A new OFRMPC formulation with on‐line synthesis of the dynamic output feedback controller

Abstract

Summary This paper proposes a new approach for the design of output feedback robust model predictive control (OFRMPC) with a dynamic output feedback controller (DOFC) for linear uncertain systems subject to input and output constraints. The main contribution of this work is the full on‐line synthesis of the DOFC as part of a convex optimization problem, with constraint satisfaction and asymptotic stability guarantees. A numerical example is employed to illustrate the advantage of the proposed control law, as compared with another OFRMPC strategy with partial DOFC synthesis. The present paper also points out an inconsistency in the mathematical development of a previous related OFRMPC formulation (‘improved dynamic output feedback RMPC for linear uncertain systems with input constraints’). Copyright © 2017 John Wiley & Sons, Ltd.

Corrigendum to: “A new OFRMPC formulation with on‐line synthesis of the dynamic output feedback controller”

“…In the paper under consideration, 1 Remark 6 should be corrected as follows. Equation (99) must be replaced with…”

An observer‐based output feedback robust MPC approach for constrained LPV systems with bounded disturbance and noise

“…However, it is known that simultaneous optimization of the observer and controller parameters in the output feedback robust MPC for LPV systems often lead to a bilinear matrix inequality problem, which is nonconvex optimization and cannot be solved in polynomial time . The authors in other works investigate dynamic output feedback robust MPC for LPV systems without bounded disturbance, in which saturated inputs are exploited to fully utilize the capability of actuators. Some parameters are predesigned in the work of Li and Xi to reformulate the problem as convex optimization, and all the controller parameters are on‐line optimized in the works of Colombo Junior et al and Shi et al by considering additional rank constraints on system parameters.…”

“…In the field of robust control, linear parameter varying (LPV) systems can approximate real nonlinear systems or represent uncertain systems by a polytopic family of linear systems, whose dynamics depend on time‐varying scheduling parameters . Over the last two decades, research activities in robust MPC for LPV systems have become an important branch, eg, state feedback robust MPC with measurable system states and output feedback robust MPC with unknown system states . In robust MPC control problems, it is important to know the current system state information to ensure reliable operations.…”

“…The authors in other works investigate dynamic output feedback robust MPC for LPV systems without bounded disturbance, in which saturated inputs are exploited to fully utilize the capability of actuators. Some parameters are predesigned in the work of Li and Xi to reformulate the problem as convex optimization, and all the controller parameters are on‐line optimized in the works of Colombo Junior et al and Shi et al by considering additional rank constraints on system parameters. For the LPV systems with bounded disturbance, the output feedback robust MPC approaches in other works employ the technique of quadratic boundedness to guarantee that the current and future augmented states are constrained in one robust positive invariant (RPI) set such that robust stability of the augmented closed‐loop system is ensured.…”