Computer Engineering and Applications ›› 2025, Vol. 61 ›› Issue (24): 322-330.DOI: 10.3778/j.issn.1002-8331.2409-0186

• Engineering and Applications • Previous Articles     Next Articles

Hierarchical Coverage Path Planning of Unmanned Aerial Vehicle Under Photogrammetric Constraints

PIAO Minnan1, ZHOU Yuhan1, LI Haifeng1+, LI Haolong1, SONG Dezhen2,3   

  1. 1.School of Computer Science and Technology, Civil Aviation University of China, Tianjin 300300, China
    2.School of Robotics, Mohamed Bin Zayed University of Artificial Intelligence, Abu Dhabi 54115, UAE
    3.Department of Computer Science and Engineering, Texas A&M University, Colic Station, Texas 77843, USA
  • Online:2025-12-15 Published:2025-12-15

摄影测量约束下无人机分层覆盖路径规划

朴敏楠1,周雨晗1,李海丰1+,李浩龙1,宋德臻2,3   

  1. 1.中国民航大学 计算机科学与技术学院,天津 300300
    2.穆罕默德·本·扎耶德人工智能大学 机器人学院,阿布扎比 54115
    3.得克萨斯A&M大学 计算机科学与工程学院,得克萨斯 科利奇站 77843

Abstract: To achieve automatic inspection of aircraft surface, an unmanned aerial vehicle-based coverage path planning method is proposed, which guarantees high image quality and path regularity. Firstly, cell decomposition method is employed to satisfy the overlapping ratio constraint, and viewpoints are generated based on the cells to realize complete coverage, avoiding the solution of set cover problem. By meeting the photogrammetric constraints of equidistance, front view, and overlapping ratio, comprehensive and precise evaluation of aircraft surface damage and the condition of its components can be conducted using the captured images. Furthermore, coverage path is planned through hierarchical solution of traveling salesman problem, and the semantic information of aircraft is incorporated, allowing the path exhibits regularity at different levels, which aids operators in monitoring task completion and enhances the readability of collected data flow. In addition, an iterative optimization strategy is employed to make a compromise between the path optimality and the calculation speed. Finally, extensive simulations and comparisons are performed to validate the effectiveness of the proposed algorithm.

Key words: photogrammetric constraints, viewpoint generation, hierarchical path planning, automatic inspection of aircraft surface, unmanned aerial vehicle

摘要: 为了实现飞机表面的自动检测,提出了一种基于无人机的覆盖路径规划方法,该方法保证了图像的高质量和路径的规律性。采用单元格分解方法来满足重叠率约束,并基于单元格生成的视点实现全覆盖,避免集合覆盖问题的求解。通过满足正视、等距和重叠率等摄影测量约束,可以根据采集到的图像对飞机表面损伤及其部件状况进行全面且精确的评价。进一步通过分层求解旅行商问题来规划覆盖路径,并结合飞机的语义信息,使得路径在不同层次上呈现规律性,有助于操作人员监控任务的完成情况,并提高采集数据流的可读性。采用了一种迭代优化策略,在路径最优性和计算速度之间取得了权衡。大量的仿真实验和对比实验验证了所提算法的有效性。

关键词: 摄影测量约束, 视点生成, 分层路径规划, 飞机表面自动检测, 无人机