State observer for semi-active suspension system based on MPA-UKF

With consideration of the nonlinearity of semi-active suspensions and to enhance the accuracy of nonlinear state observers while achieving the automatic tuning of parameters, the Marine Predators Algorithm (MPA) is applied to the noise covariance tuning process of the Unscented Kalman Filter (UKF) based on a seven-degree-of-freedom nonlinear vehicle model. Therefore, a novel MPA-UKF algorithm is proposed for the state observation of semi-active suspensions. Additionally, a semi-active suspension vehicle test platform is set up to validate the algorithm’s performance under various driving conditions. The experimental outcomes demonstrate that the proposed algorithm improves parameter tuning efficiency while enhancing the accuracy of the Unscented Kalman Filter by 10% and can serve as the observer for suspension controllers.