The Study On Urban Traffic Signal Control Method At Single Intersection

Due to the rapid development of the economy and acceleration of urbanization, the aggravation of urban traffic congestion becomes a more serious issue. It causes increased vehicle delay, frequent traffic accidents, and worsening environment and so on. Intelligent Transportation System is an important solution to traffic congestion.Urban traffic control system with randomness and heavy nonlinearity is beyond the capability of precise mathematical model. In order to relieve traffic congestion, intelligent control theory of fuzzy control, genetic algorithm and neural network are used to minimize traffic delays. Consequentially, these are not the optimal traffic controllers under exceptional traffic cases such as road accidents and roadblocks. A new traffic controller is proposed that can optimally control traffic flows under both normal and exceptional traffic conditions.First, the fuzzy control method is used for building the model for urban traffic lights at single intersection from the perspective of cybernetics. Furthermore, according to fuzzy control theory and the policeman's experience, a traffic light control strategy based on the delay is presented. In this strategy, the delay is regarded as the control object and the total delay on the contiguous phase, and the condition of the path is used to determine the green timing plan.Secondly, as for the neuro-fuzzy control, adaptive network based fuzzy inference systems (ANFIS), whose structural theory and method are described. The neural network's function of self-learning is used; it regulates fuzzy rules and membership function through neural network. This can generate fuzzy rules and adjust membership function automatically. It solves the problems that fuzzy rules are mostly designed by the policeman's experience, lack of self-study capability and low control precision.Thirdly, a multiphase FC based on GA was designed to overcome some drawbacks of classical FC. The optimization of fuzzy rules'adjusting numbers by using the new GA. And last, controls of the two methods are modeled and simulation by using the Simulink tool in Matlab.All the three simulation results show that the three control strategies can effectively reduce vehicle delay in both the normal and abnormal conditions. The application result is better. And obviously the effect of the optimization control is better than that of the traditional fuzzy control.In short, an in-depth study for urban traffic signal control strategy is carried by sufficiently synthesized fuzzy logic, genetic algorithm and neural network. And the average delay is reduced.