Analysis Method of Skill Acquisition Process using Electromyogram for Autonomous Assistive Devices

Abstract

Listen

Translate article

For development of an adaptable assist device, it requires to quantify users' skillfulness. In this paper, we propose and verify a method to index skillfulness on arm reaching tasks for motor learning. Process analysis of motor learning gives us substantial information on functional adaptability of human motion activities. Hence, we define skillfulness as reproducibility and motion accuracy of motion patterns. In our experiment, a computer monitor displays a moving cursor (e.g., cyclical and random movements), and seven subjects track the cursor with a joystick. We measure electromyogram (EMG) and cursor trajectories on skill acquisition process of the subjects. Then, we evaluate their skillfulness with the internal conditions: the reproducibility is calculated from the variance value of EMG patterns and the motion accuracy is calculated from the trajectory error. Eventually, the results indicated variety of skill acquisition processes, and proved that the proposed method quantitatively-analyzed their skillfulness.