DJI has filed a lawsuit against the US Department of Defense over its addition to the Pentagon list that designates it as a "Chinese military company." In its filing, shared by The Verge, the company said it's challenging the designation because it's "neither owned nor controlled by the Chinese military." It described itself as the "largest privately owned seller of consumer and commercial drones," mostly used by first responders, fire and police departments, businesses and hobbyists. The company claimed that because the Pentagon has officially proclaimed it as a national security threat, it has suffered "ongoing financial and reputational harm." It also said that it has lost business from both US and internal customers, which terminated contracts and refused to enter new ones, and it has been banned from signing contracts with multiple federal government agencies.

WIRED loved 2023's DJI Air 3 (9/10, WIRED Recommends). The midrange consumer drone was easy and safe to fly and compact enough to carry almost anywhere, but I found the most appealing feature to be its innovative dual-camera setup. CMOS sensor), it expanded my creative options for aerial photos and video. I could shoot wide vistas one moment, then switch to the telephoto lens to get closer to a particular feature of the landscape or compress it against the background for more dramatic framing. The new DJI Air 3S takes the concept a step further by increasing the sensor size of the wide-angle camera to a full inch, which improves dynamic range and low-light image quality.

The Israeli military, which released the video, said it shows the last moments of Mr. Sinwar, an architect of the Oct. 7 attacks, before he was killed in Rafah on Wednesday. While the room, the man's clothing and the arm injury broadly match other visual evidence of Mr. Sinwar's death, The Times could not independently verify his identity in the video. In modern warfare, militaries use drones to scope out enemy positions, and often release propaganda videos showing enemy soldiers being killed. On the battlefields in Ukraine, both sides in that conflict have released a steady stream of drone footage showing panicked soldiers moments before their deaths. But here the drone footage shows a solitary figure remarkably close-up. In a war often seen from far away, in large explosions or wide vistas of broken buildings, the moment is remarkably personal.

Existing multi-agent deep reinforcement learning (MADRL) methods for multi-UAV navigation face challenges in generalization, particularly when applied to unseen complex environments. To address these limitations, we propose a Dual-Transformer Encoder-based Proximal Policy Optimization (DTPPO) method. DTPPO enhances multi-UAV collaboration through a Spatial Transformer, which models inter-agent dynamics, and a Temporal Transformer, which captures temporal dependencies to improve generalization across diverse environments. This architecture allows UAVs to navigate new, unseen environments without retraining. Extensive simulations demonstrate that DTPPO outperforms current MADRL methods in terms of transferability, obstacle avoidance, and navigation efficiency across environments with varying obstacle densities. The results confirm DTPPO's effectiveness as a robust solution for multi-UAV navigation in both known and unseen scenarios.

When Hurricane Helene and Milton hit the Southeast US, they left a trail of devastation in their wake. Roads, homes, and chunks of towns were swept away by torrential floods. Thousands of residents were left without homes and forced to take refuge in community centers which were cut off from access to critical supplies and resources. One of those shelters, a senior center in Marion, North Carolina, has received a lifeline from an unlikely source. For a little over a week, a white, buzzing autonomous drone operated by Wing has been collecting prescription drugs, baby formula, and other critical resources from a nearby Walmart supercenter and airdropping them to the senior center.

Hamas on Friday is admitting to suffering "very painful and distressing" losses following the killing of its leader Yahya Sinwar as the Israeli military has released new drone video capturing the final moments of the terrorist's life. Footage taken of a wounded Sinwar shows him throwing a wooden board at a drone that was surveying damage inside of a building that the Israel Defense Forces targeted in Rafah, according to IDF Spokesperson Rear Adm. Daniel Hagari. Sinwar was later found dead with a gun and nearly 11,000 in his possession, he added. "Yes it's very painful and distressing to lose beloved people, especially extraordinary leaders like ours, but what we are sure of is that we are eventually victorious; this is the outcome for all people who fought for their liberty," senior Hamas official Basem Naim said Friday. "It seems that Israel believes that killing our leaders means the end of our movement and the struggle of the Palestinian people. They can believe what they want, and this is not the first time they said that," he continued.

Ukrainian troops are locked in a bitter battle for the town of Toretsk in the eastern region of Donetsk, which Russian troops entered last Friday. A spokesperson for Luhansk Technical University said Russian troops were demolishing the town as they advanced. "They erase the city with artillery. We have already seen it in other towns of Donbas. They are trying to find weak points in our defence with such small strikes," Anastasia Bobovnikova said.

The Israeli military released a video on Thursday showing a drone flying into a building in Rafah, where a man the video identifies as Yahya Sinwar is sitting on a chair. The Israeli military described the video as showing Mr. Sinwar "moments before his elimination." While the video has been edited, it is clear that the man, who is covered in dust, watches the drone for at least 20 seconds before throwing an object, possibly a stick, at the drone. The Times could not independently verify the identity of the man. The room seen in the drone video matches the location of earlier photographs obtained by The New York Times showing the corpse of a man closely resembling Mr. Sinwar.

Multi-agent systems are trained to maximize shared cost objectives, which typically reflect system-level efficiency. However, in the resource-constrained environments of mobility and transportation systems, efficiency may be achieved at the expense of fairness -- certain agents may incur significantly greater costs or lower rewards compared to others. Tasks could be distributed inequitably, leading to some agents receiving an unfair advantage while others incur disproportionately high costs. It is important to consider the tradeoffs between efficiency and fairness. We consider the problem of fair multi-agent navigation for a group of decentralized agents using multi-agent reinforcement learning (MARL). We consider the reciprocal of the coefficient of variation of the distances traveled by different agents as a measure of fairness and investigate whether agents can learn to be fair without significantly sacrificing efficiency (i.e., increasing the total distance traveled). We find that by training agents using min-max fair distance goal assignments along with a reward term that incentivizes fairness as they move towards their goals, the agents (1) learn a fair assignment of goals and (2) achieve almost perfect goal coverage in navigation scenarios using only local observations. For goal coverage scenarios, we find that, on average, our model yields a 14% improvement in efficiency and a 5% improvement in fairness over a baseline trained using random assignments. Furthermore, an average of 21% improvement in fairness can be achieved compared to a model trained on optimally efficient assignments; this increase in fairness comes at the expense of only a 7% decrease in efficiency. Finally, we extend our method to environments in which agents must complete coverage tasks in prescribed formations and show that it is possible to do so without tailoring the models to specific formation shapes.

This paper focuses on developing a bearings-only measurement-based three-dimensional window traversal guidance method for quadrotor Uninhabitated Aerial Vehicles (UAVs). The desired flight path and heading angles of the quadrotor are proposed as functions of the bearing angle information of the four vertices of the window. These angular guidance inputs employ a bearing angle bisector term and an elliptic shaping angle term, which directs the quadrotor towards the centroid of the window. Detailed stability analysis of the resulting kinematics demonstrates that all quadrotor trajectories lead to the centroid of the window along a direction which is normal to the window plane. A qualitative comparison with existing traversal methodologies showcases the superiority of the proposed guidance approach with regard to the nature of information, computations for generating the guidance commands, and flexibility of replanning the traversal path. Realistic simulations considering six degree-of-freedom quadrotor model and Monte Carlo studies validate the effectiveness, accuracy, and robustness of the proposed guidance solution. Representative flight validation trials are carried out using an indoor motion capture system.