Robust model predictive control for nonlinear systems

Abstract

On the basis of the fuzzy T-S (Takagi-Sugeno) model, we propose a robust model predictive control for a class of nonlinear systems with constraint inputs. The upper bound of predictive cost is derived; the constraints on stability and inputs are transformed into linear matrix inequalities (LMIs), which can be easily solved. Thus we adopt a parallel distributed compensation (PDC) controller in the scheme. Sufficient conditions of moving horizon optimization are derived based on LMIs and Lyapunov function, and consequently the stability of closed-loop systems is proved. The simulation results verify the effectiveness of the proposed method. © 2011 Springer-Verlag Berlin Heidelberg.