A Real-Time Stability Control Method Through sEMG Interface for Lower Extremity Rehabilitation Exoskeletons

Area of Science:

• Rehabilitation Engineering
• Robotics
• Biomechanics

Background:

• Exoskeleton rehabilitation robots aid paraplegics but struggle with system stability and wearer strength during gait transitions.
• Gait switching can destabilize the human-exoskeleton system by affecting the center of gravity.

Purpose of the Study:

• To develop a real-time, stable control gait switching method for lower limb exoskeleton rehabilitation robots.
• To improve the stability and ergonomic interaction of the human-exoskeleton system during walking.

Main Methods:

• Developed a gait planning and stability analysis using a human kinematics model and the zero moment point method.
• Constructed a neural interface using surface electromyography (sEMG) for intention recognition and muscle fatigue estimation.
• Implemented a long short-term memory (LSTM) model for intention recognition, achieving nearly 99% accuracy.

Main Results:

• Demonstrated that forming an equilateral triangle with crutch-supporting points and a supporting leg enhances walking stability.
• Validated the proposed gait switching method on the SIAT lower limb rehabilitation exoskeleton.
• Experimental results confirmed the feasibility and efficiency of the method in enhancing stability and ergonomics.

Conclusions:

• The proposed real-time stable control gait switching method significantly improves the stability and ergonomic effects of lower limb rehabilitation exoskeletons.
• This advancement offers a more effective solution for gait assistance and rehabilitation in individuals with lower limb impairments.