This paper introduces the design and fabrication of a smart and Hybrid Composite Finger (HCF) to achieve finger-like motions, such as holding and tapping motions. Bionic research on tapping motion of the index finger was conducted to obtain its structural and tapping parameters. The HCF, actuated by Shape Memory Alloy (SMA) wires, possesses a hybrid structure which is composed of a rigid structure to be its metacarpal part and a deformable structure to produce bending movement just like the function of the finger. Owing to an adhesive bonding technology, the HCF was fabricated with a composite structure which is reliable under impulsive responses, and had a worklife of more than 630000 times. A bending model was built by synthesizing the phase transformation dynamic model of the SMA wires and quasi-static analysis of the HCF. Structural optimization of the HCF was conducted by synthesizing the bending model together with experimental analyses. To produce a holding motion like as the finger, a holding heating strategy was proposed to adaptively heat the HCF to keep holding state based on the resistance feedback of SMA wires and a Proportion Differentiation (PD) algorithm. Besides, we used an impulsive heating method to heat the HCF to produce a high fidelity tapping motion with a maximum tapping force (6.83 N) at a response time (43 ms) which considerably coincided with those (about 5.8 N, 45 ms) from tapping bionics of the index finger. Finally, a soft prosthetic hand system which had a hand-like appearance was manufactured based on the HCFs and several tests like as anthropomorphic gesture motions and human-like tapping motions to tap a keyboard were conducted to prove potential application of the HCF.

中文翻译：

---

具有仿生敲击动作的智能混合手指，适用于柔软的假手

本文介绍了智能和混合复合手指（HCF）的设计和制造，以实现类似手指的动作，例如保持和敲击动作。对食指的敲击运动进行了仿生研究，以获取其结构和敲击参数。由形状记忆合金（SMA）线驱动的HCF具有混合结构，该结构由刚性结构（其掌骨部分）和可变形的结构组成，以产生弯曲运动，就像手指的功能一样。由于采用了粘合剂粘结技术，HCF采用了复合结构制造，该复合结构在脉冲响应下可靠，使用寿命超过63万次。综合SMA线材的相变动力学模型和HCF的准静态分析，建立了弯曲模型。通过综合弯曲模型和实验分析来进行HCF的结构优化。为了产生像手指一样的握持动作，提出了一种握持加热策略，基于SMA线的电阻反馈和比例微分（PD）算法，自适应地加热HCF以保持握持状态。此外，我们使用脉冲加热方法加热HCF，以在响应时间（43 ms）时以最大敲击力（6.83 N）产生高保真敲击运动，该时间与从攻食指的仿生学。最后，