Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.678010
Title: Interacting with prosthetic hands via electromyography signals
Author: Fang, Yinfeng
ISNI:       0000 0004 5369 830X
Awarding Body: University of Portsmouth
Current Institution: University of Portsmouth
Date of Award: 2015
Availability of Full Text:
Access through EThOS:
Access through Institution:
Abstract:
It is a challenge to provide robust electromyographic signals or patterns for prosthetic hand systems. This thesis proposes a comprehensive methodology to address the challenge with respect to surface electromyographic signal acquisition, electrode layouts, electromyographic features and user training strategies. A multi-channel surface electromyography acquisition platform is customised to conduct researches throughout this thesis. First of all, a zig electrode layout is proposed to provide more repeatable electromyographic signals. This electrode layout is instantiated into an electrode sleeve, which is specially presented to fix the electrodes on the forearm and acquire forearm muscular activities. Our experiments prove that zig electrode layout has better electromyographic signal repeatability than conventional parallel electrode layout in different tests. Secondly, this thesis establishes a bridge connecting forearm muscles’ functions and multi-channel electromyographic signals by means of electromyographic map and magnitude-angle feature. The electromyographic map is proposed to explore how channels of electromyographic signals correspond to individual forearm muscles. In order to understand hand motion physiological principles, magnitude-angle feature is presented to identify the most active muscles during hand motions. Thirdly, to enhance patients’ ability in generating intuitive prosthetic control commands, a training strategy based on visual trajectory feedback is proposed. In the training procedure, users are able to adjust themselves according to classifier feedback. This training procedure can significantly improve patients’s ability in generating repeatable electromyographic pattern, no matter the feedback information is able or disable.
Supervisor: Liu, Honghai ; Stevens, Brett ; Ju, Zhaojie Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Thesis
EThOS ID: uk.bl.ethos.678010  DOI: Not available
Keywords: Computing
Share: