RBFNN-based adaptive sliding mode control for nonlinear active suspension system with electromagnetic actuator

Air springs have garnered increasing attention due to their potential to significantly improve vehicle ride comfort. However, its strong nonlinearity presents substantial challenges to active control of suspension systems, especially in the absence of accurate models. To deal with this problem, this study presents an adaptive sliding mode controller that employs a radial basis function neural network (RBFNN) to estimate the nonlinear forces produced by air springs on the vehicle body. The stability of the system is ensured by constructing appropriate Lyapunov functions. The operation of electromagnetic actuator and its control via field-oriented control to track the sliding mode control force is elucidated and thus improving the suspension control performance. Furthermore, the accuracy of RBFNN in estimating air spring nonlinear force is validated through hardware-based experiments. The simulations under various road excitation conditions are carried out, which show superior control performance of the proposed method compared with traditional sliding mode controller suspension systems, particularly in enhancing ride comfort. The experiments and simulations demonstrate that the proposed method has the capability of exceptional suspension performance without requiring accurate air spring nonlinear model, which greatly simplifies controller design.