A multi-agent coordinate model for urban traffic signal control

Traffic system is time-varying and complicated so that traditional signal control methods can´t meet the demand of modern traffic management. This paper proposes a novel control model with multi-agent architecture to improve the traffic signal control efficiency and relieve traffic congestion. First, we present the structure of a single agent which consists of sense module, execution module, communication module, data center, experience base, extended base, decision logic and optimizer. And then we present the message-passing and control mechanism of the multi-agent system. Finally, the simulation results prove the feasibility and effectiveness of the signal control mechanism with multi-agent structure over traffic networks.