TY - GEN
T1 - Thumb-tip force estimation from sEMG and a musculoskeletal model for real-time finger prosthesis
AU - Park, Won Il
AU - Kwon, Sun Cheol
AU - Lee, Hae Dong
AU - Kim, Jung
N1 - Copyright:
Copyright 2009 Elsevier B.V., All rights reserved.
PY - 2009
Y1 - 2009
N2 - Due to the difficulties in measurement of muscle activities and the complex musculoskeletal structure, estimations of the thumb-tip force in real time have been a challenge for controlling artificial prosthesis naturally. This study describes an isometric thumb-tip force estimation technique based on phenomenological muscle model named Hill's model. The surface electromyogram (sEMG) signals of the muscles near surface were measured and converted to muscle activation information. The activations of deep muscles were inferred from the ratios of muscle activations from earlier study. The muscle length of each contributed muscle was obtained by using motion capture system and musculoskeletal modeling software packages. Once muscle forces were calculated, thumb-tip force was estimated based on mapping model from the muscle force to thumb-tip force. The proposed method was evaluated in comparisons with an artificial neural network (ANN) under four different thumb configurations to investigate the potential for estimations under conditions in which the thumb configuration changes. The results seem to be promising and the proposed method could be applied to predict finger-tip forces from non-invasive neurosignals with a real-time prosthesis control system.
AB - Due to the difficulties in measurement of muscle activities and the complex musculoskeletal structure, estimations of the thumb-tip force in real time have been a challenge for controlling artificial prosthesis naturally. This study describes an isometric thumb-tip force estimation technique based on phenomenological muscle model named Hill's model. The surface electromyogram (sEMG) signals of the muscles near surface were measured and converted to muscle activation information. The activations of deep muscles were inferred from the ratios of muscle activations from earlier study. The muscle length of each contributed muscle was obtained by using motion capture system and musculoskeletal modeling software packages. Once muscle forces were calculated, thumb-tip force was estimated based on mapping model from the muscle force to thumb-tip force. The proposed method was evaluated in comparisons with an artificial neural network (ANN) under four different thumb configurations to investigate the potential for estimations under conditions in which the thumb configuration changes. The results seem to be promising and the proposed method could be applied to predict finger-tip forces from non-invasive neurosignals with a real-time prosthesis control system.
UR - http://www.scopus.com/inward/record.url?scp=70449440798&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=70449440798&partnerID=8YFLogxK
U2 - 10.1109/ICORR.2009.5209518
DO - 10.1109/ICORR.2009.5209518
M3 - Conference contribution
AN - SCOPUS:70449440798
SN - 9781424437894
T3 - 2009 IEEE International Conference on Rehabilitation Robotics, ICORR 2009
SP - 305
EP - 310
BT - 2009 IEEE International Conference on Rehabilitation Robotics, ICORR 2009
T2 - 2009 IEEE International Conference on Rehabilitation Robotics, ICORR 2009
Y2 - 23 June 2009 through 26 June 2009
ER -