Robot Ants Mimic Real Colonies: How Ants Always Choose the Shortest Path to Food

Robotic ants look nothing like their biological counterparts. They're cube-shaped and instead of six legs, they have wheels powered by tiny watch motors. Yet they do have some similarities; these robots act and behave just like a real colony.

Ants usually follow scent trails, which communicate information via a chemical signal. Yet scientists have long wondered how ants find the most efficient routes to a food source in the first place. The new robots that the researchers have created could provide and answer.

The researchers programmed the sugar cube-sized robots simply to move forward toward a target and avoid obstacles. They then placed these robots within a maze. Like ants, these robots could leave a trail for others to follow; instead of a chemical trail, though, they left a trail of light.

In order to allow them to leave this light trail, researchers set up a camera to track the path of each robot. A projector connected to the camera then produced a spot of light at regular invervals along their route. The robots themselves were equipped with two "antennae" on top, which were light sensors. If a robot encountered a crossroads where there was more light on the right, then the robot would turn right.

At the beginning, the robots would choose the path that deviated the least from the path that it was already on. Once they reached their goal at the end of the maze, the mechanical ants were programmed to retrace their path to the starting point while leaving a trail of light. Their paths were extremely varied at the start; robots chose different paths to reach the same target. Yet eventually, researchers started to notice a pattern. The robots started to use the shortest, fastest route to the source.

"Because ants taking the shorter path travel faster, the amount of pheromone (or light) deposited on that path grows faster, so more ants use that path," said Simon Garnier, one of the researchers, in an interview with BBC News.

The new research shows that robots can provide new clues as to how biological organisms work. Like the robots, ants will walk forward and choose random paths. Over time, the most efficient route is created and the majority of ants follow it--you may have seen the result on a sidewalk when someone has dropped a piece of food.

"This shows that you don't need something as complex as choice to get some behavior you see in ants," said Graham in an interview with BBC News.

The findings are published in the journal PLOS Computational Biology.