Adaptive Optimization in Feature-based SLAM Visual Odometry

doi:10.16182/j.issn1004731x.joss.20-0424E

Abstract

Abstract: Aiming to reduce the impact of dynamic environments on simultaneous localization and mapping (SLAM) of mobile robots, an adaptive optimization method in a feature-based visual odometry is proposed. The method helps to improve the invariance of image feature in illumination changing situation and to extract features effectively in areas where the texture information is not sufficient to make contributions to feature matching. Meanwhile, down sampling is applied to establish image pyramids and each scaled image is divided into cells based on a defined rule. Illumination adaptive nonlinear adjustments for each cell are applied to increase the image details, and low-texture area is removed by computing the image gray level probability distribution. Based on the proposed method, a visual odometry of SLAM system is built and verified on TUM dataset. The results show that, compared with the original system, the proposed method can reduce the trajectory errors of a mobile robot and also improve the performance of robot visual odometry in the unstable dynamic environments.

Key words: mobile robot, visual odometry, SLAM, feature-based method, low-texture area

CLC Number:

TP391

Yu Yanan, Shi Dunhuang, Hua Chunjie. Adaptive Optimization in Feature-based SLAM Visual Odometry[J]. Journal of System Simulation, 2022, 34(1): 104-112.

References

[1] Fink G, Franke M, Lynch A F, et al.Visual Inertial SLAM: Application to Unmanned Aerial Vehicle[J]. IFAC-PapersOnline (S2405-8963), 2017, 50(1): 1965-1970.
[2] Ros G, Sappa A D, Ponsa D, et al.Visual SLAM for Driverless Cars: a Brief Survey[C]//IEEE Workshop on Navigation, Perception, Accurate Positioning and Mapping for Intelligent Vehicles. Los Alamitos: IEEE Computer Society Press, 2012: Article No.3.
[3] H¨oll M, Lepetit V. Monocular LSD-SLAM Integration Within AR System[EB/OL]. ArXiv:1702.02514 (2017-02-08) [2020-06-01].https://arxiv.org/pdf/1702.02514.pdf.
[4] Cadena C, Carlone L, Carrillo H, et al.Past, Present, and Future of Simultaneous Localization and mapping: Towards the Robust-perception Age[J]. IEEE Transactions on Robotics (S1552-3098), 2016, 32(6): 1309-1332.
[5] Karlsson N, Bernardo E D, Ostrowski J P, et al.The vSLAM Algorithm for Robust Localization and Mapping[C]//IEEE International Conference on Robotics and Automation. Barcelona, Spain: IEEE Press, 2005: 24-29.
[6] Zhao Y, Smith J S, Karumanchi S H, et al.A. Closed-loop Benchmarking of Stereo Visual-inertial SLAM System: Understanding the Impact of Drift and Latency on Tracking Accuracy[C]//IEEE International Conference on Robotics and Automation. Paris, France: IEEE Press, 2020: 1105-1112.
[7] 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 (S0162-8828), 2007, 29(6): 1052-1067.
[8] Klein G, Murray D.Parallel Tracking and Mapping for Small AR Workspaces[C]// 2007 6th IEEE and ACM International Symposium on Mixed and Augmented Reality. Piscataway, NJ: IEEE Press, 2007: 225-234.
[9] 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.
[10] Mur-Artal J, Montiel J M M, Tardós J D. ORB-SLAM: a Versatile and Accurate Monocular SLAM System[J]. IEEE Transactions on Robotics (S1552-3098), 2015, 31(5): 1147-1163.
[11] Mur-Artal R, Tardós J D.ORB-SLAM2: an Open-source SLAM System for Monocular, Stereo and RGB-D Cameras[J]. IEEE Transactions on Robotics (S1552-3098), 2017, 33(5): 1255-1262.
[12] Rublee E, Rabaud V, Konolige K, et al.ORB: an Efficient Alternative to SIFT or SURF[C]//IEEE International Conference on Computer Vision (ICCV). Barcelona, Spain: IEEE Press, 2011: 2564-2571.
[13] Lv Q, Lin H, Wang G, et al.ORB-SLAM-based Tracking and 3D Reconstruction for Robot Using Kinect 2.0[C]//2017 29th Chinese Control and Decision Conference (CCDC). Chongqing, China: IEEE Press, 2017: 3319-3324.
[14] Elvira R, Tardós J D, Montiel J M M. ORBSLAM-Atlas: a Robust and Accurate Multi-map System[C]//IEEE/RSJ International Conference on Intelligent Robots and System. Macau, China: IEEE Press, 2019: 6253-6259.
[15] Newcombe R A, Lovegrove S J, Davison A J.DTAM: Dense Tracking and Mapping in Real-time[C]//IEEE International Conference on Computer Vision (ICCV). Barcelona, Spain: IEEE Press, 2011: 2320-2327.
[16] Engel J, Schöps T, Cremers D.LSD-SLAM: Large-scale Direct Monocular SLAM[C]//European Conference on Computer Vision. Zurich, Switzerland: Springer, 2014: 834-849.
[17] Engel J, Koltum V, Cremers D.Direct Sparse Odometry[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence (S0162-8828), 2018, 40(3): 611-625.
[18] Forster C, Pizzoli M, Scaramuzza D.SVO: Fast Semi-direct Monocular Visual Odometry[C]//IEEE International Conference on Robotics and Automation (ICRA). Hong Kong, China: IEEE Press, 2014: 15-22.
[19] Zubizarreta J, Aguinaga I, Montiel J M M. Direct Sparse Mapping[J]. IEEE Transaction on Robotics (S1552-3098), 2020, 36(4): 1363-1370.
[20] Sturm J, Engelhard N, Endres F, et al.A Benchmark for the Evaluation of RGB-D SLAM Systems[C]//IEEE International Conference on Intelligent Robot Systems (IROS). Algarve, Portugal: IEEE Press, 2012: 573-580.