Bayesian-Monte Carlo Model for Collision Avoidance System Design of Cognitive Connected Vehicle

摘要

Cognitive connected vehicles will require a number of essential features that integrate intelligent technology and human factors, including collision avoidance advice and adaptive longitudinal control. This paper describes a self-calibrating adaptive model for aiding the design of a warning system for preventing rear and side swipe collisions. The cognitive vehicle is expected to have the capability of information on location and distance between vehicles obtained on-line. The location and distance information is used in association with a Monte Carlo simulation and Bayesian decision model to identify pre-crash condition. Here, the case of human control is covered and the system provides advice for avoiding rear or side swipe accidents while minimizing false alarms. The model structure and algorithm are presented and illustrative examples of distracted driving are provided.