English

应用科学学报 >

2026 , Vol. 44 >Issue 3: 358 - 376

DOI: https://doi.org/10.3969/j.issn.0255-8297.2026.03.002

智能信息处理

不确定性建模驱动的LiDAR-视觉自适应融合动态障碍物检测

  • 朱磊 ,
  • 钟若飞 ,
  • 袁忻泽 ,
  • 范红超 ,
  • 孙振兴
展开
  • 1. 首都师范大学三维信息获取与应用教育部重点实验室, 北京 100048;
    2. 首都师范大学资源环境与旅游学院, 北京 100048;
    3. 挪威科技大学土木与环境工程系, 挪威 特隆赫姆7491

收稿日期: 2026-02-02

  网络出版日期: 2026-06-23

基金资助

国家自然科学基金(No.U22A20568);北京市高创计划(No.202504841072)

收起

Uncertainty Modeling-Driven LiDAR-Vision Adaptive Fusion for Dynamic Obstacle Detection

  • ZHU Lei ,
  • ZHONG Ruofei ,
  • YUAN Xinze ,
  • FAN Hongchao ,
  • SUN Zhenxing
Expand
  • 1. Key Laboratory of 3D Information Acquisition and Application, Ministry of Education, Capital Normal University, Beijing 100048, China;
    2. College of Resource Environment and Tourism, Capital Normal University, Beijing 100048, China;
    3. Department of Civil and Environmental Engineering, Norwegian University of Science and Technology, Trondheim 7491, Norway

Received date: 2026-02-02

  Online published: 2026-06-23

Fold

摘要

准确可靠的动态障碍物感知是封闭有限环境中自主测绘无人机安全导航的关键前提，但现有LiDAR-视觉融合方法在光照不良、反射干扰或运动模糊等退化条件下，往往难以应对传感器可靠性发生的显著变化。针对上述问题，本文提出了一种不确定性建模驱动的LiDAR-视觉融合动态障碍物检测框架，通过显式建模传感器观测不确定性，实现传感器贡献的自适应调整。该框架基于概率模型对LiDAR点云与RGB图像进行实时不确定性量化，并引入自适应传感器可靠性评分机制，引导融合决策与后续目标追踪过程。在自主构建的多条件数据集上进行了实验，结果表明：相比现有方法，本文方法在低光照、玻璃反射及运动模糊等具有挑战性的场景下F1值提升约15%～20%，同时在嵌入式平台上保持约25 Hz的实时处理性能。实验验证了该方法在复杂退化环境下的鲁棒性与工程可行性。

关键词： 动态障碍物检测; 不确定性量化; LiDAR-视觉融合; 室内自主测绘无人机

本文引用格式

朱磊 , 钟若飞 , 袁忻泽 , 范红超 , 孙振兴 . 不确定性建模驱动的LiDAR-视觉自适应融合动态障碍物检测[J]. 应用科学学报, 2026 , 44(3) : 358 -376 . DOI: 10.3969/j.issn.0255-8297.2026.03.002

Abstract

Accurate and reliable dynamic obstacle perception is a crucial prerequisite for the safe navigation of autonomous mapping UAVs in confined and spatially constrained environments. However, existing LiDAR-vision fusion methods often struggle to cope with significant variations in sensor reliability under degraded conditions, such as poor illumination, reflection interference, or motion blur. To address these challenges, this paper proposed an uncertainty-modeling-driven LiDAR-vision fusion framework for dynamic obstacle detection, which adaptively adjusted sensor contributions by explicitly modeling sensor observation uncertainty. Based on probabilistic models, the framework performed real-time uncertainty quantification for both LiDAR point clouds and RGB images and introduced an adaptive sensor reliability score(ASRS) mechanism to guide fusion decisions and subsequent object tracking. Experiments were conducted on a self-constructed multi-condition dataset, and the results show that the proposed method improves the F1-score by approximately 15%-20% compared with existing methods in challenging scenarios involving low illumination, glass reflections, and motion blur. Furthermore, it maintains real-time processing performance at approximately 25 Hz on embedded platforms, validating the method's robustness and engineering feasibility in complex degraded environments.

Key words： dynamic obstacle detection; uncertainty quantification; LiDAR-vision fusion; indoor autonomous mapping UAV

参考文献

[1] Lin H Y, Peng X Z. Autonomous quadrotor navigation with vision based obstacle avoidance and path planning [J]. IEEE Access, 2021, 9: 102450-102459.
[2] Falanga D, Kleber K, Scaramuzza D. Dynamic obstacle avoidance for quadrotors with event cameras [J]. Science Robotics, 2020, 5(40): eaaz9712.
[3] Du Y, Kang J, Tian X, et al. Nonlinear MPC of quadrotors for path planning and dynamic obstacle avoidance [C]//2024 IEEE 14th International Conference on CYBER Technology in Automation, Control, and Intelligent Systems (CYBER), 2024: 485-490.
[4] Tranzatto M, Mascarich F, Bernreiter L, et al. CERBERUS: autonomous legged and aerial robotic exploration in the tunnel and urban circuits of the DARPA subterranean challenge [J]. Field Robotics, 2022, 2: 274-324.
[5] 董浩然. 基于深度强化学习的无人机动态避障算法设计[D]. 哈尔滨: 哈尔滨工业大学, 2025.
[6] 谭建豪, 马小萍, 李希. 无人机3D航迹规划及动态避障算法研究[J]. 仪器仪表学报, 2019, 40(12): 224-233. Tan J H, Ma X P, Li X. Research on UAV 3D flight track planning and dynamic obstacle avoidance algorithm [J]. Chinese Journal of Scientific Instrument, 2019, 40(12): 224-233. (in Chinese)
[7] Xu Z F, Han X M, Shen H Y, et al. NavRL: learning safe flight in dynamic environments [J]. IEEE Robotics and Automation Letters, 2025, 10(4): 3668-3675.
[8] Vera-Yanez D, Pereira A, Rodrigues N, et al. Optical flow-based obstacle detection for mid-air collision avoidance [J]. Sensors, 2024, 24(10): 3016.
[9] Yazdi M, Bouwmans T. New trends on moving object detection in video images captured by a moving camera: a survey [J]. Computer Science Review, 2018, 28: 157-177.
[10] Rozantsev A, Lepetit V, Fua P. Detecting flying objects using a single moving camera [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(5): 879-892.
[11] Ester M, Kriegel H P, Sander J, et al. A density-based algorithm for discovering clusters in large spatial databases with noise [C]//The Second International Conference on Knowledge Discovery and Data Mining, 1996: 226-231.
[12] Zhou Y, Tuzel O. VoxelNet: end-to-end learning for point cloud based 3D object detection [C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018: 4490-4499.
[13] Yan Y, Mao Y X, Li B. SECOND: sparsely embedded convolutional detection [J]. Sensors, 2018, 18(10): 3337.
[14] Schmid L, Andersson O, Sulser A, et al. Dynablox: real-time detection of diverse dynamic objects in complex environments [J]. IEEE Robotics and Automation Letters, 2023, 8(10): 6259- 6266.
[15] Wu H J, Li Y H, Xu W, et al. Moving event detection from LiDAR point streams [J]. Nature Communications, 2024, 15(1): 345.
[16] Tibebu H, Roche J, De Silva V, et al. LiDAR-based glass detection for improved occupancy grid mapping [J]. Sensors, 2021, 21(7): 2263.
[17] Chen X Z, Ma H M, Wan J, et al. Multi-view 3D object detection network for autonomous driving [C]//IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017: 6526-6534.
[18] Xu Z F, Zhan X Y, Xiu Y M, et al. Onboard dynamic-object detection and tracking for autonomous robot navigation with RGB-D camera [J]. IEEE Robotics and Automation Letters, 2024, 9(1): 651-658.
[19] Wang Z J, Wu Y, Niu Q Q. Multi-sensor fusion in automated driving: a survey [J]. IEEE Access, 2020, 8: 2847-2868.
[20] Xu Z, Shen H, Han X, et al. LV-DOT: LiDAR-visual dynamic obstacle detection and tracking for autonomous robot navigation [DB/OL]. (2025-12-28) [2026-02-02]. https://arxiv.org/abs/2502.20607.
[21] Ge B S, Zhang H, Jiang L Y, et al. Adaptive unscented Kalman filter for target tracking with unknown time-varying noise covariance [J]. Sensors, 2019, 19(6): 1371.
[22] Qin T, Li P L, Shen S J. VINS-Mono: a robust and versatile monocular visual-inertial state estimator [J]. IEEE Transactions on Robotics, 2018, 34(4): 1004-1020.
[23] Xu W, Cai Y X, He D J, et al. FAST-LIO2: fast direct LiDAR-inertial odometry [J]. IEEE Transactions on Robotics, 2022, 38(4): 2053-2073.
[24] Arulampalam M S, Maskell S, Gordon N, et al. A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking [J]. IEEE Transactions on Signal Processing, 2002, 50(2): 174-188.
[25] Vizzo I, Guadagnino T, Mersch B, et al. KISS-ICP: in defense of point-to-point ICP–simple, accurate, and robust registration if done the right way [J]. IEEE Robotics and Automation Letters, 2023, 8(2): 1029-1036.
[26] Malladi M V R, Guadagnino T, Lobefaro L, et al. A robust approach for LiDARinertial odometry without sensor-specific modeling [DB/OL]. (2025-09-08) [2026-02-02]. https://arxiv.org/abs/2509.06593.
[27] Gal Y, Ghahramani Z. Dropout as a Bayesian approximation: Representing model uncertainty in deep learning [C]//The 33rd International Conference on Machine Learning. PMLR, 2016: 1050-1059.
[28] Lakshminarayanan B, Pritzel A, Blundell C. Simple and scalable predictive uncertainty estimation using deep ensembles [C]//31st International Conference on Neural Information Processing Systems, 2017: 6405-6416.
[29] Zhang J, Singh S. Low-drift and real-time LiDAR odometry and mapping [J]. Autonomous Robots, 2017, 41(2): 401-416.
[30] Engel J, Koltun V, Cremers D. Direct sparse odometry [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(3): 611-625.
[31] Kalman R E. A new approach to linear filtering and prediction problems [J]. Journal of Basic Engineering, 1960, 82(1): 35-45.
[32] Mohammadikaji M, Bergmann S, Irgenfried S, et al. A framework for uncertainty propagation in 3D shape measurement using laser triangulation [C]//IEEE International Instrumentation and Measurement Technology Conference Proceedings, 2016: 1-6.
[33] 刘德儿, 朱磊, 冀炜臻, 等. 基于RGB-D相机的脐橙实时识别定位与分级方法[J]. 农业工程学报, 2022, 38(14): 154-165. Liu D E, Zhu L, Ji W Z, et al. Real-time identification, localization, and grading method for navel oranges based on RGB-D camera [J]. Transactions of the Chinese Society of Agricultural Engineering, 2022, 38(14): 154-165. (in Chinese)
[34] Tang H, Zhang T, Wang L, et al. I2Nav-robot: a large-scale indoor-outdoor robot dataset for multi-sensor fusion navigation and mapping [DB/OL]. (2025-08-15) [2026-02-02]. https://arxiv.org/abs/2508.11485.
[35] Song J, Li W, Duan C, et al. R2-GVIO: a robust, real-time GNSS-visual-inertial state estimator in urban challenging environments [J]. IEEE Internet of Things Journal, 2024, 11(12): 22269-22282.
[36] Geiger A, Lenz P, Stiller C, et al. Vision meets robotics: the KITTI dataset [J]. The International Journal of Robotics Research, 2013, 32(11): 1231-1237.
[37] Caesar H, Bankiti V, Lang A H, et al. nuScenes: a multimodal dataset for autonomous driving [C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 11621- 11631.
[38] Yin J, Li A, Li T, et al. M2DGR: a multi-sensor and multi-scenario SLAM dataset for ground robots [J]. IEEE Robotics and Automation Letters, 2022, 7(2): 2266-2273.
Options
文章导航

/

〈 〉