Multitask Learning for Simultaneous Gesture and Force Level Recognition Toward Prosthetic Hand Interaction

In the field of intelligent prosthetics (IPs), the establishment of a natural interaction between prosthetic hands and amputees holds significant importance in restoring hand functionality, enhancing quality of life, and facilitating daily activities and social engagement. Prior investigations on surface electromyographic (sEMG) signals-controlled IPs have predominantly concentrated on gesture recognition, frequently neglecting the equally significant dimension of force level. This study proposes a control strategy integrating a multitask learning (MTL) model to achieve synchronized recognition of gestures and force levels. The MTL model, incorporating shared convolutional blocks, self-attention, and multihead attention layers, enhances prosthetic hand control for seamless user-device interaction. This study consistently showcases exceptional proficiency in recognizing gestures and force levels by conducting meticulous experimentation and validating the findings using datasets from diverse participants. Comparative assessments endorse the superiority of the MTL approach, particularly in real-time testing scenarios. The findings highlight the potential of this innovative myoelectric control strategy, empowering prosthetic users for prompt, precise, and intuitive responses, significantly augmenting their autonomy and quality of life.