电动汽车充电操作机器人的视觉系统设计  

Design of vision system of electric vehicle charging operation robot

作　　者：徐建明[1] 胡弘历 XU Jianming;HU Hongli(College of Information Engineering,Zhejiang University of Technology,Hangzhou 310023,China)

机构地区：[1]浙江工业大学信息工程学院,浙江杭州310023

出　　处：《浙江工业大学学报》2021年第4期384-391,458,共9页Journal of Zhejiang University of Technology

基　　金：国家自然科学基金-浙江省自然科学基金联合基金两化融合项目(U1709213);国家自然科学基金面上项目(61374103)。

摘　　要：针对电动汽车充电过程中的充电插座识别和充电孔定位问题,设计了电动汽车充电操作机器人的双目视觉系统。首先,采用支持向量机(SVM)和线性可变形模板匹配相结合的方法识别快充和慢充插座、定位插座中心充电孔,结合双目立体匹配,获取中心充电孔的三维坐标;其次,建立超出机器人的工作范围和充电孔角度识别偏差过大情况下的异常处理机制;最后,在Halcon视觉平台编写视觉算法进行实验验证,结果表明:该视觉系统能够有效区分快充和慢充插座、定位目标插座中心充电孔,分类速度和定位精度满足充电机器人在线作业要求。In order to solve the problem of charging socket identification and charging hole positioning during the charging of electric vehicles,a binocular vision system of electric vehicle charging operation robot was designed.The method of combining SVM(Support vector machine)and linear deformable template matching was used to identify the fast and slow charging sockets,locate the central charging hole of the socket,and combine binocular stereo matching to obtain the three-dimensional coordinates of the center charging hole.An exception handling mechanism was established when the robot’s working range is exceeded and the charging hole angle recognition deviation is too large.Finally,a vision algorithm was written on the Halcon vision platform for experimental verification.The results showed that the vision system can effectively distinguish between fast charging socket and slow charging socket,and locate the center charging hole of the target socket.The classification speed and positioning accuracy can meet the requirements of online operation of charging robots.

关 键 词：双目视觉 充电机器人 SVM 模板匹配 

分 类 号：TP242.2[自动化与计算机技术—检测技术与自动化装置]