Group Thinker: Researcher Gets $2.9 Million to Further Develop Swarm Intelligence

The European Research Council opens up its checkbook to fund Marco Dorigo‘s latest efforts to get networks and robots to mimic the highly efficient behavior of ant and other colony insects.

Swarm intelligence is a branch of artificial intelligence that attempts to get computers and robots to mimic the highly efficient behavior of colony insects such as ants and bees. Ants, for example, usepheromone trails to mark the routes they use to find food. The more traversed trails develop an accumulation of pheromone that attracts new ants, whereas pheromones deposited on paths less traveled will evaporate over time.

With an eye on the potential that swarm intelligence holds for the development of information technology and robotics, the European Research Council (ERC) awarded a $2.9-million ERC Advanced Grant Tuesday to Marco Dorigo to help the research director for the Belgian Funds for Scientific Research (FNRS) and co-director of IRIDIA (the Free University of Brussels’s artificial intelligence lab) further his work engineering swarm intelligence systems.

Phone systems use a similar approach to routing calls, using bits of information as “virtual pheromones” that reinforce paths through less congested areas of a network, researchers Eric Bonabeau and Guy Théraulaz noted in Scientific American’s February 2008 special issue on robots. Dorigo and his colleagues have applied this philosophy to the Internet and managed to outperform all other data-traffic routing methods, the authors added.

Dorigo, who has been researching ant colony swarm behavior for more than a decade, will use the ERC funding to develop a universal engineering methodology for the design and implementation of artificial swarm intelligence systems. “We believe that in the future, swarm intelligence will be an important tool for researchers and engineers interested in solving certain classes of complex problems,” Dorigo wrote in an e-mail to Scientific American. “To build the foundations of this discipline and to develop an appropriate methodology, we will proceed in parallel both at an abstract level and by tackling a number of challenging problems in selected research domains. The research domains we have chosen are optimization, robotics, networks and data-mining.”

Read more . . .