Unmanned Aerial Vehicles (UAVs) are expected to transform logistics, reducing delivery time, costs, and emissions. This study addresses an on-demand delivery , in which fleets of UAVs are deployed to fulfil orders that arrive stochastically. Unlike previous work, it considers UAVs with heterogeneous, unknown energy storage capacities and assumes no knowledge of the energy consumption models. We propose a decentralised deployment strategy that combines auction-based task allocation with online learning. Each UAV independently decides whether to bid for orders based on its energy storage charge level, the parcel mass, and delivery distance. Over time, it refines its policy to bid only for orders within its capability. Simulations using realistic UAV energy models reveal that, counter-intuitively, assigning orders to the least confident bidders reduces delivery times and increases the number of successfully fulfilled orders. This strategy is shown to outperform threshold-based methods which require UAVs to exceed specific charge levels at deployment. We propose a variant of the strategy which uses learned policies for forecasting. This enables UAVs with insufficient charge levels to commit to fulfilling orders at specific future times, helping to prioritise early orders. Our work provides new insights into long-term deployment of UAV swarms, highlighting the advantages of decentralised energy-aware decision-making coupled with online learning in real-world dynamic environments.

--Mission planning for a fleet of cooperative autonomous drones in applications that involve serving distributed target points, such as disaster response, environmental monitoring, and surveillance, is challenging, especially under partial observability, limited communication range, and uncertain environments. Traditional path-planning algorithms struggle in these scenarios, particularly when prior information is not available. T o address these challenges, we propose a novel framework that integrates Graph Neural Networks (GNNs), Deep Reinforcement Learning (DRL), and transformer-based mechanisms for enhanced multi-agent coordination and collective task execution. Our approach leverages GNNs to model agent-agent and agent-goal interactions through adaptive graph construction, enabling efficient information aggregation and decision-making under constrained communication. This integration is carefully designed to address specific requirements of multi-agent navigation, such as scalability, adaptability, and efficient task execution. Experimental results demonstrate superior performance, with 90% service provisioning and 100% grid coverage (node discovery), while reducing the average steps per episode to 200, compared to 600 for benchmark methods such as particle swarm optimization (PSO), greedy algorithms and DQN.

They rise to 70ft (21 metres), tilt forward and zip away in different directions, each carrying a paper bag. On a sleepy morning in the Irish capital the takeoffs build to a steady one every few minutes, with barely anyone glancing at the constant stream of aircraft buzzing back and forth. "No one's looking up – no one ever looks up," says the man responsible, Bobby Healy, the founder of the Dublin startup Manna Aero. People probably should take notice, because the drones are part of an effort to realise an ambition shared by Amazon, the Google sister company Wing and the Californian startup Zipline: instant, autonomous home delivery. Healy and his big-tech rivals hope drone delivery will change the course of the retail industry across Ireland, and then into the UK as soon as this year.

Ukraine has opened a war crime investigation into the alleged killing of four soldiers captured by Russian forces, according to the Ukrainian parliament's human rights commissioner. Dmytro Lubinets wrote on X on Thursday that the four prisoners of war had no weapons as they walked out of a destroyed building with "their hands raised". "They were shot dead on the spot. This is a clear violation of the Geneva Convention and a grave war crime," he added. The alleged killing of the soldiers is believed to have occurred on March 13 in the southern Ukrainian village of Piatykhatky, according to The Associated Press news agency, which verified drone footage of the troops.

The Defense Ministry said Thursday that the Air Self-Defense Force scrambled fighter jets 704 times in response to possible airspace violations in fiscal 2024, up by 35 from the previous year. Of the total, scrambles against Chinese military aircraft accounted for 464, or 65.9%, down by 15. In August, Chinese military airplanes violated Japanese airspace off the Danjo Islands in Nagasaki Prefecture for the first time. The number of Chinese drones detected by the ministry more than tripled to 30, exceeding the 26 detected between fiscal 2013, when the first Chinese drone was spotted, and fiscal 2023. "China may have developed a system to (fully) operate drones, upgrading from trial flights," a ministry official said.

Reinforcement learning (RL) algorithms can enable high-maneuverability in unmanned aerial vehicles (MAVs), but transferring them from simulation to real-world use is challenging. Variable-pitch propeller (VPP) MAVs offer greater agility, yet their complex dynamics complicate the sim-to-real transfer. This paper introduces a novel RL framework to overcome these challenges, enabling VPP MAVs to perform advanced aerial maneuvers in real-world settings. Our approach includes real-to-sim transfer techniques-such as system identification, domain randomization, and curriculum learning to create robust training simulations and a sim-to-real transfer strategy combining a cascade control system with a fast-response low-level controller for reliable deployment. Results demonstrate the effectiveness of this framework in achieving zero-shot deployment, enabling MAVs to perform complex maneuvers such as flips and wall-backtracking.

Russia's territorial gains in Ukraine are slowing down dramatically, two analyses have found, continuing a pattern from 2024 at a time when both nations are trying to project strength in the face of United States-mediated negotiations aimed at ending the war. Britain's Ministry of Defence last week estimated that Russian forces seized 143sq km (55sq miles) of Ukrainian land in March, compared with 196sq km (76sq miles) in February and 326sq km (126sq miles) in January. The Institute for the Study of War, a Washington, DC-based think tank, spotted the same trend, estimating Russian gains at 203sq km (78sq miles) in March, 354sq km (137sq miles) in February and 427sq km (165sq miles) in January. These estimates are based on satellite imagery and geolocated open-source photography rather than claims by either side. Should this trend continue, Russian forces could come to a standstill by early summer, roughly coinciding with US President Donald Trump's self-imposed early deadline for achieving a ceasefire.

YellowScan's Navigator system is designed to map underwater topography in rivers, ponds and coastal areas. Below, its lidar system scans the landscape, mapping both the land and the shallow waters with pinpoint accuracy. This is precisely what YellowScan's new Navigator system is designed to do. Built specifically for mapping underwater topography in rivers, ponds and coastal areas, the Navigator is changing the game for environmental monitoring. With precision where traditional methods struggle, it's giving researchers and conservationists a whole new way to understand our planet's changing waterways.

Drones have become an important tool in a creator's bag of tricks, allowing them to capture aerial footage that elevates their videos. And nowadays, they've become more accessible as video quality and features have dramatically improved while prices have dropped. Recent budget-friendly models include DJI's Neo and Flip drones, along with the HoverAir X1 Pro lineup, all under 500. If you've got more to spend, the options are similarly plentiful with drones like the DJI Mini 4 Pro and HoverAir X1 Pro Max. And for the price of a good mirrorless camera, you can get DJI's Mavic 3 Pro that offers awesome image quality, range and other features.

In this article, we present a centralized approach for the control of multiple unmanned aerial vehicles (UAVs) for landing on moving unmanned ground vehicles (UGVs) using control barrier functions (CBFs). The proposed control framework employs two kinds of CBFs to impose safety constraints on the UAVs' motion. The first class of CBFs (LCBF) is a three-dimensional exponentially decaying function centered above the landing platform, designed to safely and precisely land UAVs on the UGVs. The second set is a spherical CBF (SCBF), defined between every pair of UAVs, which avoids collisions between them. The LCBF is time-varying and adapts to the motions of the UGVs. In the proposed CBF approach, the control input from the UAV's nominal tracking controller designed to reach the landing platform is filtered to choose a minimally-deviating control input that ensures safety (as defined by the CBFs). As the control inputs of every UAV are shared in establishing multiple CBF constraints, we prove that the control inputs are shared without conflict in rendering the safe sets forward invariant. The performance of the control framework is validated through a simulated scenario involving three UAVs landing on three moving targets.