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
- ZF Deploys Drones to Deliver Parts on Factory Grounds on November 12, 2018 at 2:07 pm
Photo: ZF ZF says it is the first company in Germany to receive official approval for automated drone flights over factory premises. A drone prototype is now flying spare parts such as sensors or cont... […]
- The Art and Science of Modern Drone Design on November 12, 2018 at 1:58 pm
The result is an elegant and highly efficient drone capable of flying over two hours ... and any other machine that has to travel through air or water, could be automated,” said Wouter Remmerie, CEO a... […]
- Komatsu bringing the Skycatch drones powering work at its automated Japanese jobsites to U.S. on November 12, 2018 at 12:46 pm
U.S. customers will soon be able to purchase the same custom drones built for Komatsu’s fully-automated site prep service currently running in Japan, here in the States. Komatsu and drone software com... […]
- Parrot Introduces End-To-End Drone Solution For Efficient Crop Assessment on November 11, 2018 at 10:37 pm
European drone group Parrot has introduced Bluegrass Fields ... mobile mapping app that simplifies the entire in-field workflow—from the planning of automatic mapping flights, to generating real-time ... […]
- New Research analysis on Drone Logistics and Transportation market expected to grow during the forecast period on November 11, 2018 at 7:10 pm
Drones can be used to transport a wide range of product from small package to medical supplies and food among other. Wireless charging, automated ground control stations and artificial intelligence ar... […]
- Japan Launches First Drone Document Delivery in Fukushima Amid Labor Shortage on November 11, 2018 at 5:28 pm
Japan to End Beyond Visual Line-of-Sight Regulations by End of 2018 BVLOS regulations have hampered commercial drone companies for years, preventing automated drone deliveries from taking off. Japan i... […]
- DOT IG audits FAA’s drone authorization system on November 9, 2018 at 11:12 am
The US Department of Transportation (DOT) inspector general (IG) has initiated an audit of the FAA and industry program to provide timely, automated authorization of drone flights near airports. […]
- This startup can shield against bad drones. And it just raised millions of dollars. on November 9, 2018 at 10:50 am
and allows customers to pinpoint the drone and its operator within two seconds. The automated technology can also form a barrier around a secure site, causing enough interference to force a drone to t... […]
- ZF is first in Germany to fly drones over plant premises on November 9, 2018 at 4:40 am
ZF is the first company in Germany to receive official approval for automated drone flights over factory premises. A drone prototype has recently started flying spare parts such as sensors or control ... […]
- Report: Robotics and Drones on November 5, 2018 at 7:47 am
Highly automated and flexible industrial robots will help fulfill this goal in warehouses and drones that improve security and maintain infrastructure will make humans safer. Ultimately, industrial, i... […]
via Google News and Bing News