Feedback control methods for task regulation by electrical stimulation of muscles

Area of Science:

• Biomedical Engineering
• Neuroscience
• Rehabilitation Engineering

Background:

• Electrical muscle stimulation (EMS) requires effective control algorithms for functional applications.
• Muscle force control is complex due to nonlinearities like recruitment, length-tension, and force-velocity relationships.

Purpose of the Study:

• To compare the performance of three feedback control algorithms with varying complexity for muscle force modulation.
• To evaluate controller efficacy across different tasks: isometric torque, position tracking, and load transitions.
• To provide guidance for selecting control strategies for neural prostheses.

Main Methods:

• Tested three controllers: first-order pulse width (PW), adaptive, and pulse width/stimulus period (PW/SP).
• Evaluated controllers in an animal model (cat ankle joint) stimulating tibialis anterior and medial gastrocnemius muscles.
• Assessed performance during isometric torque control, unloaded position tracking, and transitions between conditions.

Main Results:

• The simple PW controller demonstrated robust performance across all tested tasks.
• The PW/SP controller significantly improved isometric torque and load transition control but offered minor gains in position tracking.
• The adaptive controller did not consistently outperform the PW controller, especially during abrupt system changes.

Conclusions:

• Simpler feedback control algorithms, like the PW controller, can be highly effective for muscle stimulation tasks.
• Complex nonlinearities and sudden changes in loading conditions pose challenges for adaptive controllers.
• Results offer practical guidelines for selecting appropriate control algorithms for neural prosthetic devices.