Mobile robot operation by gesture recognition using continuous human motion

Abstract

The aim of this study is to develop the mobile robot handling system using human gesture. The gestures we use this system are some continuous motions of right arm. The sequences of 3D position data of arm motion are acquired by Kinect sensor system, and features for the gesture recognition are DC component and difference of phases estimated from the Fourier transform of arm motions. The input gesture is determined by the similarity of features considering the learning data. In the experiments, all five gestures to handle the mobile robot can be recognized correctly.