How smart do ants have to be to find their way in complicated, branching trails? Not very, it turns out. As the insects return to their nest from a food source, they follow trails that may branch in different directions. Scientists suspected that the ants had a way of calculating the angle of the trails so that they could choose the most direct way home, but a study published today in PLOS Computational Biology shows that robots can choose the same pathways as real ants without doing any math. Researchers used tiny, cube-shaped robots that were powered by watch motors and ran on dime-sized wheels. They gave the machines three rules: to walk randomly in a given direction, to turn away from obstacles they bump into, and to follow a trail of light left by other robots (as seen in the video)—similar to the way real ants use their antennae to sense chemicals left behind by other ants. These simple tenets were enough to allow the robots to copy ants' ability to find the shortest path home. Other groups, such as humans in a crowd, may also navigate by simply reacting to the shape of their environment. The findings could lead toward better design of the spaces we move through, the researchers say, such as roads and crowded buildings.
See more videos.