Boaty, the long-range autosub, has had a year. It went to, and came back from, a trip under the Antarctic ice. Shortly after, it set off on a new mission to survey the coastal ecosystems of the UK, returning back to researchers in mid-June.
Boaty McBoatface, famously named as such because of an internet contest gone awry, is one of many autonomous underwater vehicles that researchers at the UK National Oceanography Centre (NOC) use to collect data from the ocean.
The 11-foot-long sub completed a mission at the Thwaites “doomsday glacier” in late March. “My understanding is that from a technical perspective, that was highly successful. It did achieve its objectives,” says Matthew Palmer, chief scientist for Marine Autonomous & Robotic Systems (MARS) at the NOC.
Now comes the hard part: analyzing all the data it gathered and crunching the numbers. “You spend a few days in the ocean and then you spend the next three or four years trying to work out what it’s telling us,” Palmer says.
The team is not wasting any time. Boaty has a busy year ahead of it, with multiple missions lined up, some a bit closer to home.
The run from Plymouth, UK to the continental shelf south of Ireland was intended to prove that Boaty could indeed complete long-distance journeys, even in fairly unusual coastal environments. And it was the longest journey yet for the autosub. Boaty went for 5 weeks on its own to depths of over 1,000 meters (that’s more than 3,000 feet), collecting measurements on temperature, salinity, ocean physical structures, current velocities, and more. It surfaced daily to phone home sensor data and receive new piloting instructions from the team in Southampton.
“The UK is quite uniquely situated… we have a broad shelf sea. In the US you have a shallow ocean that doesn’t actually last very long before it drops away into the deep ocean. In the UK, we can travel hundreds of kilometers before we reach deep ocean,” says Palmer. “The challenge here is to go across the shallow ocean, where you’ve got ships, you’ve got fishing vessels, and it’s a hazardous environment for us.”
“The important thing scientifically is that now we have this capability on a single vehicle to track what’s going on from deep ocean all the way to the shore,” he adds. “That vehicle can operate to just a few meters deep. That’s a big advance that we’ve managed.”
Tech like Boaty is part of the NOC’s Marine Autonomous Robotic Systems (MARS) Development Group. The group also works with smaller robots that can measure more fine-scale things such as sediment mix, and more advanced systems that can contain echosounders that can survey the seabed, or measure things like what fish or phytoplankton might be in the water. Boaty can get its sensors swapped out depending on what kind of science researchers are hoping to use it for.
The team is also tweaking the onboard command and control on those vehicles, including testing giving these vehicles artificial intelligence. “It makes sense because once these vehicles go beneath the surface, we can’t talk to them. They need to operate on their own,” says Palmer. This would be an upgrade to the program Boaty used in Antarctica, where it was following a set mission, although it had to take into account environmental conditions it came into contact with like rapid, unexpected changes in density. Then, it would adjust its settings accordingly by running through a series of what-ifs, and-or scenarios. But this more old-school, rules-based approach may not work in more complex settings.
One benefits of giving it more autonomy in this way could be that it can decide to investigate something interesting it finds. “So if I’ve got a vehicle, it needs to recognize an important science feature that I might be interested in seeing. I might not want it to come all the way to the surface now and tell me about it,” Palmer says. “I might want it to stay down and say ‘ok, I’ve found a really interesting patch of phytoplankton,’ for instance. ‘I’ve got this list of ideas of what I might do, depending on its size. I might stay with it for a certain period of time. I might drift alongside it for a few days.’”
“One future aspiration is that we have fleets of these things out monitoring our oceans on an almost continuous basis,” Palmer says. A system like the Argo float program, is an example of a successful application of this capability. “I think that does help set very high aspirations on more controllable types of vehicles… It might be Boaty, it might be other vehicles such as ocean gliders. Also, we might design smaller AUVs that do shorter periods of time. They’re much cheaper, you can use them for a few hours or a few days, and that would complement the long term science that the autosubs do.”
Having fleets of these robots out near shore without the need for frequent trips by big research ships would allow the research teams to do more science with a lower carbon footprint. Palmer thinks that because there’s a lot of opportunity to grow and develop this technology globally, including in lesser developed countries that are often coastal nations, as a way to better monitor their marine environment and fisheries.
Some of the emerging opportunities for these robots have been accelerated because of the current energy crisis in Europe and the UK. One of the benefits of the UK’s shallow surrounding seas is that you can put structures like wind farms in it fairly easily.
“We’ve currently got a quite small footprint of offshore wind farms in the UK that’s starting to provide a significant amount of our energy. But what would upscaling that 10 times look like?” Palmer wonders. “What would that actually start doing to your physical ecosystems and biological structures in your oceans?” For example, if it changes the tides, it could affect the country’s current flood defenses. And, having wind farms could impact fishing, leisure, and transportation in the surrounding area. “It’s a really serious question that needs to be addressed,” he adds.