Robotic Ants Exhibit Real Colony Behavior
Michael Harper for redOrbit.com — Your Universe Online
It´s commonly known that ants navigate by scent. As each ant randomly wanders in search for food or shelter, it leaves behind a trail of pheromones which other ants will then follow. The result is that famous line of ants leading to a family picnic or a particularly sticky spill.
Though this behavior has been observed for centuries, it´s something which Simon Garnier from the New Jersey Institute of Technology and his colleagues from the Research Centre on Animal Cognition in Toulouse, France wanted to reproduce it in robots.
After building these mechanical ants and setting them loose in a maze, the robots successfully replicated this behavior, following one another´s light trails to the end of the path.
The results of this study have been published in the journal PLoS Computational Biology.
Called “Alices,” these robotic ants are no larger than a typical sugar cube and have been outfitted with two light sensors which act as their antennae. Similar to the Argentine ant´s antennae, these light sensors are used to find the light trail left behind each one of the robot ants, thus replicating the scent trail used by real ants.
The team then created a maze which mimicked the types of trails ants create to lead them to a long-term food source. When in the wild, these ants will carefully tend to this trail, removing any debris or vegetation which may get in their way. Using a simplified version of this trail, the research team created a network of three, interconnecting diamond shaped loops.
The Alices were set at the starting mark, or where a typical ant might call home. They were then set loose to navigate through the maze in the same way an ant would: randomly but in one general direction. The robots slowly began to navigate the maze successfully, and by following one another were able to eventually choose the most efficient route.
Even though the Alices were trained to navigate randomly, they knew to follow one another´s trail. So, when an Alice approached a bifurcation in a diamond shaped loop in the maze, it would generally choose the same path as the robot before it.
The research team captured all the motions with an infrared camera and were able to see how the Alices´ route choice evolved over time. In the beginning, the bots randomly wandered throughout each loop. After 30 minutes into the experiment, a stronger light path emerged, suggesting that the ants had landed on an efficient route. After one hour, the Alices had all but forgotten the other bifurcations in the maze and took only the high road with a few bends.
“This research suggests that efficient navigation and foraging can be achieved with minimal cognitive abilities in ants,” explained Garnier, lead author of the study, in a press statement.
“It also shows that the geometry of transport networks plays a critical role in the flow of information and material in ant as well as in human societies.”
Though ants are capable of working together to solve a difficult problem such as navigation, this study has found that they don´t need great eyesight or high-functioning cognitive abilities to direct them.
Image 2 (below): This image shows the robot ants (Alices) pursuing a light trail around the constructed maze. Credit: Simon Garnier: Garnier S, Combe M, Jost C, Theraulaz G (2013) Do Ants Need to Estimate the Geometrical Properties of Trail Bifurcations to Find an Efficient Route? A Swarm Robotics Test Bed. PLOS Comput Biol 9(3): e1002903. doi:10.1371/journal.pcbi.1002903. (CC BY 3.0)