Multiplexed Sensing Textiles Enabled by Reconfigurable Weaving Meso-Structures for Intricate Kinematic Posture Recognition and Thermal Therapy Healthcare

Abstract

Wearable sensing textiles with multimodal mechanical stimulation detection capabilities have broad applications, such as sports guidance and rehabilitation training. However, current mainstream multimodal sensing textiles typically rely on combining discrete sensors with single functions intensively through sewing or adhesion to detect various mechanical stimuli. Herein, an all-in-one multiplexed sensing textile (MPST) capable of simultaneously detecting pressure and tensile strain is achieved by engineering an innovative hierarchical architecture of textiles. The functionality of MPST is directly derived from the reconfigurable sensing pathways of the woven meso-structures, enabling it to exhibit excellent pressure sensitivity (0.1 kPa^–1), wide strain detection range (0–100%), superior durability, and desirable wearability. With the assistance of the Long Short-Term Memory (LSTM) algorithm, the MPST wearable system achieves a recognition accuracy of 97.5% in human kinematic postures of the elbow, knee, and foot. In addition, MPST demonstrates outstanding joule heating performance, which reaches 57.1 °C at a 2.5 V applied voltage. With its excellent multimodal sensing performance and joule heating ability, MPST holds great potential for applications in sports training guidance, assistive rehabilitation training, and soft robotics.