CaltechAUTHORS
Published March 2022 | Version Submitted + Published
Journal Article Open

Robustness and Consistency in Linear Quadratic Control with Untrusted Predictions

Creators

Abstract

We study the problem of learning-augmented predictive linear quadratic control. Our goal is to design a controller that balances consistency, which measures the competitive ratio when predictions are accurate, and robustness, which bounds the competitive ratio when predictions are inaccurate.

Additional Information

© 2022 held by the owner/author(s). Received October 2021; revised December 2021; accepted January 2022. This work is supported by the National Science Foundation, under grants ECCS1931662, CCF 1637598, ECCS 1619352, CPS 1739355, AitF-1637598, CNS-1518941, PIMCO and Amazon Web Services. Tongxin Li and Ruixiao Yang contributed equally to the paper.

Attached Files

Published - 3508038.pdf

Submitted - 2106.09659.pdf

Files

2106.09659.pdf

Files (7.1 MB)

Name Size Download all
md5:a51e45d65c7740a93fc6dc08c069b96f
3.5 MB Preview Download
md5:a73b146f22f6b9d8ec7709b6b16c5178
3.6 MB Preview Download

Additional details

Additional titles

Alternative title
Robustness and Consistency in Linear Quadratic Control with Predictions

Identifiers

Eprint ID
109906
Resolver ID
CaltechAUTHORS:20210716-225846876

Related works

Describes
https://arxiv.org/abs/2106.09659 (URL)
https://resolver.caltech.edu/CaltechAUTHORS:20220802-839213000 (URL)

Funding

NSF
ECCS-1931662
NSF
CCF-1637598
NSF
ECCS-1619352
NSF
ECCS-1739355
NSF
CCF-1637598
NSF
CNS-1518941
PIMCO
Amazon Web Services

Dates

Created
2021-07-16
Created from EPrint's datestamp field
Updated
2022-08-02
Created from EPrint's last_modified field