and its neighboring intersections were extracted. A graph convolutional neural network was then employed to aggregate these vehicle dynamic features into potential features representing regional traffic. The control strategy was derived through multiple iterations of an improved twin delayed deep deterministic policy gradient (TD3) algorithm. This control strategy was applied to the traffic phase timing of the urban road network, aiming to maximize the traffic efficiency of the road network. Simulation experiments demonstrate that the algorithm can adapt to dynamically changing and complex road network environments. Moreover, the control effect is significant under high saturation flow, effectively improving the traffic efficiency of the road network and alleviating congestion at intersections during peak hours.

Key words: traffic signal control, graph convolutional neural network, reinforcement learning, twin delayed deep deterministic policy gradient (TD3), coordinated control