School of Computing

Power independent EMG based gesture recognition for robotics

Ling Li, D. Looney, C. Park, N.U. Rehman, and D.P. Mandic

In Engineering in Medicine and Biology Society, EMBC, 2011 Annual International Conference of the IEEE, pages 182-196. IEEE EMBS, August 2011.

Abstract

A robot control system using four different gestures from an arm is presented. This is achieved based on surface Electromyograph (EMG) measurements of groups of arm muscles. The cross-information is preserved through a simultaneous processing of EMG channels using a recent multivariate extension of Empirical Mode Decomposition (EMD). Next, phase synchrony measures are employed to make the system robust to different power levels due to electrode placements and impedances. The multiple pairwise muscle synchronies are used as features of a discrete gesture space comprising four gestures (flexion, extension, pronation, supination). Simulations on real-time robot control illustrate the enhanced accuracy and robustness of the proposed methodology.

Download publication 573 kbytes (PDF)

Bibtex Record

@inproceedings{3182,
author = {Ling Li and D. Looney and C. Park and N.U. Rehman and D.P. Mandic},
title = {Power Independent {EMG} Based Gesture Recognition for Robotics},
month = {August},
year = {2011},
pages = {182-196},
keywords = {determinacy analysis, Craig interpolants},
note = {},
doi = {},
url = {http://www.cs.kent.ac.uk/pubs/2011/3182},
    publication_type = {inproceedings},
    submission_id = {25367_1320791161},
    ISBN = {9781424441228},
    publisher = {IEEE EMBS},
    booktitle = {Engineering in Medicine and Biology Society, EMBC, 2011 Annual International Conference of the IEEE},
}

School of Computing, University of Kent, Canterbury, Kent, CT2 7NF

Enquiries: +44 (0)1227 824180 or contact us.

Last Updated: 21/03/2014