A pilot study of two degrees of freedom dynamic EMG-force at the wrist using a minimum number of electrodes

Myoelectric upper-limb prostheses are generally limited to control of one degree of freedom (DoF) at a time when proportionally actuating two upper-limb devices (e.g., hand-wrist). Mode switching is then used between the two devices. Users would greatly prefer an ability to control both DoFs simultaneously, independently and proportionally (SIP control). Researchers have previously studied 2-DoF SIP control via EMG-force tasks in able-bodied subjects (as well as limb-absent), showing feasibility using high density EMG electrode systems. These high-density systems are not practical for fielded prosthetic devices, thus recent research has studied 2-DoF EMG-force using a small number of commercial electrodes. We previously reported 2-DoF EMG-force results at the wrist using a minimum number of electrodes and static contractions - constant-posture, slowly force varying. Herein, we report pilot results from five able-bodied subjects with the experimental conditions expanded to constant-posture force-varying (dynamic) conditions. We found that the minimum number of electrodes for 2-DoF EMG-force at the wrist was four, when selected using backward stepwise selection from a pool of 16 electrodes. The average RMS errors ranged from 6.0-16.3% maximum voluntary contraction, depending on the attempted motions and the training-testing strategy used. This technique is promising. Evaluation in a larger sample and by limb-absent subjects in a prosthesis control task is suggested as necessary future work.