RCM Constrained Task Space Trajectory Generation for Autonomous Robotic Surgeries

• Published in: 2024 IEEE International Conference on Cyborg and Bionic Systems (CBS) pp. 121 - 125
• Main Authors: Andal Amirthavarshini, G, Thondiyath, Asokan
• Format: Conference Proceeding
• Language: English
• Published: IEEE 20.11.2024
• Subjects: Aerospace electronics; Kinematics; Machine vision; Minimally invasive surgery; Motion control; Point cloud compression; Real-time systems; Trajectory; Trajectory planning; Visualization
• DOI: 10.1109/CBS61689.2024.10860403

As the field of Surgical Robotics progresses towards achieving autonomy, there has been a noticeable increase in the utilization of generic serial manipulators for these applications. For Minimally Invasive Surgeries, it is required to impose a motion constraint with respect to the incision point on the patient. This brings up a complexity in their motion control complying to the constraint. With the requirements of having the least violation of this Remote Centre of Motion (RCM) constraint and accuracy of the end-effector, several control strategies have been put forth and evaluated using different manipulators. However, trajectory generation in the surgical environment remains a challenge to be addressed for autonomous robotic surgeries that are minimally invasive. In this paper, we propose an RCM constrained task space trajectory generation algorithm for autonomous MIS applications. The algorithm, independent of the manipulator kinematics, utilizes unevenly distributed visual point cloud data in task space for path generation and follows a uniform displacement per unit time step approach for trajectory generation and tracking. The proposed algorithm is implemented on a KUKA LBR iiwa robot using simulated visual data that effectively illustrates the surgical scenario. The key validation metrics, RCM violation error and tool-tip tracking error are well within the acceptable range for the test case implemented.