An online learning analysis of minimax adaptive control
(2023) 2023 62nd IEEE Conference on Decision and Control (CDC) p.1034-1039- Abstract
- We present an online learning analysis of minimax adaptive control for the case where the uncertainty includes a finite set of linear dynamical systems. Precisely, for each system inside the uncertainty set, we define the model-based regret by comparing the state and input trajectories from the minimax adaptive controller against that of an optimal controller in hindsight that knows the true dynamics. We then define the total regret as the worst case model-based regret with respect to all models in the considered uncertainty set. We study how the total regret accumulates over time and its effect on the adaptation mechanism employed by the controller. Moreover, we investigate the effect of the disturbance on the growth of the regret over... (More)
- We present an online learning analysis of minimax adaptive control for the case where the uncertainty includes a finite set of linear dynamical systems. Precisely, for each system inside the uncertainty set, we define the model-based regret by comparing the state and input trajectories from the minimax adaptive controller against that of an optimal controller in hindsight that knows the true dynamics. We then define the total regret as the worst case model-based regret with respect to all models in the considered uncertainty set. We study how the total regret accumulates over time and its effect on the adaptation mechanism employed by the controller. Moreover, we investigate the effect of the disturbance on the growth of the regret over time and draw connections between robustness of the controller and the associated regret rate. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/0544f113-691c-42aa-84a7-3048c8da8ee1
- author
- Renganathan, Venkatraman LU ; Iannelli, Andrea and Rantzer, Anders LU
- organization
- publishing date
- 2023
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- An online learning analysis of minimax adaptive control
- pages
- 6 pages
- publisher
- IEEE - Institute of Electrical and Electronics Engineers Inc.
- conference name
- 2023 62nd IEEE Conference on Decision and Control (CDC)
- conference location
- Marina Bay Sands, Singapore
- conference dates
- 2023-12-13 - 2023-12-15
- external identifiers
-
- scopus:85179880001
- project
- Scalable Control of Interconnected Systems
- language
- English
- LU publication?
- yes
- id
- 0544f113-691c-42aa-84a7-3048c8da8ee1
- alternative location
- https://ieeexplore.ieee.org/abstract/document/10384114
- date added to LUP
- 2024-02-22 11:52:49
- date last changed
- 2024-02-28 13:34:45
@inproceedings{0544f113-691c-42aa-84a7-3048c8da8ee1, abstract = {{We present an online learning analysis of minimax adaptive control for the case where the uncertainty includes a finite set of linear dynamical systems. Precisely, for each system inside the uncertainty set, we define the model-based regret by comparing the state and input trajectories from the minimax adaptive controller against that of an optimal controller in hindsight that knows the true dynamics. We then define the total regret as the worst case model-based regret with respect to all models in the considered uncertainty set. We study how the total regret accumulates over time and its effect on the adaptation mechanism employed by the controller. Moreover, we investigate the effect of the disturbance on the growth of the regret over time and draw connections between robustness of the controller and the associated regret rate.}}, author = {{Renganathan, Venkatraman and Iannelli, Andrea and Rantzer, Anders}}, booktitle = {{An online learning analysis of minimax adaptive control}}, language = {{eng}}, pages = {{1034--1039}}, publisher = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}}, title = {{An online learning analysis of minimax adaptive control}}, url = {{https://ieeexplore.ieee.org/abstract/document/10384114}}, year = {{2023}}, }