Learning-Based Model Predictive Control: Toward Safe Learning in Control

TLDR

Recent advances in machine learning, sensing, and computation have spurred interest in data‑driven control, and model predictive control—an established constrained‑control framework—provides a reliable way to exploit such data while respecting safety constraints. This review summarizes and categorizes prior work on learning‑based MPC and discusses how MPC can enhance learning‑based controllers with constraint‑satisfaction guarantees. The authors classify learning‑based MPC approaches into three main categories, integrating MPC with learning methods to improve prediction models or parameterize controller cost and constraints.

Abstract

Recent successes in the field of machine learning, as well as the availability of increased sensing and computational capabilities in modern control systems, have led to a growing interest in learning and data-driven control techniques. Model predictive control (MPC), as the prime methodology for constrained control, offers a significant opportunity to exploit the abundance of data in a reliable manner, particularly while taking safety constraints into account. This review aims at summarizing and categorizing previous research on learning-based MPC, i.e., the integration or combination of MPC with learning methods, for which we consider three main categories. Most of the research addresses learning for automatic improvement of the prediction model from recorded data. There is, however, also an increasing interest in techniques to infer the parameterization of the MPC controller, i.e., the cost and constraints, that lead to the best closed-loop performance. Finally, we discuss concepts that leverage MPC to augment learning-based controllers with constraint satisfaction properties.