Optimal tracking of a sEMG based force model for a prosthetic hand

Area of Science:

• Biomedical Engineering
• Rehabilitation Engineering
• Control Systems

Background:

• Prosthetic hands lack intuitive control.
• Surface electromyography (sEMG) offers a non-invasive control signal.
• Accurate mapping of sEMG to skeletal muscle force is crucial for prosthetic control.

Purpose of the Study:

• To develop an optimal control strategy for a prosthetic hand using sEMG signals.
• To accurately model the relationship between sEMG signals and skeletal muscle force.
• To improve the dexterity and responsiveness of prosthetic hands.

Main Methods:

• System Identification (SI) was employed to determine the dynamic relationship between sEMG and muscle force.
• A Half-Gaussian filter was used for sEMG signal preprocessing.
• A fusion-based Multiple Input Single Output (MISO) model estimated skeletal muscle force.
• An optimal tracking control method was applied to follow the estimated force profile.

Main Results:

• The study successfully obtained the dynamic relation between sEMG and skeletal muscle force.
• The fusion-based MISO model provided accurate estimations of finger forces.
• The optimal tracking method demonstrated good agreement between the reference and actual force profiles.
• Simulation results validated the effectiveness of the proposed control strategy.

Conclusions:

• The developed sEMG-based optimal control strategy shows significant promise for advanced prosthetic hand control.
• Accurate sEMG-to-force modeling is key to achieving intuitive and effective prosthetic hand function.
• This approach can lead to more natural and responsive prosthetic limb performance.