Closed-Loop Force Control by Biorealistic Hand Prosthesis With Visual and Tactile Sensory Feedback

Area of Science:

• * Biomedical Engineering
• * Neuroprosthetics
• * Human-Computer Interaction

Background:

• * Developing advanced prosthetic hands requires improved human-prosthetic interaction and neural compatibility.
• * Effective grasp force control is crucial for functional prosthetic use.
• * Understanding the role of sensory feedback is key to optimizing prosthetic control.

Purpose of the Study:

• * To validate a virtual training platform for amputee subjects using a biorealistic prosthetic hand.
• * To evaluate the contributions of visual and evoked tactile sensation (ETS) feedback in grasp force control tasks.
• * To assess the integration of sensory information for improved motor performance in amputees.

Main Methods:

• * A digital twin of a tendon-driven prosthetic hand was created in the MuJoCo environment.
• * Biorealistic controllers used surface electromyography (sEMG) signals from amputees to emulate muscle activation.
• * Six forearm amputees performed force tracking and holding tasks with varying visual and ETS feedback conditions.

Main Results:

• * Visual feedback was more dominant than ETS feedback in force tracking and holding tasks.
• * ETS feedback significantly improved motor performance in force holding tasks when visual feedback was absent.
• * Sensory feedback integration was subject-specific, but amputees demonstrated good sensorimotor performance.

Conclusions:

• * Evoked tactile sensation (ETS) feedback provides reliable sensory information for stable grasp force control in prosthetic hands.
• * A virtual training platform can help amputees adapt to biorealistic hand controllers and ETS feedback.
• * The findings support the development of advanced neuroprosthetics with integrated sensory feedback systems.