Intention Detection Using Physical Sensors and Electromyogram for a Single Leg Knee Exoskeleton

Area of Science:

• Biomedical Engineering
• Robotics
• Rehabilitation Technology

Background:

• Knee exoskeletons can restrict wearer mobility, particularly during sitting.
• Existing single-leg exoskeletons face challenges in detecting user intention due to lack of bilateral leg information.

Purpose of the Study:

• To develop a knee exoskeleton with unimpeded mobility.
• To enhance user intention detection for single-leg exoskeleton use.

Main Methods:

• Designed a knee exoskeleton with a prismatic ankle joint to allow unrestricted movement.
• Utilized length between knee center of rotation and ankle (LBKA) measurements for intention detection.
• Integrated electromyogram (EMG) sensors with LBKA and encoder data.

Main Results:

• The exoskeleton design prevents limitations in wearer movement, such as during sitting.
• Intention detection accuracy reached 82.1% using LBKA and encoder sensors.
• Adding EMG sensors increased detection accuracy to 92% and improved speed by 27.1 ms.

Conclusions:

• The developed knee exoskeleton effectively maintains wearer mobility.
• Multi-sensor fusion, including LBKA and EMG, significantly enhances intention detection for single-leg exoskeleton applications.