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

By Lukas Hewing, Kim Peter Wabersich, Marcel Menner, Melanie Nicole Zeilinger. Annu. Rev. Control. Robotics Auton. Syst., 2020

910 citations recorded by the source corpus.

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.

Subfields: Control and Systems Engineering

View Paper: https://www.semanticscholar.org/paper/932709a41eee54f3eddbd3b3a7baf043c3ed33ec

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Control and Systems Engineering

1. Angela P. Schoellig
2. Mehran Mesbahi
3. M. Kanat Camlibel
4. Henk van Waarde
5. Florian Dörfler
6. Ivan Markovsky

Citations in the corpus

• From Noisy Data to Feedback Controllers: Nonconservative Design via a Matrix S-Lemma (2020, methodology): “We mention contributions to data-driven optimal control [5]–[16], PID control [17], [18], predictive control [19]–[23], and nonlinear control [24]–[28].”
• Bridging Direct and Indirect Data-Driven Control Formulations via Regularizations and Relaxations (2021, neutral): “Representative recent surveys are in [1]–[8].”
• Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learning (2021, background): “This usually yields poor performance, as described in (7).”