Abstract
<title>Abstract</title> <p>In robot-assisted telesurgery, stereo visual feedback and low end-to-end latency are critical for depth perception, hand-eye coordination, and safe remote manipulation. Visual feedback is a crucial component of teleoperated robotic systems, especially for tasks that require accurate hand-eye coordination and depth perception. In immersive teleoperation, stereo vision can improve the operator’s understanding of the remote workspace; however, visual feedback delay must be carefully characterized because it can affect task performance and user comfort. This paper presents the development and latency characterization of a distributed stereo visual feedback pipeline for virtual reality (VR)-based robotic teleoperation. A stereo camera captures left and right images at the remote side, and the video stream is transmitted through an intermediate computer to a master-side workstation. The received stereo images are then processed in Unity and displayed in a VR headset. The system latency is evaluated by separating the visual feedback path into network transmission, Unity processing, and VR headset rendering. The measured network delay is 5.59 ± 1.04 ms while the total camera-to-display visual feedback delay is 77.95 ± 10.17 ms. These results show that the main latency is introduced by the visualization and display stages rather than network communication alone. The proposed pipeline provides a foundation for future studies on stereo perception, visual delay, and operator performance in VR-based robotic teleoperation.</p>