An Estimation Strategy for Sideslip Angle Fusion Utilizing a Linear Adaptive Model

This paper introduces a robust method for estimating the sideslip angle. We propose a fusion approach that integrates an adaptive linear model and a kinematic model to enhance the accuracy of sideslip angle estimation. The adaptive linear model is constructed based on a two-degree-of-freedom vehicle dynamics model, accounting for variations in speed and wheel parameters. Dynamic estimation of the sideslip angle is achieved through the utilization of the extended Kalman filter (EKF). To determine the sideslip angle, we employ the integral method, considering the relationship between the vehicle's lateral and longitudinal acceleration and the sideslip angle. Aiming at the disadvantage that the adaptive linear model estimation method can estimate the sideslip angle accurately only when there is lateral acceleration motion, a fusion transformation mechanism of the minimum mean square error weighted average method is designed to estimate the sideslip angle. The efficacy of the proposed approach is demonstrated through Simulink-Carsim joint simulation. The results reveal that employing both algorithms for sideslip angle estimation achieves greater accuracy compared to relying on a single estimation algorithm.