Error aware multiple vertical planes based visual localization for mobile robots in urban environments

详细信息    查看全文

• 作者：HaiFeng Li (1)
  HongPeng Wang (2) (3)
  JingTai Liu (2)
  
  1. College of Computer Science and Technology ; Civil Aviation University of China ; Tianjin ; 300300 ; China
  2. Institute of Robotics and Automatic Information System ; Nankai University ; Tianjin ; 300071 ; China
  3. State Key Laboratory of Robotics ; Shenyang ; 110016 ; China
  
• 关键词：visual localization ; multiple vertical planes ; error aware ; convex optimization ; satellite images ; urban environment ; mobile robot ; 瑙嗚瀹氫綅 ; 澶氱珫鐩村钩闈?/li> 璇樊璁ょ煡 ; 鍑镐紭鍖?/li> 鍗槦鍥惧儚 ; 鍩庡競鐜 ; 绉诲姩鏈哄櫒浜?/li> 032203
• 刊名：SCIENCE CHINA Information Sciences
• 出版年：2015
• 出版时间：March 2015
• 年：2015
• 卷：58
• 期：3
• 页码：1-14
• 全文大小：1,234 KB
• 参考文献：1. Garro V, Galassi M, Fusiello A. Visual localization for mobile surveillance. In: Proceedings of the IEEE International Conference on Computer Vision, Barcelona, 2011. 80鈥?1
  2. Huang K, Tseng S, Mou W, et al. Simultaneous localization and scene reconstruction with monocular camera. In: Proceedings of the IEEE International Conference on Robotics and Automation, St. Paul, 2012. 2102鈥?107
  3. Lowe D G. Distinctive image features from scale-invariant keypoints. Intern J Comp Vision, 2004, 60: 91鈥?10 CrossRef
  4. Santosh D, Achar S, Jawahar C. Autonomous image-based exploration for mobile robot navigation. In: Proceedings of the IEEE International Conference on Robotics and Automation, Pasadena, 2008. 2717鈥?722
  5. Cummins M, Newman P. FAB-MAP: Probabilistic localization and mapping in the space of appearance. Intern J Robotics Res, 2008, 27: 647鈥?65 CrossRef
  6. Larlus D, Verbeek J, Jurie F. Category level object segmentation by combining bag-of-words models with dirichlet processes and random fields. Intern J Comp Vision, 2010, 88: 238鈥?53 CrossRef
  7. Wu L, Hoi S, Yu N. Semantics-preserving bag-of-words models and applications. IEEE Trans Image Proc, 2010, 19: 1908鈥?920 CrossRef
  8. Davison A, Reid L, Molton N, et al. Monoslam: Real-time single camera slam. IEEE Trans Pattern Anal Machine Intellig, 2007, 29: 1052鈥?067 CrossRef
  9. Gil A, Mozos O, Ballesta M, et al. A comparative evaluation of interest point detectors and local descriptors for visual SLAM. Machine Vision Appl, 2010, 21: 905鈥?20 CrossRef
  10. Lemaire T, Lacroix S. Monocular-vision based SLAM using line segments. In: Proceedings of the IEEE International Conference on Robotics and Automation, Roma, 2007. 2791鈥?796
  11. Elqursh A, Elgammal A. Line-based relative pose estimation. In: Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition, Colorado Springs, 2011. 3049鈥?056
  12. Delmerico J, David P, Adelphi M, et al. Building facade detection, segmentation, and parameter estimation for mobile localization and guidance. In: Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, San Francisco, 2011. 1632鈥?639
  13. Cham T, Ciptadi A, Tan W, et al. Estimating camera pose from a single urban ground-view omnidirectional image and a 2D building outline map. In: Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition, San Francisco, 2010. 366鈥?73
  14. Gee A, Chekhlov D, Calway A, et al. Discovering higher level structure in visual slam. IEEE Trans Robotics, 2008, 24: 980鈥?90 CrossRef
  15. Martinez-Carranza J, Calway A. Unifying planar and point mapping in monocular slam. In: Proceedings of the British Machine Vision Conference, Aberystwyth, 2010. 43.1鈥?3.11
  16. Leung K, Clark C, Huissoon J. Localization in urban environments by matching ground level video images with an aerial image. In: Proceedings of the IEEE International Conference on Robotics and Automation, Pasadena, 2008. 551鈥?56
  17. Bansal M, Sawhney H, Cheng H, et al. Geo-localization of street views with aerial image databases. In: Proceedings of the 19th ACM International Conference on Multimedia, Scottsdale, 2011. 1125鈥?128 CrossRef
  18. Selnet T, Elgammal A. Satellite image based precise robot localization on sidewalks. In: Proceedings of the IEEE International Conference on Robotics and Automation. St. Paul, 2012. 2647鈥?653
  19. Kummerle R, Steder B, Dornhege C, et al. Large scale graph-based SLAM using aerial images as prior information. Auton Robots, 2011, 30: 25鈥?9 CrossRef
  20. Muramatsu S, Tomizawa T, Matsuda H, et al. Mobile robot localization technique using Web-based aerial photos. In: Proceedings of the IEEE International Conference on Mechatronics and Automation, Chengdu, 2012. 1886鈥?891
  21. Li H, Song D, Lu Y, et al. A two-view based multilayer feature graph for robot navigation. In: Proceedings of the IEEE International Conference on Robotics and Automation, St. Paul, 2012. 3580鈥?587
  22. Li H, Liu J, Lu X. Visual localization in urban area using orthogonal building boundaries and a GIS database. ROBOT, 2012, 34: 604鈥?13
  23. Hartley R, Zisserman A. Multiple View Geometry in Computer Vision, 2nd Ed. Cambridge: Cambridge University Press, 2004. 30鈥?5 CrossRef
  
• 刊物类别：Computer Science
• 刊物主题：Chinese Library of Science
  Information Systems and Communication Service
  
• 出版者：Science China Press, co-published with Springer
• ISSN：1869-1919

文摘

A novel error-aware visual localization method is proposed that utilizes vertical planes, such as vertical building facades in urban areas as landmarks. Vertical planes, reconstructed from coplanar vertical lines, are robust high-level features if compared with point features or line features. Firstly, the error models of vertical lines and vertical planes are built, where maximum likelihood estimation (MLE) is employed to estimate all vertical planes from coplanar vertical lines. Then, the closed-form representation of camera location error variance is derived. Finally, the minimum variance camera pose estimation is formulated into a convex optimization problem, and the weight for each vertical plane is obtained by solving this well-studied problem. Experiments are carried out and the results show that the proposed localization method has an accuracy of about 2 meters, at par with commercial GPS operating in open environments.