机器人
计算机科学
开关设备
编码(集合论)
电压
实时计算
定位系统
GSM演进的增强数据速率
计算机硬件
计算机视觉
点(几何)
模拟
嵌入式系统
工程类
人工智能
电气工程
几何学
数学
集合(抽象数据类型)
程序设计语言
作者
Lan-xi Shi,Hong-yu Ni,Jinpeng Zhang,Zu-rong Lin,Wenxu Yan
标识
DOI:10.1109/ichve53725.2022.9961460
摘要
Gas Insulated Switchgear (GIS) is more and more widely used in high voltage power system. In the past, the observation of SF6 gas in substation equipment was mostly done manually. However, when the leakage concentration of SF6 gas is too high, it will pose a threat to the health of inspectors. Therefore, it is necessary to consider the robot equipped with SF6 detection sensor as the carrier to complete the detection of SF6 gas leakage in the substation. In this paper, an observation point positioning method for GIS orbital inspection robot is designed. Firstly, the application scenarios of high-voltage substations and GIS are analyzed, and actual usage requirements of orbital inspection robots in the positioning of observation points are obtained. In the actual use requirements, the angle and distance are within the controllable range. When locating observation points, assign a Quick Response (QR) code to each observation point. Secondly, a method to identify the location information of QR code is designed. QR code positioning has the advantages of high positioning accuracy, good flexibility, low cost, and small cumulative error. The method is based on OpenCV and Zbar library. The method does not need to read the distance data directly, and only obtains the position of the QR code in the coordinate system by calculating the plane angle and edge of the QR code representing the position of the observation point, thereby realizing the robot positioning. Since this method is less demanding on the camera, the cost is reduced. In the experimental verification, the hardware uses Intel realsense D455, the scene simulates the actual use scene, and multiple groups of experiments are carried out using the control variable method. Finally, the position measurement error is 1.5%, and the recognition speed is 30 frames per second. Therefore, it is proved that the method has high positioning accuracy and can solve the problem of GIS orbital inspection robot detecting and positioning observation points.
科研通智能强力驱动
Strongly Powered by AbleSci AI