Keeping track of where fish and sharks travel as well as understanding the waters in which they move is becoming easier with the help of innovative ocean-going robotic devices called Autonomous Underwater Vehicles (AUVs) that hold promise for a variety of marine research.
Marine biology Professor Christopher Lowe of California State University, Long Beach (CSULB) is collaborating with Cal Poly San Luis Obispo computer science Professor Christopher Clark and marine biology Professor Mark Moline to further develop Clark's AUV capabilities, funded by a three-year, $490,000 National Science Foundation grant. Lowe is director of CSULB's Shark Lab and Clark directs Cal Poly's Lab for Autonomous and Intelligent Robotics.
Marine biologists have attached electronic tracking devices to animals for years, but they generally need to follow the animals around in boats to listen for signals. AUVs look like small torpedoes and could make this task much easier, Lowe said.
Their collaboration came about when Clark asked Lowe for some past shark tracking data. "A couple of weeks later, he e-mailed me back and said, ‘I develop Autonomous Underwater Vehicles (AUVs) and I think that based on your tracking data that you sent me, I can design an AUV that can track sharks. It can basically look for the patterns of the sharks' movements, so in other words, I can make a smart robot that will be able to track a tagged shark,'" Lowe recalled. "When he explained to me how he how he designs these navigation systems, I realized what he was talking about was very different than what other people had talked about in the past, and it was actually brilliant."
The team spent a week in August testing an AUV in southern California waters, Lowe said. "It was interesting to see how they thought tracking works and it was interesting for us to understand how they go about the approach of, ‘How do you get this robot to basically do the same things that I go about training a student to do when I'm teaching them how to track?' I was truly impressed, because these graduate students are out in a Boston Whaler and we had three to four computers in the boat at a time. They're sitting there, programming and debugging programs on the fly to work out this AUV to be able to track a stationary and moving tag."
The team later caught an adult leopard shark and put an acoustic transmitter on it, which excited the Cal Poly students. "Then we used this prototype to follow that leopard shark all day long. To see how the software was integrated in telling the robot which way to go was amazing. In fact, one of the things that I found was that the robot could actually track the shark more accurately than I could in our boat," Lowe noted.
"I was seeing a future where we put a transmitter on an animal and drop this vehicle in the water and then 18 or 24 hours later, we pick it up and our track is done. But it's more than just getting the robot to track and figure out where the shark is going. These robots are equipped with all sorts of oceanographic sensors so we can measure water temperature, depth, clarity, salinity and different factors. Plus, it has sonar on it that can be mapping the seafloor while it's tracking. Our ultimate goal is to use teams of these AUVs that will talk to each other and give even more accurate positions. We can have one AUV to track the shark while the other can be characterizing the water column and then switch off," he said.
"Our recent test-our first field trial- was, I thought, incredibly successful," Lowe continued. "The goal is to go back and try to do this again, possibly this fall at Catalina and in deeper, clearer water so that we can videotape this," then use the system next summer to track a white shark.
"The three of us working together provides a really powerful team to not just develop the technology, but one of the things that we see in the future is using these AUVs to look at the effectiveness of marine protected areas (MPAs), for example," Lowe remarked. "A lot of my previous work has involved tagging fish in MPAs and to see how they use the habitat. Do their movements exceed the areas of the MPAs? This gets to be very labor-intensive and expensive, but what if we had AUVs that we could send out to swim along the coastline and listen for tagged fish?" on a monthly basis while also collecting other valuable water data. "By working with folks like Drs. Mark Moline and Chris Clark, we can develop cost-effective tools that state or federal agencies can use to determine whether the MPAs we're putting in are effective in protecting fish."
But their project goes beyond just the technology. "The interdisciplinary collaboration is important," Lowe said. "Quite often, biologists and computer scientists are trained completely differently and we think differently. It's great to be able to work with someone who wants to share a common goal but who is coming at it from a completely different perspective. One of the most important aspects of this project is that we're training students from two completely different disciplines who have to work together to achieve this goal, and this is going to be critical for the future. We really need to focus on training science students and engineering students to work together better, and that needs to happen earlier in their development and training.
"Of course, when you're out in a boat all day long, it becomes much easier to develop that kind of dialogue. After a couple of days, you can begin to see that kind of thought process coalesce and they begin to understand each other better," he noted. "That, I think, is probably one of the coolest things that I saw from this project so far."