CaltechAUTHORS
  A Caltech Library Service

Human Preference-Based Learning for High-dimensional Optimization of Exoskeleton Walking Gaits

Tucker, Maegan and Cheng, Myra and Novoseller, Ellen and Cheng, Richard and Yue, Yisong and Burdick, Joel W. and Ames, Aaron D. (2020) Human Preference-Based Learning for High-dimensional Optimization of Exoskeleton Walking Gaits. In: 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE , Piscataway, NJ, pp. 3423-3430. ISBN 978-1-7281-6212-6. https://resolver.caltech.edu/CaltechAUTHORS:20200526-152132290

[img] PDF - Submitted Version
See Usage Policy.

4068Kb

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20200526-152132290

Abstract

Optimizing lower-body exoskeleton walking gaits for user comfort requires understanding users’ preferences over a high-dimensional gait parameter space. However, existing preference-based learning methods have only explored low-dimensional domains due to computational limitations. To learn user preferences in high dimensions, this work presents LINECOSPAR, a human-in-the-loop preference-based framework that enables optimization over many parameters by iteratively exploring one-dimensional subspaces. Additionally, this work identifies gait attributes that characterize broader preferences across users. In simulations and human trials, we empirically verify that LINECOSPAR is a sample-efficient approach for high-dimensional preference optimization. Our analysis of the experimental data reveals a correspondence between human preferences and objective measures of dynamicity, while also highlighting differences in the utility functions underlying individual users’ gait preferences. This result has implications for exoskeleton gait synthesis, an active field with applications to clinical use and patient rehabilitation.


Item Type:Book Section
Related URLs:
URLURL TypeDescription
https://doi.org/10.1109/IROS45743.2020.9341416DOIArticle
https://ieeexplore.ieee.org/document/9341416PublisherArticle
https://arxiv.org/abs/2003.06495arXivDiscussion Paper
ORCID:
AuthorORCID
Tucker, Maegan0000-0001-7363-6809
Novoseller, Ellen0000-0001-5263-0598
Cheng, Richard0000-0001-8301-9169
Yue, Yisong0000-0001-9127-1989
Ames, Aaron D.0000-0003-0848-3177
Additional Information:© 2020 IEEE. This work was supported by NSF NRI award 1724464, NSF Graduate Research Fellowship No. DGE1745301, the Caltech Big Ideas Fund, and the ZEITLIN Fund. This work was conducted under IRB No. 16-0693. The authors would like to acknowledge the subjects who participated in exoskeleton testing, as well as the entire Wandercraft team that designed Atalante and continues to provide technical support for this project.
Funders:
Funding AgencyGrant Number
NSFIIS-1724464
NSF Graduate Research FellowshipDGE-1745301
Caltech Big Ideas FundUNSPECIFIED
ZEITLINUNSPECIFIED
Record Number:CaltechAUTHORS:20200526-152132290
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20200526-152132290
Official Citation:M. Tucker et al., "Human Preference-Based Learning for High-dimensional Optimization of Exoskeleton Walking Gaits," 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Las Vegas, NV, USA, 2020, pp. 3423-3430, doi: 10.1109/IROS45743.2020.9341416
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:103476
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:26 May 2020 22:27
Last Modified:11 Feb 2021 23:43

Repository Staff Only: item control page