Adaptive Estimation of Vehicle Dynamics Through RLS and Kalman Filter Approaches

This article presents a new methodology for estimation of vehicle's vertical forces in order to enhance road safety. Direct measurement of vertical forces requires a complex and expensive experimental set-up, which is not acceptable for ordinary passenger cars. The main contribution of this article is providing a reliable estimator of vertical tire forces by using currently available low-cost sensors. The first advantage of the proposed method is that we modified the vehicle model to take into account the roll and pitch dynamics, which makes our estimator stay robust during sharp turning or at inclined road. The other advantage is that we proposed a process to identify the vehicle parameters, in stead of regarding them as known constants. This could enable our estimator to stay reliable even when the parameters are wrongly configured. The parameter identification process is based on recursive least squares (RLS) algorithm. The state observers are based on Kalman filter. The estimation process is applied and compared to real experimental data obtained in real conditions. Experimental results validate and prove the feasibility of this approach.