Robust Shape-Distortion Prediction for a Wire-Driven Hyper-Redundant Manipulator With Engaging Rolling Joints

Abstract

In Minimally Invasive Surgery (MIS), wire-driven hyper-redundant manipulators are widely used because of their flexibility, dexterity, and miniaturization. However, due to their limited structural stiffness, wire-elongation and unwanted distortion in the shape may occur during payload handling. For example, the surgical payload at the end-effector can distort an endoscope and cause positional deviation of the end-effector. In this paper, an endoscopic manipulator with robust Engaging Rolling Joints (ERJs) is proposed, where each ERJ is deployed with a pair of Auxiliary Super-elastic Ribbons (ASRs). Based on the design, we also developed a prediction model for shape-distortion, which is able to numerically predict the shape of the manipulator under payload applied to the end-effector. Experiments were conducted to verify that the ERJ and the ASR can better reject unwanted disturbances in shape prediction. The test results showed that the predicted distorted shape is in good agreement with the actual shape; the average prediction errors for the distal angle and position deflections were measured at 8.63% and 7.71%, respectively. We believe that the presented technique will help to improve the performance of the rolling joint manipulator by compensating the predicted shape-distortion.