面向遥操作电力作业的触觉生成技术-电气工程与自动化-CSCIED科技核心评价数据库-手机版

面向遥操作电力作业的触觉生成技术

Tactile generation technology for remote operation of power operations

ES评分 0 浏览量：25 下载量：0

DOI	10.12208/j.jeea.20250107
刊名	电气工程与自动化 Journal of Electrical Engineering and Automation
年，卷(期)	2025, 4(4)
作者	李昂^1,2, 刘跟创¹,
作者单位	1 南京邮电大学通信与信息工程学院江苏南京 2 南京邮电大学空天地海通信技术一体化研究院江苏南京
摘要	融合触觉反馈的远程遥操作电力作业已成为提升高压环境作业安全性与操作精度的关键技术，但受限于触觉通信的带宽不足与信号易失真的问题。为此，本研究提出一种基于大模型的文本引导触觉生成技术，通过将高维触觉图像压缩为低维文本语义特征进行传输，显著降低了数据量，有效解决了带宽受限问题；文本特征的语义抽象与结构化编码机制显著提升了抗干扰性能，有效解决了传统触觉图像信号在传输过程中易失真的问题。
Abstract	Remote teleoperation of power operations with haptic feedback has become a key technology for enhancing safety and operational accuracy in high-voltage environments. However, it is limited by insufficient bandwidth for haptic communication and the susceptibility of signals to distortion. To address these issues, this study proposes a text-guided haptic generation technology based on large models. By compressing high-dimensional haptic images into low-dimensional text semantic features for transmission, it significantly reduces the data volume and effectively resolves the bandwidth limitation problem. The semantic abstraction and structured encoding mechanism of text features significantly enhance the anti-interference performance, effectively solving the problem of signal distortion during the transmission of traditional haptic images.
关键词	遥操作电力作业；文本特征提取；触觉图像生成
KeyWord	Teleoperation of power operations; Text feature extraction; Tactile image generation
基金项目
页码	29-33

参考文献
相关文献

[1] Elizondo D , Gentile T , Candia H ,et al.Overview of robotic applications for energized transmission line work — Technologies, field projects and future developments
[J]. IEEE, 2010.

[2] 张耀华,宋爱国,缪天缘,等.面向电力设施巡检运维的多模态遥操作机器人研究进展
[J].中国测试,2025,51(04): 18-30.

[3] Zhu J , Chen Y , Xu M ,et al.Graphical Force and Haptic Feedback Teleoperation System for Live Power Lines Maintaining Robot
[J].IEEE, 2019.

[4] Suresh S , Qi H , Wu T ,et al.NeuralFeels with neural fields: Visuotactile perception for in-hand manipulation
[J]. Science Robotics, 2024, 9(96).

[5] Yang F , Feng C , Chen Z ,et al.Binding Touch to Everything: Learning Unified Multimodal Tactile Representations
[J]. IEEE, 2024.

[6] Hu H , Chan K C K , Su Y C ,et al.Instruct-Imagen: Image Generation with Multi-modal Instruction
[J].IEEE, 2024.

[7] Zhang H , Xu T , Li H ,et al.StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networks
[J].IEEE, 2017.

[8] Hu E J, Shen Y, Wallis P, et al. Lora: Low-rank adaptation of large language models
[J]. ICLR, 2022, 1(2): 3.

引用本文

李昂^,, 刘跟创. 面向遥操作电力作业的触觉生成技术 [J]. 电气工程与自动化. 2025; 4; (4). 29 - 33.

文献评论

相关学者

相关机构