摘要
随着人工智能技术的快速发展,巡检机器人的应用已经非常普遍,它能稳定可靠地代替一些人工巡检,直接降低工人的工作量,也能实现全天不定时巡检,间接提升巡检的质量。针对那些地处偏僻,无人值守的高铁中继站机房的巡检问题,设计了一种基于SLAM激光导航的室内轮式巡检机器人,首先,基于中继站现场运行环境需求,对机器人各个硬件模块进行选型与独立设计;然后利用不同的软件平台,编写机器人底层驱动代码和上位界面代码,实现机器人各个硬件模块互相通信,以及实现机器人上下位机通过无线局域网进行数据通信;最后,基于OpenCV图像处理库处理摄像机采集到的图像,识别出图片中仪表指针的显示状态,并将结果传输到监控中心。实验结果表明,所设计的机器人定位精度高,自动运行最大偏差±10cm,图像处理算法识别准确率可达85%,后续随着训练库的扩大,正确率还会提升,最为突出的优势是设计成本低,可靠性高,在铁路行业有一定的推广价值。
With the rapid development of artificial intelligence technology,the application of inspection robots has become very common.It can stably and reliably replace some manual inspections,directly reduce the workload of workers,and can also realize unscheduled inspection throughout the day,indirectly improving the quality of inspection.This paper designs an indoor wheeled inspection robot based on SLAM laser navigation for the inspection problems of those remote and unattended high-speed railway relay stations.Firstly,based on the requirements of the relay station’s field operation environment,each hardware module of the robot is selected and independently designed.Then,different software platforms are used to write the bottom driver code and the upper interface code of the robot,so as to realize the communication between each hardware module of the robot and the data communication between the upper and lower robots through the wireless local area network.Finally,based on OpenCV image processing library,the image collected by the camera is processed,the display state of the instrument pointer in the image is recognized,and the result is transmitted to the monitoring center.The experimental results show that the robot designed in this paper has high positioning accuracy,the maximum deviation of automatic operation is ±10cm,and the recognition accuracy rate of image processing algorithm can reach 85%.With the expansion of training library,the accuracy rate will be improved.The most prominent advantages are low design cost and high reliability,which have certain promotion value in the railway industry.
作者
王泽民
林晓焕
宋扬
Wang Zemin;Lin Xiaohuan;Song Yang(Xi'an Polytechnic University,Xi'an 710048,China;Xi'an Ruixin Railway Equipment Co.,Ltd.,Xi'an 710100,China)
出处
《国外电子测量技术》
2019年第12期97-102,共6页
Foreign Electronic Measurement Technology
关键词
巡检机器人
模块集成与通信
SLAM激光导航
图像处理算法
inspection ro b o t
module integration and com m unication
slam laser navigation
image processing algorithm