Validation of Robustness of SLAM Algorithms using Deep Learning Methods in Real Conditions

0fc9e933-2aa2-47bb-afd9-da48f4db240f20250107081720009wseas:wseasmdt@crossref.orgMDT DepositWSEAS TRANSACTIONS ON INFORMATION SCIENCE AND APPLICATIONS2224-34021790-083210.37394/23209http://wseas.org/wseas/cms.action?id=40461720251720252210.37394/23209.2025.22https://wseas.com/journals/isa/2025.phpValidation of Robustness of SLAM Algorithms using Deep Learning Methods in Real ConditionsYuriiRabeshkoDepartment of Applied Mathematics, Institute of Automation, Cybernetics and Computer Engineering, National University of Water and Environmental Engineering, 11 Soborna St, 33028, Rivne, UKRAINEYuriiTurbalDepartment of Applied Mathematics, Institute of Automation, Cybernetics and Computer Engineering, National University of Water and Environmental Engineering, 11 Soborna St, 33028, Rivne, UKRAINEAs the level of modern technology development, namely autonomous robots, drones, robotics, etc., is high, the topic under study is highly relevant. Since the level of development of modern technologies, namely autonomous robots, drones, etc., is high, the topic under study is highly relevant. Due to the use of the Simultaneous Localisation and Mapping System (SLAM) in the industrial sector, ensuring and empirically verifying its robustness under challenging conditions is essential. The study aimed to evaluate and verify the reliability of the SLAM algorithm in real conditions. The following methods were used to conduct the study: deep learning methods and recurrent neural networks. ATE and RPE metrics were used to measure the accuracy of maps and trajectories. The study revealed a relatively high stability of the developed SLAM algorithm in changing lighting conditions and dynamic objects' presence. The ATE and RPE metrics were within acceptable limits. The study's scientific novelty and originality lie in considering the real conditions during the experiment, such as different lighting and dynamic objects, which were rarely considered in previous studies. The developed algorithm will be helpful for autonomous systems and in the context of the latest advanced technologies and robotics. A promising area for further research may be improving the SLAM algorithm for use in tough conditions.12320241231202456656https://wseas.com/journals/isa/2025/a125109-1130.pdf10.37394/23209.2025.22.6https://wseas.com/journals/isa/2025/a125109-1130.pdf10.1109/irc.2019.00122Singandhupe, A., and La, H. M. A review of slam techniques and security in autonomous driving. 2019 Third IEEE International Conference on Robotic Computing (IRC). IEEE, Naples, Italy, 2019. https://doi.org/10.1109/IRC.2019.00122. 10.1109/ever48776.2020.9243094Nobis, F., Papanikolaou, O., Betz, J., Lienkamp, M. Persistent map saving for visual localization for autonomous vehicles: An orb-slam 2 extension. arXiv. https://doi.org/10.48550/arXiv.2005.07429 10.1016/j.vrih.2019.07.002Jinyu, L., Bangbang, Y., Danpeng, C., Nan, W., Guofeng, Z., Hujun, B. Survey and evaluation of monocular visual-inertial SLAM algorithms for augmented reality. Virtual Reality & Intelligent Hardware, Vol.1, No.4, 2019, рр. 386–410. 10.2514/6.2020-1343Bettens, A. M., Morrell, B., Coen, M., McHenry, N., Wu, X., Gibbens, P., Chamitoff, G. UnrealNavigation: Simulation software for testing SLAM in virtual reality. AIAA Scitech 2020 Forum, Orlando, FL, USA, 2020. https://doi.org/10.2514/6.2020- 1343. 10.1109/wacv48630.2021.00180Shu, F., Lesur, P., Xie, Yа., Pagani, A., Stricker, D. SLAM in the field: An evaluation of monocular mapping and localization on challenging dynamic agricultural environment. Proceedings of the IEEE/CVF winter conference on applications of computer vision. IEEE, Waikoloa, HI, USA, 2021. https://doi.org/10.1109/WACV48630.2021.00 180. 10.1016/j.robot.2019.03.012Xiao, L., Wang, J., Qiu, X., Rong, Zh., Zou, X. Dynamic-SLAM: Semantic monocular visual localization and mapping based on deep learning in dynamic environment. Robotics and Autonomous Systems, Vol.117, 2019, рр. 1-16. https://doi.org/10.1016/j.robot.2019.03.012. 10.1088/1742-6596/2506/1/012009Zhang, J., Zhang, X., Shen, X., Wu, J., Li, Yu. A Lidar SLAM based on Improved Particle Filter and Scan Matching for Unmanned Delivery Robot. Journal of Physics: Conference Series (IOP Publishing). Vol.2506, 2023, paper 012009, https://doi.org/10.1088/1742- 6596/2506/1/012009. 10.1109/ictis.2019.8883749Wang, Wenmin, Zhao, W., Wang, X., Jin, Zh., Li, Yu., Runge, T. A low-cost simultaneous localization and mapping algorithm for last-mile indoor delivery. Proceedings from 2019 5th International Conference on Transportation Information and Safety (ICTIS). IEEE, Liverpool, UK, 2019. https://doi.org/10.1109/ICTIS.2019.8883749. 10.1017/s0263574719000675Wang, H., Zhang, Ch., Song, Y., Pang, B., Zhang, G. Three-dimensional reconstruction based on visual SLAM of mobile robot in search and rescue disaster scenarios. Robotica, Vol.38, No.2, 2020, рр. 350-373. https://doi.org/10.1017/S0263574719000675. 10.1007/978-3-031-43360-3_10Yang, Zh., Nabila Naz, N., Pengcheng Liu, P., Nazmul Huda, M. Evaluation of SLAM algorithms for search and rescue applications. Annual Conference Towards Autonomous Robotic Systems. Cham: Springer Nature Switzerland, 2023, рр. 114-125. 10.3390/drones6030079Steenbeek, A., Nex, F. CNN-based dense monocular visual SLAM for real-time UAV exploration in emergency conditions. Drones, Vol.6, No.3, 2022, рaper 79. https://doi.org/10.3390/drones6030079. 10.1016/j.neucom.2021.05.027Bruno, H. M. S., Colombini, E. L. LIFTSLAM: A deep-learning feature-based monocular visual SLAM method. Neurocomputing, Vol.455, 2021, рр. 97-110. https://doi.org/10.1016/j.neucom.2021.05.027. 10.1109/access.2022.3199350Su, P., Luo, S., Huang, X. Real-time dynamic SLAM algorithm based on deep learning. IEEE Access, Vol.10, 2022, рр. 87754-87766. https://doi.org/10.1109/ACCESS.2022.31993 50. 10.3390/s19173699Bahraini, M. S., Rad, A. B., Bozorg, M. SLAM in dynamic environments: A deep learning approach for moving object tracking using ML-RANSAC algorithm. Sensors, Vol.19, No.17, 2019, рaper 3699. https://doi.org/10.3390/s19173699. 10.1109/access.2020.2991441Ai, Y., Rui, T., Lu, M., Fu, L., Liu, S., Wang, S. DDL-SLAM: A robust RGB-D SLAM in dynamic environments combined with Deep Learning. IEEE Access, Vol.8, 2020, рр. 162335–162342. https://doi.org/10.1109/ACCESS.2020.29914 41. 10.3390/jmse10040511Burguera, A., Bonin-Font, F., Guerrero Font, E., Martorell Torres, A. Combining deep learning and robust estimation for outlierresilient underwater visual graph SLAM. Journal of Marine Science and Engineering, Vol.10, No.4, 2022, рaper 511. https://doi.org/10.3390/jmse10040511. 10.5194/isprs-archives-xliii-b1-2022-447-2022Tiozzo Fasiolo, D., Maset, E., Scalera, L., Macaulay, S. O., Gasparetto, A., Fusiello, A. Combining LiDAR SLAM and deep learningbased people detection for autonomous indoor mapping in a crowded environment. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences (рр. 447-452), Vol.43. XXIV ISPRS Congress, Nice, France, 2022, [Online]. https://pdfs.semanticscholar.org/6de8/180238 803ae6442c4108d94d3343c33e40dc.pdf (Accessed Date: January 13, 2024). 10.3390/ijgi10030163Zhang, Ch., Huang, T., Zhang, R., Yi, X. PLD-SLAM: A new RGB-D SLAM method with point and line features for indoor dynamic scene. ISPRS International Journal of Geo-Information, Vol.10, No.3, 2021, рaper 163. https://doi.org/10.3390/ijgi10030163. 10.1109/iccv51070.2023.00345Zhang, Y., Tosi, F., Mattoccia, S. Poggi, M. Go-slam: Global optimization for consistent 3D instant reconstruction. Proceedings of the IEEE/CVF International Conference on Computer Vision (рр. 3727-3737), IEEE, Paris, France 2023. https://dx.doi.org/10.1109/ICCV51070.2023.0 0345. 10.7717/peerj-cs.1628Zhang, X. Y., Rahman, A. H. A., Qamar, F. Semantic visual simultaneous localization and mapping (SLAM) using deep learning for dynamic scenes. PeerJ Computer Science, Vol.9, 2023, paper e1628. https://doi.org/10.7717/PEERJ-CS.1628. 10.3390/rs16040723Wang, B., Cheng, X., Wang, J., Jiao, L. A Bidirectional scoring strategy-based transformation matrix estimation of dynamic factors in environmental sensing. Remote Sensing, Vol.16, No.4, 2024, paper 723. https://doi.org/10.3390/rs16040723.