基于模糊神经网络的车辆跟驰建模与仿真研究
Car-following Safety Modeling and Simulation With Fuzzy Neural Network
-
摘要: 驾驶员在车辆跟驰过程中表现出来的模糊的、不确定性的行为特征,难以对驾驶员的行为进行精确的数学描述.本文在以往跟驰模型基础上,采用以实际间距与期望间距的比值和相对速度作为双输入、后车加速度作为单输出的模糊神经网络建立数学模型.结果表明,该网络能较好的反映一定道路条件下的跟驰行为.Abstract: It is hard to describe drivers' behavior with precise algorithm because of its fuzzy and uncertain minable characteristic when following a leading car. However, it is suitable to apply fuzzy neural network (FNN) to studying driving behavior. An experiment, which uses SR/Sd and VR as two inputs and the following car's acceleration as the output, is simulateed in MATLAB. Model validation results have shown that the FNN model can replicate the driving behavior of real road situation.