A New Motion Control Method for Bipedal Robot Based on Noncontact and Nonattached Human Motion Sensing Technique

Abstract

We develop a new motion control technique for the stepping motion of a bipedal robot; this technique is based on an effective noncontact and nonattached technique that is used for the detection of human stepping motion and hand moving motion. Our technique involves the generation of an electrostatic current by a change in the potential of the human body during stepping or walking motions. The waveform of the induced current contains cadence components of both feet. The knee joint angle, which is estimated from the induced current waveform, is used as a motion control signal. For example, the motion of a commercially available bipedal robot is controlled in real time by the current generated in a human body while stepping and hand moving.