Nonlinear Model Predictive Control with aggregated constraints?

This paper presents a constraint aggregation approach for Nonlinear Model Predictive Control (NMPC). Constraint aggregation functions provide an approximation of the feasible region with a reduced number of nonlinear constraints. The effect of the aggregation on the closed-loop system performance and stability is studied using tools from sensitivity analysis. Numerical results for the control of a 20th order flexible aircraft model show that significant computational savings can be achieved. The proposed method can facilitate the implementation of NMPC solutions for large-scale systems.(c) 2022 Elsevier Ltd. All rights reserved.

Automated summary

This paper presents a constraint aggregation approach for Nonlinear Model Predictive Control (NMPC) by approximating the feasible region with a reduced number of nonlinear constraints. The effect of aggregation on closed-loop system performance and stability is studied, and numerical results for a flexible aircraft model demonstrate significant computational savings and the applicability of the proposed method to large-scale systems.