|
|
Assistance localization method for mobile robot based on
monocular natural visual landmarks |
CHEN Ming-ya1,2, XIANG Zhi-yu1,2, LIU Ji-lin1,2 |
1. Department of Information Science and Electronic Engineering, Zhejiang University, Hangzhou 310027, China;
2. Zhejiang Provincial Key Laboratory of Information Network Technology, Hangzhou 310027, China |
|
|
Abstract In many occasions GPS signal may be blocked, leading the quick drop of the positioning accuracy for the robot. We present an assistance localization method for mobile robot based on the monocular natural visual landmarks. A landmark library containing the images from several scenes of the environment was set up before navigation. Each acquired image was matched with the visual landmarks while navigation, where INS positioning was used for rough localization. A fast image matching framework based on combining usage of GIST global features and SURF local features was presented. The Orientation was corrected from Structure From Motion algorithm as well. Finally a Kalman Filter was used to fuse the localization results from visual landmarks and the INS method. The results show that the proposed method improves the positioning accuracy under GPS blocked area and makes the system more robust.
|
Published: 01 February 2014
|
|
单目视觉自然路标辅助的移动机器人定位方法
针对很多场合下GPS信号会受到遮挡而无法使用,导致机器人定位精度下降很快的问题,提出一种基于单目视觉自然路标辅助的机器人绝对定位方法.在导航环境中的若干位置预先建立视觉路标库.机器人在利用惯导(INS)定位过程中,同时对采集到的单目图像和库中的视觉路标进行匹配.建立基于全局特征信息(GIST)和快速鲁棒算子(SURF)局部特征相结合的在线图像快速匹配框架,同时结合基于单目视觉的运动估计算法修正车体航向.最后利用卡尔曼滤波将视觉路标匹配获得的定位信息和INS有效地融合起来.结果表明,该方法有效地提高在GPS受限情况下惯性导航定位的精度和鲁棒性.
|
|
[1] NISTER D, NARODITSKY O, BERGEN J. Visual odometry [C]∥Pro-ceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, DC:[s.n.], 2004: 652-659.
[2] SCARAMUZZA D, SIEGWART R. Appearance-guided monocul-ar omnidirectional visual odometry for outdoor ground veh-icles[J].IEEE Transactions on Robotics, 2008,24(5): 1015-1026.
[3] BAILEY T. Mobile robot localization and mapping in exte-nsive outdoor environments[D].Sydney, Australia: Universityof Sydney,2002.
[4] FRESE U, HIRZINGER G. Simultaneous localization and mapping a discussion [C]∥In Proceedings of the IJCAI Wor-kshop on Reasoning with Uncertainty in Robotics. Seattle:[s.n.]2001: 17-26.
[5] HUSTER A, ROCK S. Relative position sensing by fusing monocular vision and inertial rate sensors [C]∥ in Proceedi-ngs of 11th International Conference Advanced Robotics. Coimbra, Portugal:[s.n.]. 2003: 1562-1567.
[6] STRELOW D, SINGH S. Motion estimation from image and inertial measurements [J]. International Journal of Robotics Research, 2004,23(12): 1157-1195.
[7] CUMMINNS M, NEWMAN P. FAB-MAP: probabilistic loca-lization and mapping in the space of appearance[J]. Inte-rnational Journal of Robotics Research, 2008,27(6): 647-665.
[8] IKEDA K, TANAKA K. Visual robot localization using co-mpact binary landmarks [C]∥Proceedings of the IEEE ICRA. Florida: [s.n]. 2010: 43974403.
[9] QUDDUS M A, OCHIENG W Y, NOLAND R B. Current ma-pmatching algorithms for transport applications: State-of-theart and future research directions [J]. Transportation Researc-h Part C, 2007,15(5): 312328.
[10] PARK C, KIM S, PARK S K. Vision-based global local-ization for mobile robots with hybrid maps of objects and spatial layouts[J]. Information Science, 2009,179(24): 41744198.
[11] LOVESKY I, SHIMSHONI I. Reliable and efficient landma-rkbased localization for mobile robots [J]. Robotics and Aut-onomous Systems, 2010,58(5): 520-528.
[12] OLIVA A. TORRALBA A. Modeling the shape of the scene: aholistic representation of the spatial envelope[J]. Internatio-nal Journal of Computer Vision,2001,42[3]: 145-175.
[13] BAY H, TUYTELAARS T, VAN GOOL L J. SURF: speeded up robust features[C]∥In Proc.ECCV. Austria:[s.n].2006,10(3): 404-417.
[14] LOWE D G. Distinctive image features from scale-invariant keypoints [J]. International Journal of Computer Vision,2004,60(2): 91-110.
[15] REMONDINO F, FRASER C. Digital camera calibration me-thods: considerations and comparisons [C]∥In Proc.ISP-RS Commission V Symposium. Dresden:[s.n.]. 2006: 266-272.
[16] KOENDERINK J, DORRN A V. Affine structure from moti-on [J]. Journal of the Optical Society of America A, 1991, 8(2): 337-385.
[17] FISCHLER M A, BOLLES R C. Random sample consens-us: a paradigm for model fitting with applications to image analysis and automated cartography [J].Communicat-ion of the ACM, 1981,24: 381-385.
[18] NOCEDAL, WRIGHT S J. Numerical optimization[M]. New York: Springer,1999: 200-263.
[19] GOLUB G H, REINSCH C. Singular value decomposition and least squares solutions [J],Numerical Mathematics, 1970,14: 403420.
[20] CHUL W K, CHAN G P. Attitude estimationwith accelerometers and gyros using fuzzy tuned kalman filter[C]∥In Proceedings of the European Control Conference. Budapest, Hungary:[s.n]. 2009: 3713-3718. |
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
|
Shared |
|
|
|
|
|
Discussed |
|
|
|
|