Sunday, August 17, 2008

Traveling Sales Ants & Swarming Robots

“Swarm intelligence” is a term used to describe the self-organized collective behavior of social insects. For instance, one ant following the trail of another eventually provides the colony the shortest route among the countless possible paths to a food source. In an article in Scientific American, Eric Bonabeau and Guy Théraulaz report on how computer scientists use these social insects as models to solve complex problems. The foraging of ants, for instance, has led to a novel method for rerouting network traffic in busy telecommunications systems.
The cooperative interaction of ants working to build their nests led to more efficient control algorithms for groups of robots. The way insects cluster their colony’s dead and sort their larvae can help analyze banking data. And lastly, the division of labor among honey bees could help streamline assembly lines in factories. Shades of biomimicry (see my last post) me thinks. Don’t you?

Traveling Sales Ants: The ant highways we see in nature (and some people’s kitchens) are created by ants depositing pheromone (a chemical attractant) for other ants to follow. The first ants to return to the nest from the food source are those that have taken the shortest path, and, because this route is the first one to be doubly marked with pheromone, it becomes the main route. Researchers used artificial ants that deposited a pheromone that would eventually decay (favoring shorter paths) to solve the traveling salesman problem: devise the most expeditious route by which to visit a given number of cities.

Swarming Robots: “Pheromone logic” can allow groups of tiny robots to navigate complex environments through “swarm intelligence”. For instance, tiny robots that mimic ants leave traces of “pheromone” light behind as they explore a network. The robots can detect and are attracted to light traces left by the other robots. Over time, the most popular path between start and goal becomes the most brightly lit. In another example, robots mimic weaver ants that cling to one another to form a living structure and allow the self-assembled colony to perform tasks that individual robots could not perform.

Recommended Reading:
Eric Bonabeau, Marco Dorigo and Guy Theraulaz. Swarm Intelligance: From Natural to Artificial Systems. Oxford University Press. 1999.
S. Camazine et al. Self-Organization in Biological Systems. Princeton University Press. 2001.
Erol Sahin, ed. William M. Spears and Alan F.T. Winfield. Swarm Robotics. Springer. 2007.

Nina Munteanu is an ecologist and internationally published author of novels, short stories and essays. She coaches writers and teaches writing at George Brown College and the University of Toronto. For more about Nina’s coaching & workshops visit Visit for more about her writing.


Jean-Luc Picard said...

Ants have always been the most fascinating creatures to me, Nina, with their societies and the way they live.

sfgirl said...

Me too, Jean-luc! When I was little and my big brother was tormenting them I used to spend hours (okay, minutes) watching them build their nests and carry materials to their nests. They are fascinating creatures. Did you know that there are over 12,000 species of ant known on our planet? While traveling in Africa, I had a close encounter with driver ants (EEK!). Not sure if you remember that post, called "The Paradox of Africa: the Little Things". LOL! I ended up sleeping on the top of the truck cap...

Anonymous said...
This comment has been removed by a blog administrator.
pheromone said...

It's interesting to see how pheromones interact with the activities of ants in colonies.