Soft robot system for heart surgery

摘要

A soft robot system for heart tracking is designed, which can take advantage of the deformation ability of the soft robot to track the target point on the heart surface during heart surgery. In contrast to rigid arm, the shape of soft arm will be affected by gravity. To solve this problem, the force deformation model of soft arms is proposed, which is based on Cosserat beam theory. Through the camera at the tip of the soft arm, the real-time three-dimensional position of the target point on heart surface can be obtained. Then, an adaptive heart motion prediction algorithm is pro-posed, which is based on auto-regression model and it can predict the heartbeat movement in the future. The para-meters of model were updated in real time according to the measured data, so the model can adapt to the change of heartbeat amplitude and frequency. Finally, a model predictive controller is designed to calculate the control quantity of the soft robot. The experimental results show that the algorithm can reduce the tracking error and adapt to the change of heartbeat amplitude and frequency.