The war between Ukraine and Russia is being fought increasingly via drone --and NATO and US military leadership is training troops for future conflicts that will pit man against machine. Subscribe to Slate Plus to access ad-free listening to the whole What Next family and all your favorite Slate podcasts. Subscribe today on Apple Podcasts by clicking "Try Free" at the top of our show page. Sign up now at slate.com/whatnextplus to get access wherever you listen.

Monocular Depth Estimation (MDE) enables the prediction of scene depths from a single RGB image, having been widely integrated into production-grade autonomous driving systems, e.g., Tesla Autopilot. Current adversarial attacks to MDE models focus on attaching an optimized adversarial patch to a designated obstacle. Although effective, this approach presents two inherent limitations: its reliance on specific obstacles and its limited malicious impact. In contrast, we propose a pioneering attack to MDE models that decouples obstacles from patches physically and deploys optimized patches on roads, thereby extending the attack scope to arbitrary traffic participants. This approach is inspired by our groundbreaking discovery: various MDE models with different architectures, trained for autonomous driving, heavily rely on road regions when predicting depths for different obstacles. Based on this discovery, we design the Adversarial Road Marking (AdvRM) attack, which camouflages patches as ordinary road markings and deploys them on roads, thereby posing a continuous threat within the environment. Experimental results from both dataset simulations and real-world scenarios demonstrate that AdvRM is effective, stealthy, and robust against various MDE models, achieving about 1.507 of Mean Relative Shift Ratio (MRSR) over 8 MDE models. The code is available at this Github Repo.

Typical LiDAR-based 3D object detection models are trained with real-world data collection, which is often imbalanced over classes. To deal with it, augmentation techniques are commonly used, such as copying ground truth LiDAR points and pasting them into scenes.

Collaborative trajectory prediction can comprehensively forecast the future motion of objects through multi-view complementary information. However, it encounters two main challenges in multi-drone collaboration settings. The expansive aerial observations make it difficult to generate precise Bird's Eye View (BEV) representations. Besides, excessive interactions can not meet real-time prediction requirements within the constrained drone-based communication bandwidth. To address these problems, we propose a novel framework named "Drones Help Drones" (DHD).

Top-down Bird's Eye View (BEV) maps are a popular perceptual representation for ground robot navigation due to their richness and flexibility for downstream tasks. While recent methods have shown promise for predicting BEV maps from First-Person View (FPV) images, their generalizability is limited to small regions captured by current autonomous vehicle-based datasets. In this context, we show that a more scalable approach towards generalizable map prediction can be enabled by using two large-scale crowd-sourced mapping platforms, Mapillary for FPV images and OpenStreetMap for BEV semantic maps. We introduce Map It Anywhere (MIA), a data engine that enables seamless curation and modeling of labeled map prediction data from existing open-source map platforms. Using our MIA data engine, we display the ease of automatically collecting a dataset of 1.2 million pairs of FPV images & BEV maps encompassing diverse geographies, landscapes, environmental factors, camera models & capture scenarios.

We now have a tentative launch date for Tesla's long-awaited robotaxi service in Austin, Texas: June 12. How long has Tesla been testing out these driverless vehicles that will soon be on the public streets of a major U.S. city? According to Tesla CEO Elon Musk, testing has been going on for "several days." "For the past several days, Tesla has been testing self-driving Model Y cars (no one in driver's seat) on Austin public streets with no incidents," Musk posted on his X account on Thursday. "A month ahead of schedule," Musk continued.

In this paper, we propose OpenSatMap, a fine-grained, high-resolution satellite dataset for large-scale map construction. Map construction is one of the foundations of the transportation industry, such as navigation and autonomous driving. Extracting road structures from satellite images is an efficient way to construct large-scale maps. However, existing satellite datasets provide only coarse semantic-level labels with a relatively low resolution (up to level 19), impeding the advancement of this field. In contrast, the proposed OpenSatMap (1) has fine-grained instance-level annotations; (2) consists of high-resolution images (level 20); (3) is currently the largest one of its kind; (4) collects data with high diversity. Moreover, OpenSatMap covers and aligns with the popular nuScenes dataset and Argoverse 2 dataset to potentially advance autonomous driving technologies. By publishing and maintaining the dataset, we provide a high-quality benchmark for satellite-based map construction and downstream tasks like autonomous driving.

Recent advances in 3D object detection leveraging multi-view cameras have demonstrated their practical and economical value in various challenging vision tasks. However, typical supervised learning approaches face challenges in achieving satisfactory adaptation toward unseen and unlabeled target datasets (i.e., direct transfer) due to the inevitable geometric misalignment between the source and target domains. In practice, we also encounter constraints on resources for training models and collecting annotations for the successful deployment of 3D object detectors. In this paper, we propose Unified Domain Generalization and Adaptation (UDGA), a practical solution to mitigate those drawbacks. We first propose Multi-view Overlap Depth Constraint that leverages the strong association between multi-view, significantly alleviating geometric gaps due to perspective view changes. Then, we present a Label-Efficient Domain Adaptation approach to handle unfamiliar targets with significantly fewer amounts of labels (i.e., 1% and 5%), while preserving well-defined source knowledge for training efficiency. Overall, UDGA framework enables stable detection performance in both source and target domains, effectively bridging inevitable domain gaps, while demanding fewer annotations. We demonstrate the robustness of UDGA with large-scale benchmarks: nuScenes, Lyft, and Waymo, where our framework outperforms the current state-of-the-art methods.

Register now free-of-charge to explore this white paper How Embodied Intelligence Enhances the Safety, Resilience, and Autonomy of UAV Systems As drones evolve into critical agents across defense, disaster response, and infrastructure inspection, they must become more adaptive, secure, and resilient. Traditional AI methods fall short in real-world unpredictability. This whitepaper from the Technology Innovation Institute (TII) explores how Embodied AI - AI that integrates perception, action, memory, and learning in dynamic environments, can revolutionize drone operations. Drawing from innovations in GenAI, Physical AI, and zero-trust frameworks, TII outlines a future where drones can perceive threats, adapt to change, and collaborate safely in real time. The result: smarter, safer, and more secure autonomous aerial systems. What Attendees will Learn: Why Embodied AI Outperforms Traditional AI The 4 Pillars of Drone Intelligence Swarm Resilience in Dynamic Environments Security Breakthroughs for Critical Missions Click on the cover to download the white paper PDF now.

Sen. Josh Hawley, R-Mo., commends President Donald Trump tearing into America's nation builders in the Middle East and weighs in on a Wisconsin judge being indicted for hiding an illegal immigrant from ICE on'The Ingraham Angle.' FIRST ON FOX – Sen. Josh Hawley, R-Mo., asked the Justice Department on Thursday to investigate a Chinese-owned self-driving trucking company, one of the largest in the U.S., citing allegations that it had shared proprietary data and other sensitive technology with state-linked entities in Beijing. The letter, sent to U.S. Attorney General Pam Bondi and previewed exclusively to Fox News Digital, asks the Justice Department to open a formal investigation into the autonomous truck company TuSimple Holdings, a Chinese-owned company and one of the largest self-driving truck companies in the U.S. In it, Hawley cites recent reporting from the Wall Street Journal that alleges that TuSimple "systematically shared proprietary data, source code, and autonomous driving technologies" with Chinese state-linked entities-- what he described as "blatant disregard" of the 2022 national security agreement with the Committee on Foreign Investment in the United States, or CFIUS. "These reports also revealed communications from TuSimple personnel inside China requesting the shipment of sensitive Nvidia AI chips and detailed records showing'deep and longstanding ties' with Chinese military-affiliated manufacturers," Hawley said. Sen. Josh Hawley, R-Mo., wants the Justice Department to investigate TuSimple Holdings, a Chinese-owned self-driving trucking company. He noted that to date, TuSimple "has not faced serious consequences" for sharing American intellectual property with China, despite having continued to share data with China after signing a national security agreement with the U.S. government in 2022, which was enforced by the Committee on Foreign Investment in the U.S. "If the reports about TuSimple are accurate, they represent not just a violation of export law, but a breach of national trust and a direct threat to American technological leadership," Hawley said.