Integration of reinforcement learning with unmanned aerial vehicles (UAVs) to achieve autonomous flight has been an active research area in recent years. An important part focuses on obstacle detection and avoidance for UAVs navigating through an environment. Exploration in an unseen environment can be tackled with Deep Q-Network (DQN). However, value exploration with uniform sampling of actions may lead to redundant states, where often the environments inherently bear sparse rewards. To resolve this, we present two techniques for improving exploration for UAV obstacle avoidance. The first is a convergence-based approach that uses convergence error to iterate through unexplored actions and temporal threshold to balance exploration and exploitation. The second is a guidance-based approach using a Domain Network which uses a Gaussian mixture distribution to compare previously seen states to a predicted next state in order to select the next action. Performance and evaluation of these approaches were implemented in multiple 3-D simulation environments, with variation in complexity. The proposed approach demonstrates a two-fold improvement in average rewards compared to state of the art.

DEME Offshore and Sabca have carried out a series of tests at the Rentel offshore wind farm with an aim to automate critical and ad hoc operations in the near future by using autonomous aerial vehicles (AAVs) and artificial intelligence (AI). The companies, which teamed up two years ago, have performed the first commercial, cross-border, "beyond visual line of sight" (BVLOS) drone operations at the wind farm 35 kilometres off the Belgian coast, where tests in Search & Rescue operations, environmental surveys, turbine and substation inspections, as well as parcel deliveries took place. During the tests, both a multicopter drone and a fixed-wing surveillance drone with a wing span of more than 3 metres were deployed in parallel. The long endurance surveillance drone took off from the Belgian coast and flew to the Rentel offshore wind farm. Meanwhile, an automated resident drone performed inspections and cargo flights from the substation and vessels.

In the last online technical talk, Adam Bry and Hayk Martiros from Skydio explained how their company tackles real-world issues when it comes to drone flying. Skydio is the leading US drone company and the world leader in autonomous flight. Our drones are used for everything from capturing amazing video, to inspecting bridges, to tracking progress on construction sites. At the core of our products is a vision-based autonomy system with seven years of development at Skydio, drawing on decades of academic research. This system pushes the state of the art in deep learning, geometric computer vision, motion planning, and control with a particular focus on real-world robustness.

SHENZHEN – Chinese drone giant DJI Technology Co. built up such a successful U.S. business over the past decade that it almost drove all competitors out of the market. Yet its North American operations have been hit by internal disturbances in recent weeks and months, with a raft of staff cuts and departures, according to interviews with more than two dozen current and former employees. The loss of key managers, including some who have joined rivals, has compounded problems caused by U.S. government restrictions on Chinese companies, and raised the once-remote prospect of DJI's dominance being eroded, said four of the people, including two senior executives who were at the company until late 2020. About a third of DJI's 200-strong team in the region was laid off or resigned last year, from offices in Palo Alto, Burbank and New York, according to three former and one current employee. In February this year, DJI's head of U.S. R&D left and the company laid off the remaining R&D staff, numbering roughly 10 people, at its flagship U.S. research center in California's Palo Alto, four people said.

DUBAI, United Arab Emirates (AP) -- A pair of B-52 bombers flew over the Mideast on Sunday, the latest such mission in the region aimed at warning Iran amid tensions between Washington and Tehran. The flight by the two heavy bombers came as a pro-Iran satellite channel based in Beirut broadcast Iranian military drone footage of an Israeli ship hit by a mysterious explosion only days earlier in the Mideast. While the channel sought to say Iran wasn't involved, Israel has blamed Tehran for what it described as an attack on the vessel. The U.S. military's Central Command said the two B-52s flew over the region accompanied by military aircraft from nations including Israel, Saudi Arabia and Qatar. It marked the fourth-such bomber deployment into the Mideast this year and the second under President Joe Biden.

For anyone who has ever misplaced their iPhone, Apple's "Find My" app is a game-changer that borders on pure magic. Sign into the app, tap a button to sound an alarm on your MIA device, and, within seconds, it'll emit a loud noise -- even if your phone is set on silent mode -- that allows you to go find the missing handset. Yeah, it's usually stuck behind your sofa cushions or left facedown on a shelf somewhere. You can think of SArdo, a new drone project created by researchers at Germany's NEC Laboratories Europe GmbH, as Apple's "Find My" app on steroids. The difference is that, while finding your iPhone is usually just a matter of convenience, the technology developed by NEC investigators could be a literal lifesaver.

Dutch cress grower Rob Baan has enlisted high-tech helpers to tackle a pest in his greenhouses: palm-sized drones seek and destroy moths that produce caterpillars that can chew up his crops. "I have unique products where you don't get certification to spray chemicals and I don't want it," Baan said in an interview in a greenhouse bathed in the pink glow of LED lights that help his seedlings grow. His company, Koppert Cress, exports aromatic seedlings, plants and flowers to top-end restaurants around the world. A keen adopter of innovative technology in his greenhouses, Baan turned to PATS Indoor Drone Solutions, a startup that is developing autonomous drone systems as greenhouse sentinels, to add another layer of protection for his plants. The drones themselves are basic, but they are steered by smart technology aided by special cameras that scan the airspace in greenhouses.

A research group has developed an autonomous robotic team of devices that can be used at hazardous or difficult-to-reach sites to make surveys and collect data--providing more and faster insights than human beings are able to deliver. These robot teams--composed of autonomous devices that gather data on the ground, in the air, and in water--would be ideally suited for hazardous environmental situations and/or for holistic environmental surveys of ecosystems. An autonomous team like this could do a survey and rapidly sample what's in the air and the water so that people could be kept out of harm's way. In another context, the robots could provide a general survey of ecosystems, or they could look at situations such as harmful algal blooms in lakes. A recent demonstration in the field showed how the autonomous robotic team can rapidly learn the characteristics of environments it has never seen before. Researchers hope the robot team prototype can be a model for changing the methods that are used to survey disaster sites, waterways, and extreme environments.

In precision crop protection, (target-orientated) object detection in image processing can help navigate Unmanned Aerial Vehicles (UAV, crop protection drones) to the right place to apply the pesticide. Unnecessary application of non-target areas could be avoided. Deep learning algorithms dominantly use in modern computer vision tasks which require high computing time, memory footprint, and power consumption. Based on the Edge Artificial Intelligence, we investigate the main three paths that lead to dealing with this problem, including hardware accelerators, efficient algorithms, and model compression. Finally, we integrate them and propose a solution based on a light deep neural network (DNN), called Ag-YOLO, which can make the crop protection UAV have the ability to target detection and autonomous operation. This solution is restricted in size, cost, flexible, fast, and energy-effective. The hardware is only 18 grams in weight and 1.5 watts in energy consumption, and the developed DNN model needs only 838 kilobytes of disc space. We tested the developed hardware and software in comparison to the tiny version of the state-of-art YOLOv3 framework, known as YOLOv3-Tiny to detect individual palm in a plantation. An average F1 score of 0.9205 at the speed of 36.5 frames per second (in comparison to similar accuracy at 18 frames per second and 8.66 megabytes of the YOLOv3-Tiny algorithm) was reached. This developed detection system is easily plugged into any machines already purchased as long as the machines have USB ports and run Linux Operating System.

Three years ago, Customs and Border Protection placed an order for self-flying aircraft that could launch on their own, rendezvous, locate and monitor multiple targets on the ground without any human intervention. In its reasoning for the order, CBP said the level of monitoring required to secure America's long land borders from the sky was too cumbersome for people alone. To research and build the drones, CBP handed $500,000 to Mitre Corp., a trusted nonprofit Skunk Works that was already furnishing border police with prototype rapid DNA testing and smartwatch hacking technology. They were "tested but not fielded operationally" as "the gap from simulation to reality turned out to be much larger than the research team originally envisioned," a CBP spokesperson says. This year, America's border police will test automated drones from Skydio, the Redwood City, Calif.-based startup that on Monday announced it had raised an additional $170 million in venture funding at a valuation of $1 billion. That brings the total raised for Skydio to $340 million.