基于模糊推理的驾驶员车速决策行为建模与仿真

Driver's Speed Decision-making Behaviour Modeling and Simulation Based on Fuzzy Inference

为了在微观交通仿真中更真实地模拟驾驶员的车速决策行为,在综合自由行驶和跟驰两种驾驶行为的基础上,采用模糊推理技术进行建模研究。选取车头时距、期望速度差、前车速度差作为模糊推理系统输入变量,加速度为系统输出,结合实际中加减速行为的不对称性及不同行驶状态下主要行驶意图的不同,构造模糊规则,建立基于模糊推理的车速决策模型。利用模糊系统的输入输出曲面进行了模型的校核,并在考虑驾驶员反应时间的基础上进行了仿真实验。仿真结果表明,车辆能够根据前车行驶状态和自身期望车速进行适当的速度决策,验证了所建模型的合理性。

To simulate the driver＇s speed decision-making behaviors in microscopic traffic simulation more credibly,synthesizing with free-travel and car-following,fuzzy inference technique was used to proceed with modeling studies. Time-headway,expected velocity difference and front-car velocity difference was selected as the fuzzy inference system input variables,acceleration as the output.Combined with the asymmetry of the actual acceleration and deceleration behavior and the driving intention changed with different driving state,the fuzzy rules were extracted and a speed decision-making model was built based on fuzzy reasoning. Fuzzy system input and output surface was used to proceed the verification of the model. Considering the driver＇s reaction time,a simulation experiments was carried out to verify the rationality of the model. The simulation result showed that the vehicle driver was able to make the proper speed decision according to the front car and his own expected velocity,which proved the rationality of the model.