摘要
危险作业环境下,巡检机器人需要依据周边环境规划无碰撞最优路线,规划导航过程中存在导航精度低、定位耗时长等问题。基于此,该文提出设计基于机器视觉的室内无轨巡检机器人导航系统。系统主要包括视觉采集模块、路线规划模块以及运动控制模块。其中,视觉采集模块通过摄像机标定、摄像机安装布局、多图像拼接、障碍物检测以及巡检机器人定位流程,得到障碍物位置信息与巡检机器人位置信息;路线规划模块以障碍物与巡检机器人位置信息为基础,利用蚁群算法规划机器人运动路线;运动控制模块通过PID控制器实现巡检机器人在直线与曲线运动路线的导航。实验结果表明,采用所提系统进行导航的误差较低,且定位耗时较短。
Under the dangerous working environment,the patrol robot needs to plan the best route without collision according to the surrounding environment.There are some problems in the process of planning navigation,such as low accuracy,long positioning time and so on.Based on this,a navigation system of indoor trackless inspection robot based on machine vision is proposed.The system mainly includes vision acquisition module,route planning module and motion control module.Among them,the vision acquisition module obtains the obstacle position information and inspection robot position information through camera calibration,camera installation layout,multi image splicing,obstacle detection and inspection robot positioning process,the route planning module uses ant colony algorithm to plan the robot motion route based on the obstacle and inspection robot position information,the motion control module uses PID controller realizes the navigation of the inspection robot in the straight line and curve motion path.The experimental results show that the proposed system has low error and short positioning time.
作者
汤义勤
高彦波
邹宏亮
叶建军
曾林
TANG Yi-qin;GAO Yan-bo;ZOU Hong-liang;YE Jian-jun;ZENG Lin(State Grid Taizhou Power Supply Company,Taizhou 318000,China;State Grid Zhejiang Taizhou Huangyan Power Supply Company,Taizhou 318020,China;Taizhou Huangyan Yongheng Power Construction Co.,Ltd.,Taizhou 318020,China)
出处
《自动化与仪表》
2020年第8期42-46,76,共6页
Automation & Instrumentation
关键词
机器视觉
室内无轨
巡检机器人
导航系统
蚁群算法
machine vision
indoor trackless
inspection robot
navigation system
ant colony algorithm
