Proceedings on Automation in Medical Engineering
Vol. 1 No. 1 (2020): Proc AUTOMED

Rehabilitation technology, ID 023

Human-in-the-loop optimization of a multi-joint wearable robot for movement assistance

Main Article Content

Florian Leander Haufe (Eidgenössische Technische Hochschule Zürich), Peter Wolf (Sensory-Motor Systems Lab, Institute of Robotics and Intelligent Systems, ETH Zurich, Zurich, Switzerland), Robert Riener (Sensory-Motor Systems Lab, Institute of Robotics and Intelligent Systems, ETH Zurich, Zurich, Switzerland and Spinal Cord Injury Center, University HospitalBalgrist, University of Zurich, Zurich, Switzerland)

Copyright (c) 2020 Proceedings on Automation in Medical Engineering

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Abstract

Soft, lightweight wearable robots can assist individuals with a lower limb weakness. However, it its challenging to identify control parameters that optimize the robotic assistance for a specific user and task. In this context, “human-in-the-loop” techniques have been suggested to automate this optimization process. Here, we show that the assistance from the Myosuit, a multi-joint wearable robot, can be optimized using a Covariance-Matrix Adaptation Evolution Strategy. The optimization resulted in a reduction of a participant’s energy expenditure that was nearly twice as large (18 % vs. 9.6 %) than when using hand-selected parameters, thereby motivating further work with impaired participants.

Article Details