Robust High-Bandwidth Preview Control Based on Integrated Estimation of State and Disturbance for Uncertain Active Suspension Systems

Safety and comfort of autonomous vehicles (AVs) are closely related to active suspension systems (ASSs). However, complex nonlinearities and actuator disturbances limit the effectiveness of perception preview control (PC) algorithms for ASS. To address this problem, a robust high-bandwidth PC method based on the integrated estimation of system state and lumped uncertainty is proposed in this article. Specifically, a novel augmented unscented Kalman filter (AUKF) considering delay is introduced to improve observation. Additionally, an innovative preview-based nonlinear disturbance observer (PNDO) facilitates the precise estimation of unknown lumped disturbance force. Then, based on multiobjective model predictive control (MPC), a robust constrained optimal preview control (COPC) framework with a disturbance compensation mechanism (DCM) is established. This framework improves ride comfort and robustness while satisfying safety constraints with reduced online computational complexity. Furthermore, the complete closed-loop system (CCLS) stability is well proved based on the Lyapunov theory. Finally, simulation and bench test results indicate that the proposed algorithm effectively handles time-varying disturbances to improve ASS performance, with the solution time remaining below 1.1ms in the bench test. These results demonstrate its strong potential for application in AVs.