Personalized Driver Car-Following Model — Considering Human's Limited Perception Ability and Risk Assessment Characteristics

Understanding and modeling driver behavior is crucial to the design, testing and evaluation of intelligent driving systems. In this study, a personalized car-following model with considering human's limited perception ability and risk assessment characteristics is proposed to enhance the accuracy and the scope of application of driver model. This paper presents a modular personalized driver following model framework to describe car-following behavior. In the perception module, human's limited perception ability and risk assessment characteristic are modeled based on normal distribution and fuzzy logic. In the decision module, we proposed a personalized spacing strategy and a speed-distance index car-following model. The simulation result indicated that the accuracy of the proposed model can meet the requirements. In addition, since the parameters of the model can be flexibly adjusted and have clear physical significance, the proposed model can describe personalized car-following behavior.