Study On Signal Control Strategy For Urban Intersection Based On Fuzzy Logic

With the rapid increase in traffic, the contradictions between urban transportation and economic development are becoming acute and the city is facing an increasingly traffic congestion problems. It has been unable to address the growing traffic problems effectively using traditional traffic control techniques, and information technology, network technology, and intelligent control technology perfected for the urban traffic control technology.In this thesis, in order to relieve traffic congestion, in-depth study on intelligent control of urban traffic signal is carried out on the basis of research on urban traffic flow. a four-phase traffic model for isolated intersection is described. Since fuzzy control does not need to establish accurate mathematical model of the object, absorbing the experience of manual control, can mimic the human brain, logical reasoning and decision-making process, it applied to random,complex urban traffic control system. A fuzzy controller for a four-phase isolated intersection is designed based on the average queue length and vehicle delay. According to the actual situation for urban traffic flow, a dual-fuzzy controller for multi-intersections was designed and a coordinating controller is added to coordinate the traffic flow of the local intersection and the adjacent intersection.Since the fuzzy control rule sets and membership functions are often pre-determined according to human experience, and they won't change in the whole control process. Therefore, when the intersection traffic conditions changed greatly, the control performance becomes unsatisfactory. Adaptively adjust the parameters of the system according to the traffic conditions will help to improve the control effect. The Genetie Algorlthm is used to optimize the fuzzy rules and membership functions of the isolated intersection and multi-intersections fuzzy controller, so that it can adapt to different changes in traffic flow, achieving the adaptive control of intersection signal. Finally, we carried out simulation of fuzzy controller at fix-time control, fuzzy control and genetic optimization of fuzzy control. Simulation results show the fuzzy control is efficient compared to time control, it is effective to reduce average vehicle delay, and fuzzy controller by Genetic algorithm optimization can adapt to different changes in traffic flow, control performance is further improved.