一种低成本的机器人室内可通行区域建模方法

doi:10.13700/j.bh.1001-5965.2019.0393

北京航空航天大学学报 ›› 2019, Vol. 45 ›› Issue (12): 2470-2478.doi: 10.13700/j.bh.1001-5965.2019.0393

一种低成本的机器人室内可通行区域建模方法

张釜恺¹, 芮挺², 何雷², 杨成松²

1. 中国人民解放军陆军工程大学研究生院, 南京 210000;
2. 中国人民解放军陆军工程大学野战工程学院, 南京 210000

收稿日期:2019-07-16 出版日期:2019-12-20 发布日期:2019-12-31
通讯作者: 芮挺 E-mail:rtinguu@sohu.com
作者简介:张釜恺男,硕士研究生。主要研究方向:计算机视觉、VSLAM;芮挺男,博士,教授,硕士生导师。主要研究方向:图像处理、模式识别、人工智能;何雷男,博士,讲师,硕士生导师。主要研究方向:系统科学,智能算法;杨成松男,博士,讲师。主要研究方向:图像处理、虚拟现实。

A low-cost indoor passable area modeling method for robots

ZHANG Fukai¹, RUI Ting², HE Lei², YANG Chengsong²

1. School of Graduate, Army Engineering University of PLA, Nanjing 210000, China;
2. School of Field Engineering, Army Engineering University of PLA, Nanjing 210000, China

Received:2019-07-16 Online:2019-12-20 Published:2019-12-31

摘要/Abstract

摘要： 基于单目视觉的同步定位与建图（SLAM）是机器人领域中的一项热门技术。然而，在场景建图方面，由于其计算量较大，各主流方法还无法在低运算能力的平台上实现实时的场景建模。针对室内环境与小型机器人的特定情况，提出了一种新的可通行区域建模方法。该方法建立在单目特征点SLAM的基础上，通过HSV色彩空间内的图像自适应阈值分割获取地面分割图像，并与SLAM生成的稀疏点云进行交叉比对，进而获取地平面与准确的地面分割区域，再将地面分割区域反投影到地平面上，获取地面的稠密建模。在室内场景的实验中，所提方法的平均运算速度能达到21帧/s，速度约为ORB-SLAM的70%，能够满足移动平台的实时性要求。对于地平面位置的还原平均误差为5.8%，地面上道路宽度的建模误差在3.5%~12.8%。

关键词: 单目同步定位与建图, 室内场景, 可通行区域, 地面建模, 图像分割

Abstract: Monocular vision-based simultaneous localization and mapping (SLAM) is a popular technology in the field of robotics in recent years. However, due to the huge computation resource required by reconstruction, mainstream methods are not able to generate meaningful reconstruction of scene in real time on platforms with low computing power. This paper proposes a new fast passable area modeling method for the specific situation of indoor environment and small robots. The method is based on the monocular feature-based SLAM. Firstly, it obtains the road segmentation image through segmentation in the HSV color space with adaptive threshold. Then, the system cross-matches the segmentation with the sparse point cloud generated by SLAM, to obtain the ground plane and accurate ground segmentation area. Finally, it projects the ground segmentation area to the ground plane for dense modeling of the floor. In the experiment of indoor scene, the average calculation speed of the proposed method can reach 21 frames per second, and the speed is about 70% of ORB-SLAM, which can meet the real-time requirements of mobile platforms. The position error for the floor plane is 5.8% on average, and the modeling error of the road width is between 3.5% and 12.8%.

Key words: monocular simultaneous localization and mapping, indoor scene, passable area, floor modeling, image segmentation

中图分类号:

TP242
TP37

张釜恺, 芮挺, 何雷, 杨成松. 一种低成本的机器人室内可通行区域建模方法[J]. 北京航空航天大学学报, 2019, 45(12): 2470-2478.

ZHANG Fukai, RUI Ting, HE Lei, YANG Chengsong. A low-cost indoor passable area modeling method for robots[J]. JOURNAL OF BEIJING UNIVERSITY OF AERONAUTICS AND A, 2019, 45(12): 2470-2478.

参考文献

[1] LI Z,TANG J.Weakly supervised deep metric learning for community-contributed image retrieval[J].IEEE Transactions on Multimedia,2015,17(11):1989-1999.
[2] LIU H,ZHANG G,BAO H.A survey of monocular simultaneous localization and mapping[J].Journal of Computer-Aided Design & Computer Graphics,2016,28(6):855-868.
[3] TURAN M,ALMALIOGLU Y,ARAUJO H,et al.A non-rigid map fusion-based direct SLAM method for endoscopic capsule robots[J].International Journal of Intelligent Robotics and Applications,2017,1(4):399-409.
[4] JAN S,GUMHOLD S,CREMERS D.Real-time dense geometry from a handheld camera[C]//Proceedings of the Pattern Recognition-32nd DAGM Symposium.Berlin:Springer,2010:22-24.
[5] ENGEL J,KOLTUN V,CREMERS D.Direct sparse odometry[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2016,40(3):611-625.
[6] KIM J H,CADENA C,REID I.Direct semi-dense SLAM for rolling shutter cameras[C]//Proceedings of the IEEE International Conference on Robotics and Automation.Piscataway,NJ:IEEE Press,2016:1308-1315.
[7] BAILEY T,NIETO J,GUIVANT J,et al.Consistency of the EKF-SLAM algorithm[C]//Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).Piscataway,NJ:IEEE Press,2006:9-15.
[8] PIRKER K,MATTHIAS R,BISCHOF H.CD-SLAM-Continuous localization and mapping in a dynamic world[C]//Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems(IROS).Piscataway,NJ:IEEE Press,2011:25-30.
[9] GEE A P.MAYOL-CUEVAS W.Real-time model-based SLAM using line segments[C]//Proceedings of the International Symposium on Visual Computing.Berlin:Springer,2006:354-363.
[10] NEWCOMBE R A,LOVEGROVE S J,DAVISON A J.DTAM:Dense tracking and mapping in real-time[C]//Proceedings of the IEEE International Conference on Computer Vision (ICCV).Piscataway,NJ:IEEE Press,2011:6-13.
[11] ENGEL J,SCHÖPS T,CREMERS D.LSD-SLAM:Large-scale direct monocular SLAM[C]//Proceedings of the European Conference on Computer Vision (ECCV).Berlin:Springer,2014:834-849.
[12] TRIGGS B,MCLAUCHLAN P F,HARTLEY R I,et al.Bundle adjustment a modern synthesis[C]//Proceedings of the Vision Algorithms:Theory and Practice.Berlin:Springer,2000:298-372.
[13] ULRICH I,NOURBAKHSH I R.Appearance-based place recognition for topological localization[C]//Proceedings of the International Conference on Robotics and Automation.Piscataway,NJ:IEEE Press,2000:1023-1029.
[14] DAVISON A J,REID I D,MOLTON N D,et al.MonoSLAM:Real-time single camera SLAM[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2007,29(6):1052-1067.
[15] KLEIN G,MURRAY D.Parallel tracking and mapping for small AR workspaces[C]//Proceedings of the IEEE and ACM International Symposium on Mixed and Augmented Reality.Piscataway,NJ:IEEE Press,2007:225-234.
[16] MUR-ARTAL R,MONTIEL J M M,TARDÓS J D.ORB-SLAM:A versatile and accurate monocular SLAM System[J].IEEE Transactions Robotics,2015,31(5):1147-1163.
[17] NEWCOMBE R A,DAVISON A J.Live dense reconstruction with a single moving camera[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Piscataway,NJ:IEEE Press,2010:1498-1505.
[18] MURARTAL R,TARDOS J D.Probabilistic semi-dense mapping from highly accurate feature-based monocular SLAM[C]//Proceedings of the Robotics Science and Systems,2015.
[19] HINZMANN T,SCHNEIDER T,DYMCZYK M,et al.Robust map generation for fixed-wing UAVs with low-cost highly-oblique monocular cameras[C]//Proceedings of the IEEE/RSJ International Conference on Intelligent Robots & Systems.Piscataway,NJ:IEEE Press,2016:3261-3268.
[20] 蒙山,唐文名.单目SLAM直线匹配增强平面发现方法[J].北京航空航天大学学报,2017,43(4):660-666.MENG S,TANG W M.Monocular SLAM linear matching enhanced plane discovery method[J].Journal of Beijing University of Aeronautics and Astronautics,2017,43(4):660-666(in Chinese).
[21] VON STUMBERG L,USENKO V,ENGEL J,et al.Autonomous exploration with a low-cost quadrocopter using semi-dense monocular SLAM[C]//Processing of the European Conference on Mobile Robots,2017.
[22] 蒋林,郭晨,朱志超.嵌入式平台上的三维重建算法研究[J].机械设计与制造,2018,330(8):264-266.JIANG L,GUO C,ZHU Z C.Research on 3D reconstruction algorithm based on embedded platform[J].Machinery Design & Manufacture,2018,330(8):264-266(in Chinese).
[23] BADRINARAYANAN V,KENDALL A,CIPOLLA R.SegNet:A deep convolutional encoder-decoder architecture for scene segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(12):2481-2495.
[24] 黄寅.基于HSV颜色空间与形态学的车辆目标分割算法[J].大庆师范学院学报,2017,37(6):11-15.HUANG Y,Vehicle target segmentation algorithm based on HSV color space and morphology[J].Journal of Daqing Normal University,2017,37(6):11-15(in Chinese).
[25] 闫敬文.数字图像处理:MATLAB版[M].北京:国防工业出版社,2011:125-127.YAN J W.Digital image processing by MATLAB[M].Beijing:National Defence Industry Press,2011:125-127(in Chinese).
[26] 李鹏飞,吴海娥,景军锋,等.点云模型的噪声分类去噪算法[J].计算机工程与应用,2016,52(20):188-192.LI P F,WU H E,JING J F,et al.Noise classification and denoising algorithm for point cloud model[J].Computer Engineering and Applications,2016,52(20):188-192(in Chinese).

一种低成本的机器人室内可通行区域建模方法

A low-cost indoor passable area modeling method for robots

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 9

编辑推荐

Metrics

本文评价

[1]	许冰, 牛燕雄, 吕建明. 复杂动态场景下目标检测与分割算法[J]. 北京航空航天大学学报, 2016, 42(2): 310-317.
[2]	钱华明, 姜波. 一种基于多通道卫星云图的台风分割方法[J]. 北京航空航天大学学报, 2011, 37(4): 466-471.
[3]	王晓亮, 李春升. 无需先验信息的水平集SAR图像分割方法[J]. 北京航空航天大学学报, 2010, 36(7): 841-844.
[4]	刘向华, 周荫清, 孙慕涵. 基于快速退火MRF的改进SAR图像分割方法[J]. 北京航空航天大学学报, 2010, 36(6): 719-722.
[5]	李苏祺, 张广军. 基于邻接表的分水岭变换快速区域合并算法[J]. 北京航空航天大学学报, 2008, 34(11): 1327-1330.
[6]	吴央, 袁运能. 基于小波包分解和FCM聚类的纹理图像分割方法[J]. 北京航空航天大学学报, 2008, 34(05): 572-575.
[7]	袁运能,吴央,成功. 核空间聚类在图像纹理分类中的简化算法[J]. 北京航空航天大学学报, 2008, 34(03): 267-270.
[8]	李超, 熊璋, 朱成军. 基于距离相关图的音频相似性度量方法[J]. 北京航空航天大学学报, 2006, 32(02): 224-227.
[9]	燕英, 周荫清. 基于分割途径的SAR单视图像斑点噪声抑制方法[J]. 北京航空航天大学学报, 2003, 29(2): 132-135.