摘要
针对工业环境中板型物体(如钢板),在基于投票策略的匹配算法基础上加入了边缘点对特征,提出了一种基于边缘点对特征的三维目标识别与定位方法,并利用该算法设计实现了一个钢板识别与定位系统,已成功应用于工业机器人自动钢板打磨项目中。另外,对匹配结果使用位姿聚类以及位姿验证与优化,进一步提高了算法的准确性和鲁棒性。根据系统在工作现场的运行统计结果得出,该方法不仅定位精度在项目容忍范围之内,而且满足项目对实时性的要求。
Aiming at common planar objects in industrial environment(e. g. steel plates),on the basis of the voting-based pose estimation algorithm,this paper proposed an efficient planar object recognition and localization algorithm based on boundary point pair feature,which was successfully applied in the automatic grinding system of steel plates with industrial robot. In addition,the using of pose clustering algorithm as well as methods of pose verification and pose refinement efficiently increased the accuracy and robustness of the system. According to the statistical results of the system in industrial environment,not only the pose localization accuracy is well within the tolerances for the grasping of steel plates,but also the processing time meets the real-time requirement of the system.
引文
[1]郑晶怡,李恩,梁自泽.基于形状先验模型的平面型工件抓取位姿检测[J].机器人,2017,39(1):99-110.(Zheng Jingyi,Li En,Liang Zizhe. Grasping posture determination of planar workpieces based on shape prior model[J]. Robot,2017,39(1):99-110.)
[2]孙庆科,何云涛,陈瑞强,等.基于结构光的三维点云重建方法研究[J].计算机应用研究,2016,33(9):2873-2877.(Sun Qingke,He Yuntao,Chen Ruiqiang,et al. Research of three-dimensional point cloud reconstruction based on structured line[J]. Application Research of Computers,2016,33(9):2873-2877.)
[3] Liu Mingyu,Tuzel O,Veeraraghavan A,et al. Fast object localization and pose estimation in heavy clutter for robotic bin picking[J]. International Journal of Robotics Research,2012,31(8):951-973.
[4]朱海波.基于视觉引导的工业机器人工件搬运技术研究[D].沈阳:沈阳工业大学,2013.(Zhu Haibo. Research of industrial robot workpiece handling technology based on visual guidance[D]. Shenyang:Shenyang University of Technology,2013.)
[5]管经纬.基于机器视觉的工件自动分拣系统研究[D].上海:东华大学,2015.(Guan Jingwei. Research on automatic workpiece sorting system based on machine vision[D]. Shanghai:Donghua University,2015.)
[6]佐立营.面向机器人抓取的散乱零件自动识别与定位技术研究[D].哈尔滨:哈尔滨工业大学,2015.(Zuo Liying. Research on automatic identification and location of scattered parts for robot binpicking[D]. Harbin:Harbin Institute of Technology,2015.)
[7] Johnson A E,Hebert M. Using spin images for efficient object recognition in cluttered 3D scenes[J]. IEEE Trans on Pattern Analysis and Machine Intelligence,1999,21(5):433-449.
[8] Drost B,Ulrich M,Navab N,et al. Model globally,match locally:efficient and robust 3D object recognition[C]//Proc of International Conference on Computer Vision and Pattern Recognition. Piscataway,NJ:IEEE Press,2010:998-1005.
[9] Rusu R B,Blodow N,Marton Z C,et al. Aligning point cloud views using persistent feature histograms[C]//Proc of IEEE/RSJ International Conference on Intelligent Robots and Systems. 2008:3384-3391.
[10]Rusu R B,Blodow N,Beetz M. Fast point feature histograms(FPFH)for 3D registration[C]//Proc of IEEE International Conference on Robotics and Automation. Piscataway,NJ:IEEE Press,2009:3212-3217.
[11] Rusu R B,Bradski G,Thibaux R,et al. Fast 3D recognition and pose using the viewpoint feature histogram[C]//Proc of IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway,NJ:IEEE Press,2010:2155-2162.
[12]Tombari F,Salti S,Di Stefano L. Unique signatures of histograms for local surface description[C]//Proc of the 11th European Conference on Computer Vision. Berlin:Springer-Verlag,2010:356-369.
[13]Guo Yulan,Sohel F,Bennamoun M,et al. Rotational projection statistics for 3D local surface description and object recognition[J]. International Journal of Computer Vision,2013,105(1):63-86.
[14]Drost B,Ilic S. 3D object detection and localization using multimodal point pair features[C]//Proc of the 2nd International Conference on3D Imaging,Modeling,Processing,Visualization&Transmission.Washington DC:IEEE Computer Society,2012:9-16.
[15]Papazov C,Burschka D. An efficient RANSAC for 3D object recognition in noisy and occluded scenes[C]//Proc of Asian Conference on Computer Vision. Berlin:Springer,2010:135-148.
[16]Papazov C,Haddadin S,Parusel S,et al. Rigid 3D geometry matching for grasping of known objects in cluttered scenes[J]. International Journal of Robotics Research,2012,31(4):538-553.
[17]Kim E,Medioni G. 3D object recognition in range images using visibility context[C]//Proc of IEEE/RSJ International Conference on Intelligent Robots and Systems. 2011:3800-3807.
[18]Choi C,Taguchi Y,Tuzel O,et al. Voting-based pose estimation for robotic assembly using a 3D sensor[C]//Proc of IEEE International Conference on Robotics and Automation. Piscataway, NJ:IEEE Press,2012:1724-1731.
[19]Tuzel O,Liu M Y,Taguchi Y,et al. Learning to rank 3D features[C]//Proc of European Conference on Computer Vision. Berlin:Springer,2014:520-535.
[20]Birdal T,Ilic S. Point pair features based object detection and pose estimation revisited[C]//Proc of International Conference on 3D Vision. Washington DC:IEEE Computer Society,2015:527-535.
[21]倪欢,张继贤,林祥国.三维点云边缘检测和直线段提取进展与展望[J].测绘通报,2016(7):1-4.(Ni Huan,Zhang Jixian,Lin Xiangguo. Edge detection and straight line segment extraction from3D-point clouds:review and prospect[J]. Bulletin of Surveying and Mapping,2016(7):1-4.)
[22]Rusu R B,Cousins S. 3D is here:point cloud library(PCL)[C]//Proc of IEEE International Conference on Robotics and Automation.Piscataway,NJ:IEEE Press,2011:1-4.
[23]薛丽霞,李涛,王佐成.一种自适应的Canny边缘检测算法[J].计算机应用研究,2010,27(9):3588-3590.(Xue Lixia,Li Tao,Wang Zuocheng. Adaptive Canny edge detection algorithm[J]. Application Research of Computers,2010,27(9):3588-3590.)
[24] Rusinkiewicz S,Levoy M. Efficient variants of the ICP algorithm[C]//Proc of the 3rd International Conference on 3-D Digital Imaging and Modeling. Piscataway,NJ:IEEE Press,2001:145-152.