CaltechAUTHORS
  A Caltech Library Service

Episodic Koopman Learning of Nonlinear Robot Dynamics with Application to Fast Multirotor Landing

Folkestad, Carl and Pastor, Daniel and Burdick, Joel W. (2020) Episodic Koopman Learning of Nonlinear Robot Dynamics with Application to Fast Multirotor Landing. In: 2020 IEEE International Conference on Robotics and Automation (ICRA). IEEE , Piscataway, NJ, pp. 9216-9222. ISBN 9781728173955. https://resolver.caltech.edu/CaltechAUTHORS:20200925-072027187

[img] PDF - Submitted Version
Creative Commons Attribution.

4Mb

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20200925-072027187

Abstract

This paper presents a novel episodic method to learn a robot’s nonlinear dynamics model and an increasingly optimal control sequence for a set of tasks. The method is based on the Koopman operator approach to nonlinear dynamical systems analysis, which models the flow of observables in a function space, rather than a flow in a state space. Practically, this method estimates a nonlinear diffeomorphism that lifts the dynamics to a higher dimensional space where they are linear. Efficient Model Predictive Control methods can then be applied to the lifted model. This approach allows for real time implementation in on-board hardware, with rigorous incorporation of both input and state constraints during learning. We demonstrate the method in a real-time implementation of fast multirotor landing, where the nonlinear ground effect is learned and used to improve landing speed and quality.


Item Type:Book Section
Related URLs:
URLURL TypeDescription
https://doi.org/10.1109/icra40945.2020.9197510DOIArticle
https://arxiv.org/abs/2004.01708arXivDiscussion Paper
Additional Information:© 2020 IEEE. Folkestad and Pastor - Both authors contributed equally. This work has been supported in part by Raytheon Company and the DARPA Physics-infused AI program. The first author is grateful for the support of Aker Scholarship Foundation. The authors would like to thank Igor Mezic, Ryan Mohr, and Maria Fonoberova for helpful discussions.
Funders:
Funding AgencyGrant Number
Raytheon CompanyUNSPECIFIED
Defense Advanced Research Projects Agency (DARPA)UNSPECIFIED
Aker Scholarship FoundationUNSPECIFIED
Record Number:CaltechAUTHORS:20200925-072027187
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20200925-072027187
Official Citation:C. Folkestad, D. Pastor and J. W. Burdick, "Episodic Koopman Learning of Nonlinear Robot Dynamics with Application to Fast Multirotor Landing," 2020 IEEE International Conference on Robotics and Automation (ICRA), Paris, France, 2020, pp. 9216-9222, doi: 10.1109/ICRA40945.2020.9197510
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:105535
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:25 Sep 2020 14:39
Last Modified:25 Sep 2020 14:39

Repository Staff Only: item control page