Design of Traffic Coordination Control Structure Based on Multi-Agent

Abstract

Urban transport, one of the city's important infrastructures, which plays an important role in the normal functioning of the city, has become the focus of attention and discussion in the whole society. Restricting to the sustainable development of urban economy, the traditional solution to the urban traffic problems can no longer meet the development requirements of existing urban traffic. Given the complexity of urban transport, the limitations of traditional traffic control methods and the popularization and application of mature artificial intelligence technology, this paper discusses the collaborative framework of traffic control and induction on the base of multi-Agent technology and make researches on the structural design of urban traffic coordination control, thus builds a simulation model for the single intersection. The simulation experiment was carried out by using the Q learning algorithm.