Constrained distributed model predictive control for linear systems with delayed input

Abstract

In this paper, an approach to design constrained distributed model predictive control (MPC) is proposed for the polytopic uncertain systems with input delays. The large-scale linear system we consider is decomposed into several subsystems with distributed control inputs. This allows us to design distributed MPC controller independently for each subsystem. An augmented polytopic uncertainty description is invoked to handle the input delays. At each time interval, each sub-controller solves a linear matrix inequality (LMI) optimization problem to obtain local control input. An iterative online algorithm for constrained distributed MPC is introduced to coordinate the distributed MPC controllers. A numerical example is simulated to demonstrate the effectiveness of the proposed algorithm.