Robust MPC for input constrained uncertain systems with nonzero reference state

摘要

In this paper, we propose an efficient receding horizon control strategy for offset free regulation of input constrained uncertain systems under the assumption that the reference state for tracking and the corresponding input are known. A proportional-integral controller is used. Off-line LMI formulations are derived to obtain a feasible and invariant set for the augmented system (contrller+plant) around the reference state. The on-line control strategy is to steer the state deep into the invariant set to have a good tracking performance. This results in not only good performance but also large domain of attraction in the presence of input constraints.