A KEY FRAME SELECTION AND LOCAL BA OPTIMISATION METHOD FOR VSLAM

Guangfeng Liu, Zhuhua Hu, Yaochi Zhao, Ruoqing Li, Kunkun Ding, and Wenlu Qi

References

  1. [1] A. Tourani, H. Bavle, J.L. Sanchez-Lopez, and H. Voos, VisualSLAM: What are the current trends and what to expect?,Sensors, 22(23), 2022, 9297.
  2. [2] W. Chen, G. Shang, A. Ji, C. Zhou, X. Wang, and C. Xu, Anoverview on visual SLAM: From tradition to semantic, RemoteSensing, 14(13), 2022, 3010.
  3. [3] Y. Fu, B. Han, Z. Hu, X. Shen, and Y. Zhao, CBAM-SLAM: A semantic SLAM based on attention module indynamic environment, Proc. 2022 6th Asian Conf. on ArtificialIntelligence Technology (ACAIT), IEEE, Changzhou, 2022,1–6.
  4. [4] H. Qi, Z. Hu, Y. Xiang, D. Cai, and Y. Zhao, ATY-SLAM:A visual semantic SLAM for dynamic indoor environments,Proc. International Conf. on Intelligent Computing, Singapore,2023, 3–14.
  5. [5] D. Cai, Z. Hu, R. Li, H. Qi, Y. Xiang, and Y. Zhao, AGAM-SLAM: An adaptive dynamic scene semantic SLAM methodbased on GAM, Proc. International Conf. on IntelligentComputing, Singapore, 2023, 27–39.
  6. [6] J. Wu, X. Wang, B. Zhang, and T. Huang, Multi-objectiveoptimal design of a novel 6-DOF spray-painting robot, Robotica,39(12), 2021, 2268–2282.
  7. [7] J. Wu, B.B. Zhang, L.P. Wang, and G. Yu, An iterative learningmethod for realizing accurate dynamic feedforward controlof an industrial hybrid robot, Science China TechnologicalSciences, 64(6), 2021, 1177–1188.
  8. [8] G. Younes, D. Asmar, E. Shammas, and J. Zelek, Keyframe-based monocular SLAM: Design, survey, and futuredirections, Robotics and Autonomous Systems, 98, 2017,67–88.
  9. [9] Y. Wang and X. Peng, New trend in back-end techniques ofvisual SLAM: From local iterative solvers to robust globaloptimization, Artificial Intelligence in China: Proc. of the3rd International Conf. on Artificial Intelligence in China,Singapore, 2022, 308–317.
  10. [10] A.J. Davison, I.D. Reid, N.D. Molton, and O. Stasse,MonoSLAM: Real-time single camera SLAM, IEEE Transac-tions on Pattern Analysis and Machine Intelligence, 29(6),2007, 1052–1067.
  11. [11] G. Klein and D. Murray, Parallel tracking and mapping for smallAR workspaces, Proc. 2007 6th IEEE and ACM InternationalSymposium on Mixed and Augmented Reality, IEEE, Nara,2007, 225–234.
  12. [12] R.A. Newcombe, S.J. Lovegrove, and A.J. Davison, DTAM:Dense tracking and mapping in real-time, Proc. 2011International Conf. on Computer Vision, Barcelona, 2011,2320–2327.
  13. [13] C. Forster, M. Pizzoli, and D. Scaramuzza, SVO: Fastsemi-direct monocular visual odometry, Proc. 2014 IEEEInternational Conf. on Robotics and Automation (ICRA),IEEE, Hong Kong, 2014, 15–22.
  14. [14] J. Engel, T. Sch¨ops, and D. Cremers, LSD-SLAM: Large-scaledirect monocular SLAM, Proc. European Conf. on ComputerVision, Cham, 2014, 834–849.
  15. [15] R. Mur-Artal, J.M.M. Montiel, and J.D. Tardos, ORB-SLAM:A versatile and accurate monocular SLAM system, IEEETransactions on Robotics, 31(5), 2015, 1147–1163.
  16. [16] R. Mur-Artal and J.D. Tard´os, ORB-SLAM2, An open-sourceslam system for monocular, stereo, and rgb-d camera, IEEETransactions on Robotics, 33(5), 2017, 1255–1262.
  17. [17] C. Campos, R. Elvira, J.J.G. Rodr´ıguez, J.M.M. Montiel,and J.D. Tard´os, ORB-SLAM3: An accurate open-sourcelibrary for visual, visual–inertial, and multimap SLAM, IEEETransactions on Robotics, 37(6), 2021, 1874–1890.
  18. [18] C. Kerl, J. Sturm, and D. Cremers, Dense visual SLAMfor RGB-D cameras, Proc. 2013 IEEE/RSJ InternationalConf. on Intelligent Robots and Systems, IEEE, Tokyo, 2013,2100–2106.
  19. [19] W. Hu, Z. Zhang, S. Ji, Q. Song, and Q. Huang,An adaptive selection algorithm of SLAM key frame forreconnaissance robots, Electronics Optics & Control, 27(6),2020, 86.10
  20. [20] L. Cui, C. Ma, and F. Wen, Direct-ORB-SLAM: Directmonocular ORB-SLAM, Journal of Physics: Conference Series,IOP Publishing, 1345(3), 2019, 032016.
  21. [21] M. Fanfani, F. Bellavia, and C. Colombo, Accuratekeyframe selection and keypoint tracking for robust visualodometry, Machine Vision and Applications, 27, 2016,833–844.
  22. [22] I. Alonso, L. Riazuelo, and A.C. Murillo, Enhancing v-SLAM keyframe selection with an efficient ConvNet forsemantic analysis, Proc. 2019 International Conf. on Roboticsand Automation (ICRA), IEEE, Montreal, QC, 2019,4717–4723.
  23. [23] G. Grisetti, R. K¨ummerle, C. Stachniss, and W. Burgard, Atutorial on graph-based SLAM, IEEE Intelligent Transporta-tion Systems Magazine, 2(4), 2010, 31–43.
  24. [24] T. Tanaka, Y. Sasagawa, and T. Okatani, Learning to bundle-adjust: A graph network approach to faster optimization ofbundle adjustment for vehicular SLAM, Proc. 2021 IEEE/CVFInternational Conf. on Computer Vision (ICCV), Montreal,QC, 2021, 6250–6259.
  25. [25] A. Belder, R. Vivanti, and A. Tal, A game of bundle adjustment-learning efficient convergence, Proc. 2023 IEEE/CVF Inter-national Conf. on Computer Vision (ICCV), Paris, 2023,8428–8437.
  26. [26] D. Ming, X. Wu, Y. Wang, Z. Zhu, H. Ge, and R. Liu, A real-time monocular visual SLAM based on the bundle adjustmentwith adaptive robust kernel, Journal of Intelligent & RoboticSystems, 107(3), 2023, 35.
  27. [27] T. Lu, X. Jin, Y. Liao, et al., Visual SLAM based on Jacobiannull-space marginalization, Automotive Engineering, 45(08),2023, 1457–1467.
  28. [28] H. Wang and B. Lei, Improved SLAM backend optimiza-tion algorithm based on trust region, Modular MachineTool & Automatic Manufacturing Technique, 01, 2021,9–12.
  29. [29] Y. Jeong, D. Nister, D. Steedly, R. Szeliski, and I.Kweon, Pushing the envelope of modern methods for bundleadjustment, IEEE Transactions on Pattern Analysis andMachine Intelligence, 34(8), 2011, 1605–1617.
  30. [30] S. Jin, X. Dai, and Q. Meng, “Focusing on the right regions”—Guided saliency prediction for visual SLAM, Expert SystemsWith Applications, 213, 2023, 119068.
  31. [31] L. Chen, Z. Ling, Y. Gao, R. Sun, and S. Jin, A real-timesemantic visual SLAM for dynamic environment based on deeplearning and dynamic probabilistic propagation, Complex &Intelligent Systems, 9(5), 2023, 5653–5677.
  32. [32] X. Gao, Fourteen lectures on visual SLAM: From theoryto practice, Publishing House of Electronics Industry,2017.
  33. [33] Y. Wang, Gauss–Newton method, Wiley Interdisci-plinary Reviews: Computational Statistics, 4(4), 2012,415–420.
  34. [34] A. Ranganathan, The Levenberg-Marquardt algorithm, Tuto-rial on LM Algorithm, 11(1), 2004, 101–110.
  35. [35] Q. Zhu, Y. Peng, and Y. Mei, Ceh∞ F-SLAM: A robustand Jacobian-Free solution to SLAM Problem, Interna-tional Journal of Robotics and Automation, 34(1), 2019,17–23.
  36. [36] H.S. Chen and M.A. Stadtherr, A modification of Powell’sdogleg method for solving systems of nonlinear equa-tions, Computers & Chemical Engineering, 5(3), 1981,143–150.
  37. [37] J.S. Dai, Euler–Rodrigues formula variations, quaternionconjugation and intrinsic connections, Mechanism and MachineTheory, 92, 2015, 144–152.
  38. [38] M. Burri, J. Nikolic, P. Gohl, T. Schneider, J. Rehder, S Omari,and M.W. Achtelik, The EuRoC micro aerial vehicle datasets,The International Journal of Robotics Research, 35(10), 2016,1157–1163.
  39. [39] D. Schubert, T. Goll, N. Demmel, V. Usenko, J. St¨uckler, andD. Cre-mers, The TUM VI benchmark for evaluating visual-inertial odometry, Proc. 2018 IEEE/RSJ International Conf.on Intelligent Robots and Systems (IROS), IEEE, Madrid,2018, 1680–1687.

Important Links:

Go Back