Publication
Title
A framework for the competitive analysis of model predictive controllers
Author
Abstract
This paper presents a framework for the competitive analysis of Model Predictive Controllers (MPC). Competitive analysis means evaluating the relative performance of the MPC as compared to other controllers. Concretely, we associate the MPC with a regret value which quantifies the maximal difference between its cost and the cost of any alternative controller from a given class. Then, the problem we tackle is that of determining whether the regret value is at most some given bound. Our contributions are both theoretical as well as practical: (1) We reduce the regret problem for controllers modeled as hybrid automata to the reachability problem for such automata. We propose a reachability-based framework to solve the regret problem. Concretely, (2) we propose a novel CEGAR-like algorithm to train a deep neural network (DNN) to clone the behavior of the MPC. Then, (3) we leverage existing reachability analysis tools capable of handling hybrid automata with DNNs to check bounds on the regret value of the controller.
Language
English
Source (journal)
Lecture notes in computer science. - Berlin, 1973, currens
Source (book)
Reachability Problems : 17th International Conference, RP 2023, Nice, France, October 11–13, 2023, Proceedings
Publication
Cham : Springer , 2023
ISSN
0302-9743 [print]
1611-3349 [online]
ISBN
978-3-031-45285-7
DOI
10.1007/978-3-031-45286-4_11
Volume/pages
p. 141-154
Full text (Publisher's DOI)
UAntwerpen
Faculty/Department
Research group
Publication type
Subject
Affiliation
Publications with a UAntwerp address
External links
Record
Identifier
Creation 12.10.2023
Last edited 17.06.2024
To cite this reference