基于神经网络的模糊推理车辆跟驰模型

Fuzzy Inference Car-following Model Based on Neural Network

根据BP神经网络自学习的特点 ,通过训练使模糊变量和隶属函数隐含在网络内部 ,并用模糊逻辑推理模拟驾驶员对车辆进行控制的过程 ,可以使模型更接近于真实的跟驰行为 ,最后用该模型进行了仿真 。

Car-following model is a basic model in traffic microscopic simulation and car-following behavior is one of the complex tasks of driving. It is hard to describe drivers behavior with precise algorithm because of the fuzzy and indetermination character and the circumstance factors which exist during the driving. In this paper,a car-following model is developed, which integrates the self-learning character of the neural network and uses the fuzzy inference theory to simulate the driver to control the vehicle. The simulation result shows the feasibility of the model.