Self-Powered Handwritten Letter Recognition Based on a Masked Triboelectric Nanogenerator for Intelligent Personal Protective Equipment

Abstract

As one of the most important ways of human−machine interfaces, the touchpad has excellent input convenience. Input devices for extreme environments require simpler structures and diverse inputs to ensure information inputs. This paper proposed a self-powered flexible input panel with single-channel output for the input recognition of all 26 letters, and a paper mask was implemented to cover the triboelectric nanogenerator (TENG) board and obtain more complicated electrical signal features. Based on the change of the triboelectric output of the mask, neural network models with different combinations of layers were designed and optimized, and the highest recognition rate of 88.7% for all letters and 100% recognition accuracy for some letters were achieved among the five testers. For letters with low recognition rates, a specific writing specification was further proposed to improve the accuracy of model recognition. These results facilitate the application of the proposed input panel as a flexible wearable device and personal protective equipment for extreme environments including chemical, biological, radiological, nuclear (CBRN) scenarios or aerospace.