Online Trajectory Planning and Force Control for Automation of Surgical Tasks

Abstract

Automation of surgical tasks is expected to improve the quality of surgery. In this paper, we address two issues that must be resolved for automation of robotic surgery: online trajectory planning and force control under dynamic conditions. By leveraging demonstrations under various conditions, we model the conditional distribution of the trajectories given the task condition. This scheme enables generalization of the trajectories of spatial motion and contact force to new conditions in real time. In addition, we propose a force tracking controller that robustly and stably tracks the planned profile of the contact force by learning the spatial motion and contact force simultaneously. The proposed scheme was tested with bimanual tasks emulating surgical tasks that require online trajectory planning and force tracking control, such as tying knots and cutting soft tissues. Experimental results show that the proposed scheme enables planning of the task trajectory under dynamic conditions in real time. In addition, the performance of the force control schemes was verified in the experiments. Note to Practitioners —This paper addresses the problem of motion planning and control for automation of surgical tasks. In surgical tasks, it is necessary to manipulate objects under conditions where positions or shapes of objects often change during the task. Thus, trajectories for surgical tasks need to be planned and updated according to the change in the conditions in real time. In this paper, we propose a framework for learning both spatial motion and force profile from human experts. The proposed system can plan and update task trajectories in real time and robustly control the contact force under dynamic conditions. On the other hand, generalization of trajectories is limited to the conditions, which are close to the conditions where the demonstrations were performed. In the future work, we will investigate reinforcement learning approaches in order to enable autonomous improvement of the performance.