Scenario-based indirect feedback SMPC of constrained uncertain systems

Abstract This paper proposes a stochastic model predictive control (SMPC) scheme using indirect feedback for constrained linear systems with uncertain parameters and additive disturbances. Using the concept of the probability reachable set, a robust model predictive control-like tightened constraint is established to handle the chance constraints. The probability reachable set is characterized by the sub-level sets derived from scenario optimization. Some sufficient conditions are derived to ensure the recursive feasibility of the proposed method. Furthermore, an online set-membership identification is adopted to estimate a non-falsified parameter set of uncertain parameters, which reduces the conservatism of SMCP. Some comparison simulations verify the effectiveness and advantages of the proposed scheme.