Hungarian researchers have developed drones that can fly as an autonomous flock, opening new possibilities for the artificial pollination of crops or search and rescue missions. The scientists originally set out to build flying robots to help study collective motion in animals, but ended up with drones that can fly cooperatively without any central control. Rob Muir reports.
BUDAPEST, HUNGARY (REUTERS) - Swarming in the evening sky over a field near Budapest, ten quadcopters are instructed to form a circle. The need no other directions. They are designed to communicate with one another to fly cooperatively, like birds in a flock, and they fall into line, each knowing its neighbour's position.
They do it, using a combination of GPS signals for navigation and radio to communicate. A small but powerful computer mounted on each quadcopter translates those signals into commands governing direction and speed, according to Gabor Vasarhelyi
GABOR VASARHELYI, ROBOTIC PROJECT LEADER , "Instead of someone holding a remote control and manually controlling the drone we have hardware that automatically calculates these steering signals."
And the team have brought an unprecedented level of sophistication to their flock. Vicsek says that when instructed to fly through an imaginary channel, his drones will line up and take their turn.
The implications of their work for industry, Vicsek says, are already becoming clear.
TAMAS VICSEK, PRINCIPAL INVESTIGATOR AND PROFESSOR OF PHYSICS AT ELTE UNIVERSITY, "One of our favourable examples for potential application is just to release a fleet of these drones over an agricultural field and tell them that 'cover the whole field and look for any problematic places or artificial pollination is a potential application so just cover everything and so the farmer can take 50 of these, release them, give them the co-ordinates of only the field, the borderline of the field, the rest is done automatically by these drones."
But that day is still some way off. The technology is not perfect and Vicsek and his team hope to refine it with camera vision, bringing another dimension of control to mechanical flocks that will one day function reliably on their own.