摘要
记录和理解外界压力刺激对于人与周围环境交互的研究和智能机器人的开发具有重要意义.现有的压力传感设备多为刚性结构难以自然贴附于物体表面,且低密度传感单元使得物体压力特征信息的获取受到限制,一种可适形、全覆盖、高密度的压力传感与分析系统亟待研究.本文采用湿法纺丝工艺连续化制备了芯层为混纺导电芯材,包层为碳纳米管掺杂的聚氨酯的核壳结构压感纤维,通过缝纫、刺绣方式在织物表面构筑经纬结构的交叉点压力传感阵列.结合阵列数据采集实时捕获压力图谱帧,基于深度学习卷积神经网络驱动的算法模型,实现了物体轮廓识别垫对环境物体的轮廓精准分类识别.该识别系统准确率高达99.4%,证明了其在提取物体的压力信息和揭示物体的形态特征的应用潜能.
Recording and understanding external pressure stimuli is of great significance for the study of human-environment interaction and the development of intelligent robots.The existing pressure sensing devices are almost rigid structures that are difficult to attach naturally to the surface of objects,and sensing units with low-density distribution have limited the acquisition of object pressure characteristic information.Therefore,a pressure sensing and analysis system with adaptability,full coverage,and high density urgently needs to be studied.This paper reports a scalable core-shell structure pressure sensing fiber with a core layer of blended conductive electrode and a cladding layer of carbon nanotube doped polyurethane continuously prepared by a wet spinning process.A cross-point pressure sensing array of warp and weft structures is constructed on the surface of the fabric through sewing and embroidery.By combining array data acquisition and real-time capture of pressure graph frames,a deep learning convolutional neural network-driven algorithm model is used to achieve precise contour classification and recognition of environmental objects on the pad.The recognition system has an accuracy of up to 99.4%,demonstrating its potential in extracting pressure information from objects and revealing their morphological features.
作者
欧阳静宇
欧阳举
胡佳雨
刘晓娟
李攀
杨麦萍
王佳希
侯冲
张其冲
陶光明
OUYANG JingYu;OUYANG Ju;HU JiaYu;LIU XiaoJuan;LI Pan;YANG MaiPing;WANG Jiaxi;HOU Chong;ZHANG QiChong;TAO GuangMing(Wuhan National Laboratory for Optoelectronics,Huazhong University of Science and Technology,Wuhan 430074,China;State Key Laboratory of Material Processing and Die&Mould Technology,School of Materials Science and Engineering,Huazhong University of Science and Technology,Wuhan 430074,China;School of Computer Science and Technology,Huazhong University of Science and Technology,Wuhan 430074,China;School of Optical and Electronic Information,Huazhong University of Science and Technology,Wuhan 430074,China;Key Laboratory of Multifunctional Nanomaterials and Smart Systems,Suzhou Institute of Nano-Tech and Nano-Bionics,Chinese Academy of Sciences,Suzhou 215123,China)
出处
《中国科学：技术科学》
EI
CSCD
北大核心
2024年第4期665-677,共13页
Scientia Sinica(Technologica)
基金
华中科技大学交叉研究支持计划(编号:2023JCYJ039)
国家重点研发计划(编号:2022YFB3805800)资助项目。
关键词
压感纤维
传感织物
阵列数据采集
深度学习卷积神经网络
pressure sensing fibers
sensing fabric
array data acquisition
deep learning conolution neural network
