Machine Learning News

© Martin Valigursky / Adobe Stock

NASA Asks Gamers, Citizen Scientists to Help Map World's Corals

.As they play the game, players' actions help train NASA's Pleiades supercomputer at Ames to recognize corals from any image of the ocean floor, even those taken with less powerful instruments. The supercomputer "learns" from the coral classifications players make by hand, using machine learning techniques to classify on its own.The more people who play NeMO-NET -- available on the Apple App store and is playable on iOS devices and Mac computers, with a forthcoming release for Android systems -- the better the supercomputer's mapping abilities become. Once it has been able to accurately

Armed and ready: an Ecotone Underwater Hyperspectral Imager, or UHI, attached to an ROV (top) and stand-alone. Photo: Courtesy Ecotone

Pipeline IMR: An All-seeing Eye

regular cameras and lights or UAVs with available sensor arrays. By 2016, the UHI had evolved for pipeline IMR into SpectraPipe with analysis and archiving software understood to be faster and safer than x-ray or acoustic methods of pipeline maintenance surveying.“It’s all automated based on machine-learning principles,” says Ecotone CEO, Oddbjoern Roedsten, adding that tweaks to the equipment to 2016 have led to today’s first commercial uses. “There has been development on the data analysis side with a higher degree of automation to cut down on the time and work involved

L3 Harris UK’s C-Worker 7 working with an ROV off the UK’s south coast. Photos from L3 Harris UK.

Marine Autonomy Above & Below the Water

is straight above the ROV making it hard to track. But, these are learnings, he adds.“At this stage, we just wanted to see it work, the next stage collaborative autonomy between vehicle and the subsurface vehicle,” he says. “So, the next step is to look at increasing automation and machine learning to generate paths to get back into its enclosure, way points, etc.”The ability to work in sea states above three will also be needed, as well as a larger ROV, with a Tiger, also from Saab Seaeye, being eyed for the next trials, as part of Phase 2 of the project, scheduled for this year

L3 Harris UK’s C-Worker 7 working with an ROV off the UK’s south coast. Photos from L3 Harris UK.

Hybrid Autonomous Systems Evolve

7 working with an ROV off the UK’s south coast. Photos from L3 Harris UK.“At this stage, we just wanted to see it work, the next stage collaborative autonomy between vehicle and the subsurface vehicle,” he says. “So, the next step is to look at increasing automation and machine learning to generate paths to get back into its enclosure, way points, etc.”The ability to work in sea states above three will also be needed, as well as a larger ROV, with a Tiger, also from Saab Seaeye, being eyed for the next trials, as part of Phase 2 of the project, scheduled for this year

Kraken President & CEO, Karl Kenny (Photo: Eric Haun)

Kraken Finalizes OceanVision Contract

develop new technologies and provide high resolution seafloor imaging and mapping covering more than 5,000 square kilometers around Atlantic Canada. Kraken and its partners will leverage key enabling technologies (unmanned maritime vehicles, advanced sensors, robotics, autonomous systems, big data machine learning and predictive analytics) to deploy innovative technology platforms across ocean sectors and extend the global reach and market opportunities for Kraken and our partners.Karl Kenny, Kraken's President & CEO said, "We are excited to finalize the contracting stage for our OceanVision

(Image: Kraken)

Kraken Wins Canadian Government Contract

Canada’s discovery of the HMS Erebus during the Franklin Expedition in Summer 2014.The SeaVision system delivers real-time, full color 3D point cloud images of subsea infrastructure with millimeter accuracy. These datasets create highly detailed models for 3D visualization to support complex machine learning and artificial intelligence algorithms used for analytics. SeaVision simplifies creating dynamic digital models of subsea physical assets and systems. This helps build and operate digital twin applications to gain knowledge and insights about performance of future operations.Karl Kenny, Kraken

NotiloPlus’ Seasam AUV has been operating around the world. Photo from NotiloPlus.

Subsea Tech's 'March of Miniaturization'

in at 9kg and is the missing link between a small and compact but piloted ROV and bigger ROVs packed with kit that cost a lot. Nicolas Gambini, the Marseille-based company’s co-founder and CEO, says NotiloPlus is enabling a small vehicle with a lot of smarts through computing power. “Machine learning and software allows you to do positioning and sensing,” he says. “For positioning, you need to use lots of hardware, but we wanted this to be as cheap as possible. So, for less than the price of one high-end positioning sensor we have a system that uses cheaper sensors but then fuses

Technology Outlook 2030 (stills from video). Image: DNV GL

10 Tech Trends to Impact Shipping Industry

and transmission will increasingly become the domain of algorithms rather than people.The next ten years will also be defined as an era of precision, particularly in healthcare, food, and materials science.  Agriculture and aquaculture will look toward robotics in combination with IoT-based and machine learning technologies to optimize crops and seafood at reduced cost.  Additionally, increasingly precise sensor data will help the food industry meet expectations from consumers for insight into the entire value chain for individual products.Traditional business models will be disrupted. Space,

(Image: FutureOn)

Subsea Digital Alliance Launched

workflows using Orcaflex integrated with FieldTwin.WINS - Operations. Delivering wireless systems for operators to retrofit live downhole data to enable a true digitalization of the field for optimized production and recovery.Digital Energy - All phases – A comprehensive suite of AI and machine learning algorithms that can be integrated with any data set to support predictive analytics and process improvements.FutureOn will also present the Subsea Digital Alliance in partnership with Mcdermott and Digital Energy on the Microsoft Azure booth at ADIPEC

Photo: EOMAP

EOMAP Wins Funding for Next-Gen Satellite-Derived Bathymetry

technologies. Their big benefits are their higher speeds and lower energy consumption compared to normal processors. However, the optimal interaction of FPGA and CPU processing requires deep knowledge of scientific programming.“By combining FPGAs, physics-based data analytics and modern machine learning procedures we are confident we can slash the processing time for SDB by 95%,” said Dr Thomas Heege, CEO of EOMAP and one of the early inventors of SDB technologies. “We believe that this R&D activity will make possible higher-quality data processing and continuous updating, and

Marine Technology Magazine Cover Mar 2020 -

Marine Technology Reporter is the world's largest audited subsea industry publication serving the offshore energy, subsea defense and scientific communities.

Subscribe
Marine Technology ENews subscription

Marine Technology ENews is the subsea industry's largest circulation and most authoritative ENews Service, delivered to your Email three times per week

Subscribe for MTR E-news