Soft constrained finite horizon model predictive control

This article addresses tuning of the finite prediction horizon of soft-constrained model predictive control. The findings presented in this paper prove that there exists a finite horizon such that the infinite horizon soft-constrained model predictive control problem can be solved as a finite horizon soft-constrained model predictive control problem. Algorithms are proposed to compute the upper-bound of the prediction horizon in online and offline modes. The effectiveness of the proposed algorithm is verified by illustrative example.