CaltechAUTHORS
  A Caltech Library Service

Meta-Adaptive Nonlinear Control: Theory and Algorithms

Shi, Guanya and Azizzadenesheli, Kamyar and O'Connell, Michael and Chung, Soon-Jo and Yue, Yisong (2021) Meta-Adaptive Nonlinear Control: Theory and Algorithms. In: 35th Conference on Neural Information Processing Systems. Neural Information Processing Systems Foundation, Inc. . https://resolver.caltech.edu/CaltechAUTHORS:20220224-200754768

[img] PDF - Published Version
See Usage Policy.

1MB
[img] PDF - Accepted Version
See Usage Policy.

994kB
[img] PDF - Supplemental Material
See Usage Policy.

461kB

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20220224-200754768

Abstract

We present an online multi-task learning approach for adaptive nonlinear control, which we call Online Meta-Adaptive Control (OMAC). The goal is to control a nonlinear system subject to adversarial disturbance and unknown environment-dependent nonlinear dynamics, under the assumption that the environment-dependent dynamics can be well captured with some shared representation. Our approach is motivated by robot control, where a robotic system encounters a sequence of new environmental conditions that it must quickly adapt to. A key emphasis is to integrate online representation learning with established methods from control theory, in order to arrive at a unified framework that yields both control-theoretic and learning-theoretic guarantees. We provide instantiations of our approach under varying conditions, leading to the first non-asymptotic end-to-end convergence guarantee for multi-task nonlinear control. OMAC can also be integrated with deep representation learning. Experiments show that OMAC significantly outperforms conventional adaptive control approaches which do not learn the shared representation, in inverted pendulum and 6-DoF drone control tasks under varying wind conditions.


Item Type:Book Section
Related URLs:
URLURL TypeDescription
https://papers.nips.cc/paper/2021/hash/52fc2aee802efbad698503d28ebd3a1f-Abstract.htmlPublisherArticle
http://arxiv.org/abs/2106.06098arXivDiscussion Paper
https://github.com/GuanyaShi/Online-Meta-Adaptive-ControlRelated ItemCode
ORCID:
AuthorORCID
Shi, Guanya0000-0002-9075-3705
Azizzadenesheli, Kamyar0000-0001-8507-1868
Chung, Soon-Jo0000-0002-6657-3907
Yue, Yisong0000-0001-9127-1989
Additional Information:© 2021 Neural Information Processing Systems Foundation, Inc. This project was supported in part by funding from Raytheon and DARPA PAI, with additional support for Guanya Shi provided by the Simoudis Discovery Prize. There is no conflict of interest.
Group:GALCIT, Center for Autonomous Systems and Technologies (CAST)
Funders:
Funding AgencyGrant Number
Raytheon CompanyUNSPECIFIED
Defense Advanced Research Projects Agency (DARPA)UNSPECIFIED
Simoudis Discovery Prize (Caltech)UNSPECIFIED
Record Number:CaltechAUTHORS:20220224-200754768
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20220224-200754768
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:113575
Collection:CaltechAUTHORS
Deposited By: George Porter
Deposited On:25 Feb 2022 23:07
Last Modified:08 Mar 2022 19:38

Repository Staff Only: item control page