unmanned aerial vehicle
PulseAugur coverage of unmanned aerial vehicle — every cluster mentioning unmanned aerial vehicle across labs, papers, and developer communities, ranked by signal.
7 天有情绪数据
-
通过解耦目标运动与相机扰动来改进无人机目标检测
研究人员开发了一种新的纯视觉框架，以改进无人机（UAV）的目标检测。该方法有效地将检测到的目标的运动与无人机自身运动和相机抖动引起的扰动分离开来。通过采用双区间运动提取策略和运动引导注意力模块，该系统增强了特征表示，以提高准确性，尤其是在动态环境中检测小目标时。
-
Robots build hut roof with drone-delivered tiles
Two robots are shown constructing a hut roof in a village setting. A third robot assists by delivering clay roof tiles using a futuristic drone. The scene depicts an integration of AI-powered robotics in construction.
-
AI applications range from search and rescue to industry savings, sparking debate
Several news items highlight the diverse applications and implications of AI. One report details a drone using AI for missing person identification, while another mentions AI solutions driving savings in industry. Conce…
-
LLM agent optimizes YOLO26-MoE for insulator fault detection
Researchers have developed a new object detection model, YOLO26-MoE, to improve the automated inspection of electrical power line insulators using UAVs. This model integrates a Mixture-of-Experts (MoE) module to better …
-
New benchmark evaluates 3D reconstruction for drone imagery
Researchers have introduced UAVFF3D, a new benchmark designed to evaluate feed-forward 3D reconstruction models specifically for Unmanned Aerial Vehicle (UAV) imagery. This benchmark includes over 170,000 real UAV image…
-
LMPath pipeline uses language models for smarter UAV search missions
Researchers have developed LMPath, a new pipeline that uses language models to generate exploration priors for Unmanned Aerial Vehicle (UAV) search missions. This approach leverages semantic context from object prompts …
-
UAV video transformed into traffic analytics with new geometry pipeline
Researchers have developed a new pipeline to convert monocular UAV traffic video into a bird's-eye-view (BEV) representation. This method uses visible road geometry, such as lane markings, to estimate a homography that …
-
UAV-assisted scan-to-simulation framework enhances landslide modeling
Researchers have developed a new framework for simulating landslides that combines drone-based imagery with physics-informed Gaussian splatting. This approach captures photorealistic scenes and integrates them with the …
-
New benchmark tackles UAV detection with event cameras in motion
Researchers have introduced M$^2$E-UAV, a new benchmark and analysis framework designed to tackle the challenge of detecting small UAVs using onboard event cameras, particularly in complex motion-on-motion scenarios. Th…
-
FCC extends software update waiver for foreign drones and routers until 2029
The Federal Communications Commission (FCC) has extended a temporary waiver, allowing certain foreign-made drones and routers already authorized for use in the U.S. to continue receiving software and firmware updates un…
-
Ukraine deploys AI turret for autonomous drone interception
Ukraine has deployed an AI-powered turret designed to autonomously intercept enemy drones. The system can detect, track, and calculate the trajectory of UAVs, with the operator confirming the engagement. This technology…
-
Memory-augmented Kalman filtering mitigates UAV navigation errors
Researchers have developed a new framework called NeuroKalman to address error accumulation in continuous navigation for Unmanned Aerial Vehicles (UAVs). This method decouples navigation into prior prediction based on m…
-
LGTrack enhances UAV tracking efficiency and occlusion robustness
Researchers have developed LGTrack, a new framework for unmanned aerial vehicle (UAV) visual object tracking that aims to improve efficiency and robustness against occlusion. The system incorporates a novel Global-Group…
-
New model and benchmark track urban construction changes with drone imagery
Researchers have introduced PTNet, a novel framework designed to improve the semantic understanding of urban construction changes from remote sensing imagery. This approach explicitly models change semantics using a lea…
-
VL-UniTrack uses visual-language prompts for unified UAV-ground object tracking
Researchers have developed VL-UniTrack, a novel framework for simultaneous tracking of objects from both UAV and ground perspectives. This unified approach encodes features from both views in a single encoder, overcomin…
-
Federated RL framework optimizes UAV teams in hazardous environments
Researchers have introduced Experience-Constrained Hierarchical Federated Reinforcement Learning (EC-HFRL) to address challenges in training large-scale UAV teams in hazardous environments. This new framework posits tha…
-
Robotic vision uses hybrid video and stills to improve detail under bandwidth limits
Researchers have proposed a novel hybrid visual telemetry scheme for bandwidth-constrained robotic vision systems. This approach combines a continuous low-bitrate video stream with selectively transmitted high-detail st…
-
Dynamic-TD3 algorithm enhances UAV path planning with obstacle trajectory prediction
Researchers have developed Dynamic-TD3, a new algorithm designed to improve Unmanned Aerial Vehicle (UAV) path planning in environments with dynamic obstacles. This framework addresses the safety-exploration dilemma in …
-
New dataset and benchmark for event-based cameras in infrastructure defect detection
Researchers have introduced the ev-CIVIL dataset, the first of its kind for detecting defects in civil infrastructure using event-based cameras. These cameras, also known as dynamic vision sensors (DVS), offer advantage…
-
强化学习、安全函数增强无人机导航
研究人员开发了一种新颖的自主无人机导航方法，可提高速度和安全性。该方法结合了强化学习、基于潜在奖励塑造、控制Lyapunov函数和控制障碍函数。该系统在一个简化的环境中进行训练，然后无需额外训练即可应用于复杂场景，在模拟中显示出缩短的任务时间和稳健的性能。