Risk indicators anticipation based on the vehicle dynamics anticipation to avoid accidents

Abstract

This article leads to the challenging problem of increasing vehicle driving security by applying on boarded intelligent diagnosis systems; it presents a methodology of evaluating, in an anticipated way, the risk of having an accident (skid and rollover). The methodology consists in adopting assumptions about the trajectory, the longitudinal velocity and the longitudinal acceleration in future instants and use these assumptions, allied to previous road information to calculate the future vehicle dynamics parameters. Once calculated, the risk indicators based on these parameters could be predicted in order to expect and avoid possible dangerous situations. These indicators are the lateral load transfer (LTR) based on vertical forces, and the lateral skid indicator (LSI) based ont the maximum friction coefficient and the used friction coefficient. A sliding window system is used to apply the method on the whole trajectory to take into account the vehicle dynamics updates by the driver.