Publication

3DKnITS: Three-dimensional Digital Knitting of Intelligent Textile Sensor for Activity Recognition and Biomechanical Monitoring

Copyright

Irmandy Wicaksono/MIT Media Lab

Irmandy Wicaksono/MIT Media Lab

Wicaksono, I., Hwang, P. G., Droubi, S., Wu, F. X., Serio, A. N., Yan, W., & Paradiso, J.A. (2022). 3DKnITS: Three-dimensional Digital Knitting of Intelligent Textile Sensor for Activity Recognition and Biomechanical Monitoring. IEEE in Medicine and Biology Society.

Abstract

We present an approach to develop seamless and scalable piezo-resistive matrix-based intelligent textile using digital flat-bed and circular knitting machines. By combining and customizing functional and common yarns, we can design the aesthetics and architecture and engineer both the electrical and mechanical properties of a sensing textile. By incorporating a melting fiber, we propose a method to shape and personalize three-dimensional piezo-resistive fabric structure that can conform to the human body through thermoforming principles. It results in a robust textile structure and intimate interfacing, suppressing sensor drifts and maximizing accuracy while ensuring comfortability. This paper describes our textile design, fabrication approach, wireless hardware system, deep-learning enabled recognition methods, experimental results, and application scenarios. The digital knitting approach enables the fabrication of 2D to 3D pressure-sensitive textile interiors and wearables, including a 45 x 45 cm intelligent mat with 256 pressure-sensing pixels, and a circularly-knitted, form-fitted shoe with 96 sensing pixels across its 3D surface both with linear piezo-resistive sensitivity of 39.4 for up to 500 N load. Our personalized convolutional neural network models are able to classify 7 basic activities and exercises and 7 yoga poses in-real time with 99.6% and 98.7% accuracy respectively. Further, we demonstrate our technology for a variety of applications ranging from rehabilitation and sport science, to wearables and gaming interfaces.

© 2022 IEEE.  Personal use of this material is permitted.  Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

Related Content