Single-task learning in artificial neural networks will be able to learn the model very well, and the benefits brought by transferring knowledge thus become limited. In this regard, when the number of tasks increases (e.g., semantic segmentation, panoptic segmentation, monocular depth estimation, and 3D point cloud), duplicate information may exist across tasks, and the improvement becomes less significant. Multi-task learning has emerged as a solution to knowledge-transfer issues and is an approach to scene understanding which involves multiple related tasks each with potentially limited training data. Multi-task learning improves generalization by leveraging the domain-specific information contained in the training data of related tasks. In urban management applications such as infrastructure development, traffic monitoring, smart 3D cities, and change detection, automated multi-task data analysis for scene understanding based on the semantic, instance, and panoptic annotation, as well as monocular depth estimation, is required to generate precise urban models. In this study, a common framework for the performance assessment of multi-task learning methods from fixed-wing UAV images for 2D/3D city modeling is presented.

A U.S.-led coalition aircraft shot down an unmanned drone in Eastern Syria's Deir al-Zour province Saturday, months after the U.S. Central Command expressed concern over the use of weaponized drones to attack U.S. forces in the region. "Coalition aircraft successfully engaged and defeated a UAS through air to air engagement in the vicinity of Mission Support Site Green Village," Reuters quoted coalition spokesperson U.S. Army Colonel Wayne Marotto. While the coalition refused to reveal the type of aircraft used or other details citing security issues, a report by Aviation Week said a U.S. Air Force F-15E Strike Eagle used an AIM-9X Sidewinder missile to hit the UAS. The report added that Brig. Gen. Christopher Sage, commander of the 332nd Air Expeditionary Wing, was piloting the F-15E that fired the missile.

Forty-seven government entities and privacy companies, including Microsoft, exposed 38 million sensitive data records online by misconfiguring the Windows giant's Power Apps, a low-code service that promises an easy way to build professional applications. Security biz UpGuard said that in May one of its analysts found that the OData API for a Power Apps portal offered anonymously accessible database records that included personal details. That led the security shop to look at other Power Apps portals and its researchers found over one thousand apps configured to make data available to anyone who asked. Among the entities identified by UpGuard are: state and municipal government bodies in Indiana, Maryland, and New York City, and private enterprises like American Airlines, Ford, JB Hunt, and Microsoft. There's no indication so far that information has been misused.

This technique offers the advantages of generalization and a fast Unmanned aerial vehicles (UAVs) are being studied convergence speed [42]. In our study, learning was first and used in various fields [1-4]. In the case of a quadcopter performed for a simple path planning to ensure that a [5-8], which is one of the most common types UAV could fly quickly to a goal point in an environment of UAV, its position can be maintained through hovering, without obstacles. After this simple task was learned, the which is not possible with a fixed-wing UAV. Various learning was performed in an environment with obstacles sensors can be mounted on the UAV and the location of to train the UAV to fly to a goal point within a short the UAV can be determined using global navigation satellite time while avoiding obstacles. Such curriculum learning systems (GNSS) [9-13], long-term evolution (LTE) is more efficient than learning a difficult task from the based positioning [14-22], enhanced long-range navigation beginning.

WASHINGTON – About a month ago, Afghanistan's Ministry of Defense posted photographs on social media of seven brand-new helicopters arriving in Kabul, delivered by the United States. "They'll continue to see a steady drumbeat of that kind of support going forward," U.S. Defense Secretary Lloyd Austin told reporters a few days later at the Pentagon. In a matter of weeks, however, the Taliban had seized most of the country, as well as any weapons and equipment left behind by fleeing Afghan forces. Video showed the advancing insurgents inspecting long lines of vehicles and opening crates of new firearms, communications gear and even military drones. "Everything that hasn't been destroyed is the Taliban's now," said one U.S. official, speaking on the condition of anonymity.

Delivery drone fleets are growing, but don't expect them to replace humans anytime soon

The Transform Technology Summits start October 13th with Low-Code/No Code: Enabling Enterprise Agility. Qualcomm is unveiling its platform that enables aerial drones to tap both 5G and AI technologies. The Qualcomm Flight RB5 5G platform aims to accelerate development for commercial, enterprise, and industrial drones to help enterprises capture data from drone cameras and process that data at the edge of the network. The platform is powered by Qualcomm's QRB5165 processor, and it builds upon the company's latest internet of things (IoT) offerings. The idea is to enable a new generation of low-power 5G drones that can capture a lot of data with cameras and transmit that data via 5G to an operator or send it longer distances over a network.

Unmanned Aerial Vehicles (UAV) can pose a major risk for aviation safety, due to both negligent and malicious use. For this reason, the automated detection and tracking of UAV is a fundamental task in aerial security systems. Common technologies for UAV detection include visible-band and thermal infrared imaging, radio frequency and radar. Recent advances in deep neural networks (DNNs) for image-based object detection open the possibility to use visual information for this detection and tracking task. Furthermore, these detection architectures can be implemented as backbones for visual tracking systems, thereby enabling persistent tracking of UAV incursions. To date, no comprehensive performance benchmark exists that applies DNNs to visible-band imagery for UAV detection and tracking. To this end, three datasets with varied environmental conditions for UAV detection and tracking, comprising a total of 241 videos (331,486 images), are assessed using four detection architectures and three tracking frameworks. The best performing detector architecture obtains an mAP of 98.6% and the best performing tracking framework obtains a MOTA of 96.3%. Cross-modality evaluation is carried out between visible and infrared spectrums, achieving a maximal 82.8% mAP on visible images when training in the infrared modality. These results provide the first public multi-approach benchmark for state-of-the-art deep learning-based methods and give insight into which detection and tracking architectures are effective in the UAV domain.

Qualcomm is showing off the type of drones that could wind up being built on its dedicated Flight RB5 5G platform. The chip-maker has released a new reference design that contains all the latest connectivity and processing tech it has been talking up since last summer. That's the exoskeleton above, which is equipped with a Qualcomm Spectra 480 Image Signal Processor that can capture 200 megapixel photos and 8K video at 30 FPS. In addition, the drone can record in 4K at 120 FPS with support for HDR. At its core, the Flight RB5 5G platform uses the QRB5165 processor and Kryo 585 CPU and Adreno 650 GPU, based on the Snapdragon 865 CPU.

Recently, needs for unmanned aerial vehicles (UAVs) that are attachable to the wall have been highlighted. As one of the ways to address the need, researches on various tilting multirotors that can increase maneuverability has been employed. Unfortunately, existing studies on the tilting multirotors require considerable amounts of prior information on the complex dynamic model. Meanwhile, reinforcement learning on quadrotors has been studied to mitigate this issue. Yet, these are only been applied to standard quadrotors, whose systems are less complex than those of tilting multirotors. In this paper, a novel reinforcement learning-based method is proposed to control a tilting multirotor on real-world applications, which is the first attempt to apply reinforcement learning to a tilting multirotor. To do so, we propose a novel reward function for a neural network model that takes power efficiency into account. The model is initially trained over a simulated environment and then fine-tuned using real-world data in order to overcome the sim-to-real gap issue. Furthermore, a novel, efficient state representation with respect to the goal frame that helps the network learn optimal policy better is proposed. As verified on real-world experiments, our proposed method shows robust controllability by overcoming the complex dynamics of tilting multirotors.