CaltechAUTHORS
  A Caltech Library Service

First steps toward translating robotic walking to prostheses: a nonlinear optimization based control approach

Zhao, Huihua and Horn, Jonathan and Reher, Jacob and Paredes, Victor and Ames, Aaron D. (2017) First steps toward translating robotic walking to prostheses: a nonlinear optimization based control approach. Autonomous Robots, 41 (3). pp. 725-742. ISSN 0929-5593. https://resolver.caltech.edu/CaltechAUTHORS:20190722-151625943

[img] PDF - Submitted Version
See Usage Policy.

4Mb
[img] Video (MPEG) - Supplemental Material
See Usage Policy.

8Mb

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20190722-151625943

Abstract

This paper presents the first steps toward successfully translating nonlinear real-time optimization based controllers from bipedal walking robots to a self-contained powered transfemoral prosthesis: AMPRO, with the goal of improving both the tracking performance and the energy efficiency of prostheses control. To achieve this goal, a novel optimization-based optimal control strategy combining control Lyapunov function based quadratic programs with impedance control is proposed. This optimization-based optimal controller is first verified on a human-like bipedal robot platform, AMBER. The results indicate improved (compared to variable impedance control) tracking performance, stability and robustness to unknown disturbances. To translate this complete methodology to a prosthetic device with an amputee, we begin by collecting reference locomotion data from a healthy subject via inertial measurement units (IMUs). This data forms the basis for an optimization problem that generates virtual constraints, i.e., parameterized trajectories, specifically for the amputee . A online optimization based controller is utilized to optimally track the resulting desired trajectories. An autonomous, state based parameterization of the trajectories is implemented through a combination of on-board sensing coupled with IMU data, thereby linking the gait progression with the actions of the user. Importantly, the proposed control law displays remarkable tracking and improved energy efficiency, outperforming PD and impedance control strategies. This is demonstrated experimentally on the prosthesis AMPRO through the implementation of a holistic sensing, algorithm and control framework, resulting in dynamic and stable prosthetic walking with a transfemoral amputee.


Item Type:Article
Related URLs:
URLURL TypeDescription
https://doi.org/10.1007/s10514-016-9565-1DOIArticle
https://link.springer.com/article/10.1007/s10514-016-9565-1PublisherArticle
https://download-tls-cdn.edge-cdn.net/videodb/5501/videodb_5501_53500_8283088_16x9_hq.mp4PublisherSupplemental Video
ORCID:
AuthorORCID
Ames, Aaron D.0000-0003-0848-3177
Additional Information:© 2016 Springer Science+Business Media New York. Received 30 March 2015; accepted 06 April 2016; first online 21 April 2016. This research is supported under: NSF CAREER Award CN-S-0953823 and Texas Emerging Technology Fund 11062013. This research has approval from the Institutional Review Board from Texas A&M University with IRB2014-0382F for testing with human subjects.
Funders:
Funding AgencyGrant Number
NSFCNS-0953823
Texas Emerging Technology Fund11062013
Subject Keywords:Prostheses, Optimization-based Control, Locomotion, Hybrid Zero Dynamics (HZD), Hybrid Systems, AMPRO
Issue or Number:3
Record Number:CaltechAUTHORS:20190722-151625943
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20190722-151625943
Official Citation:Zhao, H., Horn, J., Reher, J. et al. Auton Robot (2017) 41: 725. https://doi.org/10.1007/s10514-016-9565-1.
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:97330
Collection:CaltechAUTHORS
Deposited By: Melissa Ray
Deposited On:22 Jul 2019 23:30
Last Modified:03 Oct 2019 21:30

Repository Staff Only: item control page