Fuzzy automated braking system for collision prevention

Many automobile accidents could be avoided if the correct amount of brake pressure had been applied at the proper time. In this paper we propose the use of a fuzzy control system that could be embedded in an automobile to remove the driver´s reaction time from the situation. The system, consisting of two fuzzy controllers connected in cascade, analyzes possible accident situations based on the vehicle separation distance, their relative velocity and a static friction coefficient that represents the road conditions. The controller determines the sufficient brake pressure required to prevent collisions while providing a smooth ride for the vehicle´s passengers. This smooth performance was achieved by using genetic algorithms to tune the fuzzy controllers´ parameters.