Enabling Robotic Tele-Microsurgery: High-Resolution RGB-D Streaming Based on Swept Source OCT and Data-Driven Galvanometer Control

This article presents a real-time RGB-D imaging system for robot-assisted remote microsurgery, integrating advanced optics, mechatronics, and dynamic control theory. The system combines a swept-source optical coherence tomography (SS-OCT) probe with a micro RGB camera, exploiting the SS-OCT’s high depth resolution and fast scan rate to generate accurate depth maps co-registered with the camera’s color data. A dual-axis galvanometer scanner, controlled by a fractional-order repetitive control algorithm with a data-driven inverse filter, enables real-time imaging at 30 fps with dynamically adjustable scan areas. Experimental evaluations highlight the system’s exceptional depth-sensing capabilities, including sub-millimeter resolution, precise depth mapping, and smooth dynamic scanning. In addition, the system enhances visual guidance during robotic microsurgery, as validated in teleoperated vessel ligation on a phantom and a cadaveric porcine model.