Improving Kinematic Accuracy of Soft Wearable Data Gloves by Optimizing Sensor Locations.

Area of Science:

• Biomechanics
• Wearable Sensor Technology
• Human-Computer Interaction

Background:

• Bending sensors are crucial for compact data glove designs.
• Existing data gloves struggle with accurate thumb carpometacarpal (CMC) joint movement measurement due to sensor crosstalk.
• Optimizing sensor placement is key to overcoming these limitations.

Purpose of the Study:

• To develop a novel method for identifying optimal bending sensor locations on the hand.
• To minimize sensor crosstalk and improve the accuracy of CMC joint angle measurements.
• To validate the effectiveness of the proposed sensor placement strategy.

Main Methods:

• Collected 3D hand surface data from ten subjects across various thumb postures.
• Utilized scanned CMC joint contours to estimate flexion and abduction angles by adjusting sensor positions and orientations.
• Employed the least squares method to determine optimal sensor locations minimizing estimation errors.

Main Results:

• Optimal sensor locations were identified through computational analysis.
• Experimental validation demonstrated significant improvements in CMC joint angle accuracy.
• Achieved accuracies of 2.8° ± 1.9° for flexion and 1.9° ± 1.2° for abduction.

Conclusions:

• The proposed method effectively identifies optimal sensor locations for improved CMC joint measurement accuracy in data gloves.
• This approach enhances the precision of wearable sensing systems for hand configurations.
• The methodology can be applied to other soft wearable measurement devices.