Learning Soft Tissue Dynamics in Image Space for Automated Bimanual Tissue Manipulation with Surgical Robots

In this paper, reinforcement learning and learning from demonstration in vision strategies are proposed to automate the soft tissue manipulation task with surgical robots. A soft tissue manipulation simulation is designed to compare the performance of the algorithms, and it is found that the learning from demonstration algorithm could boost the learning policy with initialization of dynamics with given demonstrations. Furthermore, the learning from demonstration algorithm is implemented on a Raven IV surgical robotic system to show feasibility.