In a swimming pool on the University of Washington campus, engineers test a group of robotic fish designed to swim, maneuver around obstacles, communicate and navigate without the help of a human. These autonomous, fin-powered machines are the creation of Kristi Morgansen and her team at the Nonlinear Dynamics and Control Lab, who wrote a series of algorithms based on the behavior of schooling fish to power these autonomous robots.
“We would like to be able to go out on a global scale and collect information to make better models of what’s going on in the ocean,” Morgansen says in a talk at TEDxRainier. “The ways that data is being collected currently involve sending out research vessels with a number of people on them that go to specific locations; there’s some use of sensor buoys that are sent out to drift into the ocean; some use of fixed sensors and, in some cases, uses of autonomous vehicles in order to go out and collect information … What we would like to do is have a much larger scope of information we can collect than what is currently feasible,” she says. “What’s currently being collected is in a much narrower space than the entire ocean.”
One way to get more information? Fish robots. Fish robots capable of entering into months-long excursions by themselves to areas of the ocean yet unknown or poorly documented.
But with such an audacious task comes a lot of requirements: the fish must not use up too much energy, be able to navigate complex landscapes, move quickly, avoid obstacles and communicate with one another.
The best way to figure out how to do that, Morgansen and her team found, was to look at what they knew worked — live animals — so they studied how fish moved in water, how the design of their bodies contributed to the power and shape of their movements, how they flexed and propelled. To help keep the robots from colliding with one another, and to avoid foreign objects, the team studied the flight patterns of barn swallows, who avoid obstacles when moving at high speeds and within large groups.
Morgansen and her team created mathematical models built on these observations, and used these to create algorithms built for the fish robots. These algorithms iterated on the models of biological systems, to create “effects … that do work on the robots, but have never been observed in biology,” Morgansen says.
And the team continues to develop the robots, crafting man-made creatures to help us understand the biology of the sea.
Watch Morgansen’s whole talk below:
Insights from the TEDx office — why we like this talk:
The speaker is doing novel work with a university to create evidence and data-based new technology. She leads a research group at a university and has experience working on engineering projects. Her talk explains the technology her team has developed, fish robots, in detail — not only the narrative of the robot’s creation, but also how it can and will be applied in real world settings.