A safety governor for learning explicit MPC controllers from data

TL;DR

This paper introduces a safety governor for learning-based explicit MPC controllers that guarantees constraint satisfaction and recursive feasibility, simplifying implementation in high-dimensional systems.

Contribution

It presents a novel dual-mode explicit MPC structure reformulated for safety, with a safety governor ensuring constraints, and proves recursive feasibility.

Findings

Ensures constraint satisfaction in learning-based MPC

Simplifies implementation in high-dimensional systems

Proven recursive feasibility of the safety scheme

Abstract

We tackle neural networks (NNs) to approximate model predictive control (MPC) laws. We propose a novel learning-based explicit MPC structure, which is reformulated into a dual-mode scheme over maximal constrained feasible set. The scheme ensuring the learning-based explicit MPC reduces to linear feedback control while entering the neighborhood of origin. We construct a safety governor to ensure that learning-based explicit MPC satisfies all the state and input constraints. Compare to the existing approach, our approach is computationally easier to implement even in high-dimensional system. The proof of recursive feasibility for the safety governor is given. Our approach is demonstrated on numerical examples.