Wu Xiaohua, Liang Yuxuan, Lu Longsheng, Yang Shu, Liang Zhanbo, Liu Feilong, Lu Xiaoyu, Xiao Bowen, Zhong Yilin, Xie Yingxi
School of Mechanical & Automotive Engineering, South China University of Technology, Guangzhou 510641, China.
Research (Wash D C). 2025 Aug 18;8:0835. doi: 10.34133/research.0835. eCollection 2025.
The lower limb motion capture technology has garnered significant attention as a pivotal enabler in extended reality, sports science, film production, and medical rehabilitation. However, existing mature systems face critical challenges, including restricted operational environments, interference with natural activities, and cumbersome wearability. Here, a flexible insole pressure sensor array with an ultra-wide sensing range is developed using dip coating, laser cutting, and hot pressing, enabling lower limb motion capture. The developed fabric-based flexible pressure sensor exhibited an ultra-wide pressure range (3,770.9 kPa), high sensitivity (2.68 kPa), rapid response, recovery times (17.2 ms/3.5 ms), and high work life (>4 million loading/unloading cycles). The insole-shaped flexible pressure sensor array accurately measures pressure in different postures, achieving 95.5% classification accuracy across 10 dynamic and static poses. More importantly, the system achieved a joint position prediction accuracy of 7.8 pixels (~3.6 cm) in lower limb pose estimation. This high-precision lower limb motion capture system represents an ideal terminal for future extended reality applications, offering seamless integration, comfortable use, easily wearable design, and broad accessibility.
下肢运动捕捉技术作为扩展现实、体育科学、电影制作和医学康复中的关键推动因素，已引起广泛关注。然而，现有的成熟系统面临着严峻挑战，包括操作环境受限、干扰自然活动以及穿戴不便等问题。在此，通过浸涂、激光切割和热压工艺，开发出一种具有超宽传感范围的柔性鞋垫压力传感器阵列，用于下肢运动捕捉。所开发的基于织物的柔性压力传感器具有超宽压力范围（3770.9 kPa）、高灵敏度（2.68 kPa）、快速响应和恢复时间（17.2 ms/3.5 ms）以及高使用寿命（>400万次加载/卸载循环）。鞋垫形状的柔性压力传感器阵列能够准确测量不同姿势下的压力，在10种动态和静态姿势下实现了95.5%的分类准确率。更重要的是，该系统在下肢姿势估计中实现了7.8像素（约3.6厘米）的关节位置预测精度。这种高精度的下肢运动捕捉系统是未来扩展现实应用的理想终端，具有无缝集成、使用舒适、易于穿戴设计和广泛适用性等特点。
