输电线巡检机器人障碍物定位与识别  

Obstacle location and recognition method for an inspection robot based on multi-sensors

作　　者：丁平 李贞辉 张林华 杨晓龙 刘爱华[3] DING Ping;LI Zhenhui;ZHANG Linhua;YANG Xiaolong;LIU Aihua(Chuxiong Power Supply Bureau of Yunnan Power Grid Co.,Ltd.,Chuxiong 675000,China;Advanced Technology Institute,Zhejiang University,Hangzhou 310027,China;State Key Laboratory of Robotics,Shenyang Institute of Automation,Chinese Academy of Sciences,Shenyang 110016,China)

机构地区：[1]云南电网有限责任公司楚雄供电局,楚雄675000 [2]浙江大学先进技术研究院,杭州310027 [3]中国科学院沈阳自动化研究所机器人学国家重点实验室,沈阳110016

出　　处：《现代制造工程》2021年第2期36-42,共7页Modern Manufacturing Engineering

摘　　要：针对巡检机器人运行中对线路障碍物的自主定位与识别,分析了巡检机器人自主运行中对障碍物信息感知的精度和实时性要求,提出了一种用于线路障碍物定位与识别的多传感器集成结构,并介绍了各个阶段的障碍物定位与识别算法。实验表明该方法可以实现机器人稳定自主行驶、可靠定位与识别障碍物,满足各阶段机器人对障碍物信息感知的精度和实时性要求,从而为机器人制定自主越障策略提供信息。Aiming at the obstacle location and recognition when the inspection robot on high voltage transmission line is running,the accuracy and real-time performance requirements for obstacle perception information when the inspection robot is running were analyzed,and an integrated structure of multi-sensor system for obstacle location and recognition was presented.The obstacles were detected by the monocular measurement principle,located by fusing the sensors of contact sensor,encoder and current sensor.The obstacle recognition method was proposed by fusing the wavelet feature and shape characteristics of images from different cameras.The experimental results show that the aforementioned methods can control the robot to approach obstacles steadily,locate and recognize them reliably with the appropriate accuracy and real-time performance.

关 键 词：巡检机器人 障碍物定位与识别 多传感器集成 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]