The FILOSE robot mimics the geometry and shape of a rainbow trout. Tiny electronic sensors monitor pressure differences in the water flowing around it. JELENA PLJONKINA
Unlike any other animal, fish have a special sense that allows them to determine the speed and direction of currents, helps them hover in place and even swim upstream. Known as lateral line sensing, it also helps fish find the underwater “sweet spot” to catch food that may be tumbling down the river.
Now, a European engineering team based in Estonia has built an electronic version of lateral line sensing on an underwater vehicle called FILOSE (Robotic FIsh LOcomotion and SEnsing).
Around the world, underwater vehicles have been deployed for several decades to track pollution, inspect ships hulls for damage, and for surveillance. But their big drawback is limited battery life. If successful, this new device would allow vehicles to travel more efficiently through the water like a fish, saving time and battery life.
“There are 30,000 species of fish that have” lateral line sensing, said Maarja Kruusmaa, professor of bio-robotics at the Tallinn University of Technology. “If all of the fish in the sea have found it useful and none of the robots have, it makes you wonder that maybe you are missing an important piece of information.”
In fish, the lateral line runs along the skin from just behind the head to the tail. It consists of nerve cells that pick up vibrations and other data from the water, and helps them school without bumping into each other.
The FILOSE robot took four years to build. Engineers made it to mimic the geometry and shape of a rainbow trout, about 20 inches long (50 cm). To create artificial lateral line sensing, the team developed tiny electronic sensors to monitor pressure differences in the water flowing around it. Here's a video of the device in the lab.
The swimming robot can detect flow direction and swim upstream, or hold still while compensating for the downstream drift by measuring the speed of the water. It can also hover in the wake of an object to cut its energy use, according to Kruusmaa.
“It is similar to reducing your effort in the tailwind of another cyclist or reducing the fuel consumption of your car by driving behind a truck”, she said.
The next step is to take FILOSE out for a swim, something that Kruusmaa admits will be a challenge. “The sensors we have work in the lab, but the real world environment is much more complicated.”
Some researchers are trying to combine the highly-sophisticated sensing devices such as FILOSE, with robust scientific instruments that can bring back more information about the oceans.
“In the terrestrial environment, when you know the smell of food or some organic matter, you can say that’s what it looks like and I can build an electronic device that can follow that signal,” said Kanna Rajan, principal researcher for autonomy at the Monterey Bay Aquarium Research Institute (MBARI).
Rajan is designing robots with artificial intelligence to make its own decisions about how to hunt for oil spills or perform other underwater scientific tasks. It’s a lot tougher to build smart robots underwater, than a place like Mars, said Rajan, a former NASA AI researcher.
“There is no clear idea of how a chemical plume of oil or anything else to be absolutely sure you are following that line, Rajan said. “There is a huge leap in terrestrial environments and underwater environments.”
Along with Kruusmaa’s graduate student Taavi Salumae, who is first author on study, the FILOSE project was a joint effort with researchers from the University of Bath (UK), Riga Technical University (Latvia), Verona University, (Italy) and the Italian Institute of Technology. An article describing the device is appearing today in http://dx.doi.org