Robust Cooperative Distributed Model Predictive Control based on Set-membership Approach

To enhance robustness of distributed model predictive control (DMPC) and guarantee closed-loop recursive feasibility and convergence for systems suffering massive uncertainties, this work is dedicated to proposing a robust cooperative DMPC formulation for discrete-time linear systems subject to unknown-but-bounded disturbances. A robust constraint tightening technique is developed based on set-membership approach, where constraints on system states and inputs are tightened iteratively along the prediction horizon with local control gains. With this formulation, we discuss recursive feasibility and convergence of the closed-loop control system. Finally, we show simulation results with a mass-spring-damper mechanical system to illustrate the effectiveness of the proposed control strategy.