基于视觉与深度信息的无人机紧急自主降落方法研究
摘要
度信息的自主降落方法。利用单目图像构建图结构提取地貌拓扑连通性,结合深度信息过滤非降落区域,构建基于
视觉与深度信息的启发式决策网络(Visual-Depth Heuristic Network,VDHN)。该网络通过启发式策略评估降落适
宜区域,引导无人机自主降落。仿真实验中,VDHN成功率达94.5%,显著提升了无人机紧急着陆能力。
关键词
全文:
PDF参考
[1]DU W, LIN J, DU B, et al. Autonomous Emergency
Landing on 3D Terrains: Approaches for Monocular Visionbased UAVs[C]//2023 International Conference on Advanced
Robotics and Mechatronics(ICARM).2023:105-112.
[2]CHEN J, DU W, LIN J, et al. Emergency UAV
Landing on Unknown Field Using Depth-Enhanced Graph
Structure[J].Automation Science and Engineering,IEEE
Transactions on,2025,22(000):4434-4445.DOI:10.1109/
TASE.2024.3391017.
[3]马宁,曹云峰.面向无人机自主着陆的视觉感知
与位姿估计方法综述[J].自动化学报,2024,50(7):
1284-1304.DOI:10.16383/j.aas.c230557.
[4]袁泽慧,陈国栋,张亚,等.基于复合标志的四
旋翼无人机自主降落系统[J].传感器与微系统,2024,43
(3):103-107.DOI:10.13873/J.1000-9787(2024)03-
0103-05.
[5]YOO S, PARK J, CHANG D. A Two-Step
Controller for Vision-Based Autonomous Landing of
a Multirotor with a Gimbal Camera[J].Drones(2504-
446X),2024,8(8):389-405.DOI:10.3390/drones8080389.
[6]KIM D, CHEN T. Deep Neural Network for RealTime Autonomous Indoor Navigation[EB/OL].(2015-11-
18)[2026-3-16].https://arxiv.org/abs/1511.04668.
Refbacks
- 当前没有refback。
