Research on Cooperative Control of Traffic Signals based on Deep Reinforcement Learning

Research on Cooperative Control of Traffic Signals based on Deep Reinforcement Learning

연구 분야: Artificial Intelligence

학회: 2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)

With the increasing traffic pressure, the problem of road congestion needs to be solved urgently. In recent years, many researchers have used deep reinforcement learning to solve traffic signal control problems, but most of them are based on the situation of single intersection or adjacent intersection, with limitations. This paper proposes an index to evaluate the regional traffic performance, and uses this index to design a traffic signal coordination control algorithm based on deep Q network. The experimental results on the SUMO simulation platform show that the proposed algorithm has better performance in terms of average waiting time and average time loss compared with the other three signal control algorithms.