An Information-Rich and Highly Wearable Soft Sensor System Based on Displacement Myography for Practical Hand Gesture...

Area of Science:

• Biomedical Engineering
• Human-Computer Interaction
• Wearable Technology

Background:

• Wearable sensors are crucial for non-invasive human-machine interfaces.
• Existing systems face challenges balancing wearability, practicality, and performance.
• Forearm muscle deformation offers a promising, untapped signal source for gesture recognition.

Purpose of the Study:

• Introduce MyoLog, a novel wearable soft sensor system for hand gesture recognition.
• Evaluate the performance of MyoLog in recognizing a diverse range of hand gestures.
• Demonstrate the practical applicability of MyoLog in real-world scenarios.

Main Methods:

• Utilized displacement myography with magnets and tri-axis magnetometers to capture forearm muscle displacements.
• Integrated sensors into soft, flexible structures conforming to forearm anatomy.
• Collected data from 9 participants performing 44 distinct hand gestures.

Main Results:

• Achieved high classification accuracy: 97.7% for 13 gestures, 91.5% for 22 gestures, and 89.1% for 28 gestures.
• Demonstrated high signal-to-noise ratio and sensitivity, enabling rich signal acquisition.
• Validated MyoLog's performance across a wide spectrum of hand gestures.

Conclusions:

• MyoLog offers a highly wearable and practical solution for accurate hand gesture recognition.
• The system's performance supports its potential in diverse applications like virtual reality and prosthetic control.
• MyoLog advances the development of effective human-machine interactions through non-invasive sensing.