Large Aerial UAV Swarms

More goodness from the Laboratory of Intelligent Systems at EPFL in Switzerland: 10 Zagi-style UAVs autonomous flocking at the same time. From the write-up:

"Swarm Algorithms
Designing swarm controllers is typically challenging because no obvious relationship exists between the individual robot behaviors and the emergent behavior of the entire swarm. For this reason, we turn to biology for inspiration.

In a first approach, artificial evolution is used for its potential to automatically discover simple and unthought-of robot controllers. Good evolved controllers are then reverse-engineered so as to capture the simple and efficient solutions found through evolution in hand-designed controllers that are easy to understand and can be modeled. Resulting controllers can therefore be adapted to a variety of scenarios in a predictable manner. Furthermore, they can be extended to accommodate entirely new applications. Reverse-engineered controllers demonstrate a variety of behaviors such as exploration, synchronization, area coverage and communication relay.

In a second approach, inspiration is taken from ants that can optimally deploy to search for and maintain pheromone paths leading to food sources in nature. This is analogous to the deployment and maintenance of communication pathways between rescuers using the SMAVNET."

(Thanks to Sabine Hauert for the tip)

0 members like this

Moderator Comment by Gary Mortimer on September 8, 2010 at 4:23pm: Have you seen there collision detection stuff as well?? They know what they are doing!

Comment by Earl on September 8, 2010 at 4:41pm: The chance of any 2 of these being in the same space at the same time is VERY small.
Earl

Comment by Ravi Gaddipati on September 8, 2010 at 4:52pm: The GCS will be able to monitor all activity and divert all collisions.

Swarm control is not hard, the hard part is organizing them to do something useful efficiently.

Our GCS will have "swarm control" capability, and may even implement a system for the organisational aspects.

Comment by simonl on September 8, 2010 at 5:19pm: 10 of those landing in quick succession - WOW. Keep a look-out guys!

Seriously; 10 of them in the air together is quite impressive - well done.

Comment by passunca on September 8, 2010 at 5:42pm: I am curious to know how many nodes will the xbee network support. The Zigbee protocol seems to support a very interesting way of routing between the different nodes. So i could imagine a swarm of several hundred drones covering a huge area and maintaining communication between them each one. That would make a big difference in SAR missions.

Comment by Taylor Cox on September 8, 2010 at 7:01pm: What is the airframe they are using? Did they develop that one specifically for swinglet? Or where can I get a plane similar to this one?

Comment by Tedro on September 8, 2010 at 8:19pm: "embedded linux" - let the fun begin...

Comment by Gerry Lichter on September 8, 2010 at 9:11pm: http://www.sensefly.com/products/swinglet/

Is this one of those: "if you have to ask" things?

Comment by Taylor Cox on September 8, 2010 at 9:15pm: @Gerry...9k US

3D Robotics Comment by Chris Anderson on September 8, 2010 at 10:15pm: So that's $90k of planes they had in the air! It's great to be an academic on a grant ;-)

Comment

You need to be a member of DIY Drones to add comments!

Join DIY Drones

‹ Previous
1
2
3
Next ›

RSS

Welcome to
DIY Drones

Sign Up
or Sign In

Or sign in with:

Top Discussions

1	How to enable ArduIMU 3 magnetometer?
2	MyFlyDream AAT and APM/Pixhawk
3	Pixhawk MAVLINK data stream rates / Pixhawk sensor calibration
4	mission planner doesn't start after last update
5	Electric inspection plan for fixed-wing UAV
6	Help - run out of 12v connections on my PDB - options???