An adaptive robust sliding mode control strategy based on radial basis function neural networks for the magnetorheological semi-active suspension

This paper proposes an adaptive robust sliding mode control (SMC) strategy based on radial basis function neural networks (RBF-NNs) to address nonlinear issues in semi-active suspension system (SASS), such as external disturbances and uncertain parameters. Firstly, the Dahl model is selected for the parameters identification of the magnetorheological (MR) damper, followed by the establishment of a model for the SASS. Secondly, this paper proposes a control strategy that integrates robust control with SMC. The RBF-NNs are utilized to estimate the system’s uncertain parameters, thereby enhancing the robustness of the suspension system against external disturbances and ensuring its stability. The stability and controllability of the closed-loop SASS are rigorously verified through the application of Lyapunov theory. 1 Under the road excitations of B-Class and speed bump, the dynamic characteristics of the passive control, the robust SMC, and the adaptive robust SMC based on RBF-NNs applied to the MR-SASS are analyzed. The vertical acceleration of the automobile body, the suspension working space and the tire dynamic load are selected as the evaluation indices. The results demonstrate that the adaptive robust SMC significantly improves the ride comfort and handling stability of the vehicle.