考虑驾驶员行为特性的模糊推理跟驰模型改进研究  

Improving Car-following Model by Considering Driver's Behaviors with Fuzzy Inference

作　　者：钟益萍[1,2] 张存保[2] 石永辉[3] 

机构地区：[1]武汉理工大学自动化学院,武汉430063 [2]武汉理工大学水路公路交通安全控制与装备教育部工程研究中心,武汉430063 [3]武汉市公安局交通管理局,武汉430022

出　　处：《交通信息与安全》2010年第3期17-20,42,共5页Journal of Transport Information and Safety

基　　金：国家重点基础研究发展计划项目(批准号:2006CB705500)资助

摘　　要：基于模糊推理的车辆跟驰模型实现了用数学方法描述驾驶员的主观判断、推理和执行的过程,但现有的这类模型通常未考虑驾驶员行为特性的差异。针对这一问题,在基于模糊推理的跟驰模型中引入期望车头时距,不同行为特性的驾驶员具有不同的期望车头时距,在此基础上建立了改进的车辆跟驰模型。仿真结果表明,改进的跟驰模型能准确描述不同类型驾驶员的跟驰行为及其差异。Car-following models based on fuzzy inference use a mathematical method to describe the process of drivers＇ subjective judgments,reasoning and implementation.However,the existing models basically neglect drivers＇ individual differences in their driving behaviors.In this paper,a new car-following model based on fuzzy inference is proposed,which takes the desired time headway of an individual driver into account.Thus,drivers with such different behavioral selections will lead to different desired time headways. The outcome of Matlab simulation in the new model indicates that the improved approach can describe the diversity more realistically in different drivers＇ car-following behaviors.

关 键 词：跟驰模型 期望车头时距 行为特性 模糊推理 

分 类 号：U491.54[交通运输工程—交通运输规划与管理]