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
- Defending Against Drone Swarms with DroneTracker 3.5 on September 20, 2018 at 5:20 am
DroneTracker 3.5 builds upon the success of DroneTracker 3, which was released in November 2017, and introduced the industry’s first airspace security solution that includes automated summary reportin... […]
- PureTech Systems works with Skyfront to provide automated drone deployment at an alarm location on September 20, 2018 at 1:36 am
PureTech Systems announces its latest capability, which provides for automated dispatch of a security drone to an alarm location. The alarm can originate from sensor sources capable of providing a loc... […]
- If a Tree Falls in the Forest, a Drone May Be There to See It — and Seed-Bomb on September 19, 2018 at 9:19 pm
We want to take it all over India. Omkar Subbaram Jois Narasipura, a scientist working with seed-planting drones In Myanmar, the company BioCarbon Engineering is working on a plan to use six automated ... […]
- The Best Drones Under $600 on Amazon, According to Hyperenthusiastic Reviewers on September 18, 2018 at 11:53 am
“The Return Home Quadcopter Drone by Potensic was purchased for my teenage son. The main things that I really liked were the automatic return, low power return, and no signal return on this dron […]
- Domesticating Drones on September 18, 2018 at 9:32 am
But an explosion of cheaper computing power, motion sensors and batteries changed all that. Now, inexpensive four-rotor drones have automatic stabilizers that make them easier to fly, and millions are ... […]
- DJI Drone Development [Video] on September 18, 2018 at 3:34 am
The initial idea for Dronoway, the outcome of our experiment, was to conduct automated drone missions. First of all, we chose an iOS app to control our Mavic and machine learning part (embedded in ... […]
- These Drones And Humans Will Work Together In Hurricane Florence Recovery Efforts on September 16, 2018 at 2:04 pm
The drone shots will be uploaded to the platform where claims adjusters can review automated damage analysis, make notes and close out claims in one virtual location. According to Precision Hawk and E... […]
- African experiments with drone technologies could leapfrog decades of infrastructure neglect on September 16, 2018 at 12:35 am
This could set the stage for faster development of automated drone fleets for remote internet access, commercial and medical delivery, and even give Africa a lead in testing flying autonomous taxis. […]
- Drone Delivery Company Conducts City of Reno's First FAA-Approved Multi-Drone Flight on September 14, 2018 at 6:51 am
The landmark demonstration had one operator pilot multiple Flirtey drones simultaneously, simulating the delivery of automated external defibrillators for the proposed implementation of standardized a... […]
- The Drone Catcher: How A Michigan Robotics Lab Hopes To Keep The Skies Friendly on September 13, 2018 at 2:18 pm
Drones are just out there: While UAV regulations exist ... “It’s very similar to the automotive world with connected and automated vehicles, where the technology is moving faster than regulation is ke... […]
via Google News and Bing News