基于完整可见性模型的改进鲁棒OctoMap

doi:10.11772/j.issn.1001-9081.2017.05.1445

计算机应用 ›› 2017, Vol. 37 ›› Issue (5): 1445-1450.DOI: 10.11772/j.issn.1001-9081.2017.05.1445

• 计算机视觉与虚拟现实 • 上一篇下一篇

基于完整可见性模型的改进鲁棒OctoMap

刘晙¹, 袁培燕², 李永锋³

1. 河南工学院计算机科学与技术系, 河南新乡 453002;
2. 河南师范大学计算机与信息工程学院, 河南新乡 453007;
3. 96367部队, 青海德令哈 817099

收稿日期:2016-10-13 修回日期:2017-01-10 出版日期:2017-05-10 发布日期:2017-05-16
通讯作者: 刘晙
作者简介:刘晙(1981-),男,湖北天门人,讲师,硕士,主要研究方向:移动机器人视觉定位与地图创建;袁培燕(1978-),男,河南邓州人,副教授,博士,CCF会员,主要研究方向:移动机会网络;李永锋(1989-),男,河南焦作人,硕士,主要研究方向:移动机器人视觉定位与地图创建。
基金资助:
国家自然科学基金资助项目（U1404602）；河南省高等学校青年骨干教师计划项目（2011GGJS-198）；河南省教育厅科学技术重点研究项目（14A520045）。

Improved robust OctoMap based on full visibility model

LIU Jun¹, YUAN Peiyan², LI Yongfeng³

1. Department of Computer Science & Technology, Henan Institute of Technology, Xinxiang Henan 453002, China;
2. College of Computer & Information Engineering, Henan Normal University, Xinxiang Henan 453007, China;
3. Troops 96367, Delingha Qinghai 817099, China

Received:2016-10-13 Revised:2017-01-10 Online:2017-05-10 Published:2017-05-16
Supported by:
This work is partially suppotted by the National Natural Science Foundation of China (U1404602), the Funded Project of Key Teachers in Colleges and Universities of Henan Province (2011GGJS-198),the Science and Technology key research project of Education Department of Henan Province (14A520045).

摘要/Abstract

摘要： 从移动机器人自主导航对3D地图精度的需求出发，在鲁棒OctoMap的基础上提出一种基于完整可见性模型的改进鲁棒OctoMap并应用于基于Kinect的RGB-D同时定位与地图创建（SLAM）中。首先，通过考虑相机和目标体素的相对位置关系及地图分辨率进行可连通性判断，获得满足可连通性的相邻体素的个数及位置；其次，根据不同的可连通性情况分别建立目标体素的可见性模型，从而构建普适性更强的完整可见性模型，有效克服了鲁棒OctoMap可见性模型的局限性，提高了建图精度；再次，使用基于高斯混合模型的Kinect深度误差模型代替简单深度误差模型，进一步克服传感器测量误差对地图精度的影响，降低了地图的不确定性；最后，结合贝叶斯公式和线性插值算法来更新八叉树中每个节点的实际占用概率，从而构建基于八叉树的立体占用地图。实验结果表明，所提方法有效克服了Kinect传感器深度误差对地图精度的影响，降低了地图的不确定性，其建图精度较鲁棒OctoMap有明显的提高。

关键词: 3D地图, 鲁棒OctoMap, 高斯混合模型, 八叉树, 立体占用地图

Abstract: An improved robust OctoMap based on full visibility model was proposed to meet accuracy needs of 3D map for mobile robot autonomous navigation and it was applied to the RGB-D SLAM (Simultaneous Localization And Mapping) based on Kinect. First of all, the connectivity was judged by considering the the relative positional relationship between the camera and the target voxel and the map resolution to get the number and the location of adjacent voxels which met connectivity. Secondly, according to the different connectivity, the visibility model of the target voxel was built respectively to establish the full visibility model which was more universal. The proposed model could effectively overcome the limitations of the robust OctoMap visibility model, and improve the accuracy. Next, the simple depth error model was replaced by the Kinect sensor depth error model based on Gaussian mixture model to overcome the effect of the sensor measurement error on the accuracy of map further and reduce the uncertainty of the map. Finally, the Bayesian formula and linear interpolation algorithm were combined to update the occupancy probability of each node in the octree to build the volumetric occupancy map based on a octree. The experimental results show that the proposed method can effectively overcome the influence of Kinect sensor depth error on map precision and reduce the uncertainty of the map, and the accuracy of map is improved obviously compared with the robust OctoMap.

Key words: 3D map, robust OctoMap, Gaussian mixture model, octree, volumetric occupancy map

中图分类号:

TP242

刘晙, 袁培燕, 李永锋. 基于完整可见性模型的改进鲁棒OctoMap[J]. 计算机应用, 2017, 37(5): 1445-1450.

LIU Jun, YUAN Peiyan, LI Yongfeng. Improved robust OctoMap based on full visibility model[J]. Journal of Computer Applications, 2017, 37(5): 1445-1450.

参考文献

[1] KOCH R, MAY S, MURMANN P, et al. Identification of transparent and specular reflective material in laser scans to discriminate affected measurements for faultless robotic SLAM[J]. Robotics and Autonomous Systems, 2016, 87:296-312.
[2] 宋宇, 李庆玲, 康轶非, 等. 平方根容积Rao-Blackwillised粒子滤波SLAM算法[J]. 自动化学报, 2014, 40(2):357-367.(SONG Y, LI Q L, KANG Y F, et al. SLAM with square-root cubature Rao-Blackwillised particle filter[J]. Acta Automatica Sinica, 2014, 40(2):357-367.)
[3] FOSSEL J D, TUYLS K, STURM J. 2D-SDF-SLAM:a signed distance function based SLAM frontend for laser scanners[C]//Proceedings of the 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway, NJ:IEEE, 2015:1949-1955.
[4] 张文玲, 朱明清, 陈宗海. 基于强跟踪UKF的自适应SLAM算法[J]. 机器人, 2010, 32(2):190-195.(ZHANG W L, ZHU M Q, CHEN Z H. An adaptive SLAM algorithm based on strong tracking UKF[J]. Robot, 2010, 32(2):190-195.)
[5] 张毅, 汪龙峰, 余佳航. 基于深度信息的移动机器人室内环境三维地图创建[J]. 计算机应用, 2014, 34(12):3438-3440.(ZHANG Y, WANG L F, YU J H. Depth-image based 3D map reconstruction of indoor environment for mobile robots[J]. Journal of Computer Applications, 2014, 34(12):3438-3440.)
[6] HENRY P, KRAININ M, HERBST E, et al. RGB-D mapping:using depth cameras for dense 3D modeling of indoor environments[C]//Proceedings of the 12th International Symposium on Experimental Robotics. Berlin:Springer, 2010:22-25.
[7] 王可, 贾松敏, 徐涛, 等. 基于混合位姿估计模型的移动机器人三维地图创建方法[J]. 控制与决策, 2015, 30(8):1504-1508.(WANG K, JIA S M, XU T, et al. Mobile robot 3D map building based on hybrid pose estimation model[J]. Control and Decision, 2015, 30(8):1504-1508.)
[8] BRIALES J, GONZÁLEZ-JIMÉNEZ J. Fast global optimality verification in 3D SLAM[EB/OL].[2016-06-20].http://www.riuma.uma.es/xmlui/bitstream/handle/10630/12249/main.pdf?sequence=3.
[9] ENGELHARD N, ENDRES F, HESS J, et al. Real-time 3D visual SLAM with a hand-held RGB-D camera[EB/OL].[2016-06-20].http://vision.in.tum.de/_media/spezial/bib/engelhard11euron.pdf.
[10] FIORAIO N, KONOLIGE K. Real-time visual and point cloud SLAM[EB/OL].[2016-06-20].http://dosen.narotama.ac.id/wp-content/uploads/2015/02/Realtime-visual-and-point-cloud-slam.pdf.
[11] HERBERT M, CAILLAS C, KROTKOV E, et al. Terrain mapping for a roving planetary explorer[C]//Proceedings of the 1989 IEEE International Conference on Robotics and Automation. Piscataway, NJ:IEEE, 1989:997-1002.
[12] BAIRD C A. Elevation map-referenced mechanism for updating vehicle navigation system estimates:U.S. Patent 4939663[P]. 1990-07-03.
[13] MARTIN C, THRUN S. Real-time acquisition of compact volumetric 3D maps with mobile robots[C]//Proceedings of the 2002 IEEE International Conference on Robotics and Automation. Piscataway, NJ:IEEE, 2002:311-316.
[14] LI X, GUO X, WANG H, et al. Harmonic volumetric mapping for solid modeling applications[C]//Proceedings of the 2007 ACM Symposium on Solid and Physical Modeling. New York:ACM, 2007:109-120.
[15] PAILLÉ G P, POULIN P. As-conformal-as-possible discrete volumetric mapping[J]. Computers & Graphics, 2012, 36(5):427-433.
[16] WURM K M, HORNUNG A, BENNEWITZ M, et al. OctoMap:a probabilistic, flexible, and compact 3D map representation for robotic systems[EB/OL].[2016-06-20].http://first-mm.eu/files/wurm10octomap.pdf.
[17] HORNUNG A, WURM K M, BENNEWITZ M, et al. OctoMap:An efficient probabilistic 3D mapping framework based on octrees[J]. Autonomous Robots, 2013, 34(3):189-206.
[18] SCHAUWECKER K, ZELL A. Robust and efficient volumetric occupancy mapping with an application to stereo vision[C]//Proceedings of the 2014 IEEE International Conference on Robotics and Automation. Piscataway, NJ:IEEE, 2014:6102-6107.
[19] KHOSHELHAM K, ELBERINK S O. Accuracy and resolution of Kinect depth data for indoor mapping applications[J]. Sensors, 2012, 12(2):1437-1454.
[20] DRYANOVSKI I, VALENTI R G, XIAO J. Fast visual odometry and mapping from RGB-D data[C]//Proceedings of the 2013 IEEE International Conference on Robotics and Automation. Piscataway, NJ:IEEE, 2013:2305-2310.

基于完整可见性模型的改进鲁棒OctoMap

Improved robust OctoMap based on full visibility model

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	钟文彬, 孙思, 李旭瑞, 刘光帅. 结合密度阈值和三角形组逼近的点云压缩方法[J]. 计算机应用, 2020, 40(7): 2059-2068.
[2]	赵宏, 刘向东, 杨永娟. 基于RGB-D图像的室内机器人同时定位与地图构建[J]. 计算机应用, 2020, 40(12): 3637-3643.
[3]	陈聿, 田博今, 彭云竹, 廖勇. 联合手肘法和期望最大化的高斯混合聚类电力系统客户分群算法[J]. 计算机应用, 2020, 40(11): 3217-3223.
[4]	席志红, 王洪旭, 韩双全. 基于ORB-SLAM2系统的快速误匹配剔除算法与地图构建[J]. 计算机应用, 2020, 40(11): 3289-3294.
[5]	彭磊, 杨秀云, 张裕飞, 李光耀. 基于全局与局部相似性测度的非刚性点集配准[J]. 计算机应用, 2019, 39(10): 3028-3033.
[6]	吕宏武, 付俊强, 王慧强, 李冰洋, 袁泉, 陈诗军, 陈大伟. 基于八叉树的三维室内地图数据快速检索方法[J]. 计算机应用, 2019, 39(1): 82-86.
[7]	林朗, 王让定, 严迪群, 李璨. 基于修正倒谱特征的回放语音检测算法[J]. 计算机应用, 2018, 38(6): 1648-1652.
[8]	喻新荣, 李志华, 闫成雨, 李双俐. 云数据中心高效的虚拟机整合方法[J]. 计算机应用, 2018, 38(2): 550-556.
[9]	陶志勇, 刘晓芳, 王和章. 融合密度峰值的高斯混合模型聚类算法[J]. 计算机应用, 2018, 38(12): 3433-3437.
[10]	陈文兵, 管正雄, 陈允杰. 基于条件生成式对抗网络的数据增强方法[J]. 计算机应用, 2018, 38(11): 3305-3311.
[11]	陈艳, 严腾, 宋俊芳, 宋焕生. 基于高斯混合模型和AdaBoost的夜间车辆检测[J]. 计算机应用, 2018, 38(1): 260-263.
[12]	黄亮, 潘平, 周超. 基于量子隧穿效应的说话人真伪鉴别方法[J]. 计算机应用, 2017, 37(9): 2617-2620.
[13]	杜振林, 唐卫清, 覃力, 李士才. 基于八叉树的大规模流程工厂模型细节裁剪算法[J]. 计算机应用, 2017, 37(9): 2626-2630.
[14]	李俊山, 杨亚威, 朱子江, 张姣. 基于自然图像块相似性和稀疏先验性的图像复原[J]. 计算机应用, 2017, 37(8): 2319-2323.
[15]	张海艳, 高尚兵. 图像分割中改进空间约束贝叶斯网络模型的应用[J]. 计算机应用, 2017, 37(3): 823-826.