Nonlinear Optimal Control for Mismatched Disturbance Rejection and Its Application to AGV

This paper addresses the challenge of mismatched disturbance rejection in nonlinear discrete-time systems by proposing a novel model predictive control algorithm based on optimal control theory, which significantly enhances disturbance rejection performance. The algorithm constructs a concise yet effective performance index capable of handling both matched and mismatched disturbances by incorporating a disturbance observer, thereby simultaneously fulfilling the dual objectives of trajectory tracking and disturbance attenuation. To improve computational efficiency, a solution method grounded in optimal control theory is devised, enabling rapid online controller computation. This approach ensures superior real-time performance while delivering more efficient tracking and disturbance rejection. The effectiveness of the proposed algorithm is validated through simulations and experimental tests on an automated guided vehicle platform. Results show that the algorithm achieves outstanding performance in both disturbance rejection and computational efficiency, markedly surpassing conventional methods. It can effectively withstand unknown disturbances while meeting the real-time and reliability demands of practical engineering applications, demonstrating broad prospects for deployment in real-world systems.