MPC for stable linear systems with model uncertainty (Q1868068)

The output tracking problem for systems with model uncertainties is under investigation. A stable robust controller based on the infinite horizon model predictive control (IHMPC) is extended to plants with model uncertainties. The state-space model, which is tailored to predict the system output as a continuous function of time is used. The method was developed for two classes of model uncertainty: (i) the uncertainty is approximated by a finite set of plant models corresponding to the possible operating points of the process, and (ii) the uncertainty concentrates on the system gain matrix that is assumed to lie in a polytopic set. Taking advantage of this particular model form, the cost function is defined in terms of the integral of the output error along an infinitive prediction horizon. Simulation examples demonstrate the usefulness of the developed IHMPC controller in practical applications.