EXPERIMENTAL CONSIDERATION ON THE FACTORS WHICH CAUSES VARIATION IN FITTING SURFACE EMG INTERFACE

Abstract

Electricalmyography or myoelectric signal is used to voluntarily control current commercial electric prosthetic modules in the upper limb prosthetics. One of the technological questions that are rarely discussed is the assessment of fitting or tuning the EMG sensors to gain suitable and robust control signal. Such methods are mostly neglected by laboratory researchers because workable signals can be obtained through trial and error for short-period experiments, and for clinical use, humans are more adaptable then machines. However, the reliability of the interface is a key feature for promising everyday use of a device, and therefore, an assessment strategy and a design method for assembling a durable myoelectrically controlled prosthetic arm is expected. Furthermore, quantitative data is advisable for product design which requires robustness in long life use. In this paper, we applied a Quality Engineering technique [1] to investigate the factors in installing EMG sensors for generating activation (ON/OFF) control signal. Eight influential factors on fitting surface EMG electrodes for prosthetic hand control were selected based on heuristics and De Luca’s relation diagram [2], and a multifactor experiment was conducted as a pilot test on a single ablebodied subject.