NUWC Keyport exploring, developing AI technologies to help Navy operations
KEYPORT — No longer just a buzzword, Artificial Intelligence has become a hot topic between the Navy and its civilian partners across the Puget Sound region and beyond.
The Navy has been exploring artificial intelligence for quite a while, but not until October did the Keyport Division — the Naval Undersea Warfare Center Division in Kitsap which has engineers and scientists to develop advanced technologies to maintain the country's undersea warfare systems — create a Digital Transformation Division to consolidate its research and development resources on several kernels of AI, like machine learning, data science, and Large Language Models.
"The creation of the Digital Transformation Division allowed us to better organize and provide a clear strategic leadership to those teams, so they can move together under a similar path," Jeremy Landis, NUWC Keyport’s Digital Transformation Division Technology Manager said.
The Kitsap Sun interviewed Landis to discuss Keyport's research and implementation of Artificial Intelligence.
Keyport's Use of AI: business analysis, obsolescence management, and more
NUWC Keyport is implementing AI machine-learning technologies. The formation of Digital Transformation Division brings various technical teams together so that Keyport can delve deeper into those topics and continue to increase its rate of innovation in the right way, Landis said.
Current Keyport projects that incorporate elements of artificial intelligence, machine learning, and data science include the Obsolescence Management Information System (OMIS) and Data Aggregation and Research Tool (DART), according to NUWC Keyport.
“OMIS uses applied machine learning to build a model using data to predict when parts used in Navy systems will become obsolete, or when the sources providing a part will change,” Landis said. “This knowledge is used to influence our decision-making process, allowing us to efficiently support our existing Navy systems while reducing obsolescence challenges in new systems.”
“DART is an innovative business intelligence tool that uses data science and predictive analytics across multiple Navy databases to combine information on configuration, schedules, inventory, demand, attributes, and the parts obsolescence,” Landis said. “The information is then aggregated into one user interface to easily identify proactive solutions to support a wide array of Navy systems.”
These systems, combined with others, provide essential support for fleet readiness and future sustainability, Landis said.
Overall, NUWC Keyport is building smarter intelligence into all the systems so that it can better support its customers, Landis said.
Without getting into details, Landis said Keyport has projects to invest in Large Language Model technologies and evaluate where these models can best apply to the fleet.
There are many ways that Large Language models and natural language processing can be implemented, and the division is developing use cases and working with different organizations "to find out the best ways in which we can bring forth those technologies in a safe and reliable manner," Landis said.
How to ensure safety and security when using AI?
Last fall, Landis attended a three-day AI workshop hosted by the Defense Advanced Research Projects Agency's (DARPA) AI Forward workshop, where Navy experts from the industry, academia, and the government brainstormed new research directions about using AI to address national security challenges, according to a Navy's press release in September. DARPA is the central research and development organization of the Department of Defense.
In the event, Landis collaborated with experts in various fields — AI theory, ethics of AI, knowledge representation and reasoning, computer science and nuclear physics, and more — to work on documents for potential future DARPA projects.
Landis said the workshop focused specifically on how to build trustworthiness and reliability into the systems so that the military can ensure safety and security both for those interacting with the systems and people who are involved indirectly.
"Ultimately, when you're dealing with artificial intelligence and the information that it is providing you, you want to ensure that it is accurate, reliable. And part of the process is that we have our experts working with that data so they understand it," Landis said.
"As part of the software, you build in explainability, so that when the software developers are looking at the back end of the software, they can understand what the artificial intelligence is, how it's working in an explainable way so that we can ensure safety and security," Landis explained.
Eventually, the systems remain human-centered, Landis said.
Partnership with universities, private corporations across the nation
NUWC Keyport is working with its partners, including universities, government agencies, and private companies to develop AI together.
Through a Navy program called the Naval Engineering Education Consortium (NEEC), NUWC Keyport supports colleges and universities across the country with grants for research on a variety of topics related to AI and machine learning, including obsolescence forecasting and data analysis, maintenance sustainability, and intelligent maneuvering of autonomous unmanned vehicles, according to NUWC Keyport.
For example, NUWC Keyport has projects with the University of Washington Tacoma to work through specific items that the Navy passes to the university as related to obsolescence management and making AI-assisted decisions based on predictive analytics, Landis said.
"Those types of realms are the areas that we're working with universities," Landis said.
Keyport's current NEEC partnerships include the University of Washington, Washington State University, Embry Riddle Aeronautical University, Virginia Tech University and the South Dakota School of Mines.
The trend of developing AI may bring new job opportunities to Keyport, as the NEEC partnerships include potential job opportunities for engineering students involved in Navy-funded projects, the Navy said.
“There’s a push to embrace emerging technologies across the Department of Defense, and there may be opportunities for new careers at NUWC Division, Keyport that focus on these scientific innovations,” said Aaron Darnton, NUWC Keyport's Chief Technology Officer.
The Keyport Division also has the Northwest Tech Bridge to support the development of AI. The Tech Bridge builds partnerships with private companies or government agencies to enable greater private industry collaboration and the commercialization of cutting-edge technology, including artificial intelligence, the Navy said.
Some of Tech Bridge’s key partners include Amazon, Boeing, Microsoft, the Washington State Department of Commerce, the Department of Energy Pacific Northwest National Laboratory, Impact Washington, the Washington Autonomous Vehicle-Cluster, and the Pacific Northwest Defense Coalition.
Vision for the future
Landis highlights the work Keyport personnel are doing in developing trustworthy and reliable new technologies and their partnership with the other universities and the Tech Bridge.
"Keyport is made up of a lot of great individuals who have been members of the community, in some cases, (for) a long time," Landis said. "It's good for the community to know that we're working in a, not only innovative way but also in a trustworthy and reliable way as well."
When asked if NUWC Keyport will use more AI technologies to help with the Navy's operations in the future, Landis simply agreed.
"There are different types of AI, whether it's deep learning, machine learning, computer vision, or natural language processing. There's various ways in which the overarching technology can be implemented in the various projects," Landis said.
"The short answer to that is yes. I think how that's implemented across the command will vary depending on the project or the area in which that project supports."
This article originally appeared on Kitsap Sun: NUWC Keyport facility creates team to develop AI tech Navy may use
