Assist Control for Master-Slave System with Operator’s Eye Gaze Tracking

In recent years, because of many natural disasters, the demand for robots that can perform various tasks instead of humans has increased. In irregular works under a complicated environment, high ability for recognition and judgment is often necessary. A master-slave system which can be controlled with intuitive operations are suitable for such works. However, a master system and a slave robot generally have different structures, and there is a communication delay between them. Therefore, it is difficult for an operator to perform complicated operations in conventional master-slave systems. In this paper, we proposed a assist system that predicts the intention of an operator using an eye tracking device and corrects the motion of a slave robot by the prediction. As a result of experiments, it was verified that the proposed method is useful to predict operator’s intention and assist their motions.