STOCHASTIC POLICY GRADIENT-BASED TRAFFIC SIGNAL CONTROL METHOD AND SYSTEM, AND ELECTRONIC DEVICE

摘要

A stochastic policy gradient-based traffic signal control method and system, and an electronic device. The method comprises: acquiring static road network data (step S100); visualizing a traffic simulation road network (step S200); acquiring real-time traffic state data (step S300); obtaining an optimized traffic simulation road network (step S400); obtaining an evaluation value of a signal control scheme, and updating parameters of a value network (step S500); obtaining a probability value of each signal control scheme, and perform random sampling to obtain one signal control scheme (step S600); and updating the parameters of a policy network by means of stochastic policy gradients and based on the evaluation value of each signal control scheme in the traffic state and the signal control scheme obtained by sampling (step S700). The method can solve the problem of the curse of dimensionality related to signal controls.