航空学报 > 2025, Vol. 46 Issue (23): 632761-632761   doi: 10.7527/S1000-6893.2025.32761

干扰环境下无人机多源感知专栏

基于深度学习的无人机搜救方法研究进展

彭勃1,2, 白吉康2,3, 陈伟文3, 郑向涛3(), 雷建军1, 卢孝强3   

  1. 1.天津大学 电气自动化与信息工程学院,天津 300072
    2.天津大学 福州国际联合学院,天津 300072
    3.福州大学 物理与信息工程学院,福州 350108
  • 收稿日期:2025-09-08 修回日期:2025-09-24 接受日期:2025-10-22 出版日期:2025-11-14 发布日期:2025-11-13
  • 通讯作者: 郑向涛 E-mail:xiangtaoz@gmail.com
  • 基金资助:
    国家自然科学基金(62271484)

Research progress for UAV search and rescue methods based on deep learning

Bo PENG1,2, Jikang BAI2,3, Weiwen CHEN3, Xiangtao ZHENG3(), Jianjun LEI1, Xiaoqiang LU3   

  1. 1.School of Electrical and Information Engineering,Tianjin University,Tianjin 300072,China
    2.The International Joint Institute of Tianjin University,Tianjin University,Tianjin 300072,China
    3.College of Physics and Information Engineering,Fuzhou University,Fuzhou 350108,China
  • Received:2025-09-08 Revised:2025-09-24 Accepted:2025-10-22 Online:2025-11-14 Published:2025-11-13
  • Contact: Xiangtao ZHENG E-mail:xiangtaoz@gmail.com
  • Supported by:
    National Natural Science Foundation of China(62271484)

摘要:

随着人工智能技术的发展,无人机(UAV)可通过搭载不同光电载荷对感兴趣目标进行自动识别定位,用于灾害环境下的人员搜救(SaR)任务。无人机搜救任务需要在复杂危险的搜救环境中快速找到失联人员,面临着探测手段单一、成像过程易干扰、背景错综复杂、无人机平台资源受限等挑战。近年来,将无人机系统与深度学习算法相结合已成为研究热点,为了展现无人机搜救的研究进展,对无人机搜救方法进行了综述。首先,针对无人机搜救任务的特点对无人机搜救面临的难点进行了分析;然后,对近年来各类基于深度学习的无人机搜救方法进行了梳理;此外,根据不同的数据类型,整理了与无人机搜救任务相关的目标检测数据集;最后,总结了目前研究存在的问题及未来的发展方向。

关键词: 无人机搜救, 深度学习, 目标检测, 探测手段, 成像环境干扰

Abstract:

With the development of artificial intelligence technology, Unmanned Aerial Vehicles (UAVs) can automatically identify and locate objects of interest by equipping different optoelectronic payloads, and are used for Search and Rescue (SaR) missions in disaster environments. UAV search and rescue missions require the rapid location of missing persons in complex and dangerous environments, facing challenges such as single detection technique, susceptibility to interference during imaging, complex background, and limited UAV platform resources. In recent years, the integration of UAV systems with deep learning algorithms has emerged as a prominent research focus. To demonstrate the advancement of UAV search and rescue techniques, the review of these methodologies has been conducted. Firstly, the difficulties faced by UAV search and rescue are analyzed according to the characteristics of UAV search and rescue missions. Subsequently, various UAV search and rescue methods based on deep learning in recent years are sorted out. Additionally, object detection datasets relevant to UAV search and rescue missions have been sorted out according to different data types. Finally, the problems of current research and future development directions are summarized.

Key words: UAV search and rescue, deep learning, object detection, detection technique, imaging environment interference

中图分类号: