By observing the behavior of ants, including the way they sometimes stop and visually scan the world, scientists at Sussex last year were, for example, able to understand the nature of the special “learning walks” ants engage in when exploring new terrain. Then using these “very efficient and simple view-based methods” they were able to come up with a biologically plausible algorithm that could provide robots with “a highly robust and minimal method for navigation in difficult environments like deep space.”
“If we think like a human then it’s going to be very hard work to solve some of these challenges,” according to Husbands. “Instead ants are optimized for interacting with their environment. Their resources are limited but they are very sophisticated.
“So with a very small brain they can do very simple things in very efficient ways which can then be implemented very economically” in robots and artificial intelligence. “It’s very illuminating and chastening to think about insects,” he adds. “It’s a reminder of a very different view of the world.”