New Algorithm for Unmanned Aerial Vehicle Path Planning Based on Sparse Bridge Point Cloud Structure Mapping and Convenient and Efficient Bridge Inspection Practice
Currently, bridge inspections mainly employ inspection vehicles and directed drone inspections, which present issues such as inspection gaps and imbalanced monitoring areas. For the inspection requirements of different key components of bridges and, based on the symmetric structure of sparse bridge point clouds, was a UAV path-planning algorithm driven by structural mapping of sparse bridge point clouds was proposed. Firstly, utilizing the bridge point cloud model, a point cloud completion algorithm leveraging structural symmetry was designed to achieve the completion of missing parts of the bridge model. Secondly, the Sanjiang Bridge was divided into the pier and the main girder, and feature points were extracted for each module to obtain the relative positions of the bridge components and to determine the inspection range. Finally, according to photogrammetry and inspection task requirements, optimal waypoints were interpolated and expanded between the feature points, and all waypoints were connected to generate a feasible flight path. An overall coverage rate of 91.3% for Sanjiang Bridge is achieved, which is about 9.4% higher than the A* algorithm, improving the efficiency of bridge monitoring and strongly supporting refined bridge detection. This method effectively reduces the existing loopholes in bridge detection and provides a convenient and efficient new idea for unmanned aerial vehicle bridge inspection route planning.
ZHOU Lian-jie, ZHANG Feng, HE Qing-lin, DU Jia-kuan, ZHANG Xiang.
New Algorithm for Unmanned Aerial Vehicle Path Planning Based on Sparse Bridge Point Cloud Structure Mapping and Convenient and Efficient Bridge Inspection Practice[J].
Science Technology and Engineering, 2026, 26(7): 2750-2758 DOI:10.12404/j.issn.1671-1815.2502541
桥梁信息模型包含建筑信息模型(building information modeling, BIM),点云、摄影测量模型等[8]。Almadhoun等[9]在初始飞行获取粗略点云模型后,根据该模型的完整性评估结果自动补充相机视点,从而进一步完善缺失区域并提升重建精度。点云模型以精度高的优点被广泛应用在无人机巡检中。对于有残缺的点云模型,已形成若干补全方法。刘青青等[10]提出基于多视角点云配准的三维重建方法,通过粗配准以及基于权重的精配准进行点云拼接。郑立华等[11]结合随机采样一致性和迭代最近点算法实现对非完整点云的三维重建。刘心溥等[12]提出一种嵌入Transformer结构的多尺度点云补全方法。然而,目前点云补全方法仍缺少以桥梁巡检特征视角进行补全的机制。
通过与实际建筑信息模型(building information modeling,BIM)桥梁模型进行比较,如图7所示为A*算法、CNN、蚁群算法与本文方法在补全精度、补全计算时间上比较。其中,A*算法、CNN、本文方法的补全精度较高,达到88.3%、87.5%、89.3%,蚁群算法精度补全精度达到78.3%。在补全时间上,本文方法补全计算时间较短达到89 s,基于CNN方案的残缺点云补全时间最长,达到335 s。因此,本方法在保证精度的前提下,计算时间较短。
ChenKen, ZhongAiping, WenYuxuan, et al. Research on data compression algorithm for bridge health monitoring[J]. Science Technology and Engineering, 2025, 25(8): 3304-3315.
[3]
FiandacaD, Di MatteoA, PatellaB, et al. An integrated approach for structural health monitoring and damage detection of bridges: an experimental assessment[J]. Applied Sciences, 2022, 12(24): 13018.
LiLingxing, ZhangXiaojun. Lidar road edge detection algorithm based on object detection[J]. Science Technology and Engineering, 2024, 24(9): 3519-3525.
[6]
FerozS, Abu DabousS. UAV-based remote sensing applications for bridge condition assessment[J]. Remote Sensing, 2021, 13(9): 1809.
[7]
DuqueL, SeoJ, WackerJ. Bridge deterioration quantification protocol using UAV[J]. Journal of Bridge Engineering, 2018, 23(10): 04018080.
[8]
ZhangY, ZhaoP, LiX, et al. Retrieval of terrain surface elevation in mountainous areas with ICESat-2/ATLAS[J]. Remote Sensing of Environment, 2025, 327(1): 114823.
LiuJunchen, ZhuWeizhu, ZhouZhixiang, et al. Bridge structure displacement monitoring method based on camera pinhole model[J]. Science Technology and Engineering, 2025, 25(1): 383-393.
[11]
BolourianN, HammadA. LiDAR-equipped UAV path planning considering potential locations of defects for bridge inspection[J]. Automation in Construction, 2020, 117(10): 103250.
[12]
AlmadhounR, AbduldayemA, TahaT, et al. Guided next best view for 3D reconstruction of large complex structures[J]. Remote Sensing, 2019, 11(22): 2622.
LiuQingqing, WuTao. Research on RealSense based multi view 3D reconstruction method for human body[J]. Foreign Electronic Measurement Technology, 2021, 40(10): 77-83.
ZhengLihua, WangLuhan, WangMinjuan, et al. Automated 3D point cloud reconstruction of oilseed rape based on Kinect camera[J]. Journal of Agricultural Machinery, 2021, 52(7): 159-168.
LiuXinbo, MaYanxin, XuKe, et al. Multi scale point cloud completion embedded with Transformer structure[J]. Journal of Image and Graphics, 2022, 27(2): 538-549.
[19]
AlmadhounR, TahaT, SeneviratneL, et al. A survey on multi-robot coverage path planning for model reconstruction and mapping[J]. SN Applied Sciences, 2019, 1(8): 847.
ZhangRongxia, WuChangxu, SunTongchao, et al. Deep reinforcement learning and its research progress in path planning[J]. Computer Engineering and Applications, 2021, 57(19): 44-56.
XiaoYao, WangQiang, JinZhongping, et al. Path planning of quadruped gas micro leak inspection robot based on improved A * algorithm[J]. Science Technology and Engineering, 2024, 24(13): 5421-5426.
ZhangHao, LiuWei. Robot path planning based on improved ant colony algorithm[J]. Science Technology and Engineering, 2025, 25(3): 1142-1149.
[26]
HuangS, TianJ, QiaoL, et al. Unmanned aerial vehicle path planning based on improved genetic algorithm[J]. Journal of Computer Applications, 2021, 41(2): 390.
[27]
Puente-CastroA, RiveroD, PazosA, et al. A review of artificial intelligence applied to path planning in UAV swarms[J]. Neural Computing, 2022, 34(1): 1-18.
[28]
SaleemM R, ParkJ W, LeeJ H, et al. Instant bridge visual inspection using an unmanned aerial vehicle by image capturing and geo-tagging system and deep convolutional neural network[J]. Structural Health Monitoring, 2021, 20(4): 1760-1777.
XuXiaoling, LeiGaowei, LiuMei. Multi UAV smooth patrol path planning algorithm[J]. Science Technology and Engineering, 2024, 24(19): 8346-8355.
[31]
Lozano-GalantF, MontalvoA, AtencioE, et al. An open-source GIS app for digitalising bridge inspections and automating BIM model integration[J]. Structure and Infrastructure Engineering, 2025, 21(2): 1-12.