Hierarchical Fuzzy Control for Urban Traffic Trunk Roads and Its Neural Network Implementation
城市交通干线递阶模糊控制及其神经网络实现

This paper uses the principle of decomposition-coordination for large-scale systems, fuzzy theory and neural networks technique to solve the real time arterial coordinated control problem. The urban traffic trunk road is regarded as a large system, and the subsystems are the intersections in the trunk road. A neural network implemented two-level coordinated fuzzy control method is presented. The second layer adjusts on-line signal cycle and splits of every intersection. Based on the traffic volume data measured from every intersection the first layer coordinates the number vehicle between the adjacent intersections, the object of the controller is to make the traffic trunk road unblocked and the average vehicle delay time shorten. The simulation shows the proposed method has better performances than the vehicle actuated method.