Joint Modeling and Closed-Loop Control of a Robotic Hand Driven by the Tendon-Sheath

Abstract

Robotic dexterous operation is a long-term challenging problem. Although recent advances in robotic hand have successfully realized some complex operations, its high dexterity and high precision control are still challenges for robotic hands to replace the human hand. Here, this letter developed a tendon-sheath transmission-based robotic hand with decoupling joint design and high-precision closed-loop control strategy. Firstly, the designed hand is introduced, and the nonlinear joint transmission model is established. Then, the transfer function of the joint is established based on the dynamic model, and an adaptive controller with neural network compensation is designed to realize the high-precision joint control. The simulation and experiment results show that it is feasible to apply the tendon-sheath transmission method to the robotic hand, and the proposed closed-loop control strategy can significantly improve the joint control accuracy. This research opens new prospects for the novel tendon-sheath driving method applied in the design of robotic hand, which is expected to have a long-term impact for the design and control of dexterous hand.