微型张力传感器实现高精度力反馈，适合机器人手部紧凑集成。
A Miniature High-Resolution Tension Sensor Based on a Photo-Reflector for Robotic Hands and Grippers
- 基于光反射器与弹性结构设计，提升灵敏度和耐久性。
- 分辨率达9.9毫牛，力控误差低至0.073牛，性能超旧方案十倍。
- 结构简单轻便，适合集成于机械手、假肢等需要精细触觉的系统。
本文提出一种基于光反射器的微型张力传感器，专为紧凑型肌腱驱动机械手和夹持器设计。传感器尺寸仅为13~mm × 7~mm × 6.5~mm，可测量最高200~N的拉力。其对称弹性体结构结合圆角与柔性铰链，基于铁莫申科梁理论设计，并通过有限元分析验证，有效提升灵敏度与机械耐久性，同时最小化扭转变形。采用小型光反射器（VCNT2020）在近场区域测量位移，无需吸光材料或几何改造，克服了传统光断续器设计的局限。配备16位模数转换器与CAN-FD通信协议，支持最高5~kHz采样率。校准实验显示分辨率达9.9~mN（对应14位以上精度），均方根误差（RMSE）为0.455~N；使用扭线驱动器与PI控制的力控实验中，RMSE低至0.073~N。相比以往光断续器方案，本方法分辨率提升逾十倍，同时显著降低非线性与迟滞。该设计机械结构简单、重量轻、易装配，适用于需高精度力反馈的机器人及假肢系统。
原文摘要 · Abstract (English)
This paper presents a miniature tension sensor using a photo-reflector, designed for compact tendon-driven grippers and robotic hands. The proposed sensor has a small form factor of 13~mm x 7~mm x 6.5~mm and is capable of measuring tensile forces up to 200~N. A symmetric elastomer structure incorporating fillets and flexure hinges is designed based on Timoshenko beam theory and verified via FEM analysis, enabling improved sensitivity and mechanical durability while minimizing torsional deformation. The sensor utilizes a compact photo-reflector (VCNT2020) to measure displacement in the near-field region, eliminating the need for light-absorbing materials or geometric modifications required in photo-interrupter-based designs. A 16-bit analog-to-digital converter (ADC) and CAN-FD (Flexible Data-rate) communication enable efficient signal acquisition with up to 5~kHz sampling rate. Calibration experiments demonstrate a resolution of 9.9~mN (corresponding to over 14-bit accuracy) and a root mean square error (RMSE) of 0.455~N. Force control experiments using a twisted string actuator and PI control yield RMSEs as low as 0.073~N. Compared to previous research using photo-interrupter, the proposed method achieves more than tenfold improvement in resolution while also reducing nonlinearity and hysteresis. The design is mechanically simple, lightweight, easy to assemble, and suitable for integration into robotic and prosthetic systems requiring high-resolution force feedback.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。