An aid ship heading to Gaza has sent out a distress signal after crew members say it was hit in a drone attack and has caught fire. There are 30 aid workers on board the Freedom Flotilla ship, which was attempting to break Israel's 2-month aid blockade.

Vance made the comments in a wide-ranging interview, in which he defended Trump's approach to the war in Ukraine. "Yes, of course, [the Ukrainians] are angry that they were invaded," Vance added. "But are we going to continue to lose thousands and thousands of soldiers over a few miles of territory this or that way?" Trump this week suggested that Ukraine might be willing to cede Crimea - which Russia invaded in 2014 - in order to reach a truce settlement. But Ukraine's President Volodymyr Zelensky had earlier implied that he would be unable to accept Russian control of the peninsula, citing the Ukrainian constitution. In a separate interview with Fox News on Thursday, US Secretary of State Marco Rubio said there needed to be a "breakthrough" in the conflict soon, otherwise Trump "will have to decide how much time to dedicate to this". Russian president Vladimir Putin this week announced a temporary three-day ceasefire from 8 May, to coincide with anniversary celebrations marking the end of World War Two.

Russia accused Ukraine of deliberately targeting civilians during a recent drone attack that killed at least seven people and wounded more than 20 on Thursday morning in partially occupied Kherson. The drone strike hit a market in the town of Oleshky in Russian-controlled Kherson at approximately 9:30am local time, when many people were outdoors due to the May 1 public holiday, the region's Moscow-appointed governor said. Ukraine's military said the attack targeted Russian troops, and only military personnel were killed, although the claims by either side have not been independently verified. A Russian strike on Ukraine's Odesa killed two people, and a Russian drone attack in the southeastern Ukrainian city of Zaporizhzhia set a building on fire on Thursday night, injuring 14 people, with no fatalities. Ukraine's SBU Security Service said it has thwarted the attempted murder of Sergiy Sternenko, a prominent activist and video blogger, and also detained a suspect.

As machine learning is increasingly deployed in autonomous systems, verification of neural network controllers is becoming an active research domain. Existing tools and annual verification competitions suggest that soon this technology will become effective for real-world applications. Our application comes from the emerging field of microflyers that are passively transported by the wind, which may have various uses in weather or pollution monitoring. Specifically, we investigate centimetre-scale bio-inspired gliding drones that resemble Alsomitra macrocarpa diaspores. In this paper, we propose a new case study on verifying Alsomitra-inspired drones with neural network controllers, with the aim of adhering closely to a target trajectory. We show that our system differs substantially from existing VNN and ARCH competition benchmarks, and show that a combination of tools holds promise for verifying such systems in the future, if certain shortcomings can be overcome. We propose a novel method for robust training of regression networks, and investigate formalisations of this case study in Vehicle and CORA. Our verification results suggest that the investigated training methods do improve performance and robustness of neural network controllers in this application, but are limited in scope and usefulness. This is due to systematic limitations of both Vehicle and CORA, and the complexity of our system reducing the scale of reachability, which we investigate in detail. If these limitations can be overcome, it will enable engineers to develop safe and robust technologies that improve people's lives and reduce our impact on the environment.

Ukraine's parliament is expected to ratify a controversial minerals deal with the United States in a decisive step towards securing the latter's long-term commitment to the war-battered country amid stalled efforts to strike a Ukraine-Russia ceasefire. The deal, signed by Kyiv and Washington on Wednesday, pushed by US President Donald Trump and after protracted negotiations, marks an inflection point of sorts in the war, granting the US priority access to Ukraine's critical minerals as a means of deterring future Russian aggression. However, it stops short of offering specific security guarantees and questions remain over accessing minerals in areas under Russian control. Ukraine's Minister of Foreign Affairs Andrii Sybiha said on Thursday that the deal "marks an important milestone in Ukraine–US strategic partnership aimed at strengthening Ukraine's economy and security". "We're expecting it to be discussed and ratified by Ukraine's parliament later today," said Al Jazeera's Zein Basravi, reporting from Kyiv.

Following several months of numerous high-profile aviation accidents, new data suggest pilots are facing a specific threat when it comes to keeping airline passengers safe in the skies. Last year, drones accounted for approximately two-thirds of reported near-midair collisions with commercial aircraft taking off or landing within the country's 30 busiest airports, according to the Associated Press. The findings come as aviation safety data indicate drones accounted for the highest number of near-misses since 2020, with the first reports dating back to 2014. "The rise in recreational and commercial drone use has simply outpaced education and enforcement," aviation attorney Jason Matzus told Fox News Digital. "More people are flying drones without fully understanding the rules or the risks."

The enemy is attacking Odesa with strike drones. There is considerable damage to civilian infrastructure, particularly dwellings,

Russian drones attacked Ukraine's Black Sea port of Odesa early on Thursday, killing at least two people and injuring five, the regional governor said. The attack sparked fires and damaged residential dwellings and infrastructure. In Kharkiv, Ukraine's second-largest city in the northeast, the mayor said another Russian drone had struck a petrol station in the city centre, triggering a fire. Ukraine's SBU security agency claimed responsibility for a drone strike on a defence manufacturing facility in Russia. The strike on Murom Instrument-Building Plant, 300km (186 miles) east of Moscow, sparked a fire and damaged two buildings, the region's governor reported.

Ego-motion estimation is vital for drones when flying in GPS-denied environments. Vision-based methods struggle when flight speed increases and close-by objects lead to difficult visual conditions with considerable motion blur and large occlusions. To tackle this, vision is typically complemented by state estimation filters that combine a drone model with inertial measurements. However, these drone models are currently learned in a supervised manner with ground-truth data from external motion capture systems, limiting scalability to different environments and drones. In this work, we propose a self-supervised learning scheme to train a neural-network-based drone model using only onboard monocular video and flight controller data (IMU and motor feedback). We achieve this by first training a self-supervised relative pose estimation model, which then serves as a teacher for the drone model. To allow this to work at high speed close to obstacles, we propose an improved occlusion handling method for training self-supervised pose estimation models. Due to this method, the root mean squared error of resulting odometry estimates is reduced by an average of 15%. Moreover, the student neural drone model can be successfully obtained from the onboard data. It even becomes more accurate at higher speeds compared to its teacher, the self-supervised vision-based model. We demonstrate the value of the neural drone model by integrating it into a traditional filter-based VIO system (ROVIO), resulting in superior odometry accuracy on aggressive 3D racing trajectories near obstacles. Self-supervised learning of ego-motion estimation represents a significant step toward bridging the gap between flying in controlled, expensive lab environments and real-world drone applications. The fusion of vision and drone models will enable higher-speed flight and improve state estimation, on any drone in any environment.

In high-speed quadcopter racing, finding a single controller that works well across different platforms remains challenging. This work presents the first neural network controller for drone racing that generalizes across physically distinct quadcopters. We demonstrate that a single network, trained with domain randomization, can robustly control various types of quadcopters. The network relies solely on the current state to directly compute motor commands. The effectiveness of this generalized controller is validated through real-world tests on two substantially different crafts (3-inch and 5-inch race quadcopters). We further compare the performance of this generalized controller with controllers specifically trained for the 3-inch and 5-inch drone, using their identified model parameters with varying levels of domain randomization (0%, 10%, 20%, 30%). While the generalized controller shows slightly slower speeds compared to the fine-tuned models, it excels in adaptability across different platforms. Our results show that no randomization fails sim-to-real transfer while increasing randomization improves robustness but reduces speed. Despite this trade-off, our findings highlight the potential of domain randomization for generalizing controllers, paving the way for universal AI controllers that can adapt to any platform.