Control of an optimal finger exoskeleton based on continuous joint angle estimation from EMG signals

Area of Science:

• Biomedical Engineering
• Rehabilitation Technology
• Neuroprosthetics

Background:

• Loss of hand function due to stroke and spinal cord injuries necessitates advanced assistive devices.
• Wearable exoskeletons offer a promising solution for restoring hand mobility and independence.

Purpose of the Study:

• To develop and evaluate a novel, low-profile finger exoskeleton system.
• To enable intuitive control of the exoskeleton using surface electromyographic (sEMG) signals.
• To restore hand function for patients with neurological impairments.

Main Methods:

• Designed an optimally configured four-bar linkage finger exoskeleton to mimic natural finger trajectory.
• Developed an artificial neural network model to predict joint angles from EMG signals, incorporating electromechanical delay (EMD).
• Implemented a real-time control system using sEMG signals from the forearm to actuate the exoskeleton on the contralateral hand.

Main Results:

• The artificial neural network accurately predicted multiple finger joint angles.
• The sEMG-based control strategy effectively actuated the finger exoskeleton to achieve intended positions.
• Subjects reported appropriate motion support and intuitive control of the device.

Conclusions:

• The developed sEMG-controlled finger exoskeleton shows efficacy in restoring hand function.
• This system represents a significant advancement in wearable assistive technology for individuals with hand impairments.
• Future work may focus on refining control algorithms and expanding applications for diverse patient populations.