A dissipativity approach to robustness in constrained model predictive control

In this paper, we propose a novel closed-loop stability test applicable to a broad class of model predictive control (MPC) policies for discrete-time systems having model uncertainty described by given sum quadratic constraints. The proposed stability test is well suited for both analysis and design. In particular, robust closed-loop stability can be ensured by choosing the cost function parameters and the constraints in the MPC algorithm so as to satisfy, respectively, a linear matrix inequality condition and a set invariance condition.