CaltechAUTHORS
  A Caltech Library Service

Learning to Control an Unstable System with One Minute of Data: Leveraging Gaussian Process Differentiation in Predictive Control

Jimenez Rodriguez, Ivan D. and Rosolia, Ugo and Ames, Aaron D. and Yue, Yisong (2021) Learning to Control an Unstable System with One Minute of Data: Leveraging Gaussian Process Differentiation in Predictive Control. In: 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE , Piscataway, NJ, pp. 3896-3903. ISBN 978-1-6654-1714-3. https://resolver.caltech.edu/CaltechAUTHORS:20210510-095024771

[img] PDF - Submitted Version
See Usage Policy.

3MB

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20210510-095024771

Abstract

We present a straightforward and efficient way to control unstable robotic systems using an estimated dynamics model. Specifically, we show how to exploit the differentiability of Gaussian Processes to create a state-dependent linearized approximation of the true continuous dynamics that can be integrated with model predictive control. Our approach is compatible with most Gaussian process approaches for system identification, and can learn an accurate model using modest amounts of training data. We validate our approach by learning the dynamics of an unstable system such as a segway with a 7-D state space and 2-D input space (using only one minute of data), and we show that the resulting controller is robust to unmodelled dynamics and disturbances, while state-of-the-art control methods based on nominal models can fail under small perturbations. Code is open sourced at https://github.com/learning-and-control/core.


Item Type:Book Section
Related URLs:
URLURL TypeDescription
https://doi.org/10.1109/IROS51168.2021.9636786DOIArticle
https://arxiv.org/abs/2103.04548arXivDiscussion Paper
ORCID:
AuthorORCID
Rosolia, Ugo0000-0002-1682-0551
Ames, Aaron D.0000-0003-0848-3177
Yue, Yisong0000-0001-9127-1989
Alternate Title:Learning Unstable Dynamics with One Minute of Data: A Differentiation-based Gaussian Process Approach
Additional Information:© 2021 IEEE. This work was supported by NSF awards 1637598, 1645832, 1932091, 1924526, and 1923239, and funding from AeroVironment, JPL and BMW. We would like to thank Andrew Singletary and Ellen Novoseller for their help in setting up the robotics and the mathematical derivations, respectively. We would also like to thank the follwing software packages: PyTorch [43], CVXPY [44] and Gurobi [45].
Funders:
Funding AgencyGrant Number
NSFCCF-1637598
NSFCNS-1645832
NSFCNS-1932091
NSFECCS-1924526
NSFCMMI-1923239
AeroVironmentUNSPECIFIED
JPLUNSPECIFIED
BMWUNSPECIFIED
DOI:10.1109/IROS51168.2021.9636786
Record Number:CaltechAUTHORS:20210510-095024771
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20210510-095024771
Official Citation:I. D. J. Rodriguez, U. Rosolia, A. D. Ames and Y. Yue, "Learning to Control an Unstable System with One Minute of Data: Leveraging Gaussian Process Differentiation in Predictive Control," 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2021, pp. 3896-3903, doi: 10.1109/IROS51168.2021.9636786
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:109029
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:10 May 2021 17:22
Last Modified:04 Apr 2022 19:04

Repository Staff Only: item control page