Biomimetic Tactile Data Driven Closed-loop Control of Myoelectric Prosthetic Hand

摘要

The dexterous prosthetic hand is capable of grasping many objects with different size and shape such as precision, power, and lateral grips. However, most of these prosthetic hands haven't the ability of the tactile. Recent advances in EMG pattern recognition techniques are useful for creating hand prostheses with more intuitive control, but amputees still face the problem of lost tactile feedback. In this paper, a biomimetic tactile data driven closed-loop control of myoelectric prosthetic hand is proposed. The problem addressed in this paper is myoelectric prosthetic hand how to consistently grasp delicate objects without causing damage and dropping. The peculiarity of the proposed controller will offer the hand prostheses the ability to make rapid and accurate adjustments the grasping force based on the tactile feedback, which is similar to our very own reflex pathway. The experiments results show that the proposed controller can significantly improve the grasping function.