Motion Recognition Based Cooperation between Human Operating Robot and Autonomous Assistant Robot.

Abstract

The purpose of this research is to make multiple robots perform the task according to the intention of an operator. As it is difficult for a human to operate all robots simultaneously, we propose the cooperation style in which the human operates one robot and the other autonomous robots assist it. In order to assist the human operating robot autonomously, it is necessary for the assistant robots to recognize the motion of the human operating robot in real time. For this purpose, we developed the motion recognition based cooperation system between the human operating robot and the autonomous assistant robot. In this paper, we proposed the motion recognition mechanism referring to the human cognition model. We implemented the motion symbolization method using the “Characteristic Matrix” and the motion recognition method using the “Feature Pattern” . As the assistance mechanism by the autonomous robot, we implemented “Task-Operation Model” to describe the motion and “Event Driven Method” to manage the execution. We indicated the effectiveness of these methods by the experiment using two hexapod robots which lifted up a box in cooperation.