Resumo

QUINAYAS-BURGOS, César Augusto  e  GAVIRIA-LOPEZ, Carlos Alberto. Movement Intention Detection System for Myoelectric Control of a Prosthetic Robotic Hand. Ing. Univ. [online]. 2015, vol.19, n.1, pp.27-50. ISSN 0123-2126.  https://doi.org/10.11144/Javeriana.iyu19-1.siim.

This paper presents an embedded system that detects in real time the movement intention to control a prosthetic hand. This work shows that using temporal characteristics of simple calculation can provide subsets of feature vectors discernible enough as to use simple pattern classifiers. Thus, this paper proposes a classifier, which is based on the minimum distance from the centroid of the groups characterizing the movements to identify, by modifying the known algorithm K-nearest neighbors. Movement intention classification results obtained from the developed system are shown: using the percentage of success as an effectiveness measurement, by conducting tests over three persons with healthy muscles. The experimental results show that this system can be used effectively for the control of execution of four motor primitives on a prosthetic robotic hand.

Palavras-chave : electromyography (EMG); pattern recognition; K-nearest neighbors; prosthetic robotic hand.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )