CSE DSI Machine Learning Seminar with Kshitiz Upadhyay (AME, UMN)

Data-Driven Mechanics for Soft Materials

Soft materials—including polymers, foams, hydrogels, and biological tissues—exhibit large, nonlinear deformations, rate-dependent dissipation, and damage under diverse thermo-mechanical loading conditions. Understanding and predicting their behavior is central to applications ranging from aerospace structures and soft robotics to tissue biomechanics and injury prevention. However, the complexity of these materials challenges traditional constitutive modeling, computational simulation, and uncertainty quantification (UQ) frameworks. In this talk, I will present our recent efforts toward building a data-driven mechanics framework that integrates traditional theoretical and computational physics with machine learning to overcome these challenges.

In the first part, I will introduce a physics-informed, data-driven framework for constitutive model discovery, which leverages Gaussian Process Regression (GPR) to identify interpretable constitutive relations directly from experimental data while embedding thermodynamic consistency. This approach enables accurate and generalizable predictions of stress and dissipation behavior for soft polymers and biological tissues with minimal training data.

In the second part, I will discuss our work on surrogate modeling and UQ of computational head-injury models. By combining manifold learning with Gaussian Process supervised regression, we construct efficient and reliable surrogates for high-fidelity material point method simulations of human head biomechanics. These surrogates enable probabilistic prediction of full-field strain and injury metrics under realistic head motions—capabilities not feasible with conventional UQ methods.

Together, these efforts highlight how machine learning can accelerate advances in mechanics and materials—from constitutive modeling to simulation-based prediction—while retaining interpretability and physical grounding. I will conclude by outlining emerging directions at the intersection of mechanics, data science, and artificial intelligence that aim to transform material modeling, design, and discovery for both engineering and biomedical applications.

Kshitiz Upadhyay is an Assistant Professor in the Department of Aerospace Engineering and Mechanics at the University of Minnesota Twin Cities, where he directs the Soft Materials Mechanics Laboratory. His research focuses on the mechanics of soft materials, with emphasis on constitutive modeling, data-driven methods, experimental solid mechanics, and injury biomechanics. Dr. Upadhyay’s work has been supported by the National Science Foundation (NSF), the Office of Naval Research (ONR), and the National Aeronautics and Space Administration (NASA). His research has been recognized through multiple honors, including the Early Career Research Award from the World Council of Biomechanics (2022), the Best Dissertation Award from the Department of Mechanical and Aerospace Engineering at the University of Florida (2020), and the Gator Engineering Attribute Award for Professional Excellence (2020). Before joining UMN, Dr. Upadhyay served as an Assistant Professor at Louisiana State University (2022–2025) and as a Postdoctoral Fellow at the Hopkins Extreme Materials Institute at Johns Hopkins University (2020–2022). He earned his Ph.D. and M.S. in Mechanical Engineering from the University of Florida (2020, 2019) and his B.Tech. in Mechanical Engineering from the National Institute of Technology, Bhopal, India (2014).