基于模糊神经网络的交通干线分层递阶控制

针对城市交通干线协调控制的要求，提出了利用模糊神经网络分层递阶控制的方法。采用两层结构，第一层为控制层。针对单个路口，对下一时间段内路口各个方向的车流量进行预测。并在此基础上计算出下一时间段内各个路口的周期、相序、各个方向上的绿信比；第二层是协调层，综合主干方向的车流状况及各个路口的情况，采用模糊神经网络对各个路口的周期、相位及主干方向的绿信比进行调整。仿真结果表明，该方法优于定时控制，达到了减少车辆的停车次数和延误时间的目的。

Traffic Control for Trunk Roads Based on Fuzzy Neural Network and Hierarchy Control

The fuzzy neural network and hierarchy control is used to solve the problem of traffic control for the trunk roads. The first layer of the network is manipulative layer that forecast the traffic flow of single intersection, and based on that, the signal cycle, phase and split of each direction are calculated. The second layer is coordinated layer. Fuzzy neural network is used to correct the cycle, phase and split of intersections by considering the traffic of trunk roads and intersections. The simulation result shows that the proposed method can reduce the vehicle stop times and delay time.