The Proceedings of JSME annual Conference on Robotics and Mechatronics (Robomec)
Online ISSN : 2424-3124
2023
Session ID : 2P2-B22
Conference information

Robotic Fundamentals of Laparoscopic Surgery with Constrained Imitation Learning
*Kento KAWAHARAZUKAKei OKADAMasayuki INABA
Author information
CONFERENCE PROCEEDINGS RESTRICTED ACCESS

Details
Abstract

Toward robotic laparoscopic surgery, we develop a robot that can perform a peg transfer task in Fundamentals of Laparoscopic Surgery (FLS) using imitation learning. There are two problems: (1) the forceps must be operated using laparoscopic ports as fulcrums, and (2) the depth information cannot be recognized due to a monocular endoscope. We solve (1) by generating arm motions with laparoscopic port constraints, and (2) by collecting data with constraints generated from one exemplary motion for imitation learning. We implemented these methods with two arm robots and confirmed the effectiveness.

Content from these authors
© 2023 The Japan Society of Mechanical Engineers
Previous article Next article
feedback
Top