Research on Fuzzy Inference of Driver´s Risk Perception of Rear-End Collision on Freeway

The purpose of this research is to develop a microscopic indicator of risk perception of rear-end collision based on fuzzy inference system in order to effectively surrogate the traditional evaluation method based on the accident data. The factors that result in the rear-end collision such as the relative speed, the acceleration of leader vehicle, car-following safety distance are taken as the fuzzy input variables and the value of risk perception of rear-end collision is regarded as the output variable. Meanwhile, considering the effect of the weather, the perception reaction time can be also obtained by the fuzzy inference system so as to calculate car-following safety distance. Finally, the research results can provide a support for the early warning technology of traffic accidents.