Controle preditivo robusto via desigualdades matriciais lineares aplicado a sistemas dinâmicos com comutação

Abstract

This work proposes the application of robust model predictive control for discrete linear systems subject to switching under uncertainty and induced norm bounded disturbances. The proposed control was modeled via linear matrix inequalities from Lyapunov inequality and aims to minimize the upper bound of the quadratic performance index to a horizon of infinite prediction. The proposed control strategy can be implemented in the online and offline designs and in the state feedback and output feedback configurations, the latter being associated with a state estimator. To demonstrate the effectiveness of the control strategy, a comparison between the implemented designs is performed through the graphical analysis of the response in time and non-intrusive performance indices. Based on the results, it can be affirmed that the application of the proposed control was successful in both models approached and that the offline design proved to be more promising than the online for these applications