Robust Model Predictive Control Based on Nominal System Optimization and Control Input Saturation

Abstract

The paper presents in detail alternative robust model predictive control approach based on the optimization of convergence rate subject to nominal system, and additional saturation of control inputs. The approach is a compromise between guaranteed convergence rate and high computational complexity on the one hand, and larger set of feasible initial conditions and lower computational burden on the other hand. The applicability of the proposed strategy is verified using a case study of uncertain chemical reactor stabilization.