Artificial intelligence and machine learning show promise for materials-by-design approach
ABERDEEN PROVING GROUND, Md. -- The U.S. Army signed a new cooperative agreement with Johns Hopkins University Sept. 30, 2020, to advance materials research using artificial intelligence and machine learning.
Researchers from the U.S. Army Combat Capabilities Development Command (DEVCOM) Army Research Laboratory will collaborate with Johns Hopkins faculty and students on four focused projects:
- Using artificial intelligence to accelerate the iterative materials design cycle by high-throughput microstructural characterization and rapid processing
- Acoustic signature and reconstruction of defect avalanches in metals
- Real-time monitoring of laser-material interactions
- Toward self-repairing devices: Data-directed design of active, hierarchical colloidal assembly and reconfiguration
Dr. Sikhanda Satapathy, from DEVCOM ARL, and Prof. K.T. Ramesh, director of the Hopkins Extreme Materials Institute, will lead the research activities. To launch these projects, the partners held a joint virtual kickoff meeting Nov. 4.
“This collaborative agreement will enable and accelerate intelligent design of materials for extreme dynamic environments to support our Soldiers and address Army’s future material needs,” Satapathy said.
Over the next two years, researchers will explore the use of artificial intelligence and machine learning to accelerate materials development. One of the projects is focused using artificial intelligence techniques to accelerate the processing and characterization of new materials.
“This has been a traditional bottleneck in the process,” said Prof. Mark Foster, Johns Hopkins.
Foster’s team will employ thermo-mechanical processing and diffractive imaging to create and characterize new materials. The experimentally and computationally generated data will be used to train neural networks which will be used to accelerate the materials design process.
Another project will incorporate machine learning of acoustic emission measurements to characterize materials deformation mechanisms. These measurements are non-trivial and require expertise in both instrumentation and data analysis.
“This collaboration will benefit both current and future ARL programs,” said Dr. Daniel Magagnosc, DEVCOM ARL scientist. “For instance, AE characterization of ductile fracture mechanisms in metals would aid in the development of new models for metal deformation, which incorporates microstructural information, by discerning the role of various features of the microstructure.”
The proposed signal analysis algorithms could also assist in the identification on active deformation mechanisms in ARL-developed advanced high strength steels, he said.
“Newly developed expertise AE is expected to benefit the study of brittle fracture in hard ceramic systems by providing additional insight into the early stages of fracture,” Magagnosc said. “All of these benefits will provide ARL the scientific understanding to improve protection materials for military armor applications.”
In the real-time monitoring of laser-material interactions project, DEVCOM ARL scientist Dr. Debjoy Mallick will use the novel diagnostics to directly support improved additively manufacturing techniques. The machine-learning guided techniques to minimize defects in production parts have the potential to improve part reliability/durability and reduce the need for excessive quality control.
Mallick’s research on shockwave-material interaction will also benefit from the multi-disciplinary scope of all projects. Understanding the equation-of-state, or the interplay of temperature, pressure, density, and volume in a material at the extremes (like when experiencing a blast wave or an impact), requires diagnostics with an especially granular resolution in time, while understanding shock-induced heating of an armor or deflagration and initiation of an explosive/propellant requires granular resolution in temperature and velocity.
According to Satapathy, the projects have the potential to improve the state-of-the-art of time-resolved detection in these types of experiments. Mallick is also exploring the ability to generate shockwaves in armors and energetics by both direct laser ablation and the laser-driven acceleration of micro-projectiles.
These projects will accelerate material development for Army’s emerging needs by integrating artificial intelligence and machine learning into materials science, Satapathy said. Advances in this area directly supports ARL’s mission of disruptive foundational research which will lead to new material solutions for Army applications.
DEVCOM Army Research Laboratory is an element of the U.S. Army Combat Capabilities Development Command. As the Army’s corporate research laboratory, ARL is operationalizing science to achieve transformational overmatch. Through collaboration across the command’s core technical competencies, DEVCOM leads in the discovery, development and delivery of the technology-based capabilities required to make Soldiers more successful at winning the nation’s wars and come home safely. DEVCOM is a major subordinate command of the Army Futures Command.