Winning the subsea space race
Image credit: Oceaneering
The race is on to introduce advanced robotics and autonomy into subsea oil and gas production operations.
Robotics and autonomy may be key for the future of subsea oil and gas production operations. They will play an important role in the drive to reduce the costs and carbon emissions related to production of oil and gas, as well as to reduce the number of people who have to work offshore, on ships or on platforms.
It’s a unique challenge. As well as dealing with the obvious physical limitations of operating robotic systems in the marine environment, there are two main challenges: navigation and communication. There’s no GPS or 4G network subsea with which systems can navigate and use to communicate. But many think that surmounting these issues is worth the effort.
Rune Aase, vice president at Norwegian energy firm Equinor, which promotes use of underwater drones, says: “This technology is an enabler. It’s an enabler for unlocking new ways of working, transforming the way we’re working and, not least, reducing CO2 footprint and increasing competitiveness on the Norwegian Continental Shelf and internationally. With drones, we can move more of the task onshore, move people onshore closer to their homes.”
While offshore oil and gas production is often associated with oil platforms standing out in the sea, there are also thousands of wellheads and manifolds on the seabed, covered in actuators that control production from reservoirs under the seabed. These are connected by hundreds of thousands of pipelines and umbilicals and it all needs inspecting and sometimes intervention – and much of it is out of reach of divers.
Today, vessels, complete with crew, are sent offshore. They deploy remotely operated vehicles (ROV), which are connected to the vessel by an umbilical that provides power, communications and control of complex manipulators and other tooling. It’s an unwieldy package.
The idea is to replace this package with a so-called resident ROV or autonomous underwater vehicle (AUV), that’s at least part-autonomous, part remote-controlled from shore, and that lives at the seabed in its own garage – free of the need for a support vessel. “Basin-wide, companies [could] invest in resident robots that everyone can then pull up on an app and see which are available,” Chevron’s John Brian suggested to an industry event in Aberdeen last year.
While an Uber AUV service is still some way off, significant steps have been made. Since early 2018, a more traditional ROV has been used on three-month-long support missions underneath Equinor’s Snorre platform offshore Norway. When it’s needed, pilots offshore and in an onshore control room can remotely control it. It’s still connected to an umbilical, but instead of a being deployed from a vessel, it’s connected to a subsea garage that, in turn, is connected to the platform, providing power, communications and control. This meant making the Merlin UCV R-ROV (‘R’ for resident) operated by IKM Subsea, a Norwegian company, as reliable as possible and enabling remote control, says Ments Tore Møller, IKM Subsea’s engineering manager. These vehicles are usually serviced weekly – that had to change to every three months.
What if there’s no nearby platform? US firm Oceaneering has been covering that base. In 2019 it’s been deploying its E-ROV, another cabled electric ROV system, which also works out of a subsea garage, on temporary missions. The garage, which comes with batteries for power, is dropped off where it’s needed then left for weeks at a time. For communications and control, it comes with a surface buoy, which provides a communications link to shore using the 4G networks that are now common in most busy offshore oilfield areas, including the North Sea and US Gulf of Mexico, or Iridium satellite communications where 4G isn’t available.
The next step was taken last year when tetherless vehicle control and docking was demonstrated. A Sabertooth AUV, built by Saab Seaeye, part of Sweden’s Saab group, wirelessly docked, charged and downloaded data, all inductively, with remote part-automated control and live visual control during a demonstration in a lake in Sweden. Oceaneering did similar with a prototype of its new hybrid vehicle Freedom during a demonstration event in Norway later in the year.
These feats were part-driven by Norwegian energy firm Equinor, which has been pushing for what it calls underwater intervention drones, or UiDs. Equinor wants to see many of these vehicles on the seabed so that they’re available whenever needed – like the Uber AUV vision. The idea is that one could cover a number of fields, each of which could be operated by a different company or business unit (Equinor’s fields are operated by different business units).
“Today we’re doing IMR [inspection, maintenance and repair] on subsea wells by an IMR vessel with ROVs with 70-80 people on board,” says Aase. “The benefit of drones is that we can move more technology onshore and move people onshore to a safer workplace close to home.” But, he says: “We haven’t had the option to just wait for industry and see what’s showing up. We have been discussing and calibrating with industry and trying to be the front runner in this area. We think that’s important, not just to use what’s in market but set up an operating model across assets [that] can standardise and have economy of scale and skill so need to set some standards and requirements.”
To that end, Equinor commissioned an ‘open-standard’ standardised docking station (SDS) to be designed and built. This was used by Saab Seaeye and Oceaneering for their demonstrations. Key elements of the SDS are a number of other technologies, some of which are more established than others. For long-distance homing in, vehicles can use acoustic signals, which travel out to multiple kilometres. Closer in, they can use machine-vision technology to detect so-called AruCo and ChaRuCo markers (similar to QR codes). They’re also using BlueComm, a through-water modem, which uses light to wirelessly transmit live high-definition video, so pilots can steer the vehicle towards its target, or high-bandwidth data, provided by UK firm Sonardyne. Once docked, they’re using inductive power and communications connectors, variants of which are produced by Norwegian firms WiSub and Blue Logic, to recharge these vehicles.
‘This technology is an enabler. It’s an enabler for unlocking new ways of working, transforming the way we’re working and, not least, reducing CO2 footprint and increasing competitiveness.’
For Jan Siesjö, Saab Seaeye’s chief engineer, a key enabler was surface communications – the 4G network. Then, bringing together docking, charging and data download was the final piece in the puzzle to bring this capability into the field. “With that, all of a sudden you can have a docking station with a communications buoy or a USV supporting an AUV with communication. All of the pieces are there.”
There’s also been a lot of work behind the scenes. “Remote control over long distances might seem simple but to make it reliable you need a lot of stuff in place,” says Siesjö. “It’s not just sending commands over the internet, it’s having systems that can keep themselves safe, can be maintained so they don’t go wrong and if something does go wrong it’s not so complex you need a university-grade engineer to fix it.” That includes station keeping, waypoint navigation and obstacle avoidance.
On the communications side, Saab Seaeye worked with Boeing to prove the ability to run an ROV and its manipulators safely over a satellite link. “We had some pretty strict limitations, just 1.6MB/s, and a latency that we pushed up to three seconds. We also intentionally messed with the data quality,” Siesjö adds. “Despite that, we were able to mate flying lead connectors and do a lot of other things.”
Another challenge is navigation. Underwater vehicles use inertial navigation systems (INS), which contain inertial measurement units, complete with gyroscopes, accelerometers, pressure sensor, etc, and a computer uses these and Doppler velocity logs to calculate the vehicle’s movement. But, these sensors do drift and the vehicles are operating in a dynamic environment, so external aiding from ultra-short baseline (USBL) positioning systems can also be used, but that usually means using a ship.
So, systems that build, in real-time, 3D maps of the environment the ROV is in, i.e. 3D simultaneous localisation and mapping (SLAM), are also being developed. This will let the vehicle navigate through and measure what it sees relative to itself, as well as to maps or plans of structures it may already have been given. Subsea contractor Subsea 7, through its i-Tech 7 business, has been developing technology in this space. It’s been building relocalisation capability into its autonomous inspection vehicle (AIV), which is deployed from a subsea basket, so that the vehicle doesn’t need regular position updates from a USBL system. “The development was triggered by a decision to move to autonomous hovering vehicles, focusing the capability at infield subsea infrastructure inspection,” says Jim Jamieson, strategy & technology development manager at i-Tech 7. “The equipment itself could be tracked to provide the positions needed to navigate without updates from a surface vessel.”
Herriot-Watt University technology spin-out SeeByte has been working with Subsea 7 and others in this area. Chris Haworth, the company’s commercial manager, says, due to the challenges around navigation and communication underwater, these systems need adaptive autonomy, where the AUV monitors its surroundings with a range of sensors (video, sonar, navigation, laser, for example), interpreting the incoming information, matching this against mission goals and making real-time decisions about the best course.
Advances in artificial intelligence (AI) are helping here, improving machine-learning algorithms. Deep learning, which uses extensive data sets, tagged with features that are fed into multi-layer (deep) neural networks, is an example of this, says Haworth. A challenge of this approach is the amount of data needed to feed the algorithms; data sets from subsea operations are not necessarily that easy to get hold of. To address this, SeeByte is using a smaller amount of data to train a Deep Fake system. “The Deep Fake technology is then used to produce a much larger data set, allowing the Deep Learning target system to be trained on a mix of both the real and fake data,” says Gavin Irvine, senior project lead at SeeByte.
It’s not just about the vehicle and its behaviour. Infrastructure needs to be in place. This includes subsea power distribution, which could include renewable energy sources, which some in the UK are working on, or even subsea replaceable fuel cells, proposed by US technology firm Teledyne, so that there’s power available for recharging. Subsea infrastructure also needs to be designed to be vehicle-friendly.
“For this to be economical, you need to increase the scope of work of drones,” says Helge Sverre Eide, business manager at Blue Logic. “You need to change both sides of the equation.” This also means new tooling – most tooling today is hydraulic; electric systems are needed and are now being developed. Subsea maintenance also needs to be about having smaller pieces to change out, like modern cars. “It’s a different mindset – a new philosophy,” he adds.
Then, thinking ahead to when different vendor vehicles can dock in different docking stations, how will the way they connect into different data and control networks be managed? asks Jan Christian Torvestad from Equinor. “If I have cell phone with a Norwegian subscription I can still travel to America and use it, even if the service provider is not there – we have agreements. With a standardised docking station, I can get power and communication and then a service provider making sure that a drone docking on an Equinor docking station is connected to the correct control room,” he says. “The service IT and architecture in the background needs to be considered. It’s part of the puzzle.”
How do you then prioritise bandwidth and ensure data security and how does this work commercially? Answer these questions and “then you can have the Uber AUV and high utilisation because you can switch from one mission to another and between companies”, says Torvestad.
This year, the first commercial contract for use of a resident subsea robot is due to come into operation. Italian company Saipem has been contracted to deploy a vehicle on the seabed at Equinor’s Njord field, off Norway, where it will be available for use by operators based onshore.
Success here could result in significant changes in how subsea operations are done. “We are on the edge of some really big changes in our industry,” said Stephen Gray, CEO at UK-based ROV services firm ROVOP, during a Subsea UK Underwater Robotics conference in Aberdeen. Gray suggests that the change will reflect a change that’s already happened in other industries, such as telecommunications (think mobile phones and what happened to Nokia when Apple came out of nowhere).
“It’s not the technology development that’s lagging, it’s the commercial development to exploit it. The struggle is finding commercial value,” says Steffan Lindsø, director of emerging technology, Europe, at Oceaneering. This year could well be a decider for this technology.
Sign up to the E&T News e-mail to get great stories like this delivered to your inbox every day.