Knitted Inductive Flex Sensors for Wearable Applications

Paper Title

Knitted Inductive Flex Sensors for Wearable Applications

Publication Info

• Topic area: Wearable sensing technologies using textile-based inductive sensors.
• Keywords: Knitted sensors, inductive sensing, LC circuits, wearable technology, healthcare applications, rehabilitation, textile electronics, bend sensors, motion capture, interactive garments.

Background and Problem

• Problem / challenge: Existing bend sensors, such as resistive flex sensors and rigid inductive sensors, lack textile qualities like softness, elasticity, and washability, making them unsuitable for wearable applications.
• Significance: Wearable sensors are critical for healthcare, rehabilitation, and interactive garments, where comfort, unobtrusiveness, and reliable sensing are essential.
• Motivation and related work: Prior work demonstrated textile-based inductive sensors using embroidered coils but did not integrate a full LC circuit into the fabric. This paper builds on these efforts by embedding both a knitted coil and capacitor into a seamless textile structure.

Solution

• Proposed approach: Fully knitted resonant LC circuit integrating a coil and capacitor into a tubular textile structure for sensing joint bending angles.
• Novelty:
  1. Development of a fully knitted LC circuit with integrated coil and capacitor.
  2. Demonstration of wire-pair capacitors knitted through plaiting insulated wires.
  3. Evaluation of sensor performance over 100 bending cycles, showing stability, accuracy, and low hysteresis.
  4. Applicability of the sensors across joints of varying radii for versatility in wearable applications.
• Procedure and key techniques:
  • Fabrication of tubular knitted coils using insulated copper wires and polyamide/spandex yarns to ensure flexibility and robustness.
  • Integration of parallel-wire capacitors using plating techniques to minimize mechanical stress and optimize capacitance.
  • Assembly of the LC circuit by soldering the coil and capacitor ends to form a continuous wire.
  • Optimization of sensor parameters (coil turns, spacing, capacitor length) to achieve desired resonance frequencies within the sensing hardware range.

Results

• Concrete findings:
  • Knitted coils achieved a mean inductance of 7.163 µH with a maximum standard deviation of 0.138 µH across samples, ensuring manufacturing repeatability (<2% deviation).
  • Knitted capacitors showed a capacitance variance of 22.5 pF across samples, confirming consistent fabrication.
  • Sensor sensitivity measured at -2.39 kHz/deg between 180° and 140° bending angles.
  • Maximum hysteresis observed at 16° during bending tests.
• Advantage over baselines:
  • Improved wearability and washability compared to foil-based resistive sensors and rigid inductive sensors.
  • Stable sensing performance unaffected by touch, thanks to the integrated capacitor reducing parasitic capacitance effects.
• Experiments / evaluation:
  • 100 bending cycles tested using a 3D-printed hinge with controlled angles (180° to 90°).
  • Resonance frequency measured at 2.66 MHz, aligning closely with theoretical calculations (2.59 MHz).
  • Demonstrations included knee and elbow sleeve prototypes for motion capture applications.
• Limitations and future work:
  • Influence of elastic materials like Spandex on sensor performance needs further exploration.
  • Feasibility of smaller sensors operating at higher resonance frequencies should be investigated.
  • Parasitic effects in higher frequency bands require study for improved sensor designs.

Summary

This paper introduces a fully knitted resonant LC circuit for wearable bend sensing applications, combining textile softness and elasticity with reliable inductive sensing. By integrating a knitted coil and capacitor, the sensor achieves high stability, accuracy, and low hysteresis over repeated bending cycles. Demonstrations on knee and elbow sleeves highlight its adaptability for healthcare, rehabilitation, and interactive garments. Future work aims to expand the dynamic range, explore smaller sensors, and advance contactless sensing methods for electronic textiles.