Suppr超能文献

一种用于计算机触觉的仿生触觉扫描仪。

A bioinspired tactile scanner for computer haptics.

机构信息

Guangdong Province Key Laboratory of Display Material and Technology, State Key Laboratory of Optoelectronic Materials and Technologies, School of Electronics and Information Technology (School of Microelectronics), Sun Yat-sen University, Guangzhou, 510275, China.

Shenzhen Chipwey Innovation Technologies Co. Ltd., Shenzhen, 518100, China.

出版信息

Nat Commun. 2024 Sep 2;15(1):7632. doi: 10.1038/s41467-024-51674-2.

Abstract

Computer haptics (CH) is about integration of tactile sensation and rendering in Metaverse. However, unlike computer vision (CV) where both hardware infrastructure and software programs are well developed, a generic tactile data capturing device that serves the same role as what a camera does for CV, is missing. Bioinspired by electrophysiological processes in human tactile somatosensory nervous system, here we propose a tactile scanner along with a neuromorphically-engineered system, in which a closed-loop tactile acquisition and rendering (re-creation) are preliminarily achieved. Based on the architecture of afferent nerves and intelligent functions of mechano-gating and leaky integrate-and-fire models, such a tactile scanner is designed and developed by using piezoelectric transducers as axon neurons and thin film transistor (TFT)-based neuromorphic circuits to mimic synaptic behaviours and neural functions. As an example, the neuron-like tactile information of surface textures is captured and further used to render the texture friction of a virtual surface for "recreating" a "true" feeling of touch.

摘要

计算机触觉(CH)是关于元宇宙中触觉感知和渲染的整合。然而,与计算机视觉(CV)不同,CV 的硬件基础设施和软件程序都得到了很好的发展,而缺少一种通用的触觉数据捕捉设备,它的作用类似于相机对于 CV 的作用。受人类触觉躯体感觉神经系统的电生理过程的启发,我们在这里提出了一种触觉扫描仪,以及一个神经形态工程系统,初步实现了闭环触觉采集和渲染(再创造)。基于传入神经的架构和机械门控和漏电积分和放电模型的智能功能,使用压电换能器作为轴突神经元和基于薄膜晶体管(TFT)的神经形态电路来模拟突触行为和神经功能,设计和开发了这种触觉扫描仪。例如,表面纹理的类神经元触觉信息被捕获,并进一步用于渲染虚拟表面的纹理摩擦,以“再现”真实的触感。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/db99/11369279/af729e25fde2/41467_2024_51674_Fig1_HTML.jpg

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验