'Robofish' makes friends with biological cousins
By Ben Coxworth
June 30, 2010
Scientists seem to like the idea of robotic fish, and why not? They have all sorts of potential applications including exploration, pollution-detection, communications, or just for quiet contemplation. A team from the University of Leeds, however, have created a robotic fish that can do something no previous effort has laid claim to – fool other fish into thinking it’s one of them.
The computer-controlled replica, dubbed Robofish, is a plaster cast model of a three-spined stickleback fish. It has an acetate tail, and is painted to match the stickleback’s natural coloration. A rigid clear tube runs from its underside to a magnet. When placed in an aquarium, that magnet sits against the bottom glass, and is drawn to a corresponding magnet on the outside of that glass. By moving the outside magnet with an electric motor, the team was able to create the illusion of an independently-swimming fish.
The model was placed in a tank with either single sticklebacks, or groups of ten. In both cases, it was programmed to “swim” along a set path at a speed slightly higher than average, to see if it could coax other fish out of the tank’s refuge area, and into a 90-degree turn. While the researchers initially wondered if Robofish’s lack of stickleback odor would be an issue, it turned out not to be a problem.
Individual fish followed the model most readily, although even the groups took to it pretty quickly. Interestingly, though, the longer the real fish had been in the tank, the less likely they were to follow their phony leader. "Because Robofish moved faster and without stopping and tended to be at the front or on the edge of the shoal, the other fish saw it as bold and definite in its actions, which encouraged them to follow," said PhD student Jolyon Faria. "The fish were more easily influenced by a bolder member when they were nervous in new surroundings. Once they'd got used to the tank, they moved round quite happily to explore the tank, rather than moving in unison as they did at the start."
It was also noted that the fish closest behind Robofish was always the first one to follow it into the turn, regardless of how close behind it was following. This apparently addresses a long-standing question in the field of group dynamics – is a leader’s effect on a group influenced by how physically-close that leader is to the group? This experiment would seem to indicate not.
The Leeds team believes that knowledge gleaned from robotic fish could be used in the field of freshwater and marine environmental management, to predict migration routes and the effects of human interference. Given Ms. Faria’s observations about the stickleback experiment, however, it looks like they could also teach us something about ourselves.