Immersive Teleoperation Platform Based on Eye Tracker

摘要

When humans remotely control robots, it is often difficult for them to observe the robot's working environment from the robot's first-person perspective, which invisibly increases the uncertainty in the teleoperation process. Therefore, this paper developed an immersive teleoperation platform based on an eye tracker. The system mainly controlled the rotation of the robot's head joints through the operator's eye movement. The camera on the robot's head can transmit the robot's first-person perspective to the operator, allowing the operator to observe the environment in the direction the robot's head is facing while moving his eyes. This paper designed a mapping algorithm between the horizontal coordinate data of the eyeball and the rotation angle of the robot's head joint, and built an experimental platform based on communication between the upper and lower computers. In addition, in order to avoid the situation where the operator's eyeballs shake or move over a large range instantaneously, which may cause the head joints to shake violently, this paper added an exponentially weighted moving average filter to the control algorithm to denoise the eye movement data. Finally, experiments verified the effectiveness and practicality of the immersive teleoperation platform developed in this paper.