Real-time dense SLAM algorithm for autonomous vehicles based on 3D reconstruction priors

doi:10.13474/j.cnki.11-2246.2025.1013

Abstract

Abstract: In response to the challenges faced by autonomous vehicles in achieving accurate localization and dense mapping in complex environments, this paper proposes a real-time monocular dense SLAM algorithm based on 3D reconstruction priors.By incorporating robust geometric priors, the algorithm demonstrates exceptional robustness in unstructured environments and does not rely on predefined camera models, making it adaptable to various general time-varying camera models.The algorithm's architecture consists of four core modules:point-to-map matching, tracking and local fusion, map construction and loop closure detection, a second-order global optimization mechanism.Through adaptive parameter calibration, the algorithm achieves leading performance in multiple benchmark tests under complex scenarios such as dynamic lighting and weak textures.Additionally, the algorithm is capable of operating in real-time.

Key words: autonomous vehicle, 3D reconstruction, accurate localization, dense mapping, SLAM

CLC Number:

P237

ZHANG Hongwei, Lü Yunfei, GAO Haikuan, YANG Pengxin, WU Wenjun, OU Weiming. Real-time dense SLAM algorithm for autonomous vehicles based on 3D reconstruction priors[J]. Bulletin of Surveying and Mapping, 2025, 0(10): 76-81.

References

[1] 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,2007,29(6):1052-1067. [2] KLEIN G,MURRAY D. Parallel tracking and mapping for small AR workspaces[C]//Proceedings of 2007 IEEE and ACM international symposium on mixed and augmented reality. [S. l]:IEEE,2007:225-234. [3] MUR-ARTAL R,MONTIEL J M M,TARDOS J D. ORB-SLAM:a versatile and accurate monocular SLAM system[J]. IEEE Transactions on Robotics,2015,31(5):1147-1163. [4] NEWCOMBE R A,LOVEGROVE S J,DAVISON A J. DTAM:dense tracking and mapping in real-time[C]//Proceedings of 2011 International Conference on Computer Vision. Barcelona:IEEE,2011:2320-2327. [5] EIGEN D,PUHRSCH C,FERGUS R. Depth map prediction from a single image using a multi-scale deep network[J]. Advances in neural information processing systems,2014:1406. 2283. [6] KE Bingxin,OBUKHOV A,HUANG Shengyu,et al. Repurposing diffusion-based image generators for monocular depth estimation[C]//Proceedings of 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle:IEEE,2024:9492-9502. [7] RANFTL R,LASINGER K,HAFNER D,et al. Towards robust monocular depth estimation:mixing datasets for zero-shot cross-dataset transfer[J]. IEEE transactions on pattern analysis and machine intelligence,2020,44(3):1623-1637. [8] YANG Lihe,KANG Bingyi,HUANG Zilong,et al. Depth anything:unleashing the power of large-scale unlabeled data[C]//Proceedings of 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle:IEEE,2024:10371-10381. [9] KOESTLER L,YANG N,ZELLER N,et al. Tandem:tracking and dense mapping in real-time using deep multi-view stereo[C]//Proceedings of the Conference on Robot Learning. [S. l. ]:PMLR,2022:34-45. [10] SAYED M,GIBSON J,WATSON J,et al. SimpleRecon:3D reconstruction without 3D convolutions[M]//Computer Vision - ECCV 2022. Cham:Springer Nature Switzerland,2022:1-19. [11] ZHOU Huizhong,UMMENHOFER B,BROX T. DeepTAM:deep tracking and mapping[M]//Computer Vision-ECCV 2018. Cham:Springer International Publishing,2018:851-868. [12] MILDENHALL B,SRINIVASAN P P,TANCIK M,et al. NeRF:representing scenes as neural radiance fields for view synthesis[C]//Proceedings of 2020 Computer Vision. Cham:Springer,2020:405-421. [13] MATSUKI H,MURAI R,KELLY P H J,et al. Gaussian splatting SLAM[C]//Proceedings of 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle:IEEE,2024:18039-18048. [14] HARTLEY R,ZISSERMAN A. Multiple view geometry in computer vision[M]. Cambridge:Cambridge University Press,2004. [15] KEIVAN N,SIBLEY G. Constant-time monocular self-calibration[C]//Proceedings of 2014 IEEE International Conference on Robotics and Biomimetics (ROBIO 2014). Bali:IEEE,2014:1590-1595. [16] JIN Linyi,ZHANG Jianming,HOLD-GEOFFROY Y,et al. Perspective fields for single image camera calibration[C]//Proceedings of 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Vancouver:IEEE,2023:17307-17316. [17] WANG Shuzhe,LEROY V,CABON Y,et al. DUSt3R:geometric 3D vision made easy[C]//Proceedings of 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle:IEEE,2024:20697-20709. [18] LEROY V,CABON Y,REVAUD J. Grounding image matching in 3D with MASt3R[M]//Computer Vision -ECCV 2024. Cham:Springer Nature Switzerland,2024:71-91. [19] DUISTERHOF B,ZUST L,WEINZAEPFEL P,et al. MASt3R-SfM:a fully-integrated solution for unconstrained structure-from-motion[EB/OL]. [2025-02-01]. https//arxiv. org/abs/2409. 19152. [20] STURM J,ENGELHARD N,ENDRES F,et al. A benchmark for the evaluation of RGB-D SLAM systems[C]//Proceedings of 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems. Vilamoura-Algarve:IEEE,2012:573-580. [21] SHOTTON J,GLOCKER B,ZACH C,et al. Scene coordinate regression forests for camera relocalization in RGB-D images[C]//Proceedings of 2013 IEEE Conference on Computer Vision and Pattern Recognition. Portland:IEEE,2013:2930-2937. [22] BURRI M,NIKOLIC J,GOHL P,et al. The EuRoC micro aerial vehicle datasets[J]. The International Journal of Robotics Research,2016,35(10):1157-1163. [23] LIPSON L,TEED Z,DENG Jia. Deep patch visual SLAM[M]//Computer Vision-ECCV 2024. Cham:Springer Nature Switzerland,2024:424-440. [24] ZHANG Youmin,TOSI F,MATTOCCIA S,et al. GO-SLAM:global optimization for consistent 3D instant reconstruction[C]//Proceedings of 2023 IEEE/CVF International Conference on Computer Vision (ICCV). Paris:IEEE,2023:3704-3714. [25] TEED Z,DENG J. Droid-SLAM:deep visual SLAM for monocular,stereo,and rgb-d cameras[J]. Advances in neural information processing systems,2021,34:16558-16569. [26] ZHU Z,PENG S,LARSSON V,et al. Nicer-slam:neural implicit scene encoding for RGB SLAM[C]//Proceedings of 2024 International Conference on 3D Vision (3DV). [S. l. ]:IEEE,2024:42-52.