A new acquisition in the drone services sector may be instrumental in moving the commercial and recreational drone industries significantly toward broader adoption. DroneUp, a delivery drone provider, is acquiring AirMap, which makes the most comprehensive airspace management software solution in the world. The news is important as drone operations will soon hit tens of thousands of flights per day, which is beyond human scale air traffic control operations. As airspace becomes more congested, autonomous drones need to be navigated separately, safely, and at broad scale to deconflict with one another and with manned aviation. DroneUp, which focuses on drone delivery services, is particularly interested in offering streamlined air traffic control for its delivery partners and to aid regulators in adopting drone delivery infrastructure.
Engineers at Stanford University have built a robotic bird to understand how birds are able to fly and perch on branches. The robot has a pair of snatching talons that attach to a circular flat base; that is then attached to a quadcopter drone to fly around. To account for the size of the drone that allows it to fly, the avian robot is based on the legs of a peregrine falcon. In place of bones, the machine has a 3D-printed structure with motors and fishing line for muscles and tendons. Each leg has its own motor to move back and forth, another for grasping, and a mechanism to absorb impact energy when it lands.
"By understanding how people enjoy the Space Needle's observation decks, food and beverage experiences, and amenities, we can better provide both a safe and enjoyable experience," said Luis Quintero, senior operations manager at the Space Needle. "Through Veovo's crowd management solution, we can reduce and prevent overcrowding, while understanding trends over time will allow us to optimise our operations and resourcing." London Gatwick Airport will use Passenger Predictability solution to optimise security operations and improve passenger flow. The partnership will allow the airport to efficiently handle increasing passenger numbers and build back better for a more sustainable, passenger-centred travel experience. The AI-powered technology gives Gatwick real-time awareness of people's movement and experiences in the North and South terminal security areas.
Prof Marc Stettler, transport and environment lecturer at Imperial College London, says changing the altitude of fewer than 2% of flights could potentially reduce contrail-linked climate change by a staggering 59%. "Tweaking the flight elevation by just a thousand feet can stop some contrails from forming," he explains.
The Cincinnati/Northern Kentucky International Airport (CVG) is expanding its long-standing relationship with TaskWatch to help automate manual processes and gain insight into its complex operations. Adding TaskWatch's computer vision platform that works in conjunction with AWS Panorama, the Airport sees, adds, and explores distinct use cases for advanced artificial intelligence (AI). TaskWatch is leading the way for continuous improvement in capacity management and productivity, which are vital contributors to the overall customer experience. Brian Cobb, CIO at CVG, explains: "CVG Airport is committed to providing a world-class traveler experience through continuous innovation and strategic partnerships. Using TaskWatch's application on AWS Panorama, we can bring computer vision to our existing IP cameras to automatically monitor congestion for over 70,000 square feet of airport traffic lanes.
Traffic state prediction is necessary for many Intelligent Transportation Systems applications. Recent developments of the topic have focused on network-wide, multi-step prediction, where state of the art performance is achieved via deep learning models, in particular, graph neural network-based models. While the prediction accuracy of deep learning models is high, these models' robustness has raised many safety concerns, given that imperceptible perturbations added to input can substantially degrade the model performance. In this work, we propose an adversarial attack framework by treating the prediction model as a black-box, i.e., assuming no knowledge of the model architecture, training data, and (hyper)parameters. However, we assume that the adversary can oracle the prediction model with any input and obtain corresponding output. Next, the adversary can train a substitute model using input-output pairs and generate adversarial signals based on the substitute model. To test the attack effectiveness, two state of the art, graph neural network-based models (GCGRNN and DCRNN) are examined. As a result, the adversary can degrade the target model's prediction accuracy up to $54\%$. In comparison, two conventional statistical models (linear regression and historical average) are also examined. While these two models do not produce high prediction accuracy, they are either influenced negligibly (less than $3\%$) or are immune to the adversary's attack.
To efficiently navigate their surrounding environments and complete missions, unmanned aerial systems (UASs) should be able to detect multiple objects in their surroundings and track their movements over time. So far, however, enabling multi-object tracking in unmanned aerial vehicles has proved to be fairly challenging. Researchers at Lockheed Martin AI Center have recently developed a new deep learning technique that could allow UASs to track multiple objects in their surroundings. Their technique, presented in a paper pre-published on arXiv, could aid the development of better performing and more responsive autonomous flying systems. "We present a robust object tracking architecture aimed to accommodate for the noise in real-time situations," the researchers wrote in their paper.
A two-legged robot inspired by birds can walk, skateboard, fly and balance on a slackline, which is like a loose tightrope. It could potentially become a new tool to monitor infrastructure in hard-to-reach environments. The robot, named LEONARDO by its creators at the California Institute of Technology (Caltech) and Northeastern University in Boston, is a human-like machine with knee, hip and ankle joints, but rotor blades for arms that give it upward thrust. LEONARDO is 75 centimetres tall, weighs 2.6 kilograms and walks at up to 20cm per second. It is "the first robot to achieve seamless integration of walking …
Leica Geosystems has two new autonomous hardware products that are pushing robots to bold new places. Announced recently is a flying UAV laser scanning sensor and a reality capture product for robots, including SPOT from Boston Dynamics. The flying sensor is called BLK2FLY, and Leica Geosystems, which is a brand of global information systems company Hexagon AB, calls it the world's first autonomous UAV laser scanning sensor. The user sets up a flight path, taps a tablet, and it flies off to accurately scan and capture the dimensions of an area or building. It's best for inaccessible or hard to reach areas, such as facades or rooftops, or to document site conditions after a disaster.
The Automatic Dependent Surveillance Broadcast protocol is one of the latest compulsory advances in air surveillance. While it supports the tracking of the ever-growing number of aircraft in the air, it also introduces cybersecurity issues that must be mitigated e.g., false data injection attacks where an attacker emits fake surveillance information. The recent data sources and tools available to obtain flight tracking records allow the researchers to create datasets and develop Machine Learning models capable of detecting such anomalies in En-Route trajectories. In this context, we propose a novel multivariate anomaly detection model called Discriminatory Auto-Encoder (DAE). It uses the baseline of a regular LSTM-based auto-encoder but with several decoders, each getting data of a specific flight phase (e.g. climbing, cruising or descending) during its training.To illustrate the DAE's efficiency, an evaluation dataset was created using real-life anomalies as well as realistically crafted ones, with which the DAE as well as three anomaly detection models from the literature were evaluated. Results show that the DAE achieves better results in both accuracy and speed of detection. The dataset, the models implementations and the evaluation results are available in an online repository, thereby enabling replicability and facilitating future experiments.