Vehicular Vertical Tire Forces Estimation Using Unscented Kalman Filter

Abstract

In this paper, we propose a vertical forces estimator based on Unscented Kalman Filter (UKF). The vehicle dynamics is so complicated that it is common to estimate its motion using its reduced model based on a decoupled model consisting of a lateral dynamic model and a longitudinal dynamic model on an even road. The decoupled model is not effective in estimating vehicle motion, when the vehicle is severely maneuvered causing high rolling motion. In this paper, we propose a new estimator to estimate the roll angle and lateral acceleration in using UKF algorithm for 3-Degree-Of-Freedom Model. Then a lateral load transfer model and a longitudinal load transfer model are used to estimate the vertical load force on each tire. From the numerical simulation study using MATLAB/CarSim, we observed that there were good agreements between ground truth and the estimated vertical tire forces even the vehicle was driven at 60 km/hour speed by a sine wave steering wheel angle with ± 60 degrees at 0.1Hz on various driving roads.