当特定任务的真实(Real)数据稀缺时,合成(Synthetic)无人机(UAV)图像可扩充部署前的测试证据,但仅靠视觉逼真度或语义标签对齐不足以确立模型选择的有效性。我们在固定评估协议下,通过两个互补案例研究合成评估何时能保留从真实无人机数据得到的模型选择结论。在特定需求的道路 debris 基准中,四个语言条件感知系统在合成数据上的绝对 F1 值更低,但点估计排序和排名第一的模型得以保留;bootstrap 分析未发现支持的成对反转,且大小匹配的重采样保留了排名第一模型的决策。相比之下,在独立实现的业务语料库上进行的闭集压力测试中,使用四个 VisDrone 兼容检测器针对八个经实证支持的类别,改变了点估计排名第一的检测器,并反转了六个成对点估计偏好方向中的三个。对五个无人机业务场景的补充审计进一步揭示,真实-合成在类别构成、目标密度和目标规模上存在显著不匹配,包括小目标的系统性代表性不足。这些结果表明,合成无人机基准的可信度是有条件的:语义覆盖度不足,必须针对感兴趣的任务、运行模式和模型面板实证确立其有效性。
英文摘要:
Synthetic UAV imagery can expand pre-deployment test evidence when task-specific Real data are scarce, but visual plausibility or semantic label alignment alone does not establish validity for model selection. We study when Synthetic evaluation preserves the model-selection conclusions obtained from Real UAV data through two complementary cases under fixed evaluation protocols. In a requirement-specific road-debris benchmark, four language-conditioned perception systems obtain lower absolute F1 on Synthetic data, yet the point-estimate ordering and top-ranked model are preserved; bootstrap analysis yields no supported pairwise reversal, and size-matched resampling preserves the top-model decision. In contrast, a closed-set stress test on a separately materialized business corpus, using four VisDrone-compatible detectors over eight empirically supported classes, changes the point-estimate top-ranked detector and reverses three of six pairwise point-estimate preference directions. A complementary audit of five UAV business scenarios further reveals substantial Real-Synthetic mismatch in class composition, object density, and target scale, including systematic under-representation of small targets. These results show that Synthetic UAV benchmark trust is conditional: semantic coverage is insufficient, and validity must be established empirically for the task, operating regime, and model panel of interest.
Visual-based Safe Landing for UAVs in Populated Areas: Real-time Validation in Virtual Environments
Universidad Autónoma de Ciudad Juárez UACJ(华雷斯自治大学);Center for Research in Mathematics CIMAT AC(数学研究中心CIMAT AC);University of Texas at El Paso(德克萨斯大学埃尔帕索分校)
RT-SHCUA: Real-Time Self-Hosted Computer-Use Agent for UAV Control
School of Computer Science and Technology, Xidian University(西安电子科技大学计算机科学与技术学院);Shaanxi Key Laboratory of Network and System Security(陕西省网络与系统安全重点实验室);College of Cyber Security, Jinan University(广州大学网络安全学院);School of Cyber Engineering, Shaanxi Key Lab of Network and System Security(陕西省网络与系统安全重点实验室)
M$^2$E-UAV: A Benchmark and Analysis for Onboard Motion-on-Motion Event-Based Tiny UAV Detection
Fujian Key Laboratory of Urban Intelligent Sensing and Computing, School of Informatics, Xiamen University, Xiamen, China(福建城市智能感知与计算重点实验室,厦门大学信息学院,厦门,中国);Meta, Menlo Park, USA(Meta,Menlo Park,美国)
SinD 2.0: A Multi-City UAV Dataset with Semantic Risk Annotations for SOTIF-Oriented Safety Validation at Signalized Intersections
Tsinghua University(清华大学);Beijing Institute of Technology(北京理工大学);Guangzhou Automobile Group Co., Ltd.(广州汽车集团股份有限公司);Jilin University(吉林大学);Chang’an University(长安大学);Chongqing University(重庆大学);Southwest University(西南大学);Wuhan University of Technology(武汉理工大学);Virginia Tech Transportation Institute(弗吉尼亚理工大学交通研究所);Virginia Tech(弗吉尼亚理工大学)
Does it work outside this benchmark? Introducing the Rigid Depth Constructor tool, depth validation dataset construction in rigid scenes for the masses
U2IS, ENSTA Paris, Institut Polytechnique de Paris(巴黎综合理工学院,ENSTA巴黎,U2IS)
我们提出了一个协议来构建自己的深度验证数据集导航。该协议被称为RDC for Rigid Depth Constructor,旨在比现有技术更容易获得和更便宜,只需要一个摄像头和一个激光雷达传感器即可开始。我们还开发了一个测试套件,以获得有见地的信息,从评估的算法。最后,我们以无人机视频为例,测试了最初在KITTI上测试的两种深度算法,并表明无人机上下文与车内视频有很大不同。这表明,单个上下文基准不应被认为是可靠的,并且在开发深度估计算法时,应该在最适合特定需求的数据集上对其进行基准测试,这通常意味着创建一个全新的数据集。在本文中,我们提供了一个开源实现的工具,并计划使其尽可能用户友好,即使是小团队也可以创建深度数据集。我们的主要贡献如下:我们提出了一个完整的,开源的,几乎完全自动化的软件应用程序,用于创建具有密集注释深度的验证数据集,适用于各种图像,视频和范围数据。它包括用于使数据集适应特定验证需求的选择工具,以及用于其他数据集格式的转换工具。使用这个应用程序,我们提出了两个新的真实数据集,室外和室内,易于使用的无人机导航上下文。最后,作为例子,我们展示了两个深度预测算法的评估,使用一个全面的(例如,基于分布的)指标的集合。
Droneulator: A Portable UAV Simulator for Agricultural Workflows with RotorPy and Godot 4
Lincoln Centre for Autonomous Systems (L-CAS), University of Lincoln, Lincoln, UK(林肯自主系统中心(L-CAS)、林肯大学、林肯,英国);Faculty of Aerospace Engineering, MAVLab, Delft University of Technology (TU Delft), Delft, Netherlands(航空航天工程学院、MAVLab、代尔夫特理工大学(TU Delft)、代尔夫特,荷兰)
BIM-Discrepancy-Driven Active Sensing for Risk-Aware UAV-UGV Navigation
Department of Electrical and Computer Engineering, University of California San Diego, La Jolla, CA, USA(电气与计算机工程系,加州大学圣地亚哥分校);Department of Civil, Construction, and Environmental Engineering, San Diego State University, San Diego, CA, USA(土木、建设与环境工程系,圣地亚哥州立大学)
SAGE-LLM: Towards Safe and Generalizable LLM Controller with Fuzzy-CBF Verification and Graph-Structured Knowledge Retrieval for UAV Decision
School of Artificial Intelligence, OPtics and ElectroNics (iOPEN), Northwestern Polytechnical University, Xi’an 710072, China(人工智能学院、光学与电子学(iOPEN)、西北工业大学,西安 710072,中国);Institute of Artificial Intelligence (TeleAI), China Telecom, China(人工智能研究所(TeleAI)、中国电信,中国)
Flying by Inference: Active Inference World Models for Adaptive UAV Swarms
Department of Engineering and Naval Architecture (DITEN), University of Genoa(工程与 naval 架构系(DITEN),热那亚大学);Intelligent Systems Laboratory, Department of Systems Engineering and Automation, Carlos III University of Madrid(智能系统实验室,系统工程与自动化系,马德里卡洛斯三世大学)
UAV and Machine Learning Based Refinement of a Satellite-Driven Vegetation Index for Precision Agriculture
Politecnico di Torino(都灵理工大学);Politecnico Interdepartmental Centre for Service Robotics(都灵理工大学跨部门服务机器人中心);Università degli Studi di Torino(都灵大学);Institute of Electronics, Computer and Telecommunication Engineering of the National Research Council of Italy(意大利国家研究委员会电子、计算机与电信工程研究所)
精准农业被认为是追求低投入、高效率和可持续农业的基本途径。为了实现这一目标,需要对当地作物状况进行可靠和最新的描述。遥感,特别是卫星图像,已证明是作物制图、监测和疾病评估的宝贵工具。然而,免费提供的低分辨率或中等分辨率的卫星图像在具体的农业应用方面显示出一些限制,例如,农作物是一排排种植的。事实上,在这个框架内,卫星的输出可能会受到行内覆盖的影响,从而提供关于作物状况的不准确信息。本文提出了一种新的卫星图像细化框架,基于深度学习技术,该技术充分利用了无人机(UAV)机载多光谱传感器获取的高分辨率图像中的信息。为了训练卷积神经网络,仅需要单个无人机驱动的数据集,使得所提出的方法简单且具有成本效益。Serralunga d 'Alba(意大利北部)的葡萄园被选为验证目的的案例研究。细化卫星驱动的归一化差异植被指数(NDVI)地图,在葡萄生长季节的四个不同时期获得的,被证明是更好地描述作物状况相对于原始数据集的相关性分析和方差分析。此外,使用基于K-均值的分类器,3级葡萄园活力地图是有益的,从NDVI地图,这是一个有价值的工具,为种植者。
A UAV-assisted Wireless Localization Challenge on AERPAW
School of Electrical and Computer Engineering, University of Georgia(佐治亚大学电气与计算机工程学院);University of North Texas(北得克萨斯大学);NYU Tandon School of Engineering(纽约大学坦登工程学院);Delmont Systems LLC(德尔蒙特系统有限责任公司);North Carolina State University(北卡罗来纳州立大学)
A UAV-Based VNIR Hyperspectral Benchmark Dataset for Landmine and UXO Detection
Rochester Institute of Technology, Chester F. Carlson Center for Imaging Science(罗切斯特理工学院,切斯特·F·卡尔森成像科学中心);Lamont Doherty Earth Observatory, Columbia University(拉蒙特-多赫蒂地球观测站,哥伦比亚大学);Demining Research Community(排雷研究社区)
Affordable Data Collection System for UAVs Taxi Vibration Testing
Department of Aerospace Engineering, Universidad Carlos III de Madrid(航空航天工程系,卡洛斯三世大学);Electronic Technology Department, Universidad Carlos III de Madrid(电子技术系,卡洛斯三世大学)
Curriculum-Based Reinforcement Learning for Autonomous UAV Navigation in Unknown Curved Tubular Conduit
DGA Techniques Navales - Direction Générale de l’Armement(法国国防总局海军技术部);LIRMM, TETIS, CNRS, Université de Montpellier Paul-Valéry(蒙彼利埃大学保罗-瓦莱里耶大学);LIS, CNRS, Université de Toulon(土伦大学)
UniRef-UAV: A Multimodal Benchmark for Universal Referring in UAV Imagery
School of Automation, Northwestern Polytechnical University(自动化学院,西北工业大学);College of Missile Engineering, Rocket Force University of Engineering(导弹工程学院,火箭兵工程大学);School of Artificial Intelligence, Chongqing University of Posts and Telecommunications(人工智能学院,重庆邮电大学)
A Multimodal Large Language Model-Driven Framework for Context-Aware UAV Emergency Landing Site Selection
School of Information Science and Engineering, Southeast University(信息科学与工程学院,东南大学);LASER, International Digital Economy Academy(LASER,国际数字经济学院);Department of Electronic Engineering, East China Normal University(电子工程系,华东师范大学)
Agricultural Plant Cataloging and Establishment of a Data Framework from UAV-based Crop Images by Computer Vision
Fraunhofer Institute for Intelligent Analysis and Information Systems IAIS(弗劳恩霍夫智能分析与信息系统研究所(IAIS));University of Bonn(波恩大学);Institute for Sugar Beet Research (IfZ)(甜菜研究所(IfZ));Technical University of Munich(慕尼黑工业大学)