A Real-Time Robot Motion Generation System Based on Human Gesture

Abstract

When a communication robot conveys some messages to human users, showing some motions matching with the messages is an effective way in human robot interaction. Designing and implementing a set of useful robot motions, however, is a difficult problem. It requires developers to describe instructions for accurately controlling motors embedded in the robot. In this paper, we propose a method for generating robot motions by using human gestures as input. The proposed method captures human gestures in real time by using motion sensors, converting the acquired data to robot motion instructions, and applying them to a physical robot. We developed a prototype system targeted at supporting different motion sensors and robots with lightweight data communication between a control node and a robot. We conducted experiments for controlling the robots by human gestures to verify the effectiveness of the proposed method. It can promote better human robot communication environments with reducing labors in robot motion development.