摘要
为实现板式换热器在氦气检漏工序的自动上料,设计了基于RealSense 3D相机的视觉引导系统。采用深度信息与连通域分析结合的方法找出板式换热器的最小外接矩形,并根据深度信息微调最小外接矩形的4个顶点,使其在同一深度距离平面上,减小了误差;将最小外接矩形的中心坐标作为工件坐标发送给可编程逻辑控制器(PLC)进行坐标转换,PLC将实际偏移值发送给机器人完成抓取。经实验验证,抓取4种不同型号、分布于4个工位且摆放层为1~4层的板式换热器的平均抓取成功率为98.3%;在机械手速度设定为AUTO模式下最高速的50%时,每抓取一个工件耗时59.29 s,能够满足氦气检漏工序上料的节拍需求。
In order to realize the automatic feeding of the plate heat exchanger in the helium leak detection process,a visual guidance system based on the RealSense 3D camera was designed.The combination of depth information and connected domain analysis was used to find the smallest circumscribed rectangles of the plate heat exchanger,the four vertices of the smallest circumscribed rectangles were adjusted according to the depth information,so that the four vertices were on the same depth distance plane,which could reduce the error.The center coordinates of the smallest circumscribed rectangles as product coordinates were sent to the PLC for coordinate conversion.The PLC sent the deviate value to the manipulator to complete the grab.After experimental verification,the success rate of grab was 98.3%when the manipulator grabbed 4 different types of plate heat exchangers from 1~4 layers distributing in 4 stations.When the robot speed was set to 50%of the highest speed in AUTO mode,each grab takes 59.29 s,which can meet the production beat demand of helium leak detection.
作者
王浩
曹松晓
谢代梁
刘铁军
徐志鹏
徐雅
WANG Hao;CAO Songxiao;XIE Dailiang;LIU Tiejun;XU Zhipeng;XU Ya(College of Metrology and Measurement Engineering,China Jiliang University,Hangzhou 310018,China)
出处
《现代制造工程》
CSCD
北大核心
2021年第10期65-73,32,共10页
Modern Manufacturing Engineering
基金
浙江省自然科学基金项目(LY18E050009)。
关键词
上料机器人
深度相机
视觉引导
连通域分析
feeding robot
depth camera
visual guidance
connected domain analysis
