MUET logo

Robust outdoor trajectory mapping using CNN features and loop closure optimization

Authors

  • Kamran Kazi

    Institute of Information & Communication Technologies, Mehran University of Engineering & Technology, Jamshoro, Pakistan, Department of Electronic Engineering, Mehran University of Engineering & Technology, Jamshoro, Pakistan
    Author

  • Arbab Nighat

    Institute of Information & Communication Technologies, Mehran University of Engineering & Technology, Jamshoro, Pakistan, Department of Electronic Engineering, Mehran University of Engineering & Technology, Jamshoro, Pakistan
    Author

  • Farida Memon

    Department of Electronic Engineering, Mehran University of Engineering & Technology, Jamshoro, Pakistan
    Author

  • Tarique Rafique

    Department of Electronic Engineering, Quaid-e-Awam University of Engineering, Science & Technology, Nawabshah
    Author

  • Azam Rafique Memon

    College of Engineering, Prince Sultan University, Riyadh, Saudi Arabia
    Author

DOI:

https://doi.org/10.22581/muet1982.0351

Keywords:

Visual SLAM, CNN-based Feature Extraction, , Outdoor Cluttered Environments, Trajectory Optimization, Odometry Error

Abstract

Autonomous navigation in cluttered outdoor environments remains a significant challenge for robotic systems due to dynamic obstacles, occlusions, and feature degradation. Traditional Simultaneous Localization and Mapping (SLAM) systems, reliant on low-level geometric features, often fail to accurately estimate trajectory in such conditions. This paper introduces a robust framework for outdoor trajectory mapping that synergizes deep learning-based feature extraction with loop closure detection and optimization to address these limitations. To find a suitable layer to obtain features from the CNN model, 313,746 filters were checked to find a filter that has the least odometry error from three pre-trained CNN models, the ConvNeXtXLarge is leveraged to extract high-level semantic features from monocular images, enabling resilient optical flow estimation even in scenes with transient objects and lighting variations. These features are fused with inertial measurement unit (IMU) data to compute initial trajectories, minimizing scale ambiguity. To counteract accumulated drift, a hybrid optimization strategy is proposed: Bag of Words (BoW)-based loop closure detection identifies revisit locations, while an algorithm is developed to optimize the map. Experimental validation on the KITTI dataset demonstrates a significant reduction in offset error compared to state-of-the-art methods like DeepVO and TartanAir. 

Downloads

References

R. Mur-Artal and J. D. Tardos, “ORB-SLAM2: An Open-Source SLAM System for Monocular, Stereo, and RGB-D Cameras,” IEEE Transactions on Robotics, vol. 33, no. 5, pp. 1255–1262, 2017, doi: 10.1109/TRO.2017.2705103.

S. Wang, R. Clark, H. Wen, and N. Trigoni, “DeepVO: Towards end-to-end visual odometry with deep Recurrent Convolutional Neural Networks,” in 2017 IEEE International Conference on Robotics and Automation (ICRA), IEEE, May 2017, pp. 2043–2050. doi: 10.1109/ICRA.2017.7989236.

D. Ming and X. Wu, “Research on Monocular Vision SLAM Algorithm for Multi-map Fusion and Loop Detection,” in 2022 6th International Conference on Automation, Control and Robots, ICACR 2022, 2022. doi: 10.1109/ICACR55854.2022.9935516.

Z. Xu, Z. Rong, and Y. Wu, “A survey: which features are required for dynamic visual simultaneous localization and mapping?,” Vis Comput Ind Biomed Art, vol. 4, no. 1, p. 20, Dec. 2021, doi: 10.1186/s42492-021-00086-w.

A. Hussain, A. R. Memon, H. Wang, Y. Wang, Y. Miao, and X. Zhang, “S-VIT: Stereo Visual-Inertial Tracking of Lower Limb for Physiotherapy Rehabilitation in Context of Comprehensive Evaluation of SLAM Systems,” IEEE Transactions on Automation Science and Engineering, vol. 18, no. 4, pp. 1550–1562, Oct. 2021, doi: 10.1109/TASE.2020.3010415.

A. Mahmoud and M. Atia, “Improved Visual SLAM Using Semantic Segmentation and Layout Estimation,” Robotics, vol. 11, no. 5, p. 91, Sep. 2022, doi: 10.3390/robotics11050091.

D. Li et al., “DXSLAM: A Robust and Efficient Visual SLAM System with Deep Features,” in 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), IEEE, Oct. 2020, pp. 4958–4965. doi: 10.1109/IROS45743.2020.9340907.

W. Wang et al., “TartanAir: A Dataset to Push the Limits of Visual SLAM,” in 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), IEEE, Oct. 2020, pp. 4909–4916. doi: 10.1109/IROS45743.2020.9341801.

D. Galvez-López and J. D. Tardos, “Bags of Binary Words for Fast Place Recognition in Image Sequences,” IEEE Transactions on Robotics, vol. 28, no. 5, pp. 1188–1197, Oct. 2012, doi: 10.1109/TRO.2012.2197158.

A. Geiger, P. Lenz, and R. Urtasun, “Are we ready for autonomous driving? The KITTI vision benchmark suite,” in 2012 IEEE Conference on Computer Vision and Pattern Recognition, IEEE, Jun. 2012, pp. 3354–3361. doi: 10.1109/CVPR.2012.6248074.

R. Elvira, J. D. Tardos, and J. M. M. Montiel, “ORBSLAM-Atlas: a robust and accurate multi-map system,” in 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), IEEE, Nov. 2019, pp. 6253–6259. doi: 10.1109/IROS40897.2019.8967572.

A. Memon, F. K. Shaikh, and E. Felemban, “Experimental Evaluation of Vehicle-to-Vehicle based Data Transfer,” in International Conference on Information and Communication Technology Research (ICTRC2015), 2015, pp. 274–277. doi: 10.1109/ICTRC.2015.7156475.

J. Engel, V. Koltun, and D. Cremers, “Direct Sparse Odometry,” IEEE Trans Pattern Anal Mach Intell, vol. 40, no. 3, pp. 611–625, Mar. 2018, doi: 10.1109/TPAMI.2017.2658577.

M. Zins, G. Simon, and M.-O. Berger, “OA-SLAM: Leveraging Objects for Camera Relocalization in Visual SLAM,” in 2022 IEEE International Symposium on Mixed and Augmented Reality (ISMAR), IEEE, Oct. 2022, pp. 720–728. doi: 10.1109/ISMAR55827.2022.00090.

Q. Peng, Z. Xiang, Y. Fan, T. Zhao, and X. Zhao, “RWT-SLAM: Robust Visual SLAM for Highly Weak-textured Environments,” Jul. 2022, doi: https://doi.org/10.48550/arXiv.2207.03539.

R. Giubilato et al., “Relocalization With Submaps: Multi-Session Mapping for Planetary Rovers Equipped With Stereo Cameras,” IEEE Robot Autom Lett, vol. 5, no. 2, pp. 580–587, Apr. 2020, doi: 10.1109/LRA.2020.2964157.

X. Hu, J. Wu, J. Jiao, W. Zhang, and P. Tan, “MS-Mapping: Multi-session LiDAR Mapping with Wasserstein-based Keyframe Selection,” 2024, doi: 10.48550/arXiv.2406.02096.

M. Menezes et al., “A Multisession SLAM Approach for RatSLAM,” J Intell Robot Syst, vol. 108, no. 4, p. 61, Aug. 2023, doi: 10.1007/s10846-023-01816-3.

M. Cummins and P. Newman, “FAB-MAP: Probabilistic Localization and Mapping in the Space of Appearance,” Int J Rob Res, vol. 27, no. 6, pp. 647–665, Jun. 2008, doi: 10.1177/0278364908090961.

M. J. Milford and G. F. Wyeth, “SeqSLAM: Visual route-based navigation for sunny summer days and stormy winter nights,” Proc IEEE Int Conf Robot Autom, pp. 1643–1649, 2012, doi: 10.1109/ICRA.2012.6224623.

X. Hu et al., “PALoc: Advancing SLAM Benchmarking With Prior-Assisted 6-DoF Trajectory Generation and Uncertainty Estimation,” IEEE/ASME Transactions on Mechatronics, vol. 29, no. 6, pp. 4297–4308, Dec. 2024, doi: 10.1109/TMECH.2024.3362902.

H. Matsuki, R. Murai, P. H. J. Kelly, and A. J. Davison, “Gaussian Splatting SLAM,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2024, pp. 18039–18048. doi: 10.48550/arXiv.2312.06741.

Z. Xing, X. Zhu, and D. Dong, “DE‐SLAM: SLAM for highly dynamic environment,” J Field Robot, no. May 2021, pp. 1–15, Feb. 2022, doi: 10.1002/rob.22062.

A. Li, J. Wang, M. Xu, and Z. Chen, “DP-SLAM: A visual SLAM with moving probability towards dynamic environments,” Inf Sci (N Y), vol. 556, pp. 128–142, May 2021, doi: 10.1016/j.ins.2020.12.019.

Z. Xing, X. Zhu, and D. Dong, “DE‐SLAM: SLAM for highly dynamic environment,” J Field Robot, vol. 39, no. 5, pp. 528–542, Aug. 2022, doi: 10.1002/rob.22062.

“ethz-asl/kalibr: The Kalibr visual-inertial calibration toolbox.” Accessed: Jun. 21, 2025. [Online]. Available: https://github.com/ethz-asl/kalibr

K. Kazi, A. N. Kalhoro, F. Memon, A. R. Memon, and M. Iqbal, “Beyond Handcrafted Features: A Deep Learning Framework for Optical Flow and SLAM,” J Imaging, vol. 11, no. 5, p. 155, May 2025, doi: 10.3390/jimaging11050155.

M. Iqbal, A. R. Memon, and D. J. Almakhles, “Accelerating Resource-Constrained Swarm Robotics With Cone-Based Loop Closure and 6G Communication,” IEEE Transactions on Intelligent Transportation Systems, pp. 1–11, 2025, doi: 10.1109/TITS.2025.3526819.

H. Sun, P. Wang, C. Ni, and J. ming Li, “Loop closure detection based on image semantic feature and bag-of-words,” Multimed Tools Appl, 2022, doi: 10.1007/s11042-022-13353-6.

A. R. Memon, H. Wang, and A. Hussain, “Loop closure detection using supervised and unsupervised deep neural networks for monocular SLAM systems,” Rob Auton Syst, vol. 126, p. 103470, Apr. 2020, doi: 10.1016/j.robot.2020.103470.

A. R. Memon, Z. Liu, and H. Wang, “Viewpoint-Invariant Loop Closure Detection Using Step-Wise Learning With Controlling Embeddings of Landmarks,” IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 11, pp. 20148–20159, Nov. 2022, doi: 10.1109/TITS.2022.3173681.

Published

2025-07-01

Issue

Vol. 44 No. 3 (2025): July Issue

Section

Articles

License

Copyright (c) 2025 Mehran University Research Journal of Engineering and Technology

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite

Robust outdoor trajectory mapping using CNN features and loop closure optimization. (2025). Mehran University Research Journal of Engineering and Technology, 44(3), 174-186. https://doi.org/10.22581/muet1982.0351

Similar Articles

1-10 of 94

You may also start an advanced similarity search for this article.