FUZZY APPROACH TO THE REAL-TIME LONGITUDINAL VELOCITY ESTIMATION OF AFWD CAR IN CRITICAL SITUATIONS

This paper presents a new approach to the fuzzy estimation of the vari ables of complex, fast, closed-loop systems. It is used to develop an original real-time longitudinal velocity estimator for FWD cars. Its a pplication covers highly critical driving situations and avoids the us e of an expensive optical cross-correlation sensor. The aim is to prov ide vehicle monitoring processes with a reliable value of the longitud inal velocity. Fuzzy aggregate indicators are used to identify and det ect the different ways a vehicle behaves. Then, a fuzzy expert system with rules based on these indicators decides which values should be us ed among those which allow the estimation of the longitudinal velocity .