Autonomous Drones Navigate Disaster Zones Using Computer Vision

With their potential to navigate quickly through unstructured environments and fly through collapsed buildings, autonomous drones are set to play a major role in search-and-rescue and remote-inspection missions, where a fast response is crucial.

However, speed and manouverability in drones is still far from that of birds. Agile navigation through unknown, indoor environments poses a number of challenges for robotics research in terms of perception, state estimation, planning, and control. At the RE•WORK Machine Intelligence Summit in Berlin next week, Davide Scaramuzza, will give an overview of his research on navigation of quadrotors, and how machine intelligence improves this.

Davide is Professor of Robotics at the University of Zurich, where he does research at the intersection of robotics, computer vision, and neuroscience. From 2009 to 2012, he led the European project sFly, which introduced the world’s first autonomous navigation of micro drones in GPS-denied environments, using vision as the main sensor modality. I asked him a few questions to learn more.

Tell us more about your work at the University of Zurich.
In my research lab, we develop autonomous machines that can navigate all by themselves using only onboard cameras, without relying on external infrastructure, such as GPS, radio beacons or motion capture systems. Our interests encompasses both ground and micro flying robots but has an emphasis on flying robots because of their increased agility and maneuverability. More specifically, we aim to advance the perception capabilities of flying robots to make them fly like or better than birds. Only by achieving this goal, we will be able to have one day drones that can be deployed to search and rescue victims in disaster zones, by entering and exiting collapsed buildings or by navigating quickly through complex unstructured environments, like only birds can do for now.

What are the leading factors enabling recent advancements of mobile robots?
Availability of commodity hardware with increasing CPU and GPU computing power, better sensors, better actuators, better algorithms.

What present or potential future applications of robotics and computer vision excite you most?
Tele-robotics and prostheses. Although I dont directly work with these, I think that they are the most exciting and useful ones in the short term. Unfortunately, we are still far from having rescue robots that can be controlled remotely similar to the ones in the movie "Avatar" or robotic prostheses that can be integrated and felt as natural parts our body (with touch feedback) like in the TV series "Extant"

Which industries will be most disrupted by mobile robots in the future?
Medicine, agriculture, transportation, manufacture, service robotics.

What developments can we expect to see in machine intelligence in the next 5 years?
AI algorithms that can better understand our needs and assist us in our daily activities, much similar to the AI character in the movie "Her". Regarding robotics, within 5 years we will see more perception capabilities integrated in cars and drones.

Davide Scaramuzza will be speaking at the RE•WORK Machine Intelligence Summit, in Berlin on 29-30 June 2016. Other speakers include Axel Koehler, Principal Solution Architect at NVIDIA; Polina Mamoshina, Research Scientist, Insilico Medicine; Kilian Koepsell, Co-Founder & CTO, Bay Labs; Eli David, CTO, Deep Instinct; and Sébastien Bratières, PhD Researcher, University of Cambridge. For more information and to register, please visit the website here.

Tickets are limited for this event, book now to avoid disappointment. We are holding summits focused on AI, Deep Learning and Machine Intelligence in London, Boston, San Francisco and Singapore - see the full events list here.

UAVs Dangerous Environments Drones Deep Learning Robotics Machine Intelligence Summit Machine Intelligence Computer Vision


    As Featured In


    Partners & Attendees

    Ibm watson health 3.001
    Rbc research.001
    Mit tech review.001
    Kd nuggets.001
    Maluuba 2017.001
    This website uses cookies to ensure you get the best experience. Learn more