Developed by UZH researchers, the algorithm DroNet allows drones to fly completely by themselves through the streets of a city and in indoor environments. Therefore, the algorithm had to learn traffic rules and adapt training examples from cyclists and car drivers.
All today’s commercial drones use GPS, which works fine above building roofs and in high altitudes. But what, when the drones have to navigate autonomously at low altitude among tall buildings or in the dense, unstructured city streets with cars, cyclists or pedestrians suddenly crossing their way? Until now, commercial drones are not able to quickly react to such unforeseen events.
Integrate autonomously navigating drones
Researchers of the University of Zurich and the National Centre of Competence in Research NCCR Robotics developed DroNet, an algorithm that can safely drive a drone through the streets of a city. Designed as a fast 8-layers residual network, it produces two outputs for each single input image: a steering angle to keep the drone navigating while avoiding obstacles, and a collision probability to let the drone recognise dangerous situations and promptly react to them. “DroNet recognises static and dynamic obstacles and can slow down to avoid crashing into them. With this algorithm we have taken a step forward towards integrating autonomously navigating drones into our everyday life”, says Davide Scaramuzza, Professor for Robotics and Perception at the University of Zurich.
Powerful artificial intelligence algorithm
Instead of relying on sophisticated sensors, the drone developed by Swiss researchers uses a normal camera like that of every smartphone, and a very powerful artificial intelligence algorithm to interpret the scene it observes and react accordingly. The algorithm consists of a so-called Deep Neural Network. “This is a computer algorithm that learns to solve complex tasks from a set of ‘training examples’ that show the drone how to do certain things and cope with some difficult situations, much like children learn from their parents or teachers”, says Prof. Scaramuzza.
Cars and bicycles are the drones’ teachers
One of the most difficult challenges in Deep Learning is to collect several thousand ‘training examples’. To gain enough data to train their algorithms, Prof. Scaramuzza and his team collected data from cars and bicycles, that were driving in urban environments. By imitating them, the drone automatically learned to respect the safety rules, such as “How to follow the street without crossing into the oncoming lane”, and “How to stop when obstacles like pedestrians, construction works, or other vehicles, block their ways”. Even more interestingly, the researchers showed that their drones learned to not only navigate through city streets, but also in completely different environments, where they were never taught to do so. Indeed, the drones learned to fly autonomously in indoor environments, such as parking lots and office’s corridors.
Toward fully autonomous drones
This research opens potential for monitoring and surveillance or parcel delivery in cluttered city streets as well as rescue operations in disaster urban areas. Nevertheless, the research team warns from exaggerated expectations of what lightweight, cheap drones can do. “Many technological issues must still be overcome before the most ambitious applications can become reality,” says PhD student Antonio Loquercio.
The Latest on: Automated drones
via Google News
The Latest on: Automated drones
- Drone Charging for DJI Matrice 300/210: Heisha Tech Releases D135on February 25, 2021 at 6:51 am
Heisha Tech‘s new D135 dock provides automated drone charging for DJI Matrice series and other larger enterprise drone solutions. Heisha is focused on enabling automated drone operations by prov ...
- Pointivo Lands Eighth Patent Increasing Coverage of the Drone Analytics Marketon February 23, 2021 at 10:50 am
Pointivo is a pioneer in the use of AI and computer vision to extract information from imagery derived from various capture methods. This patent augments Pointivo’s broad IP portfolio covering its ...
- Drone Soccer Might Be the Best Way Yet to Get Kids Into Droneson February 23, 2021 at 5:50 am
Dawn M.K. Zoldi, Guest Contributor Drone soccer. It’s a thing. And it’s not just a sport that could be featured on ESPN 8, “The Ocho” (like Dodg ...
- Swoop Aero to trial medical delivery drones in regional Queenslandon February 21, 2021 at 9:30 pm
Drone company Swoop Aero, in partnership with Australian healthcare wholesaler Symbion and pharmacy retailer TerryWhite Chemmart, has announced plans to begin trialling the delivery of medication ...
- New Applications for Drones Could Mean Big Changes to Drone Lawson February 19, 2021 at 1:38 pm
There’s really not that much legal work concerning drone laws, but that could change, says David T. Norton, a partner and head of the aviation practice at Shackelford, Bowen, McKinley & Norton LLP.
- It's a Bird, it's a Plane, it's a Major Innovation: How Drones are Impacting the Mining, Energy and Agriculture Industrieson February 18, 2021 at 11:52 pm
How Drones are Impacting the Mining, Energy and Agriculture Industries By Bethany Taylor - Industries are constantly evolving, benefiting from dozens of little innovations every year. That’s not what ...
- Avidrone Aerospace Exhibits Flagship 210TL Tandem Drone Integrated with Iris Automation's Casia Onboard Detect and Avoid Technology at IDEX 2021on February 18, 2021 at 10:28 pm
ABU DHABI, UAE, /PRNewswire/ -- Abu Dhabi National Exhibition Center, Canada Pavilion stand 11-C11 - Iris Automation is showcasing its innovative Detect-and-Avoid (DAA) Casia system with partner and ...
- Edgybees raises $9.5 million for AI that augments drone camera footageon February 17, 2021 at 2:00 pm
Edgybees, a startup developing a platform that augments drone camera footage using AI, has raised $9.5 million.
- Drones with facial recognition are closer to becoming a reality thanks to a patent from an Israeli companyon February 17, 2021 at 2:39 am
AnyVision, an Israeli firm that has worked with Mercadona, patents a system that brings the use of drones with facial recognition cameras closer to reality. The patented technology will help the ...
via Bing News