Johns Hopkins and Stanford Teach Da Vinci Robot to Perform Surgical Tasks via Imitation Learning

Jul 19, 2024, 05:31 AM

Researchers have made significant progress in robotic surgery by teaching the Da Vinci robot to perform surgical manipulation tasks through imitation learning. The robot successfully executed three fundamental surgical tasks: tissue manipulation, needle handling, and knot-tying. This breakthrough was achieved despite the unique challenges posed by the Da Vinci system's inconsistent forward kinematics. The project, named Surgical Robot Transformer (SRT), is a collaboration between Johns Hopkins and Stanford, with contributions from Ji Woong (Brian) Kim, Tony Z. Zhao, Samuel Schmidgall, and Anton Deguet, and advisors Marin Kobilarov, Chelsea Finn, and Axel Krieger. The research aims to transform surgery through robotics with end-to-end imitation learning and has been highlighted by Vicarious Surgical Inc.