Stereoscopic low-latency vision system via ethernet network for Humanoid Teleoperation

Despite the recent rise of artificial intelligence, the automatic operation of the robot in an unstructured environment is still insufficient to perform complicated and detailed tasks and requires remote operation by a human. The most important thing to improve the efficiency of such Teleoperation is the vision system that enables the human operator to recognize the surrounding situation of the robot. While constructing the system, immersive experience for the operator was considered. Low latency and curved plane rendering are proposed to provide an immersive experience for operators. Moreover, considering the operator’s exhaustion during the teleoperation that comes from Virtual Reality sickness, the reduction of Virtual Reality sickness is also suggested. Finally, to check the entire system’s visual feedback latency, display-to-display latency is quantitatively measured. This paper presents an implementation of a vision system for the Teleoperation of humanoid robots through Virtual Reality.