Swarm of micro FLYing robots
The European project sFly aims at developing the technology for a swarm of small helicopters to collaboratively sense their workspace. The outlook is to use such a system for assisting tasks like surveillance, inspection and search & rescue operations. sFly is funded by the European Community's Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 231855.
The sFly team at ASL is mainly working on visual navigation using a single camera researching avenues for fusing additional feeds for sensors like Inertial Measurement Units and GPS. More details about the project itself can be found on http://www.sfly.org while, for more information on the work we do at the Autonomous Systems Lab (ASL) you are welcome to explore our Research pages.
Our paper on Real-time onboard visual-inertial state estimation and self-calibration of MAVs in unknown environments (can be found
here) was a
finalist for the Best Vision Paper Award at ICRA 2012.
On the 29th of February we had the final review meeting of the project, including an impressive outdoor demonstration in a simulated disaster area. The evaluation committee (
Simon Lacroix,
Anibal Ollero and
Olivier Da-Costa) has given excellent reviews, giving special praise on the vision-based hovering and flight control developed in ASL.
Check out the newest demo proposal video for a search-and-rescue scenario with three helicopters
here.
Publications: Two papers have been accepted for publication in ICRA 2012. Check out the papers and the related video
here.
Publications: Two papers have been accepted in IROS 2011 which had a record acceptance rate of
32% this year and another paper has been accepted for oral presentation in BMVC 2011 (oral acceptance rate
8%).
We organised the
Autonomous MAVs Summer School here in Zurich together with Dr Davide Scaramuzza, aimed at keen graduate students of all levels. The
program covered aspects in design, perception and control of MAVs featuring distinguished speakers from the Computer Vision and Robotics communities.
Publications: We have a paper accepted in ICCV 2011, one of the top-tier conferences in Computer Vision with acceptance rate of
24% this year.