Interdigitated capacitive strain sensor enables precise yoga-inspired motion tracking.

Area of Science:

• Materials Science
• Biomedical Engineering
• Wearable Technology

Background:

• Wearable sensors are crucial for remote health monitoring and digital fitness.
• Developing comfortable, durable, and washable sensors remains a challenge.

Purpose of the Study:

• To develop and evaluate a fully textile-based interdigitated capacitive (IDC) strain sensor for wearable posture recognition.
• To assess the sensor's performance in capturing multidimensional electrical signals during distinct yoga postures.
• To investigate the efficacy of a machine-learning pipeline for accurate posture classification.

Main Methods:

• Fabrication of a lightweight, washable textile-based IDC strain sensor.
• Integration of the sensor into clothing for unobtrusive monitoring.
• Collection of capacitance, resistance, and phase signals during four yoga postures.
• Application of a supervised machine-learning pipeline with handcrafted and MiniROCKET features for classification.

Main Results:

• The sensor demonstrated stable electromechanical performance over 6,500 strain cycles (gauge factor 0.67–0.81).
• Accurate classification of four yoga postures (High Lunge, Lunge Forward, Squat, Tree) with 94.4% accuracy and 94.2% F1 score.
• Capacitance-derived features were identified as the most informative for posture recognition.
• Robust and distinct signal patterns were observed for each posture.

Conclusions:

• Textile-based IDC sensing is a feasible technology for wearable movement monitoring.
• The developed sensor system shows high potential for applications in digital fitness and rehabilitation.
• Machine learning enhances the accuracy and reliability of posture recognition using wearable sensors.