Flocking drones, swarming pigeons
19 May 2015

Not only living organisms, but also robots can move in certain collective patterns. But what rules underlie such behaviour? The study of the patterns of group motion in pigeons motivated Professor Tamás Vicsek to develop a swarm of drones being able to reproduce analogous collective motion behaviour.

 

Cover image of Flocking drones, swarming pigeons

Collective motion is one of the most spectacular forms of collective behaviour. To better understand its main patterns, Prof. Tamás Vicsek and his team recorded tracks of pigeons, using miniature GPS-devices that were attached to the birds. It appeared that pigeons flying in a flock demonstrate a delicate collective decision making process. The team discovered a well-defined hierarchy among flock members, and the average spacial position of a pigeon within the flock strongly correlated with its place in this hierarchy.

Based on the hierarchy structures observed in the flocking of pigeons, the team, funded by the ERC, created an array of autonomous flying robots, each one equipped with a little ‘brain’. These helicopter-like quadcopter drones, propelled by four motors, were able to make decisions concerning direction, flight and positions entirely by themselves. When the drones were instructed to form a circle, for instance, they collectively decided which position each of them should take within that circle. In certain experiments, the team also introduced intelligent leaders into the flock of drones to mimic the hierarchy within pigeon flocks.

These autonomous drones will make possible a wide range of future applications. Hence, a self-organised flock of drones could function as a monitoring system to follow up on environmental and agricultural  changes. Moreover, a single person interacting - by means of simple commands - with the flock, opens a new dimension of human-machine cooperation.

Project information

COLLMOT
Complex structure and dynamics of collective motion
Researcher:
Tamás Vicsek
Host institution:
Eotvos Lorand Tudomanyegyetem
,
Hungary
Call details
ERC-2008-AdG, PE3
ERC funding
1 248 000 €