Open access
Date
2015-04Type
- Journal Article
Abstract
This paper presents a motion intention estimation algorithm that is based on the recordings of joint torques, joint positions, electromyography, eye tracking and contextual information. It is intended to be used to support a virtual-reality-based robotic arm rehabilitation training. The algorithm first detects the onset of a reachingmotion using joint torques and electromyography. It thenpredicts the motion target using a combination of eyetracking and context, and activates robotic assistance to-ward the target. The algorithm was first validated offlinewith 12 healthy subjects, then in a real-time robot controlsetting with 3 healthy subjects. In offline crossvalidation,onset was detected using torques and electromyography116msprior to detectable changes in joint positions. Fur-thermore, it was possible to successfully predict a majorityof motion targets, with the accuracy increasing over thecourse of the motion. Results were slightly worse in onlinevalidation, but nonetheless show great potential for real-time use with stroke patients. Show more
Permanent link
https://doi.org/10.3929/ethz-b-000100242Publication status
publishedExternal links
Journal / series
at - AutomatisierungstechnikVolume
Pages / Article No.
Publisher
De GruyterSubject
Intention detection; Physical human-robot interaction; Rehabilitation robotics; Sensor fusion; Intentionsdetektion; Mensch-Maschine Interaktion; Rehabilitationsrobotik; SensorfusionOrganisational unit
03654 - Riener, Robert / Riener, Robert
Notes
Published in the Special Issue "Zum 80. Geburtstag von Günther Schmidt / Martin Buss" and also published in "Bestimmung des Bewegungsbeginns und Bewegungsziels bei der roboterunterstützten Armrehabilitation".
It was possible to publish this article open access thanks to a Swiss National Licence with the publisher.More
Show all metadata