A Data-driven Riccati Equation
Rantzer, Anders LU
(2024)
6th Annual Learning for Dynamics and Control Conference, L4DC 2024
242.
p.504-513
- Abstract
Certainty equivalence adaptive controllers are analysed using a “data-driven Riccati equation”, corresponding to the model-free Bellman equation used in Q-learning. The equation depends quadratically on data correlation matrices. This makes it possible to derive simple sufficient conditions for stability and robustness to unmodeled dynamics in adaptive systems. The paper is concluded by short remarks on how the bounds can be used to quantify the interplay between excitation levels and robustness to unmodeled dynamics.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/8b0f08a1-afce-4fca-a3fd-fe21c2652e69
- author
- Rantzer, Anders
LU
- organization
- publishing date
- 2024
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- adaptive control, dual control, linear quadratic control, online learning
- host publication
- Proceedings of Machine Learning Research
- editor
- Abate, Alessandro ; Cannon, Mark ; Margellos, Kostas and Papachristodoulou, Antonis
- volume
- 242
- pages
- 10 pages
- publisher
- ML Research Press
- conference name
- 6th Annual Learning for Dynamics and Control Conference, L4DC 2024
- conference location
- Oxford, United Kingdom
- conference dates
- 2024-07-15 - 2024-07-17
- external identifiers
-
- scopus:85203676904
- language
- English
- LU publication?
- yes
- additional info
- Publisher Copyright: © 2024 A. Rantzer.
- id
- 8b0f08a1-afce-4fca-a3fd-fe21c2652e69
- alternative location
- https://proceedings.mlr.press/v242/rantzer24a.html
- date added to LUP
- 2024-12-04 10:29:13
- date last changed
- 2025-04-04 14:15:50
@inproceedings{8b0f08a1-afce-4fca-a3fd-fe21c2652e69,
abstract = {{<p>Certainty equivalence adaptive controllers are analysed using a “data-driven Riccati equation”, corresponding to the model-free Bellman equation used in Q-learning. The equation depends quadratically on data correlation matrices. This makes it possible to derive simple sufficient conditions for stability and robustness to unmodeled dynamics in adaptive systems. The paper is concluded by short remarks on how the bounds can be used to quantify the interplay between excitation levels and robustness to unmodeled dynamics.</p>}},
author = {{Rantzer, Anders}},
booktitle = {{Proceedings of Machine Learning Research}},
editor = {{Abate, Alessandro and Cannon, Mark and Margellos, Kostas and Papachristodoulou, Antonis}},
keywords = {{adaptive control; dual control; linear quadratic control; online learning}},
language = {{eng}},
pages = {{504--513}},
publisher = {{ML Research Press}},
title = {{A Data-driven Riccati Equation}},
url = {{https://proceedings.mlr.press/v242/rantzer24a.html}},
volume = {{242}},
year = {{2024}},
}