TY - GEN
T1 - Hand Prosthesis Control using Electromyographic Signal Trained Neural Network for Live Gesture Classification
AU - Minshew, Zackary
AU - Alba-Flores, Dr Rocio
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/4
Y1 - 2019/4
N2 - - This work addresses the use of a user-friendly app to implement the surface Electromyographic (sEMG) sensors from the Myo Armband by Thalmic Labs and a pattern recognition neural network to control a robotic hand. The Myo Armband consists of eight electrodes that are placed on the forearm of the subject that performs wrist and finger motions. The ANN was trained to recognize five hand gestures, named Fist, Rest, Spread, Wave In, and Wave Out. The output of the ANN is then used to control a dexterous robotic hand, 6 DOF, that was 3D printed using ABS plastic material. The robotic hand was able to perform in real time the same hand gestures that the subject performed. The pattern recognition system was able to classify the hand motions with an accuracy of 81.4%. Presented in this document is a simple project design that demonstrates how current affordable technologies could be used in applications such as human-robot interfaces, and in the rehabilitation and solutions for amputees.
AB - - This work addresses the use of a user-friendly app to implement the surface Electromyographic (sEMG) sensors from the Myo Armband by Thalmic Labs and a pattern recognition neural network to control a robotic hand. The Myo Armband consists of eight electrodes that are placed on the forearm of the subject that performs wrist and finger motions. The ANN was trained to recognize five hand gestures, named Fist, Rest, Spread, Wave In, and Wave Out. The output of the ANN is then used to control a dexterous robotic hand, 6 DOF, that was 3D printed using ABS plastic material. The robotic hand was able to perform in real time the same hand gestures that the subject performed. The pattern recognition system was able to classify the hand motions with an accuracy of 81.4%. Presented in this document is a simple project design that demonstrates how current affordable technologies could be used in applications such as human-robot interfaces, and in the rehabilitation and solutions for amputees.
KW - Artificial Neural Network
KW - Myo armband
KW - Pattern Recognition
KW - prosthetic control
KW - sEMG
UR - http://www.scopus.com/inward/record.url?scp=85082398639&partnerID=8YFLogxK
U2 - 10.1109/SoutheastCon42311.2019.9020485
DO - 10.1109/SoutheastCon42311.2019.9020485
M3 - Conference article
AN - SCOPUS:85082398639
T3 - Conference Proceedings - IEEE SOUTHEASTCON
BT - 2019 IEEE SoutheastCon, SoutheastCon 2019
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2019 IEEE SoutheastCon, SoutheastCon 2019
Y2 - 11 April 2019 through 14 April 2019
ER -