Evoked electromyography-based closed-loop torque control in functional electrical stimulation.

Area of Science:

• Biomedical Engineering
• Rehabilitation Technology
• Control Systems

Background:

• Functional electrical stimulation (FES) systems face challenges with time-variant muscle dynamics, such as fatigue, impacting restored motion accuracy.
• Lack of implantable sensors for direct torque feedback in human FES systems necessitates alternative control strategies.
• Electromyography (EMG) signals reflect stimulated muscle activity and can be used for joint torque prediction.

Purpose of the Study:

• To propose and evaluate a closed-loop torque control strategy for FES using EMG feedback.
• To develop an adaptive controller capable of compensating for muscle fatigue and predicting muscle responses.
• To achieve accurate, safe, and robust FES-driven motion restoration.

Main Methods:

• Development of an EMG-feedback predictive controller for FES.
• Utilizing FES-evoked EMG signals for joint torque prediction and adaptive control.
• Experimental and simulation studies to evaluate control performance, fatigue compensation, and control suppression.

Main Results:

• The proposed EMG-feedback predictive controller demonstrated effective adaptive joint torque control.
• The system successfully compensated for muscle fatigue, maintaining accuracy despite physiological changes.
• The controller showed capabilities in aggressive control suppression, enhancing system robustness.

Conclusions:

• EMG-feedback predictive control offers a viable solution for accurate and robust FES torque control.
• This strategy effectively addresses the challenge of time-variant muscle dynamics in FES applications.
• The developed controller enhances FES system performance by incorporating muscle state information.