Estimation of Vehicle State Using Robust Cubature Kalman Filter

Abstract

Accurate estimation of the vehicle state is quite significant for the efficacy of advanced active safety systems. To improve the active safety of vehicles, this paper focuses on the estimation of the vehicle state, in which a robust cubature Kalman algorithm is proposed to estimate yaw rate, sideslip angle, and vehicle speed. A three-degree-of-freedom (TDOF) vehicle dynamics model is first established. Based on the developed vehicle model, a robust cubature Kalman filter is used to estimate the vehicle state. Finally, both simulation and experimental tests are carried out to demonstrate the effectiveness of the proposed method. The test results indicate that the proposed estimation method can obtain more accurate estimators of the vehicle state than the existing methods, the estimation accuracy of the vehicle state is at least improved by more than 46.4%.