Robots go their own way deep in the ocean
“It’s very common,” says Jess Hanham casually, when asked how often he finds suspicious unexploded bombs.
Mr. Hanham is a co-founder of Spectrum Offshore, a maritime survey company that works extensively in the Thames Estuary.
His company does all kinds of marine surveys, but working on sites for new offshore wind farms has become big business for him.
Works in the Thames Estuary and other areas that were bombed in World War II are likely to involve the collection of unexploded ammunition signals.
“You can find a significant amount of leads that need further investigation and for a wind farm that will be established in the first pre-engineering survey,” he says.
With this information, project managers can decide whether to place turbines and other equipment at a safe distance from suspected bombs or blow them up by a specialized firm.
At the moment the marine survey is carried out by teams that go out on the boats, collect the data and report them for analysis.
Sometimes this will involve a relatively small ship with two crew members, a surveyor and his kit. But larger offshore inspection projects can involve much larger boats, with dozens of crew, costing in the region of £ 100,000 a day.
Sensor equipment varies depending on the job. Sometimes it could be a sonar array towed behind the boat, for other jobs it could be an unmanned underwater vehicle, which can be controlled by surface surveyors.
Bad weather can interrupt work and make life uncomfortable. “I’ve been at sea in force nine and force 10 and they’re not good places to work,” says Brian Allen, Rovco’s chief executive.
More technology for business
His company is one of many seeking to revolutionize that market by using artificial intelligence (AI) systems. They see a future where underwater robots, known as autonomous underwater vehicles (AUVs), will be able to do survey work without much human supervision and send data to surveyors in the office.
Rovco, based in Bristol, is working on key elements of the technology. He trained an artificial intelligence system to recognize objects on the sea floor from data collected at sea, a process that took four years.
The addition of AI means that the data does not need to be analyzed by a human on the ship or brought back to shore for evaluation. That work is actually done there and then by the AI, which can operate on the ship or soon on the underwater robot itself.
“Without artificial intelligence, autonomous underwater robots are pretty dumb – they can only follow pipelines and cables in preprogrammed lines,” says Allen.
“Allowing the AUV to analyze the data in real time means that you can actually train the robot to do other things. If you encounter a problem, the survey can be stopped and more data collected, with the robot making the decisions for itself.” ha says.
So, for example, if the AI reports something that looks like an unexploded bomb, it could stop, go back and do further analysis.
For some jobs, such as the decommissioning of subsea oil and gas infrastructure, engineers need to know the exact dimensions and locations of the equipment.
To help with this, Rovco has also developed a vision system that produces accurate maps of the underwater infrastructure.
The system generates a 3D cloud of individual data points, a format used in modeling software such as CAD. Combine these points with camera images to generate a realistic 3D reconstruction.
Rovco is currently bringing together the vision system, the AI and the underwater vehicle into one package.
Other companies are also racing to introduce AI to the industry.
Jake Tompkins is the managing director of UK-based Modus, which owns a fleet of 12 unmanned underwater vehicles. It is about to begin a two-year program with Durham University to develop an artificial intelligence control system that would allow some of its underwater vehicles to recognize their location, objects and anomalies during an investigation.
He says the combination with Durham is a very efficient way to develop the technology, because they already have proven AI systems for the automotive and aerospace industries.
Using autonomous submarine robots to survey the sea floor and inspect undersea structures would be a “game changer”, according to Tompkins, and should “significantly” reduce costs.
He thinks it won’t be long before submarine robots will be stationed at sea, perhaps in an offshore wind farm or an oil or gas plant.
When needed, they will be woken up and sent to collect data, which will be sent back to a ground control center for processing.
“I think we’re probably about two years away from the first commercial deployment of autonomous field resident vehicles, but that’s certainly where we’re headed,” says Tompkins.
His company is currently working on ways to keep AUVs charged while at sea and on technology that allows them to return the data.
There is a juicy prize for companies that can make such smart underwater robots work. According to Søren Lassen, head of offshore wind research at consultancy Wood Mackenzie, the offshore wind market is expected to see “rather extraordinary” growth over the next decade.
Currently only six countries have a commercial-scale offshore wind industry. In 10 years, he expects 20 countries to join that club.
Last year 29 gigawatts of offshore wind capacity was connected to power grids around the world. In 2029 Wood Mackenzie predicts the number will reach 180 gigawatts.
This will involve building thousands of wind turbines and laying thousands of miles of cables to connect those wind farms, all of which will require the services of underwater surveyors.
By 2029, much of that work at sea could be done by autonomous systems, with humans in the office.
Jess Hanham will keep her business up-to-date with the latest technology, but fears the work will become less rewarding.
“I love variety. For me being stuck in the office – I hate it. Going out and polling, coming back and seeing everything from start to finish – gives you ownership of the job. I really like That. If we were to lose some of that, I think that would be a real shame “.