US researchers have made a robotic version of an Amazonian fish that can move from swimming forward and backward to vertically almost instanteously, as a result of which they hope to improve our understanding how the nervous system sends messages throughout the body to make it move.
They also hope their research will pave the way for nimble underwater robots that assist in recovery operations or long-term monitoring of coral reefs.
After making computer simulations of the movements of the Amazonian black ghost knifefish, the team, led by Dr Malcolm MacIver, associate professor of mechanical and biomedical engineering at Northwestern University’s McCormick School of Engineering and Applied Science in Evanston, Illinois, gave their robotic fish a sophisticated, ribbon-like fin so it could have the same agility as the knifefish.
They published a paper on their work in the online before print December 2010 issue of the Journal of the Royal Society Interface.
The black ghost knifefish, works at night in rivers of the Amazon basin. It hunts its prey using a weak electric field that surrounds its whole body, and it has a ribbon-like fin on its underside that helps it move both backward and forward.
MacIver, a robotics expert and recent recipient of the prestigious Presidential Early Career Award for Scientists and Engineers, has been studying this particular fish for years.
Together with co-author Dr Neelesh Patankar, who is also associate professor of mechanical engineering, he has been building mechanical models of the fish in the hope of better understanding how the nervous system sends signals to various parts of the body to make them move.
It was when graduate student Oscar Curet, a co-author of the paper, saw one suddenly move in the vertical direction in a tank in the lab, that they got the idea to create what they called the “GhostBot”.
“We had only tracked it horizontally before”, said MacIver in a statement, explaining that they wondered “How could it be doing this?”
After observing the fish again, they notice that for horizontal motion it only made one wave ripple along the fin (forward or backward), but for vertical motion it used two waves: one going from head to tail, and other going from tail to head. The two waves collide and stop in the middle of the fin.
The team then simulated this effect on the computer.
They found that when these “inward counterpropagating waves” meet they cancel horizontal thrust and funnel the fluid motion of the two waves into a downward jet from the center of the fin, pushing the body upwards. The shape of the flow looks like a mushroom cloud with an inverted jet.
MacIver said it was interesting that the force was coming from the fish in a completely unexpected direction, allowing it to perform the acrobatics necessary for its agile lifestyle of hunting and maneuvering among tree roots.
To take the concept further, the team hired Kinea Design, a design firm that specializes in human interactive mechatronics and was co-founded by Michael Peshkin, professor of mechanical engineering, who worked closely with the team to design and build a robot that seven months and 200,000 dollars later, became the GhostBot.
GhostBot, which is about the length of a human forearm, has 32 motors that give independent control of the 32 artificial fin rays of its artificial fin. This means it has 32 degrees of freedom: quite a few more than the average industrial robot arm which has only 10.
The team then tested it in a flow tunnel in the lab of co-author Dr George V Lauder, professor of ichthyology at Harvard University.
They observed how water flowed around GhostBot by adding reflective particles to the water and then radiating it with a laser sheet. The flow travelled exactly as predicted by the computer simulations.
“It worked perfectly the first time,” said MacIver, “We high-fived. We had the robot in the real world being pushed by real forces.”
As well as the motorised fin to propel it through the water, the GhostBot is fitted with electrosensory system similar to that of the knifefish, and the team are planning to improve the design so the robot can sense an object and then autonomously “decide” to use its mechanical system to approach it.
Humans are very good at making airplanes, cars and other high-speed but not highly maneuvrable devices, but studying animals gives scientists a platform from which to design low-speed highly maneuvrable devices, a technology that doen’t yet exist, said MacIver.
Such devices would be invaluable in underwater recovery operations, such as plugging a leaking oil pipe, or in long-term monitoring of various environments under the sea, such as coral reefs.
MacIver and colleagues are also trying to answer other, more basic, science questions as they pursue more lab studies on the robot.
“The robot is a tool for uncovering the extremely complicated story of how to coordinate movement in animals,” said MacIver.
He said by simulating and then performing the movements of the fish, they are increasing their understanding of how this non-visual creature has such remarkable, acrobatic agility.
The next step is to bring together the sensory work with the mechanical movement work, he added.
“Aquatic manoeuvering with counter-propagating waves: a novel locomotive strategy.”
Oscar M. Curet, Neelesh A. Patankar, George V. Lauder, and Malcolm A. MacIver
Journal of the Royal Society Interface, Published online before print 22 December 2010.
DOI: 10.1098/ rsif.2010.0493
Additional source: Northwestern University News Center (article 18 Jan 2011).
Written by: Catharine Paddock, PhD