Real-time EMG-based Human Machine Interface Using Dynamic Hand Gestures Conference Paper uri icon

abstract

  • 2017 American Automatic Control Council (AACC). This study proposes a myoelectric Human Machine Interface (HMI) to control a 6-DOF robotic manipulator with a 1-DOF gripper. Previous study has shown that using dynamic gestures such as 'snapping fingers' is more reliable in the limb position changes than using static gestures such as 'closed hand'. This work utilizes dynamic gestures and additionally infers muscle forces from the EMG signals to activate/inactivate a myoelectric HMI system. In order to estimate the performance of the myoelectric interface, real-time classification accuracy, path efficiency, and time-related measures are introduced. For comparison purposes, the performance of a GUI button-based jog interface was also measured. The average real-time classification accuracy of the myoelectric interface is approximately 95%. The path efficiency of the myoelectric interface also appears to be similar to that of the jog interface reflecting the utility of this approach for HMI applications in robotics.

name of conference

  • 2017 American Control Conference (ACC)

published proceedings

  • 2017 AMERICAN CONTROL CONFERENCE (ACC)

author list (cited authors)

  • Shin, S., Tafreshi, R., & Langari, R.

citation count

  • 5

complete list of authors

  • Shin, Sungtae||Tafreshi, Reza||Langari, Reza

publication date

  • January 2017