Resumo

SANDOVAL RODRIGUEZ, Camilo; VILLAMIZAR MEJIA, Rodolfo; DELGADO VELOSA, Eusebio  e  CORDERO CARDOZO, Andrés. HAND MOVEMENTS SPEED ESTIMATION BY MEANS OF ARTIFICIAL NEURAL NETWORKS AND ELECTROMIOGRAPHICAL MEASUREMENTS. Rev. ing. biomed. [online]. 2010, vol.4, n.8, pp.41-56. ISSN 1909-9762.

In this paper an algorithm for estimating the speed of the basic hand movements using artificial neural networks based on recorded electromyographic activity at the forearm is presented. To implement this algorithm it was necessary to adapt a model for measuring the speed, using digital image processing, which presented a low error rate measurement. Likewise, for speed estimation, a NARX network (network nonlinear autoregressive with exogenous inputs) was chosen after comparing different dynamic neural network topologies. The results shown demonstrated a suitable approach to the estimation of speed, which serves as a comparison to the different methodologies used to obtain the velocity profiles.

Palavras-chave : Artificial neural networks (ANN); Arm prosthesis; Electromyography (EMG); Speed estimation.

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