Autonomous Underwater Vehicles for deep-ocean and under ice surveys

The Autosub family of autonomous underwater vehicles, designed and built at the National Oceanography Centre, Southampton have an excellent track record in undertaking seabed surveys in challenging environments.

Autosub6000

With its maximum diving depth of 6000m, and a range of up to 300km offers an unrivalled specification. With its industry standard EM2000 swath bathymetry sounder, it delivers excellent quality, high spatial resolution mapping of the deep ocean.

We recognise that accurate geo-referenced navigation is essential if the digital terrain maps are to be of most use. To this end, Autosub6000 uses an inertial navigation system (INS) comprising a fibre-optic gyrocompass and a longrange bottom-tracking Doppler velocity log. The difficult task of updating the vehicle position once it has reached the seabed is done using a proven proprietary procedure that makes use of the high grade INS.

Launch and recovery is effective in sea states up to sea state 5, using a dedicated gantry system. This, together with the vehicle and support systems, is housed within two standard 20’ shipping containers.

EM2000 bathymetric map of the seabed at a depth of ~4200m near the Canary Islands, showing scour from turbidite flow, July 2008 An indication of the data quality is seen in these chevronshaped bedforms at a depth of ~4560m in the Horseshoe Abyssal Plain. The relief here is ~1m, suggesting cohesive sediments abraded by the large flows coming down the canyons.

Autosub3

Operated by Sea Systems within the National Marine Facilities Division is a resource available to the UK marine science community. Following from its predecessor, Autosub2, it has obtained stunning sonar imagery of the under side of ice in the polar regions. As exploiting hydrocarbon resources in currently ice-covered regions becomes economically and technically feasible, autonomous underwater vehicles can contribute in several areas. In survey operations, for example, AUVs can extend the operating season for geophysical investigations of the seabed.

EM2000 bathymetric map of the margin of the Courtauld Glacier margin, Greenland, from Autosub2. (From Dowdeswell et al., Proc. Workshop on AUV science in extreme environments, SUT, 2008).

Obtaining information on sea ice scour is important in preliminary studies for pipeline routes subject to seasonal ice cover. Upwardlooking bathymetric sonar on AUVs provides a wealth of quantitative information on ice draft. In principle, such surveys could be carried out throughout the year.
A swath sonar image of the underside of first year and multiyear ice off Greenland from Autosub2. The areas in red bottom right first year ice, cut through with a ridge (from yellow to blue), indicating ice keels to 25+m. (From Wadhams et al., Geophysical Research Letters, Vol. 33, L04501, February 2006).
Navigating with change within a sea bounded above by sea ice, below by the seabed, with the likelihood of deep ice keels in contact with the seabed, and with the possibility of icebergs moving through the ice is a great challenge. Autosub3 is one of the very few AUVs with a sufficiently advanced collision avoidance system to ensure mission completion in such complex environments.

Proven in the Arctic and Antarctic, control algorithms and software autonomously replan the mission to navigate around obstacles whilst maintaining headroom and height off bottom.

Risk Management for Autonomous Underwater Vehicles

The offshore hydrocarbon industry is making increasing use of autonomous underwater vehicles in several areas of operation. Some of the companies that provide AUV services to the industry have developed procedures to evaluate the reliability of the vehicles they use. At the National Oceanography Centre, Southampton, the reliability of our AUVs is important to us - we send them into situations where, if something goes wrong, recovery will not be possible - under Antarctic Ice Shelves, for example.

Consequently we have developed a Risk Management Process for AUVs that is based on targeted reliability. Furthermore, we have developed extensions to existing analysis methods that allow us to assess the risk of loss for an AUV undertaking a mission in a defined environment, under sea ice for example.

To date we have studied the operational risk and reliability of Autosub3 and REMUS 100 AUVs, and offer a service tailored to meet client needs. To inform this process we use the technique of eliciting expert judgment. Here, we gather together the fault and incident history of a vehicle, assemble a team of international experts in AUV technology and operations, and elicit from them, in a structured way, their judgements on the probability of loss given each fault or incident.

Using statistical techniques we apply these expert judgments to the mission scenarios. This presents the responsible owner with an informed opinion on the probability of loss of an expensive asset on a particular set of missions. As the offshore industry seeks to use AUVs in harsh environments, unforgiving of faults, such as under Arctic ice, techniques such as the NOCS Risk Management Process are essential.

We recognise that it is not feasible to seek expert judgments for each and every environment the AUV will encounter. Instead, we have developed an approach based on Bayesian Belief Networks that allows us to update the initial judgments of experts with sets of relevant data on environmental and other conditions that will affect the risk. For under sea ice for example, our procedure takes in information on ice thickness, extent and the characteristics of the support vessel. More complex networks can be built to suit the application.

Consultancy in the Technology and Applications of Autonomous Underwater Vehicles

As designers, builders and operators of advanced autonomous underwater vehicles we are well placed to offer consultancy on all aspects of the technology and applications of AUVs. We can also draw upon experience with commercial underwater gliders and ROV technology where appropriate. Recent clients for our services have been from the hydrocarbons and defence sectors.

In particular we offer:

Technology intelligence – tailored assessments based on our wide knowledge of and contacts with the international community to answer questions such as: What are leading research and development groups working on in the area of AUVs? How effective are emerging technologies in areas such as AUV navigation, energy systems, communications? What has been the track record of using AUVs in particular geographical contexts?

AUV Campaign Design – you have an option to use AUVs for a set of tasks, perhaps it would be the first time such a set of tasks was asked of an AUV. What would an AUV campaign to deliver the results look like? What missions would be feasible? What equipment would you need to fit to the AUV? How would this affect the performance of the vehicle? What would be a sensible schedule? How should expectations be managed? Who are credible service providers?

Use of underwater gliders – with a fleet of four Teledyne Webb Research Slocum electric 1000m gliders NOCS’ experience dates back to 2006. Having operated these AUVs on missions totalling over 15000km we have a wealth of knowledge on how these vehicles can be used effectively for ocean data gathering.

Today, gliders can provide realtime environmental data in support of offshore operations, for example on the physical structure of the water column and on currents. When equipped with sensors in development, they will be able to function as long endurance sentinels (up to one year), making measurements of dissolved hydrocarbons, for example.

We can assist you in assessing whether underwater gliders might be a solution to your marine environmental data gathering needs. We can advise on additional sensors, and on implications for endurance and reliability. Careful experiment design is essential with gliders, hence we have developed a technique of ‘flying’ virtual gliders in an ocean model to test scenarios before committing to operations at sea.