Composite anti-disturbance control with disturbance prediction and utilization: A tube MPC approach

This paper proposes a composite anti-disturbance control scheme with disturbance prediction and utilization. Part of the disturbances is estimated together with the system state via a full-order disturbance observer (FODO). Then, these portions of disturbances and all states are predicted based on partially known dynamics, and updated with the estimation information at the beginning of each optimization. The rest of the disturbances are handled via compensation and attenuation with the composite tube model predictive control (MPC). Such predictions of both disturbances and states require less conservative assumptions on prior information compared with existing disturbance prediction/preview works. In addition, compared to existing anti-disturbance control (ADC) algorithms, the control energy consumption is reduced by introducing predicted disturbance into the optimization for utilization. Furthermore, rigorous theoretical analyses are provided for the feasibility of the proposed algorithm and the stability of the closed-loop system. Finally, numerical examples are presented to validate the effectiveness of the proposed scheme.