Meta-Adaptive Nonlinear Control: Theory and Algorithms
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.
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.Attached Files
Published - NeurIPS-2021-meta-adaptive-nonlinear-control-theory-and-algorithms-Paper.pdf
Accepted Version - 2106.06098.pdf
Supplemental Material - NeurIPS-2021-meta-adaptive-nonlinear-control-theory-and-algorithms-Supplemental.pdf
Files
Additional details
- Eprint ID
- 113575
- Resolver ID
- CaltechAUTHORS:20220224-200754768
- Raytheon Company
- Defense Advanced Research Projects Agency (DARPA)
- Simoudis Discovery Prize (Caltech)
- Created
-
2022-02-25Created from EPrint's datestamp field
- Updated
-
2023-06-02Created from EPrint's last_modified field
- Caltech groups
- GALCIT, Center for Autonomous Systems and Technologies (CAST)