CAMBRIDGE, Mass., Aug. 15 (UPI) -- One thousand small and extremely simple robots, each only a few centimeters across and balancing on three tiny legs, assemble autonomously -- the world's first robot flash-mob. Asked to form a "K" or "star-shape" by their programmers and they obey, slowly and seamlessly marching into position.
Researchers built and programmed the so-called kilobots to demonstrate the ways simple organisms (or machines) can combine to form impressively complex systems -- like cells that organize to form organs or ant colonies that efficiently locate and retrieve food or built elaborate shelters from dirt and sand.
The self-organizing swarm or kilobats was designed by a team of researchers lead by Radhika Nagpal, an engineering and applied sciences professor at Harvard University's Wyss Institute.
"The beauty of biological systems is that they are elegantly simple -- and yet, in large numbers, accomplish the seemingly impossible," explained Nagpal. "At some level, you no longer even see the individuals; you just see the collective as an entity to itself."
Once the robots are programmed with an initial algorithm they are completely autonomous, able to follow simple commands and form specified shapes by communicating with their neighbors. On their own, their clumsy little machines, but when they cooperate they are much more efficient.
Nagpal says this is the way of the future.
"Increasingly, we're going to see large numbers of robots working together, whether it's hundreds of robots cooperating to achieve environmental cleanup or a quick disaster response, or millions of self-driving cars on our highways," she said. "Understanding how to design 'good' systems at that scale will be critical."
Other researchers are impressed too. Roderich Gross, a researcher at the University of Sheffield, told National Geographic that he was so impressed he bought 900 of the mini-bots to use in his own experiments.
"This is not only the largest swarm of robots in the world but also an excellent test bed, allowing us to validate collective algorithms in practice," Gross said.
Nagpal's work was assisted by Harvard researchers Mike Rubenstein and Alejandro Cornejo. Their efforts are detailed in the latest edition of the journal Science.