1 Flares Twitter 1 Facebook 0 Google+ 0 Reddit 0 Pin It Share 0 Email -- 1 Flares ×

antrobots

Nobody would disagree that ants have their stuff together when it comes to efficiency; they're models for success. One of their more remarkable traits is the ability to choose the most efficient route to their food, even when the route is far from obvious. In order to find out how ants accomplish this, a team of scientists built a robotic ant army.

The scientists in question programmed ant behaviors into a tiny, wheeled cube robot name Alice, which was developed years ago by a Swiss team. The robots possessed the capabilities to avoid obstacles and followed light trails in the same way that ants follow pheromone trails. Using only these few simple behaviors, the robots were able to the find the shortest path through a maze over seventy percent of the time.

So what did they learn? Ants don't have any sort of "complex cognitive process," but instead rely on an obstacle-avoidance system and directions from their peers. The next step is to learn how ants devise their mazes in the first place.

via The Verge