Constrained nonlinear filter for vehicle sideslip angle estimation with no a priori knowledge of tyre characteristics

The vehicle sideslip angle is one of the most functional feedbacks for the actual control systems of vehicle dynamics. The measurement of the sideslip angle is expensive and unsuitable for common vehicles. Consequently, its estimation is nowadays an important task. This paper focuses on the vehicle sideslip angle estimation adopting a constrained unscented Kalman filter (CUKF) that takes into account state constrains during the estimation process. State boundaries are useful in real-world applications to prevent unphysical results and to improve the estimator robustness. The proposed technique fully takes into account the measurement noise and nonlinearities. A vehicle model with single track has been adopted for the design of the estimator. Simulations have been carried out and comparisons with the unscented Kalman filter (UKF) are illustrated. Performance of the estimators have been checked through the application to experimental data. The results show the goodness of the CUKF, able to give an estimate fully in accordance with the measurement. Moreover, the results show that the CUKF, due to the presence of the boundaries, outperforms the UKF. (C) 2017 Elsevier Ltd. All rights reserved.