Abstract:

Objective Integrating rigid conductive materials with textile-based flexible structures still encounters significant challenges in design compatibility, structure-performance coupling, and device reliability. Departing from conventional strategies that rely on intricate functional materials, this study fabricates an all-textile flexible strain-sensing fabric by tuning multiscale parameters, including yarn wrapping mode, twist, and fabric structure. The switchable structure textile strain sensor is expected to the performance requirements in practical scenarios such as fetal-movement simulation and gait recognition, highlighting its broad potential in intelligent wearable systems.

Method The spandex core yarn was fed into the braider under 100 cN pre-tension. Two silver-nylon and two spandex yarns were mounted counter-directionally and braided around the core into an X-interlocked structure at a 0.2 mm pitch. The X-braided yarn was then twisted at 50 turns/m in S and Z directions. These twisted yarns were arranged in an "SSZZ" pattern as warp (8 counts/cm) on a loom, interwoven with nylon weft (6 counts/cm), forming a stable switchable structure textile strain sensor with high strength and durability.

Results This study developed a highly sensitive and wide range flexible wearable device based on innovative design of sensing yarn wrapped in the opposite direction and switchable structure textile strain sensor. Sensing yarn wrapped in the opposite direction featured a counter-directional wrapping structure. At 0% strain, two silver-plated nylon yarns were in close contact, yielding low resistance. At 100% strain, the spandex acted as an isolation layer, causing complete separation of the filaments and maximum resistance. In contrast, sensing yarn wrapped in the same direction lacked this mechanism, reaching its maximum the relative resistance at only 20%-60% strain and showing no further increase beyond 60% strain due to identical wrapping direction and the absence of an isolation layer. The sensing yarn wrapped in the opposite direction exhibited a gauge factor (G_F) of 1.04 (R²= 0.999) at 0%-60% strain. At 60% - 100% strain, its G_F further increased to 2.21, while that of sensing yarn wrapped in the same direction dropped to 0.04. The sensing yarn wrapped in the opposite direction also demonstrated a wider response range (0%-100%), faster response/recovery (0.65 s/0.75 s), and excellent durability over 500 cycles. Furthermore, the yarn was woven into a switchable structure textile strain sensor. The fabric exhibited segmented sensitivity, where G_F reached 3.47 (R²= 0.995) at 0%-60% strain, 2.21 at 60%-100%, and 0.87 at 100%- 140%, achieving both high sensitivity and a broad sensing range. The structure remained stable under stepped strain (20%-140%), with no significant signal drift. It also showed rapid response (100 ms) and recovery (150 ms), consistent performance across frequencies (0.2-0.7 Hz), and outstanding durability over 1 500 cycles. The fabric was applied in fetal movement monitoring, offering stable signal output under simulated conditions. Additionally, when integrated into smart shoe uppers, it captured complex gait signals. Using a convolutional neural network (CNN) algorithm, the sensing fabric achieved 97.14% accuracy in classifying seven gait types.

Conclusion Through weaving technology and textile topological structure design, the coordinated effect of deformation between yarns and the textile structure is achieved. This enables the sensing textile to maintain high sensitivity while realizing a wide detection range of 0%-140%. Based on the wide detection range of the sensing fabric, as well as the comfort and electromagnetic shielding effect of the fabric itself, the feasibility of its application in fetal movement detection for pregnant women is verified. Leveraging the high sensitivity of the sensing fabric, it is integrated into shoe uppers, through signal acquisition and the combination of convolutional neural network, accurate recognition and prediction of highly complex gait patterns are realized. It is anticipated that this work inspires the development of flexible strain sensors with textile structures and provides an effective and cost-efficient design strategy for the next generation of intelligent robot systems.

Key words: sensing fabric, sensing performance, switchable-structure fabric, flexible electronic wearable device, strain sensor, gait detection, fetal movement monitoring