Surgical Systems

Real-time Path Planning of Dual-Arm Surgical Robots via Reinforcement Learning

The main objective of this project is to develop a reinforcement learning algorithm for the real-time trajectory planning of dual-arm surgical robots using stereo camera. We hypothesize that adopting and modifying an RL model architecture will significantly improve the performance of autonomous robotic resection surgery while using vision-based methods as input data. Furthermore, dual-arm manipulators in comparison with single-arm robot and manual surgery will operate the resection surgery with the least error and in shorter time through the collaboration of both arms.