Integrated ASMO-DCKF state observer for distributed drive vehicles

To address the issue of inadequate information correction capabilities between the vehicle tyre force observer and state estimator, this study introduces an estimation algorithm that combines an adaptive sliding-mode observer with a double-cubature Kalman filter state estimator (ASMO-DCKF). Firstly, a seven-degree-of-freedom vehicle model is established, and a sliding-mode observer (SMO) is used to estimate tyre forces. A saturation function is utilised as a substitute for singular values to address system oscillation issues. Subsequently, double-cubature Kalman filter state estimators are used - one for vehicle state estimation and the other for road adhesion coefficient estimation. These two estimators mutually correct each other to enhance accuracy in estimating vehicle and road conditions. The Speedgoat-CarSim hardware-in-the-loop simulation platform is established to examine the proposed algorithm under typical operating conditions and compare it with the extended Kalman filter (EKF) algorithm. Results show a notable enhancement in the robustness and accuracy of the proposed algorithm over the EKF algorithm.