Research on flexible tactile end for visual-tactile fusion operations of power inspection robots

Journal Information

J. Hefei Univ. Tech. (Nat. Sci.), Sept.2025, Vol.48 No.9: 1153-1160

DOI: 10.3969/j.issn.1003-5060.2025.09.001

Authors

HU Qiusheng $ ^{1} $, LIU Ping $ ^{1} $, HU Ruohai $ ^{1} $, TENG Fei $ ^{2} $, YANG Qianqian $ ^{2} $, MA Yuanming $ ^{1} $

(1. School of Microelectronics, Hefei University of Technology, Hefei 230601, China; 2. Chuzhou Power Supply Company, State Grid Anhui Electric Power Co., Ltd., Chuzhou 239000, China)

Abstract and Keywords

Abstract: The power inspection robot can transmit real-time information and provide intelligent analysis and early warning, strengthening the management ability of power equipment. The existing inspection robots are mainly based on visual technology, with a focus on inspection, and lack the capability for intelligent operation under problematic conditions. In this paper, a capacitive flexible pressure sensor based on a composite material of polyvinyl alcohol (PVA) and $ LiFePO_{4} $ was developed. The dielectric film was subjected to microstructure treatment using sandpaper inversion to improve the sensitivity of the capacitive sensor. The relative capacitance change of the sensor reached 2 700 times (0-50 N), and it had fast response speed (47 ms) and high stability (1 000 cycles). Meanwhile, based on the working mechanism of self-capacitance, a proximity sensor with a maximum detection distance greater than 5 mm was designed. Finally, a flexible tactile end with a $ 4 \times 4 $ capacitive pressure array was constructed and applied to the power inspection robot for the plug and network cable extraction tasks. By analyzing the success rate of 100 operations, it was found that the success rate of high-precision operations was significantly improved in the scenario of visual-tactile fusion, indicating that the tactile end can provide physical support for high-precision tasks in power inspection.

Keywords: tactile end; pressure sensing; proximity sensing; visual-tactile fusion; robot

作者信息

胡秋生 $ ^{1} $，刘平 $ ^{1} $，胡若海 $ ^{1} $，滕飞 $ ^{2} $，杨茜茜 $ ^{2} $，马渊明 $ ^{1} $

(1. 合肥工业大学微电子学院，安徽合肥 230601；2. 国网安徽省电力有限公司滁州供电公司，安徽滁州 239000)

摘要和关键词

摘要: 电力巡检机器人具备实时信息传输和智能分析预警功能,能有效提升电力设备管理水平,然而现有的巡检机器人主要依赖视觉技术,侧重于“巡检”功能,缺乏问题工况智能作业的能力。文章首先制备一种基于聚乙烯醇(polyvinyl alcohol, PVA)和磷酸铁锂 $ \left(\mathrm{LiFePO}_{4}\right) $复合材料的电容式柔性压力传感器,并采用砂纸倒模对介电膜进行微结构处理,提升电容式传感器的灵敏度,使其电容相对变化达2700倍(0~50N),并具备快速响应(47ms)和高稳定性(1000次循环)特性;然后基于自电容工作机理,设计最远探测距离大于5mm的接近觉传感器;最后构建 $ 4\times4 $电容压力阵列的柔性触觉末端。将该柔性触觉末端应用于电力巡检机器人进行插头及网线拔取作业,并统计100次作业的操作成功率,结果表明,视触觉融合模式下的高精度作业成功率显著提升,证明该触觉末端能够为电力巡检中高精度作业任务提供物理支撑。

关键词: 触觉末端；压力感知；接近感知；视触觉融合；机器人

Funding

基金项目：浙江省“领雁”研发攻关计划资助项目（2022C03052；2024C01150）；国家电网有限公司科技资助项目（B312M023000R）

合肥工业大学学报(自然科学版)

Navigation