基于激光雷达点云补全的飞机停泊引导定位研究

魏宁; 李明磊; 陈广永; 叶方舟

doi:10.12000/JR25002

基于激光雷达点云补全的飞机停泊引导定位研究

DOI: 10.12000/JR25002 CSTR: 32380.14.JR25002

魏宁^{1, 2},
李明磊^1, ,,
陈广永²,
叶方舟²

1.
南京航空航天大学电子信息工程学院南京 211106
2.
中国航空无线电电子研究所上海 200241

基金项目: 国家自然科学基金(42271343)

详细信息

作者简介:
魏　宁，硕士生，研究方向为三维计算机视觉、点云补全技术

李明磊，博士，副教授，研究方向为三维计算机视觉、先进遥感技术、面向自动驾驶/无人机的三维环境感知、激光雷达数据处理等

陈广永，硕士，高级工程师，主要研究方向为光电融合、多源环境感知

叶方舟，硕士，高级工程师，主要研究方向为电信技术、航路算法规划等

通讯作者:
李明磊 minglei_li@nuaa.edu.cn

责任主编：杨星 Corresponding Editor: YANG Xing
中图分类号: TN958.98
计量
- 文章访问数:
- HTML全文浏览量:
- PDF下载量:
- 被引次数: 0
出版历程
- 收稿日期: 2025-01-03
- 修回日期: 2025-04-01

Research on Aircraft Docking Guidance Localization Based on LiDAR Point Cloud Completion

1.
College of Electronic and Information Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China
2.
China National Astronautical Radio Electronics Research Institute, Shanghai 200241, China

Funds: The National Natural Science Foundation of China (42271343)

More Information

Corresponding author: LI Minglei minglei_li@nuaa.edu.cn

摘要

摘要: 机场泊位引导系统对提高机场安全性和运行效率有着重要作用，为了利用激光雷达精确获取飞机停泊位置，该文提出了一种基于深度学习的点云补全网络并通过点云配准的方式定位飞机中心坐标。首先，参考真实场景中飞机停泊过程进行仿真得到模拟激光雷达点云。接着对遮挡等原因造成残缺的模拟点云进行补全，恢复出完整结构。最后将补全后的点云与飞机模型点云配准，坐标转换后计算出飞机中心点在模拟激光雷达坐标系中的准确位置。实验表明，提出的点云补全网络能够完整地恢复出模拟点云中缺失部分，从而计算出模拟点云的飞机中心坐标，实现了对飞机泊位引导过程中飞机位置的精确检测。为了便于研究人员评估和使用，文中算法可通过https://www.scidb.cn/anonymous/UXZFZkFm开源获取。
- 泊位引导系统 /
- 三维激光扫描 /
- 点云补全 /
- 点云配准 /
- 深度学习
Abstract: The airport docking guidance system is essential for enhancing airport safety and operational efficiency. This study introduces a deep learning-based point cloud completion network designed for accurate aircraft localization using LiDAR technology. Initially, the aircraft parking process is simulated in a realistic virtual environment to generate complete point cloud data. Subsequently, partial point clouds caused by occlusions or sensor limitations are processed through the proposed network to reconstruct their complete geometric structures. Then the restored point cloud is aligned with a predefined aircraft model, enabling precise calculation of the aircraft’s center coordinates in the simulated coordinate system through spatial transformation. Experimental results demonstrate that the network effectively recovers structural details from incomplete point clouds, enabling accurate computation of aircraft centroid coordinates. This approach achieves high-precision position detection for aircraft during docking, showing significant potential for practical airport applications. The codes are available at: https://www.scidb.cn/anonymous/UXZFZkFm.
- Docking Guidance System (DFS) /
- Three-dimensional laser scanning /
- Point cloud completion /
- Point cloud registration /
- Deep learning

HTML全文

图 1 飞机泊位引导系统示意图

Figure 1. Structure of airport docking guidance system

下载: 全尺寸图片幻灯片

图 2 PointSimCompletiton网络结构图

Figure 2. Architecture of PointSimCompletiton network

下载: 全尺寸图片幻灯片

图 3 几何感知的Transformer模块

Figure 3. Geometry-aware Transformer block

下载: 全尺寸图片幻灯片

图 4 基于仿真数据的激光雷达点云补全结果

Figure 4. Completion result of LiDAR point cloud based on simulation data

下载: 全尺寸图片幻灯片

图 5 坐标系示意图

Figure 5. Diagram of coordinates

下载: 全尺寸图片幻灯片

图 6 飞机停泊过程模拟扫描点云示意图

Figure 6. LiDAR simulation for aircraft docking process

下载: 全尺寸图片幻灯片

图 7 不同型号飞机扫描点云补全结果

Figure 7. Completion results of simulation point clouds for different aircraft models

下载: 全尺寸图片幻灯片

图 8 停机处雷达扫描点云

Figure 8. Scanning point cloud of parking area

下载: 全尺寸图片幻灯片

图 9 停机处采集点云补全结果

Figure 9. Scanning point cloud and completion result

下载: 全尺寸图片幻灯片

图 10 预测扫描点云中心

Figure 10. Predicted centroid of simulation point cloud

下载: 全尺寸图片幻灯片

表 1 点云中心预测误差分析

Table 1. Error analysis of predicted centroids

飞机型号	预测扫描点云中心	补全点云中心	误差(m)
A330	(–5.144 80.475 10.478)	(–5.507 80.760 10.506)	0.49
A330	(–5.044 50.565 10.678)	(–5.647 50.157 10.622)	0.56
A340	(–5.089 80.796 10.748)	(–5.468 80.725 10.771)	0.74
A340	(–5.061 50.843 10.236)	(–5.856 50.175 10.208)	0.42
波音737	(12.406 –11.009 2.973)	(12.625 –11.132 3.106)	0.28
	(12.473 –11.055 2.846)	(12.694 –11.377 2.694)	0.41
波音747	(10.703 –8.746 3.066)	(10.795 –8.429 3.124)	0.36
	(10.215 –8.488 3.131)	(10.659 –8.737 3.207)	0.51

下载: 导出CSV

参考文献(32)

[1]	韩万鹏, 蒙文, 李云霞, 等. 机场泊位引导系统的发展现状及关键技术分析[J]. 激光与红外, 2012, 42(3): 244–249. doi: 10.3969/j.issn.1001-5078.2012.03.002. HAN Wanpeng, MENG Wen, LI Yunxia, et al. Development status and key technical analysis of airport docking guide system[J]. Laser & Infrared, 2012, 42(3): 244–249. doi: 10.3969/j.issn.1001-5078.2012.03.002.
[2]	任宁, 徐乃付. 泊位引导系统与A-SMGCS系统运行方式探究[C] 中国计算机用户协会网络应用分会2021年第二十五届网络新技术与应用年会论文集, 北京. 2021: 4. doi: 10.26914/c.cnkihy.2021.047813. REN Ning and XU Naifu. Study of operation mode of visual docking guidance system and A-SMGCS[C]. The 25th Annual Conference on Network New Technologies and Applications of the China Computer User Association Network Application Branch (p. 4). Beijing, China, 2021: 4. doi: 10.26914/c.cnkihy.2021.047813.
[3]	张积洪, 李兴旺. 双目激光泊位引导系统[J]. 机械设计, 2015, 32(11): 88–91. doi: 10.13841/j.cnki.jxsj.2015.11.018. ZHANG Jihong and LI Xingwang. Binocular laser docking guidance system[J]. Machine Design, 2015, 32(11): 88–91. doi: 10.13841/j.cnki.jxsj.2015.11.018.
[4]	魏红波. 组合精密进近着陆技术研究[J]. 现代导航, 2017, 8(01): 5–8. doi: 10.3969/j.issn.1674-7976.2017.01.002. WEI Hongbo. Technology research on integrated precise approach and landing[J]. Modern Navigation, 2017, 8(01): 5–8. doi: 10.3969/j.issn.1674-7976.2017.01.002.
[5]	南晓虎, 丁雷. 深度学习的典型目标检测算法综述[J]. 计算机应用研究, 2020, 37(S2): 15–21. NAN Xiaohu and DING Lei. Review of typical target detection algorithms for deep learning[J]. Application Research of Computers, 2020, 37(S2): 15–21.
[6]	MITRA N J, PAULY M, WAND M, et al. Symmetry in 3d geometry: Extraction and applications[J]. Computer Graphics Forum.2013, 32(6): 1–23. doi: 10.1111/cgf.12010.
[7]	SARKAR K, VARANASI K, STRICKER D. Learning quadrangulated patches for 3d shape parameterization and completion[C]//2017 International Conference on 3D Vision (3DV). IEEE, 2017: 383–392. doi: 10.1109/3DV.2017.00051.
[8]	SUNG M, KIM V G, ANGST R, et al. Data-driven structural priors for shape completion[J]. ACM Transactions on Graphics, 2015, 34(6): 175. doi: 10.1145/2816795.2818094.
[9]	QI C R, SU H, MO K, et al. Pointnet: Deep learning on point sets for 3d classification and segmentation[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 652–660. doi: 10.1109/CVPR.2017.16.
[10]	QI C R, YI L, SU H, et al. Pointnet++: Deep hierarchical feature learning on point sets in a metric space[J]. Advances in neural information processing systems, 2017, 30. doi: 10.48550/arXiv.1706.02413.
[11]	YUAN W, KHOT T, HELD D, ET al. Pcn: Point completion network[C]//2018 international conference on 3D vision (3DV). IEEE, 2018: 728–737. doi: 10.1109/3DV.2018.00088.
[12]	CHANG A X, FUNKHOUSER T, GUIBAS L, et al. Shapenet: An information-rich 3d model repository [EB/OL]. https://arxiv.org/abs/1512.03012, 2015.
[13]	HUANG Zitian, YU Yikuan, XU Jiawen, et al. Pf-Net: Point fractal network for 3D point cloud completion[C]. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, USA, 2020: 7659–7667. doi: 10.1109/CVPR42600.2020.00768.
[14]	PROKOP M, SHAIKH S A and KIM K S. Low overlapping point cloud registration using line features detection[J]. Remote Sensing, 2019, 12(1): 61. doi: 10.3390/rs12010061.
[15]	BESL P J and MCKAY N D. Method for registration of 3-D shapes[C]. SPIE 1611, Sensor Fusion IV: Control Paradigms and Data Structures, Boston, USA, 1992: 586–606. doi: 10.1117/12.57955.
[16]	HAN Jianda, YIN Peng, HE Yuqing, et al. Enhanced ICP for the registration of large-scale 3D environment models: An experimental study[J]. Sensors, 2016, 16(2): 228. doi: 10.3390/s16020228.
[17]	李仁忠, 杨曼, 田瑜, 等. 基于ISS特征点结合改进ICP的点云配准算法[J]. 激光与光电子学进展, 2017, 54(11): 111503. doi: 10.3788/LOP54.111503. LI Renzhong, YANG Man, TIAN Yu, et al. Point cloud registration algorithm based on the ISS feature points combined with improved ICP algorithm[J]. Laser & Optoelectronics Progress, 2017, 54(11): 111503. doi: 10.3788/LOP54.111503.
[18]	BIBER P, STRAßER W. The normal distributions transform: A new approach to laser scan matching[C]//Proceedings 2003 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2003)(Cat. No. 03CH37453). IEEE, 2003, 3: 2743–2748. doi: 1 0.1109/IROS.2003.1249285.
[19]	杨宜林, 李积英, 王燕, 等. 基于NDT和特征点检测的点云配准算法研究[J]. 激光与光电子学进展, 2022, 59(8): 0810016. doi: 10.3788/LOP202259.0810016. YANG Yilin, LI Jiying, WANG Yan, et al. Point cloud registration algorithm based on NDT and feature point detection[J]. Laser & Optoelectronics Progress, 2022, 59(8): 0810016. doi: 10.3788/LOP202259.0810016.
[20]	荆路, 武斌, 李先帅. 基于SAC-IA和NDT融合的点云配准方法[J]. 大地测量与地球动力学, 2021, 41(4): 378–381. doi: 10.14075/j.jgg.2021.04.010. JING Lu, WU Bin, and LI Xianshuai. Point cloud registration method based on SAC-IA and NDT fusion[J]. Journal of Geodesy and Geodynamics, 2021, 41(4): 378–381. doi: 10.14075/j.jgg.2021.04.010.
[21]	FEI Ben, YANG Weidong, CHEN Wenming, et al. Comprehensive review of deep learning-based 3D point cloud completion processing and analysis[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(12): 22862–22883. doi: 10.1109/TITS.2022.3195555.
[22]	严璐, 顾昕. 激光雷达三维点云目标补全算法[J]. 电子技术与软件工程, 2022(5): 101–104. doi: 10.20109/j.cnki.etse.2022.05.025. YAN Lu and GU Xin. Deep learning on radar centric 3D object completion[J]. Electronic Technology & Software Engineering, 2022(5): 101–104. doi: 10.20109/j.cnki.etse.2022.05.025.
[23]	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]. The 31st International Conference on Neural Information Processing Systems, Long Beach, USA, 2017: 6000–6010.
[24]	GUO Menghao, CAI Junxiong, LIU Zhengning, et al. PCT: Point cloud transformer[J]. Computational Visual Media, 2021, 7: 187–199. doi: 10.1007/s41095-021-0229-5.
[25]	PAN Xuran, XIA Zhuofan, SONG Shiji, et al. 3D object detection with pointformer[C]. 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Nashville, USA, 2021: 7459–7468. doi: 10.1109/CVPR46437.2021.00738.
[26]	DOSOVITSKIY A, BEYER L, KOLESNIKOV A, et al. An image is worth 16x16 words: Transformers for image recognition at scale[J]. arXiv preprint arXiv: 2010. 11929, 2020.
[27]	WANG Yue, SUN Yongbin, LIU Ziwei, et al. Dynamic graph CNN for learning on point clouds[J]. ACM Transactions on Graphics (TOG), 2019, 38(5): 146. doi: 10.1145/3326362.
[28]	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[J]. Advances in neural information processing systems, 2017, 30. doi: 10.48550/arXiv.1706.03762.
[29]	YANG Yaoqing, FENG Chen, SHEN Yiru, et al. FoldingNet: Point cloud auto-encoder via deep grid deformation[C]. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2018: 206–215. doi: 10.1109/CVPR.2018.00029.
[30]	李文博, 王琦, 高尚. 基于深度学习的红外小目标检测算法综述[J]. 激光与红外, 2023, 53(10): 1476–1484. doi: 10.3969/j.issn.1001-5078.2023.10.003. LI Wenbo, WANG Qi, GAO Shang. A review of infrared small target detection algorithms based on deep learning[J]. Laser & Infrared, 2023, 53(10): 1476–1484. doi: 10.3969/j.issn.1001-5078.2023.10.003.
[31]	FISCHLER AND M A. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography[J]. Commun. ACM, 1981, 24(6): 381–395.
[32]	LIU Jiang, ZHU Jiwen, YANG Jinling, et al. Three-dimensional point cloud registration based on ICP algorithm employing K-D tree optimization[C]. SPIE 10033, Eighth International Conference on Digital Image Processing (ICDIP 2016), Chengu, China, 2016: 100334D. doi: 10.1117/12.2248362.