Horizon Adaptation for Nonlinear Model Predictive Controllers with guaranteed Degree of Suboptimality

TL;DR

This paper introduces adaptation strategies for nonlinear model predictive controllers to ensure a guaranteed minimum suboptimality level by adjusting the optimization horizon length, with proven stability and suboptimality bounds.

Contribution

It develops and proves horizon shortening and prolongation strategies for nonlinear MPC that guarantee suboptimality bounds and stability, extending existing results to variable horizon lengths.

Findings

Proposed horizon adaptation strategies with proven suboptimality guarantees

Effective implementation methods for horizon adjustment

Extended stability and suboptimality results to variable horizon MPC

Abstract

We propose adaptation strategies to modify the standard constrained model predictive controller scheme in order to guarantee a certain lower bound on the degree of suboptimality. Within this analysis, the length of the optimization horizon is the parameter we wish to adapt. We develop and prove several shortening and prolongation strategies which also allow for an effective implementation. Moreover, extensions of stability results and suboptimality estimates to model predictive controllers with varying optimization horizon are shown.