Towards an Autonomous Robot-Assistant for Laparoscopy Using Exteroceptive Sensors: Feasibility Study and Implementation

Abstract

A new autonomous camera-holder robotic system for Laparoscopic Surgery (LS) is presented in this paper. In the proposed system, a 7-DoF collaborative robot holds and moves the laparoscope, or surgical camera, so that it permanently points towards the instruments' tips. A motion capture (MoCap) system, tracking online the surgical instruments, provides the robot's controller with instrument's motion data. Unlike the existing systems, this paper proposes an autonomous robot-assistant, freeing up surgeons from the laparoscope motion control, fully responding to the surgical community requirements. Moreover, a fully compliance control law is implemented. In task-space, it is suitable for smoothing robot movements and reducing the contact forces generated at the trocar. Null-space is also exploited through a compliance control law restricting the elbow's robot motion into a desired range, defined to avoid undesired collisions with the rest of the medical equipment. Robot Operating System (ROS) framework has been used to establish the communication between the robot and MoCap system, using the UDP protocol for data exchange. Training tasks, such as suturing training, have been performed by surgeons to validate the usefulness of the proposed platform, proving that a collaborative robot can autonomously comply with camera-holder assistant tasks.