Understanding the geometric and semantic properties of the scene is crucial in autonomous navigation and particularly challenging in the case of Unmanned Aerial Vehicle (UAV) navigation. Such information may be by obtained by estimating depth and semantic segmentation maps of the surrounding environment and for their practical use in autonomous navigation, the procedure must be performed as close to real-time as possible. In this paper, we leverage monocular cameras on aerial robots to predict depth and semantic maps in low-altitude unstructured environments. We propose a joint deep-learning architecture that can perform the two tasks accurately and rapidly, and validate its effectiveness on MidAir and Aeroscapes benchmark datasets. Our joint-architecture proves to be competitive or superior to the other single and joint architecture methods while performing its task fast predicting 20.2 FPS on a single NVIDIA quadro p5000 GPU and it has a low memory footprint. All codes for training and prediction can be found on this link: https://github.com/Malga-Vision/Co-SemDepth

With the United States set to meet delegations from Russia Ukraine separately in Saudi Arabia on Monday in an ongoing bid to halt the three-year war, Russia has launched a drone attack on Friday night on the Ukrainian city of Zaporizhzhia, killing three people and wounding 12, Ukrainian officials said. The city was hit by 12 drones, police said. Regional Governor Ivan Fedorov said residential buildings, cars and communal buildings were set on fire. Photos from the scene showed emergency services scouring the rubble for survivors. Ukraine and Russia agreed this week in principle to a limited ceasefire after US President Donald Trump held separate calls on consecutive days with the countries' leaders, but what actual targets would be off limits to attack remains contentious.

There's now an entire subcategory of small, pocketable camera drones built just for capturing selfies, and ZeroZero Robotics' HoverAir X1 ProMax might be the most assured, impressive model in it. It's effortlessly portable, making even the likes of the DJI Mini 4 Pro seem bulky. Weighing just 6.79 ounces with a clever folding design and all-over cage to keep the propellers out of harm's way, the X1 ProMax can be tossed into a coat pocket or backpack without fear of it getting damaged. It's the kind of thing you can casually bring along on any day trip or vacation alongside your sunglasses and water bottle, just on the off chance it comes in useful. Pull it out, unfold it, hit the power button, select your preferred selfie video style with the left and right buttons, place the drone on your outstretched palm facing you, tap the power button again, and it'll take off, capture your selected shot, and return to land on your palm, ready to be put away again.

As far as we know, Bigfoot isn't in therapy, so it's unlikely he's going to find himself. That means it's time to do this right. No more grainy video or photos that turn out to just be some random guy in the woods. No, we're doing this with drones, specifically this one because it has a good camera and it's pretty cheap. The Ninja Dragon Phantom is an HD dual-camera drone, and instead of paying 199 for one, it's only 79.97 right now.

Additive manufacturing (AM) has transformed the production landscape by enabling the precision creation of complex geometries. However, AM faces limitations when applied to challenging environments, such as elevated surfaces and remote locations. Aerial additive manufacturing, facilitated by drones, presents a solution to these challenges. However, despite advances in methods for the planning, control, and localization of drones, the accuracy of these methods is insufficient to run traditional feedforward extrusion-based additive manufacturing processes (such as Fused Deposition Manufacturing). Recently, the emergence of LLMs has revolutionized various fields by introducing advanced semantic reasoning and real-time planning capabilities. This paper proposes the integration of LLMs with aerial additive manufacturing to assist with the planning and execution of construction tasks, granting greater flexibility and enabling a feed-back based design and construction system. Using the semantic understanding and adaptability of LLMs, we can overcome the limitations of drone based systems by dynamically generating and adapting building plans on site, ensuring efficient and accurate construction even in constrained environments. Our system is able to design and build structures given only a semantic prompt and has shown success in understanding the spatial environment despite tight planning constraints. Our method's feedback system enables replanning using the LLM if the manufacturing process encounters unforeseen errors, without requiring complicated heuristics or evaluation functions. Combining the semantic planning with automatic error correction, our system achieved a 90% build accuracy, converting simple text prompts to build structures.

Detecting and tracking multiple unmanned aerial vehicles (UAVs) in thermal infrared video is inherently challenging due to low contrast, environmental noise, and small target sizes. This paper provides a straightforward approach to address multi-UAV tracking in thermal infrared video, leveraging recent advances in detection and tracking. Instead of relying on the YOLOv5 with the DeepSORT pipeline, we present a tracking framework built on YOLOv12 and BoT-SORT, enhanced with tailored training and inference strategies. We evaluate our approach following the metrics from the 4th Anti-UAV Challenge and demonstrate competitive performance. Notably, we achieve strong results without using contrast enhancement or temporal information fusion to enrich UAV features, highlighting our approach as a "Strong Baseline" for the multi-UAV tracking task. We provide implementation details, in-depth experimental analysis, and a discussion of potential improvements. The code is available at https://github.com/wish44165/YOLOv12-BoT-SORT-ReID .

Russian bombardments in eastern Ukraine ramped up overnight, killing two people, as Ukraine hit Russia's Engels military airfield in the country's southwest region of Saratov with drones. Both Russia and Ukraine stepped up aerial attacks in the early hours of Thursday as United States President Donald Trump pushes both sides to agree to a ceasefire after more than three years of fighting. Ukrainian officials in the northeastern Sumy and Kharkiv regions said two people were killed and several others injured after Russia dropped more than three dozen glide bombs on the towns in the border regions. Russian drone attacks on the town of Kropyvnytskyi, hundreds of kilometres from the front line, wounded 14 people and damaged rail infrastructure. "Kropyvnytskyi underwent the most massive enemy attack. Peaceful residential buildings were destroyed," regional governor Andriy Raikovych said.

At Fort Sill, service members from across the military are undergoing counter-drone training at the Joint C-sUAS (Counter small Unmanned Aircraft System) University (JCU), also known as "drone university." The program has become a critical part of the Military's efforts to combat the rapidly growing use of unmanned aerial systems (UAS) by adversaries. "It's the Army's premier Counter-Small UAS training institution," said Col. Moseph Sauda, the program's director. "Our mission is to prepare and train the joint force to counter the threat, to be able to understand that threat, how they operate, and how they attack us… We can then develop not only tactics, techniques, and procedures, but also the employment methodology that maximizes the capabilities of our existing systems." A 3D-printed drone flies above from Oklahoma's Fort Sill at the U.S. Army's Joint C-sUAS University.

Regional authorities in northeast Ukraine's Sumy region said Russian drone attacks damaged two hospitals there, while a 29-year-old man was killed and three others were injured in a separate attack on a residential building. Kyiv region's Governor Mykola Kalashnyk said Kremlin drones damaged several houses near the Ukrainian capital, resulting in a 60-year-old man being injured. Ukraine's state railway network Ukrzaliznytsia said Moscow's forces attacked its power system twice in the city of Dnipro, with the first strike hitting just hours after Russian President Vladimir Putin committed to a 30-day pause on attacks on Ukraine's energy infrastructure. The second attack injured four people. "Russia is attacking civilian infrastructure and people – right now," said Andriy Yermak, Ukrainian President Volodymyr Zelenskyy's chief of staff.

We reappraise the idea of colliding with robots, moving from a position that tries to avoid or mitigate collisions to one that considers them an important facet of human interaction. We report on a soma design workshop that explored how our bodies could collide with telepresence robots, mobility aids, and a quadruped robot. Based on our findings, we employed soma trajectories to analyse collisions as extended experiences that negotiate key transitions of consent, preparation, launch, contact, ripple, sting, untangle, debris and reflect. We then employed these ideas to analyse two collision experiences, an accidental collision between a person and a drone, and the deliberate design of a robot to play with cats, revealing how real-world collisions involve the complex and ongoing entanglement of soma trajectories. We discuss how viewing collisions as entangled trajectories, or tangles, can be used analytically, as a design approach, and as a lens to broach ethical complexity.