Title :
14. Grasping-posture classification using myoelectric signal on hand pre-shaping for natural control of myoelectric hand
Author :
Suzuki, Daiki ; Yamanoi, Yusuke ; Yamada, Hiroshi ; Wakita, Ko ; Kato, Ryu ; Yokoi, Hiroshi
Author_Institution :
Dept. of Inf. & Eng., Univ. of Electro-Commun., Tokyo, Japan
Abstract :
A stationary grasping posture is classified in the control method of an electromyogram prosthetic hand. This grasping posture is static, such as an open hand posture, and one in which the operator of an electromyogram prosthetic hand intentionally continues muscular contraction. In classifying the stationary grasping posture, a movement delay of the robot hand occurs, which feels unnaturally to the operator. To solve these problems, authors propose a method that predicts a grasping posture using the surface electromyogram (sEMG) of low muscle contraction power in hand pre-shaping. In this paper, our research on the performance of grasping posture classification using sEMG for naturally reaching for and grasping an object is presented. Experimental results demonstrate that when the sEMG amplitude peaks in hand pre-shaping, it is useful in classifying the grasping posture.
Keywords :
dexterous manipulators; electromyography; medical signal processing; position control; prosthetics; signal classification; electromyogram prosthetic hand; grasping-posture classification; muscular contraction; myoelectric hand control; myoelectric signal; robot hand movement delay; sEMG; stationary grasping posture; surface electromyogram; Electromyography; Feature extraction; Grasping; Muscles; Prosthetic hand; Thumb; Electromyogram classification; artificial neural networks; human-robot interface; medical robotics; prosthetic hand;
Conference_Titel :
Technologies for Practical Robot Applications (TePRA), 2015 IEEE International Conference on
Conference_Location :
Woburn, MA
DOI :
10.1109/TePRA.2015.7219657
