Multi-intersection self-adaptive phase difference coordinated-control system and method based on Q learning

Abstract

The invention discloses a multi-intersection self-adaptive phase difference coordinated-control system and method based on Q learning. The system comprises an intersection control module, a coordinated-control module, a Q learning control module, a regulation module and an output execution module. The intersection control module is used for providing a reasonable single-intersection traffic timingplan for a current phase according the traffic state of a local interaction. The coordinated-control module is used for judging whether the current phase needs phase difference coordination or not byanalyzing the traffic states of local interaction and adjacent intersections. The multi-intersection self-adaptive phase difference coordinated-control method can effectively shorten the response time for traffic jam, quickly coordinate signal control of the intersections and improve the intersection passing efficiency and has very good universality in the application of traffic signal self-adaptive control. The system can give accurate and reasonable green lamp timing plans through phase coordination and is more suitable for large-scale intersections having larger vehicle flow compared withcoordination control free of accurate time.

Description

technical field [0001] The invention relates to a multi-intersection adaptive phase difference coordination control system and method based on Q learning, which belongs to the technical field of intelligent transportation. Background technique [0002] Compared with the independent control of single intersection, regional traffic control can obtain the overall optimized control effect, so it has been paid more and more attention by people. Urban traffic system is a typical nonlinear, dynamic time-varying, uncertain complex system, and it is very difficult to establish its precise mathematical model. The traffic signal coordination control method based on Q-learning does not need a model and can realize learning in the environment, which is an effective technical means to solve the above problems. The existing Q-learning regional traffic control methods have problems such as slow learning speed and inability to respond to changing traffic flow conditions in a timely manner d...

AI Technical Summary

Problems solved by technology

[0003] Aiming at the problems existing in the existing regional self-adaptive traffic control, the present invention proposes a multi-intersection self-adaptive phase difference coordinated control system and method based on Q-learning

Images