3D Simultaneous Localization and Mapping Based on RGB-D Images

doi:10.16182/j.issn1004731x.joss.201711033

Abstract

Abstract: To reduce the accumulated pose error of robots during the three-dimensional simultaneous localization and mapping, a global optimization method is proposed to improve the positioning accuracy and the quality of the map. This method, which is based on the visual odometry of the frame and frame registration model, adds the pose-constraints by closed-loop detection based on image matching. Local loop is combined with random loop to improve the optimization efficiency. The general graph optimization algorithm is used to globally optimize the robot poses. A key-frame selection strategy is also proposed to decrease the consumption of the computing resources and memory footprint. The experiment results show that this method can reduce the root mean square error to only 8.7mm with a 3.96 m path and generate 3D map of indoor scenes accurately.

Key words: SLAM, visual odometry, loop-detection, key-frame, general graph optimization

CLC Number:

TP391.41

Hu Lingyan, Cao Lu, Xiong Pengwen, Xin Yong, Xie Zekun. 3D Simultaneous Localization and Mapping Based on RGB-D Images[J]. Journal of System Simulation, 2017, 29(11): 2840-2846.

References

[1] 刘艳丽. 融合颜色和深度信息的三维同步定位与地图构建研究[D]. 中南大学, 2014.
Liu Yanli.Research on 3D Simultaneous Localization and Mapping Fusing Color and Depth Information[D]. Changsha, China: Central South University, 2014.
[2] 梁明杰, 闵华清, 罗荣华. 基于图优化的同时定位与地图创建综述[J]. 机器人, 2013, 35(4): 500-512.
LIANG Mingjie, MIN Huaqing, LUO Ronghua.A summary of simultaneous location and map creation based on graph optimization[J]. Robot, 2013, 35(4):500-512.
[3] Dryanovski I, Valenti R G, Xiao J Z.Fast visual odometry and mapping from RGB-D data[C]//IEEE International Conference on Robotics and Automation. Piscataway, USA: IEEE, 2013: 2305-2310.
[4] Schleicher D, Bergasa L M, Ocaña M, et al.Real-Time Stereo Visual SLAM in Large-Scale Environments Based on SIFT Fingerprints[C]//IEEE International Symposium on Intelligent Signal Processing. IEEE, 2007: 684-691.
[5] Endres F, Hess J, Sturm J, et al.3-D Mapping With an RGB-D Camera[J]. IEEE Transactions on Robotics (S1552-3098), 2014, 30(1): 177-187.
[6] Lowe D G.Distinctive Image Features from Scale- Invariant Keypoints[C]//International Journal of Computer Vision. 2004: 91-110.
[7] Bay H, Tuytelaars T, Gool L V.SURF: Speeded Up Robust Features[J]. Computer Vision & Image Understanding (S1077-3142), 2006, 110(3): 404-417.
[8] Rublee E, Rabaud V, Konolige K, et al.ORB: An efficient alternative to SIFT or SURF[C]//International Conference on Computer Vision. IEEE Computer Society, 2011: 2564-2571.
[9] Besl P J, Mckay N D.Method for registration of 3-D shapes[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence (S0162-8828), 1992, 14(2): 239-256.
[10] Angeli A, Filliat D, Doncieux S, et al.A Fast and Incremental Method for Loop-Closure Detection Using Bags of Visual Words[J]. IEEE Transactions on Robotics (S1552-3098), 2008, 24(5): 1027-1037.
[11] Labbe M, Michaud F.Online global loop closure detection for large-scale multi-session graph-based SLAM[C]//Ieee/rsj International Conference on Intelligent Robots and Systems. 2014: 2661-2666.
[12] Kümmerle R, Grisetti G, Strasdat H, et al.G 2 o: A general framework for graph optimization[C]//IEEE International Conference on Robotics & Automation. 2011: 3607-3613.