Shark Scientists Pit Robot Against Human
Autonomous underwater robots have the ability to track tagged sharks at a smart distance and measure all kinds of crucial data using sophisticated sensors. But how does it compare to a human in a boat who’s trained to track sharks? Two scientists in California pitted them against each other to find out.
California State University Long Beach marine biology professor Chris Lowe has been collaborating with Harvey Mudd College associate engineering professor Chris Clark on autonomous underwater shark-tracking robots for several years now. Last summer they used the robots to track tagged leopard sharks for the first time and gathered a bunch of data for ongoing research. Lowe told me he thinks the technology will revolutionize marine science. Now he just needs to convince his colleagues.
To that end, he and Clark recently set up a test he dubbed “Man vs. Robot.” They had a student in a boat who had the best tracking skills Lowe could teach. The student’s goal was to get the most accurate positions on a tagged leopard shark without influencing the shark’s behavior. Leopard sharks are slender, spotted sharks known for being calm around humans. They like the California coast. At the same time, the team placed a robot in the water to track at a distance.
“Because we knew exactly where the boat was and the tracker in the boat estimated where the shark was, we could actually compare man versus robot,” Lowe said. So, how’d it go? “As you can imagine,” he said, “the robot kicked the man’s butt.”
The sensor-laden robots can pick up all kinds of data, including depth, salinity, pH, chlorophyll and pressure — all while staying the right distance from the shark. These robots are also equipped with Wi-Fi and programmed to come up to the surface periodically so the scientists can download the data easily from a nearby boat.
Lowe and his team hope that this head-to-head test — and all the detailed data they’re gathering — will lead scientists to use fleets of high-tech robots in the future. The bots could help us better understand how environmental conditions influence marine animals’ behavior and migratory patterns. They have implications beyond sharks. Lowe said the robots could be deployed to track and monitor fish, lobsters and even crab as long as they’re large enough to be tagged.
“If there are commercially important species and fishers go out and they’re not catching them, does that mean they caught them all or does that mean the animals simply moved somewhere else?” Lowe wondered. The agencies responsible for monitoring fish populations could use autonomous robot fleets, which Lowe said would be more cost-effective than sending out ships full of people.
Next, Lowe, Clark and their team plan to use the robots to track a baby white shark. Unlike gregarious leopard sharks that grow to around four or five feet, great whites can be 15 to 20 feet in size. They’re extremely challenging to study. Lowe has tracked white sharks using satellite telemetry in the past, but it has limitations. The tagged shark needed to be near the surface to get a good GPS reading. Sending a robot after a tagged baby white shark will represent a first, and should reveal new information about the shark’s fine-scale movement.
At the moment the scientists are waiting for the permits, which they hope will come through in the next couple weeks. Lowe said he thinks smart autonomous underwater robots are the wave of the future. “It’s exciting to be at the front of the wave,” he said.
Photo: A leopard shark like this one was tagged for the robot vs. human test. Credit: Scott via Flickr.