Catalog product

Navy Surface Fleet AI Task Force Develops Comprehensive Data Catalog

Written by Brandi Vincent

A Navy task force formed to operationalize artificial intelligence and machine learning across the surface fleet is leading efforts to correct data and lay the groundwork for associated emerging technology applications in the near future.

Effective AI depends on “clean” or sufficiently consistent data to create algorithms. Naval Surface Force components, which outfit and equip warships prior to deployment to respective fleet commands for military operations, create and use heaps of data – but currently everything is quite complicated from a point of view. organizational view.

“Our data landscape is so vast and complex. There is no common data ecosystem, no data catalog and not enough clean data,” Capt. Pete Kim, director of Task Force Hopper, told FedScoop in a July 29 interview.

Kim has led the task force since its launch last summer to pilot AI capabilities through surface force. He and his team made progress in shaping a nascent approach, strategy, and implementation plan to guide this large-scale effort. While these documents are being prepared for publication, the task force is also working to design and refine a digital and conceptual hub that makes sense of organizations’ multitudes of data and helps staff better analyze and apply it. for AI and ML.

“As you can imagine, it’s quite difficult to set up this infrastructure,” Kim said. “I think it’s because of the nature of our different classifications, roles and security environments. It’s not as simple as, say, getting an app on your iPhone and updating quickly.

“Crack this code”

Kim now leads both the Surface Analytics Group (SAG) and Task Force Hopper. The experience was revealing.

Capt. Pete Kim, then commanding officer of the Ticonderoga-class guided-missile cruiser USS Princeton (CG 59), uses the ship’s general announcement system to speak to the crew, Aug. 14, 2020. (U.S. Navy photo by Mass Communication Specialist 2nd Class Logan C. Kellums)

“I’ve always been in the operational fleet – so the one that provides the data – and I didn’t realize how much data we have in the Navy that is not being used,” he said. Explain. “I think we tend to look for the data we need to respond to mail and things like that, but until the last few years I don’t think we’ve had the ability to really deal with big data. . And we do now. So that’s probably the coolest part.

Yet there is still a long way to go before the service’s ambitious goals of applying AI and ML at scale are fully realized.

One of Task Force Hopper’s first priorities is to help the surface force identify and cleanse the data, so that the various parts of the sprawling enterprise can collectively take full advantage of it. To help demonstrate the vast breadth of information, Kim noted that the SAG focuses on readiness-related data.

“Spectrum contains data from authoritative databases, where it’s very structured data and we extract all the unique datasets we need – until it’s on someone’s desk or on a drive shared hidden somewhere you have got to find the right person or the right manager to access that data,” he said.

In addition to data availability, quality and governance challenges, Kim noted that technology-driven work in the Navy has traditionally been organized and structured around platforms and supporting program offices. But AI development cuts across many different silos and organizations.

“That’s why this federated model is so essential to cracking this code,” he explained.

This nascent approach he alluded to was recently conceptualized by the task force and will be detailed in a soon-to-be-released data and AI strategy and implementation plan. The general idea is to have more centralized data governance and a one-stop-shop data catalog, combined with “decentralized analytics and AI development nodes in different parts of the business”, where staff knows and uses data best.

Each node will focus on certain categories associated with artificial intelligence and machine learning, such as maintenance or lethality. The SAG, for example, is considered a readiness-focused AI node.

“I think every node is going to be a little bit different, and it really depends on the problem, the use case. And then, again, what’s the state of the data?” Kim said.

If nodes have mature, high-quality datasets, they will likely develop AI models fairly quickly. But if they’re starting near or from square one, they’ll likely have to spend more time on data collection, cleaning, and tagging in the early part of the journey.

“I know data management isn’t the sexiest topic, but we think it’s one of the significant steps towards accelerating AI and ML in a large organization” , added Kim.

Enter a new era

Task Force Hopper is named for computer pioneer and naval officer Grace Hopper, who rose to the rank of rear admiral (lower half) before her retirement.

A group of key players in AI and data across the surface force — one of the largest businesses in the Navy — has met every two weeks for the past year. Kim said they started “this data governance process” and identified many datasets for their respective areas to prioritize.

Developing clearly defined use cases for the many surface force data sources is also currently a priority for Task Force Hopper.

“When it comes to analytics and AI, we’re sort of entering a new era where the operator, fighter, or maintainer has to be involved at every stage of development,” Kim said. “I think it’s a change from the past where we just give requirements to a contractor and then they come back in two years with the product.”

In his view, the task force and SAG find success by “having the right subject matter experts sitting side-by-side with data scientists, with AI model developers to produce truly valuable products.”

Task Force Hopper has also made progress working with the office of the Navy’s director of information, according to Kim, to apply a Platform called Advana-Jupiter as a common development environment.

“It has data warehousing tools and all the apps you need to visualize data and build AI models,” he explained. “We’re using this platform as a place to have a single catalog so that if people are working on a project and they’re looking for certain datasets to move forward, they’re not stuck because they don’t can’t find it or it’s unavailable.”

As a scalable part of Advana, the Pentagon’s largest enterprise data center, Jupiter will allow surface force members to seamlessly access data, then build AI algorithms and ML informed by them.

“On the preparedness side, we are turning to predictive and prescriptive maintenance to support our vessels and increase reliability at sea,” Kim said.

Another focus area of ​​the Readiness Node is condition-based maintenance. “As we begin to use unmanned surface vehicles, we will need these types of CBM models to support these ships at sea, as they will not have maintenance personnel on board,” Kim noted.

He added that while Jupiter doesn’t need to host every dataset, “that’s where we want to catalog it so that if someone is working on a project, it’s like a menu” where they can see the contact point and data details.

“We’re going to use Advana-Jupiter as a platform where we can kind of integrate different datasets, because as we start to build more advanced AI models, it’s not just going to be about… ‘a sensor data source, it’s going to be multiple things,’” Kim said.

One of the main goals of the task force is to help the surface force prepare for AI by 2025.

“I think with new technologies, you always feel like you’re behind. That’s why we put so much intelligence behind it. But as you know, having this high quality dataset, the tools, the right people for the project — I mean it’s like 80% of the journey. So if we get that part of the infrastructure right, the last 10% of producing that widget or what have you is the easy part,” Kim said. “And we can really partner with industry to take full advantage of existing technology and develop these unique tools that we need.”

-In this story-

Advana, artificial intelligence (AI), data, data analytics, emerging technology, Jupiter, machine learning, Naval Surface Force, Pete Kim, Surface Analytics Group, Task Force Hopper