ABERDEEN PROVING GROUND, Md. -- Researchers from the U.S. Army Research, Development, and Engineering Command Research Laboratory, the Army's corporate research laboratory, recently partnered with Texas A&M University to work on artificial intelligence and machine learning as applied to material informatics (and genome).
1st Lt. Levi McClenny, a doctoral candidate in the university's Department of Electrical and Computer Engineering and an active member of the U.S. Army Reserve serving as a platoon leader and Black Hawk helicopter pilot in an aviation battalion in Conroe, Texas, recently completed a two-week internship at the lab's Vehicle Technology Directorate at Aberdeen Proving Ground.
At Texas A&M University, McClenny and his adviser Dr. Ulisses Braga-Neto support the development of an AI agent to determine the internal state of various materials and systems using microscopic images and deep machine learning techniques.
Researchers want to understand how materials fracture and break so they can potentially predict when a component will break in an aircraft, for instance, to help with maintenance and operational requirements. The idea is to engineer vehicles that can begin to detect their own deterioration.
"We are applying machine learning techniques to better understand what is happening at the microstructure level in materials," said Dr. Mulugeta Haile, research aerospace engineer at VTD. "We want to have a complete understanding of how materials behave during normal usage or in extreme conditions from the day they are put there until they are removed."
McClenny said coming to the Army's corporate research laboratory and working in its facilities allowed him to interact with some brilliant and experienced materials scientists that can not only shed some light on the work he's done, but also pave a way forward.
"The new AI lab is absolutely incredible," McClenny said. "I was able to use the supercomputer facilities to generate products that I will be taking back to Texas A&M with me for future projects that would not be possible without the facilities Dr. Haile and Mr. Ed Zhu put together."
According to Haile, the new AI/ML lab was conceived to facilitate research in artificial intelligence and machine learning to focus on vehicle technology and maneuver sciences. The lab, not only hosts state-of-the-art GPU accelerated high performance computing resources, it makes these resources highly available and easily configurable to users in an open and collaborative space.
"I was able to get these products, as well as develop a plan of action for the microstructure research in the two weeks I was here," McClenny said. "I was also able to sit down with numerous researchers from the VTD to see their data and see how we could apply machine learning approaches to learn more from it. We always say that models are only as good as the data, and here we can generate some top-notch data."
The directorate was pleased to host McClenny and found his mix of skills to add to the overall research.
"As a PhD student and an Army Black Hawk pilot, Levi brings to the research environment a unique mix of skills and understanding," said Dr. Jaret Riddick, director of VTD. "The unique mix of scientist and end user gives Levi a perspective that can be key to enabling the Army Futures Command's objective of incorporating warfighter feedback into advancing science and technology for the modernization process."
McClenny said working at the Army's corporate research laboratory was an incredible experience and absolutely surpassed his expectations. He also said being a member of the military and a researcher offered some unique perspective.
"Throughout all the conversations and ideas, I have tried to remember the 'why' for these projects," he said. "This is important to me, potentially more so than the average researcher, because I can directly impact the Soldiers in my own unit, and future units, with this work. The facilities and expertise offered at this facility, not only by Dr. Mulugeta Haile, my mentor, but others in the group like Dr. Dan Cole and Dr. John Chen, really helped to expand my understanding of why we are researching the topics we are."