机载线阵激光雷达动态成像建模与验证

doi:10.12382/bgxb.2024.0836

摘要/Abstract

摘要：

针对无人机成像场景典型目标点云数据集稀缺的问题,建立机载线阵激光雷达动态成像模型,提出一种基于柱体素的点云目标检测算法对仿真数据真实性进行验证。基于虚拟仿真平台建立典型目标模型与崎岖地形、伪装覆盖和植被遮蔽等典型场景,结合激光点云获取仿真模型、机载平台运动模型、拼接成像及畸变校正方法,获取激光点云数据集,基于重叠面积的完整度判别标注方法进行数据集标注;采用柱体素特征提取模块处理目标顶部特征,通过标注的仿真数据集训练点云目标检测算法,并在基于等效实验获取的真实数据集上进行算法评价。检测算法在实测数据集上的识别准确率为93.2%,评价结果表明机载线阵激光雷达动态成像模型具有较高的可信度,仿真数据能够反映真实目标特征。

关键词: 线阵激光雷达, 推扫成像, 激光点云, 无人机, 目标识别

Abstract:

A dynamic imaging model for UAV(Unmanned Aerial Vehicle)-borne line-array LiDAR is proposed to address the scarcity of point cloud datasets in UAV imaging scene.A pillar-voxel-based point cloud target detection algorithm is proposed to verify the authenticity of the simulation data.The typical target models and the typical scenarios with rugged terrain,camouflage,and vegetation cover are established based on a virtual simulation platform.A laser point cloud dataset is generated by using the point cloud simulation model,UAV motion model,stitching imaging and distortion correction methods.The dataset is annotated by using a completeness judgment method based on overlapping areas.A pilar-voxel feature extraction module is used to process the target’s top features.The point cloud target detection algorithm is trained using the annotated simulation dataset,and evaluated on a real dataset obtained from equivalent experiments.The proposed algorithm achieves an accuracy rate of 93.2% on the real dataset.The evaluated result indicates that the simulated data effectively reflect the true characteristics of the targets,and the dynamic imaging model has high credibility.

Key words: line-array LiDAR, pushbroom imaging, laser point cloud, unmanned aerial vehicle, target recognition

中图分类号:

TJ43⁺9.2

王伟翰, 高铭泽, 施小龙, 胡诗苑, 吴沿江, 陈慧敏. 机载线阵激光雷达动态成像建模与验证[J]. 兵工学报, 2025, 46(6): 240836-.

WANG Weihan, GAO Mingze, SHI Xiaolong, HU Shiyuan, WU Yanjiang, CHEN Huimin. Modeling and Verification of Dynamic Imaging of UAV-borne Line-array LiDAR[J]. Acta Armamentarii, 2025, 46(6): 240836-.

图/表 18

图1 线阵激光推扫成像原理示意图

Fig.1 Schematic diagram of pushbroom imaging

图2 无人机与目标交会场景示意图

Fig.2 Scene of UAV and target rendezvous

图3 典型目标模型

Fig.3 Typical target model

图4 崎岖地形场景构建

Fig.4 Construction of rugged terrain scenes

图5 伪装覆盖与植被遮蔽场景构建

Fig.5 Construction of camouflage and vegetation cover scene

图6 发射方向向量簇示意图

Fig.6 Emission direction vector cluster

图7 激光雷达坐标系下不同攻击角的点云

Fig.7 Point clouds with different attack angles in the coordinate system of LiDAR

图8 目标坐标系下畸变校正后的点云

Fig.8 Point cloud after distortion correction in the target coordinate system

图9 柱体素特征提取方法

Fig.9 Pillar-voxel-based feature extraction method

图10 骨干网络示意图

Fig.10 Schematic diagram of backbone network

表1 线阵激光雷达成像仿真参数

参数	数值	参数	数值
脱靶量/m	5	激光重频/Hz	360
脱靶方位/(°)	0~180	起始位置/m	10
交会角度/(°)	0~360	水平视场角/(°)	60
飞行速度/(m·s^-1)	10	激光线数	300
攻击角/(°)	0~60	探测距离/m	20

图11 典型目标及参数下的仿真结果

Fig.11 Simulated results under typical targets and parameters

图12 典型环境干扰下的仿真点云

Fig.12 Simulation point cloud under typical environmental interference

图13 坦克与家用汽车的识别准确率迭代曲线

Fig.13 Iterative curves of recognition accuracy for tank and car

图14 各典型场景与参数下的识别效果

Fig.14 Recognition effects under various typical scenarios and parameters

图15 线阵激光雷达成像等效实验场景

Fig.15 Equivalent experimental imaging scene of line-array LiDAR

图16 实测数据及仿真数据

Fig.16 Measured and simulated data

图17 实测点云数据的识别效果

Fig.17 Recognition effect of measured point cloud

参考文献 25

[1]	张宇, 张焱, 石志广, 等. 基于图像衍生的红外无人机图像仿真方法研究[J]. 光学学报, 2022, 42(2):99-112.
	ZHANG Y, ZHANG Y, SHI Z G, et al. Image simulation method of infrared UAV based on image derivation[J]. Acta Optica Sinica, 2022, 42(2):99-112. (in Chinese)
[2]	陈杉杉, 张合, 徐孝彬. 脉冲激光引信探测平面目标的回波特性研究[J]. 兵工学报, 2018, 39(6):1095-1102. doi: 10.3969/j.issn.1000-1093.2018.06.008
	CHEN S S, ZHANG H, XU X B. Echo characteristic of planar target in pulsed laser fuze detection[J]. Acta Armamentarii, 2018, 39(6):1095-1102. (in Chinese)
[3]	陈慧敏, 贾晓东, 蔡克荣. 激光引信技术[M]. 北京: 国防工业出版社, 2016.
	CHEN H M, JIA X D, CAI K R. Laser fuze technology[M]. Beijing: National Defense Industry Press, 2016. (in Chinese)
[4]	霍健, 陈慧敏, 马云飞, 等. 基于MEMS激光雷达的车辆目标识别算法[J]. 兵工学报, 2023, 44(4):940-948.
	HUO J, CHEN H M, MA Y F, et al. Vehicle target recognition algorithm based on MEMS LiDAR[J]. Acta Armamentarii, 2023, 44(4):940-948. (in Chinese) doi: 10.12382/bgxb.2021.0822
[5]	邓涛, 刘双, 魏汉迪. 基于视觉的无人机艇系统姿态感知研究[J]. 舰船科学技术, 2024, 46(16):81-89.
	DENG T, LIU S, WEI H D. Research on visual-based motion sensing for unmanned drone-vessel system[J]. Ship Science and Technology, 2024, 46(16):81-89. (in Chinese)
[6]	任娜, 张玉, 王洪江, 等. 基于二维正态分布的无人机激光雷达点云匹配研究[J]. 激光杂志, 2024, 45(4):265-270.
	REN N, ZHANG Y, WANG H J, et al. Research on UAV LiDAR point cloud matching based on two-dimensional normal distribution[J]. Laser Journal, 2024, 45(4):265-270. (in Chinese)
[7]	FANG J, ZHOU D F, YAN F L, et al. Augmented LiDAR simulator for autonomous driving[J]. IEEE Robotics and Automation Letters, 2020, 5(2):1931-1938.
[8]	RICHA J P, DESCHAUD J E, GOULETTE F, et al. AdaSplats:adaptive splatting of point clouds for accurate 3D modeling and real-time high-fidelity LiDAR simulation[J]. Remote Sensing, 2022, 14(24):6262.
[9]	LI C Q, REN Y, LIU B B. Pcgen:point cloud generator for LiDAR simulation[C]// Proceedings of 2023 IEEE International Conference on Robotics and Automation.London,UK:IEEE, 2023:11676-11682.
[10]	YANG T, LI Y, RUICHEK Y, et al. Performance modeling a near-infrared Tof LiDAR under fog:a data-driven approach[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 23(8):11227-11236.
[11]	LEE J, SHIOTSUKA D, NISHIMORI T, et al. GAN-based lidar translation between sunny and adverse weather for autonomous driving and driving simulation[J]. Sensors, 2022, 22(14):5287.
[12]	MARCUS R, GABEL F, KNOOP N, et al. GAN-based LiDAR intensity simulation[C]// Proceedings of the 4th International Conference on Deep Learning Theory and Applications. Rome,Italy: Springer, 2023:419-433.
[13]	LI M L, HU Y H, ZHAO N X, et al. Modeling and analyzing point cloud generation in missile-borne LiDAR[J]. Defence Technology, 2020, 16(1):69-76. doi: 10.1016/j.dt.2019.10.003
[14]	张文豪, 李松, 马跃, 等. 星载光子计数激光雷达海面点云仿真方法[J]. 红外与毫米波学报, 2020, 39(4):483-490.
	ZHANG W H, LI S, MA Y, et al. Photon-counting lidar simulation method based on three-dimensional sea surface[J]. Journal of Infrared and Millimeter Waves, 2020, 39(4):483-490. (in Chinese)
[15]	陈剑彪, 孙华燕, 赵延仲. 空中目标激光雷达一维距离像仿真及实验研究[J]. 激光与光电子学进展, 2017, 54(7):324-331.
	CHEN J B, SUN H Y, ZHAO Y Z. Simulation and experimental research on one-dimensional Lidar range profile of airborne target[J]. Laser & Optoelectronics Progress, 2017, 54(7):324-331. (in Chinese)
[16]	武军安, 郭锐, 刘荣忠, 等. 用于弹载线阵红外与激光扫描成像引信的轻量化卷积神经网络目标识别方法[J]. 弹道学报, 2021, 33(3):89-96. doi: 10.12115/j.issn.1004-499X(2021)03-014
	WU J A, GUO R, LIU R Z, et al. A lightweight convolutional neural network target recognition method for missile-borne fuse linear array infrared and LiDAR-scanning image[J]. Journal of Ballistics, 2021, 33(3):89-96. (in Chinese)
[17]	张笑宇, 王凤香, 郭颖, 等. 基于InGaAs单光子探测器的线阵扫描激光雷达及其光子信号处理技术研究[J]. 红外与激光工程, 2023, 52(3):186-194.
	ZHANG X Y, WANG F X, GUO Y, et al. Research on linear array scanning LiDAR and photon signal processing technology based on InGaAs single-photon detector[J]. Infrared and Laser Engineering, 2023, 52(3):186-194. (in Chinese)
[18]	陈慧敏, 李铁, 刘锡民, 等. 近程激光探测技术[M]. 北京: 北京理工大学出版社, 2018.
	CHEN H M, LI T, LIU X M, et al. Short range laser detection technology[M]. Beijing: Beijing Institude of Technology Press, 2018. (in Chinese)
[19]	GAUDFRIN F, PUJOL O, RIVIERE N, et al. A new lidar technique based on supercontinuum laser sources for aerosol soundings:simulations and measurements—the PERFALIS code and the COLIBRIS instrument[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61(1):1-21.
[20]	LIAO Y P, SHANGGUAN M J, YANG Z F, et al. GPU-accelerated Monte Carlo simulation for a single-photon underwater lidar[J]. Remote Sensing, 2023, 15(21):1-9.
[21]	黄泽青, 贺伟. 基于激光引信的回波仿真及抗干扰研究[J]. 激光技术, 2020, 44(6):716-720.
	HUANG Z Q, HE W. Echo simulation and anti-jamming research based on laser fuze[J]. Laser Technology, 2020, 44(6):716-720. (in Chinese)
[22]	LIN Z D, CHEN X, DI C A, et al. Modeling and analysis of light detection and ranging point cloud error in vibration state of airborne platform[J]. Optical Engineering, 2022, 61(2):024103.
[23]	QI C R, YI L, SU H, et al. PointNet++:deep hierarchical feature learning on point sets in a metric space[C]// Advances in Neural Information Processing Systems 30.Long Beach,CA,US:NeurIPS, 2017:5099-5108.
[24]	LANG A H, VORA S, CAESAR H, et al. Pointpillars:fast encoders for object detection from point clouds[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Long Beach,CA,US:IEEE, 2019:12697-12705.
[25]	CHEN A, ZHANG K, ZHANG R, et al. PiMAE:point cloud and image interactive masked autoencoders for 3d object detection[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Vancouver,Canada:IEEE, 2023:5291-5301.