Review of sEMG for Exoskeleton Robots: Motion Intention Recognition Techniques and Applications

Area of Science:

• Biomedical Engineering
• Rehabilitation Robotics
• Human-Computer Interaction

Background:

• Global aging population increases demand for assistive technologies.
• Exoskeleton robots offer solutions for limb movement assistance and rehabilitation.
• Surface electromyography (sEMG) and intention recognition are crucial for advanced human-robot interaction.

Purpose of the Study:

• To review research on upper limb exoskeleton robots, sEMG, and intention recognition.
• To analyze the application of sEMG, machine learning, and deep learning in exoskeleton control.
• To identify key challenges and future directions in human movement intention recognition.

Main Methods:

• Literature review and keyword clustering analysis.
• Comprehensive discussion of sEMG technology and AI algorithms for intention recognition.
• Analysis of traditional machine learning and deep learning approaches.

Main Results:

• sEMG technology combined with AI is vital for exoskeleton robot control.
• Current research focuses on algorithms with high adaptability and accuracy.
• Deep learning methods show promise for improved intention recognition.

Conclusions:

• Accurate human movement intention recognition is essential for effective exoskeleton robot control.
• Future research should explore multi-information fusion (e.g., EEG, sEMG) for enhanced generalization.
• Advancements support human-machine fusion-embodied intelligence in exoskeleton systems.