Advanced Braille recognition based on protein biomimetic skin mechanical sensors

Abstract

Tactile perception, dominated by human skin, plays an important role in human-nature interaction, especially when visual perception is limited. The key to mimicking skin tactile perception is to address skin-like properties of materials and integration of biomimetic tactile functions. To be competent in bionic skin, composite silk fibroin (SF) films possessing high tensile properties (159.71 %) and hydrogels with elasticity (54.42 %) were prepared through a mesoscopic reconstruction strategy of SF materials, using fish scale gelatin molecules as mesoscopic templates to regulate the nucleation and crystallization kinetics of SF molecules. In addition, inspired by the mechanoreceptors, a bimodal protein bionic skin (BPBS) was prepared by horizontally integrating a single-electrode triboelectric sensor and a capacitive sensing array, which can work in sliding and pressing modes to mimic the sliding touch and pressing touch of the finger, respectively. The system achieves a 99 % Braille recognition accuracy in sliding mode through a deep learning algorithm, and Bluetooth technology enables real-time wireless Braille recognition in pressing mode, synergistically enhancing the robustness and practicality of BPBS. This research provides novel insights into enhancing human touch perception, human-computer interaction, and the advancement of intelligent prosthetics, marking a significant stride in the development of bionic skins with multimodal sensing capabilities.