Project Grant 2622123

The National Science Foundation Division of Computing and Communication Foundations awarded the University of Southern California $525,000 on October 1, 2026, under the Computer and Information Science and Engineering program (CFDA 47.070) to develop highly accurate digital twins of soft materials used in medical devices, robotics, manufactured products, and virtual reality.

The project will measure how real elastic materials deform internally using an industrial computed tomography scanner and use those measurements to build improved computer models that combine state-of-the-art artificial intelligence with physics-based simulation. The work addresses the limitation that existing computer models rely on simple formulas unable to accurately predict behavior under large deformations, resulting in products that underperform, cost more than necessary, or are poorly designed. The developed computer models and simulation software will be publicly released for use by researchers, educators, students, and companies.

The project will train graduate students in computer graphics, artificial intelligence, and physical simulation; incorporate research findings into university courses; and support outreach activities for high school students in Los Angeles. Work is performed at the University of Southern California in Los Angeles, California, with a period of performance from October 1, 2026, through September 30, 2029. The assistance type is a Project Grant.