基于模糊神经理论的驾驶员控制模型的研究

把横向预瞄偏差作为模糊神经驾驶员控制模型的输入变量之一，解决了由于输入变量较多，神经网络节点数巨大的问题，从而设计并实现了模糊神经驾驶员控制模型网络拓扑结构，并用BP网络实现了控制规则的模糊映射。仿真的结果表明：所建立的模糊神经驾驶员模型是正确和可行的。这种模型的核心思想是利用神经网络对汽车驾驶员驾驶知识的表达机理，通过学习训练来实现控制规则的记忆，在一定程度体现了驾驶员的智能行为。

A Study on Driver Control Model Based on Fuzzy-Neural Technology

Based on fuzzy-neural network theory,a driver control model is set up,in which only 2 variables,i.e.lateral preview yaw and yaw rate,are taken as inputs,so the problem of too many input variabiles resulting in a huge number of neural-network node and making the simulation impossible is eliminated.The network topology of fuzzy-neural driver control model is designed and the fuzzy mapping of driver control rule is realized with back propagation neural networks.The simulation result shows that the driver control model presented is feasible.