Road Slope and Vehicle Dynamics Estimation

Driving safely can be achieved by the prevention of risky situations which may require the knowledge of vehicle dynamic state as well as road geometry. It is thus essential to have in real-time a good estimation of the related variables and parameters. Among the parameters of the road that are influencing vehicle longitudinal motion one can find the slope which can not be measured with reduced cost sensor. Vehicle lateral motion is mainly affected by the value of the lateral speed which can not simply measured too. In this paper, an observer based method for the estimation of the vehicle dynamics using a nonlinear vehicle model is proposed. It uses a combination of Extended Kalman filter (EKF) and Luenberger Observer (LO). The observers use several standard measurements such as : the yaw rate, the steering angle and the rotational velocity of the four tires. Experimental tests conducted with a prototype vehicle prove the effectiveness of the proposed approach.