基于机器视觉避障的理论研究
|
摘要
有效解决机器人怎样"看"的问题,就是我们经常能够听到的机器视觉。而针对这一问题进行解决的基础就是怎样基于二维图像来获取三维信息,进而了解我们所处于的这个三维世界,规划导航路线。而对于机器人的导航问题通常都是关乎到三个方面:身处何处、要往哪去、要怎样去。第一个问题就是需要对系统中的定位问题采取解决措施,确定好机器人在工作环境中的相对位置;后边两个问题指的是对导航系统的路径实施规划和跟踪。本文主要从图像处理、障碍物识别和单目、双目视觉测距四个方面,对机器视觉避障相关的部分理论知识进行了研究。
The problem that effectively solves the problem of how robots "look" is the machine vision that we often hear. The basis for solving this problem is how to obtain three-dimensional information based on two-dimensional images,and then understand the three-dimensional world we are in, and plan the navigation route. The navigation problems for robots are usually related to three aspects: where they are, where they are going, and how to go. The first problem is that it is necessary to take corrective measures for the positioning problem in the system to determine the relative position of the robot in the working environment; the latter two problems refer to the planning and tracking of the path of the navigation system. In this paper, from the four aspects of image processing, obstacle recognition and monocular and binocular vision ranging, some theoretical knowledge related to machine vision obstacle avoidance is studied.
The problem that effectively solves the problem of how robots "look" is the machine vision that we often hear. The basis for solving this problem is how to obtain three-dimensional information based on two-dimensional images,and then understand the three-dimensional world we are in, and plan the navigation route. The navigation problems for robots are usually related to three aspects: where they are, where they are going, and how to go. The first problem is that it is necessary to take corrective measures for the positioning problem in the system to determine the relative position of the robot in the working environment; the latter two problems refer to the planning and tracking of the path of the navigation system. In this paper, from the four aspects of image processing, obstacle recognition and monocular and binocular vision ranging, some theoretical knowledge related to machine vision obstacle avoidance is studied.
引文
[1]魏小邦.基于计算机视觉的内河船舶导航技术研究[D].大连海事大学,2018.WEI Xiao-bang.Research on navigation technology of inland navigation based on computer vision[D].Dalian Maritime University,2018.
[2]范莹莉.基于机器视觉的AGV动态路径识别算法研究[D].兰州交通大学,2011.FAN Ying-li.Research on AGV dynamic path recognition algorithm based on machine vision[D].Lanzhou Jiaotong University,2011.
[3]杨欢.一种基于视觉避障及导航功能的机器人设计[J].机电技术,2015(03):40-43.YANG Huan.A Robot Design Based on Visual Obstacle Avoidance and Navigation Function[J].Electromechanical Technology,2015(03):40-43.
[4]朱凯凯,初阳,华维超.机器视觉在无人机智能避障的应用[J].通讯世界,2016(21):282.ZHU Kai-kai,CHU Yang,HUA Wei-chao.Application of Machine Vision in UAV Intelligent Obstacle Avoidance[J].Communication World,2016(21):282.
[5]陈高攀,徐美华,王琪,等.一种基于单目视觉的前方车辆检测算法[J].上海大学学报(自然科学版),2019,25(01):56-65.CHEN Gao-pan,XU Mei-hua,WANG Qi,et al.AForward Vehicle Detection Algorithm Based on Monocular Vision[J].Journal of Shanghai University(Natural Science),2019,25(01):56-65.
[6]闵小,李迎,李蒙,等.智能小车单目视觉障碍检测及避障系统设计[J].电子技术,2018,47(12):91-94+90.MIN Xiao,LI Ying,LI Meng,et al.Design of monocular vision obstacle detection and obstacle avoidance system for intelligent car[J].Electronic technology,2018,47(12):91-94+90.
[7]李海军.康复理疗系统中基于平行双目视觉自动识别系统的研究[D].昆明理工大学,2006.LI Hai-jun.Research on automatic binocular vision recognition system in rehabilitation physiotherapy system[D].Kunming University of Science and Technology,2006.
[8]王铮,赵晓,佘宏杰,等.基于双目视觉的AGV障碍物检测与避障[J].计算机集成制造系统,2018,24(02):400-409.WANG Zheng,ZHAO Xiao,SHE Hong-jie,et al.AGV obstacle detection and avoidance based on binocular vision[J].Computer Integrated Manufacturing System,2018,24(02):400-409.
[9]徐杰,陈一民,史志龙.双目视觉变焦测距技术[J].上海大学学报(自然科学版),2009,15(02):169-174.XU Jie,CHEN Yi-min,SHI Zhi-long.The Binocular Vision Zoom Distance Measurement Technology[J].Journal of Shanghai University(Natural Science),2009,15(02):169-174.
[10]李伯平.机器视觉的双工业机器人协调作业分析研究[J].电子元器件与信息技术,2018.LI Bo-ping.Research on Coordination Work Analysis of Dual Industrial Robots Based on Machine Vision[J].Electronic Components and Information Technology,2018.
[2]范莹莉.基于机器视觉的AGV动态路径识别算法研究[D].兰州交通大学,2011.FAN Ying-li.Research on AGV dynamic path recognition algorithm based on machine vision[D].Lanzhou Jiaotong University,2011.
[3]杨欢.一种基于视觉避障及导航功能的机器人设计[J].机电技术,2015(03):40-43.YANG Huan.A Robot Design Based on Visual Obstacle Avoidance and Navigation Function[J].Electromechanical Technology,2015(03):40-43.
[4]朱凯凯,初阳,华维超.机器视觉在无人机智能避障的应用[J].通讯世界,2016(21):282.ZHU Kai-kai,CHU Yang,HUA Wei-chao.Application of Machine Vision in UAV Intelligent Obstacle Avoidance[J].Communication World,2016(21):282.
[5]陈高攀,徐美华,王琪,等.一种基于单目视觉的前方车辆检测算法[J].上海大学学报(自然科学版),2019,25(01):56-65.CHEN Gao-pan,XU Mei-hua,WANG Qi,et al.AForward Vehicle Detection Algorithm Based on Monocular Vision[J].Journal of Shanghai University(Natural Science),2019,25(01):56-65.
[6]闵小,李迎,李蒙,等.智能小车单目视觉障碍检测及避障系统设计[J].电子技术,2018,47(12):91-94+90.MIN Xiao,LI Ying,LI Meng,et al.Design of monocular vision obstacle detection and obstacle avoidance system for intelligent car[J].Electronic technology,2018,47(12):91-94+90.
[7]李海军.康复理疗系统中基于平行双目视觉自动识别系统的研究[D].昆明理工大学,2006.LI Hai-jun.Research on automatic binocular vision recognition system in rehabilitation physiotherapy system[D].Kunming University of Science and Technology,2006.
[8]王铮,赵晓,佘宏杰,等.基于双目视觉的AGV障碍物检测与避障[J].计算机集成制造系统,2018,24(02):400-409.WANG Zheng,ZHAO Xiao,SHE Hong-jie,et al.AGV obstacle detection and avoidance based on binocular vision[J].Computer Integrated Manufacturing System,2018,24(02):400-409.
[9]徐杰,陈一民,史志龙.双目视觉变焦测距技术[J].上海大学学报(自然科学版),2009,15(02):169-174.XU Jie,CHEN Yi-min,SHI Zhi-long.The Binocular Vision Zoom Distance Measurement Technology[J].Journal of Shanghai University(Natural Science),2009,15(02):169-174.
[10]李伯平.机器视觉的双工业机器人协调作业分析研究[J].电子元器件与信息技术,2018.LI Bo-ping.Research on Coordination Work Analysis of Dual Industrial Robots Based on Machine Vision[J].Electronic Components and Information Technology,2018.