Abstract

The hierarchical and heterogeneous nature of materials drives their complex deformation and failure mechanisms across multiple length and time scales. Understanding how the microstructural features of polycrystalline metals (e.g. grain orientations, texture evolution, defect interactions) govern their macroscopic mechanical response remains a key challenge in mechanics.

Recent advances in experimental techniques, such as those in scanning electron microscopy (SEM) and in-situ characterization, now generate massive, high-resolution datasets that capture deformation processes as they unfold. However, extracting meaningful physics from these unprecedented volumes of multi-modal data requires more than traditional analysis. Conventional machine learning approaches often fail to capture the underlying physics and scale-bridging relationships that are essential for reliable predictive modeling.

This talk explores how scientific machine learning can bridge these scales when built on trustworthy foundations. We will discuss why enforcing known physical constraints, particularly material symmetries ranging from crystallographic symmetries at the grain level to texture-induced anisotropy at larger scales, is essential for creating trustworthy models for scientific discovery. By embedding these intrinsic symmetries, models become more interpretable, data-efficient, and physically consistent, directly addressing critical limitations of black-box approaches.

Through examples of SEM-based datasets, this talk will demonstrate how integrating rich experimental data with physics-informed architectures can enable trustworthy, interpretable models that respect the fundamental principles governing material behavior and reveal new insights into multi-scale deformation mechanisms.

Bio

Samantha (Sam) Daly is a Professor of the Department of Mechanical Engineering at the University of California, Santa Barbara. She earned her Ph.D. from the California Institute of Technology in 2007 and subsequently joined the University of Michigan, where she was on the faculty until 2016 prior to her move to UCSB. Her research interests lie at the intersection of experimental mechanics and scientific artificial intelligence, with the goal of advancing the understanding of deformation and failure mechanisms in advanced metallic and composite materials. She currently serves on the Executive Committees of the U.S. National Committee for Theoretical and Applied Mechanics (USNC/TAM), the Applied Mechanics Division of the American Society of Mechanical Engineers (ASME), and the Society for Experimental Mechanics (SEM).

Abstract

Nature is in a perpetual state of reorganization. In animal groups on the move, such as in fish schools and bird flocks, local social interactions result in cohesive global patterns. These cohesive patterns are regularly documented in groups of moderate size occupying open spaces. However, less is known about how these patterns scale with increasing group size and in complex environments. Here, I address these issues in computational models of large schools of fish, reaching up to 50,000 swimmers that interact through self-generated flows and follow behavioral rules inferred directly from experimental data in shallow water environments. The results reveal a “more is different” transition: smaller groups remain cohesive and polarized whereas larger groups spontaneously reorganize, repeatedly fragmenting, scattering and reassembling. I will analyze the spatiotemporal correlations in these emergent patterns and discuss their implications for information transfer, group size regulation, and the dynamics of collective behavior in living and engineering systems.

Bio

Eva Kanso is a professor and the Z.A. Kaprielian Fellow in Aerospace and Mechanical Engineering at the University of Southern California, with courtesy appointment in the department of Physics and Astronomy. While at USC, Kanso served as a program director at the National Science Foundation (2021-2023) and held visiting positions at leading research institutions in the United States and France. Kanso was a postdoctoral researcher at Caltech (2003-2005). She received a Ph.D. (2003) and an M.S. (1999) degrees in Mechanical Engineering, and an M.A. (2002) in Mathematics, all from the University of California at Berkeley. She earned a Bachelor of Engineering degree (1997) from the American University of Beirut with distinction. Kanso’s research interests concern fundamental problems in the biophysics of cellular and subcellular processes and the physics of animal behavior, both at the individual and collective levels. A central theme in her work is the role of the mechanical environment, specifically the fluid medium and fluid-structure interactions, in shaping and driving biological functions. 

Abstract

In this work we first use explainable deep learning based on Shapley explanations to identify the most important regions for predicting the future states of a turbulent channel flow. The explainability framework (based on gradient SHAP) is applied to each grid point in the domain, and through percolation analysis we identify coherent flow regions of high importance. These regions have around 70% overlap with the intense Reynolds-stress (Q) events in two-dimensional vertical planes. Interestingly, these importance-based structures have high overlap with classical turbulence structures (Q events, streaks and vortex clusters) in different wall-normal locations, suggesting that this new framework provides a more comprehensive way to study turbulence. We also discuss the application of deep reinforcement learning (DRL) to discover active-flow-control strategies for turbulent flows, including turbulent channels, three-dimensional cylinders and turbulent separation bubbles. In all the cases, the discovered DRL-based strategies significantly outperform classical flow-control approaches. We conclude that DRL has tremendous potential for drag reduction in a wide range of complex turbulent-flow configurations.

Bio

Dr. Ricardo Vinuesa is an Associate Professor at the Department of Aerospace Engineering, University of Michigan. He studied Mechanical Engineering at the Polytechnic University of Valencia (Spain), and he received his PhD in Mechanical and Aerospace Engineering from the Illinois Institute of Technology in Chicago. His research combines numerical simulations and data-driven methods to understand, control and predict complex wall-bounded turbulent flows, such as the boundary layers developing around wings and urban environments. Dr. Vinuesa has received, among others, an ERC Consolidator Grant, the Harleman Lecture Award, the TSFP Kasagi Award, the MST Emerging Leaders Award, the Goran Gustafsson Award for Young Researchers, the IIT Outstanding Young Alumnus Award, the SARES Young Researcher Award and he is also a member of the Young Academy of Science of Spain.

Abstract

In recent years, architected materials and structures have gained significant popularity due to their ability to reach enhanced performance for use in multifunctional and multidisciplinary applications. Among numerous architected materials, those based on triply periodic minimal surface (TPMS) exhibit a favorable integration of properties and the resulting performance merits. However, because of the complexities involved in the geometry representation and mechanical response of these structures, physics-based and data-driven modeling for TPMS engenders a set of challenges. To address the challenges, we developed a modeling framework for architected materials and structures focusing on sheet TPMS-based designs and applications involving protective structures that can mitigate the effect of extreme loading. To handle the geometric complexity and advanced nature of the mechanics involved, we developed a modeling framework based on Isogeometric Analysis (IGA). The key strength of IGA is the inherently tight integration between geometric design and physics-based modeling, which enables efficient simulation-based design of TPMS structures for a desired set of performance characteristics, before these are handed off to manufacturing (typically through additive manufacturing techniques) for further experimental testing and production. In this work we model TPMS structures that are fabricated using polymers, for which we envision several advantages. These advantages come with technical challenges, which we to address in this work. Constitutive modeling of polymeric materials presents a key challenge in this research. Typical stress-strain response obtained from standard tensile tests demonstrates a highly complicated constitutive relation, characterized by pseudo-elasticity at small strains, strain softening after “yielding”, strain hardening at large strains, and a strain-rate dependence across the entire regime of deformations. A comprehensive constitutive model that can capture this full spectrum of mechanical behavior at the material-point level is fundamental to understanding and accurately predicting the mechanical response at the structural level, which we simulate using IGA.

Bio

Yuri Bazilevs is the E. Paul Sorensen Professor in the School of Engineering at Brown University, where he is the inaugural Director of the Mechanics of Undersea Science and Engineering (MUSE) center and also served as the Lead and Executive Committee representative of the Mechanics of Solids and Structures group. Yuri’s research interests are in computational mechanics, with emphasis on the modeling and simulation in solids and structures, fluids, and their coupling in HPC environments. For his research contributions Yuri received many awards and honors, including the 2018 Walter E. Huber Research Prize from the ASCE, the 2020 Gustus L. Larson Award from the ASME, and the 2022 Computational Mechanics Award from the International Association for Computational Mechanics (IACM). He is included in the lists of Highly Cited Researchers, both in the Engineering (2015-2018) and Computer Science (2014-2019) categories. Yuri recently completed his service as the President of the US Association for Computational Mechanics (USACM) and as the Chairman of the Applied Mechanics Division of the ASME. He currently serves on the US National Committee for Theoretical and Applied Mechanics (USNCTAM) as a Member-at-Large, the Board of Directors of the USACM, and the Executive Council of the IACM.

Abstract

The exceptional advances in computational mechanics since the 1950s are intrinsically linked to the exponential increase in computing power. As current digital computers approach their physical limits, further increases without escalating energy and infrastructure costs appear unlikely.  Quantum computing has emerged as a new paradigm for performing certain calculations far more efficiently. To date, numerous foundational quantum algorithms with fundamentally different scaling properties have been proposed for encoding and solving problems with vector- and tensor-valued data. There is currently intense research to identify which engineering problems may benefit from quantum computing and which quantum algorithms can be integrated into computational mechanics. This presentation will examine how quantum computing can be utilised to achieve potentially orders-of-magnitude performance gains in selected computational mechanics problems. Specific problems to be explored include quantum-accelerated multiscale analysis, quantum encoding of finite-element systems, and the solution of high-dimensional probabilistic problems.

Bio

Prof. Fehmi Cirak is Professor of Computational Mechanics at the University of Cambridge and Head of the Computational Structural Mechanics Laboratory (CSMLab). His current research is focused on statistical finite elements, isogeometric analysis, and emerging quantum-computing approaches. He holds a PhD from the University of Stuttgart. Before joining Cambridge, he spent seven years at Caltech, and his past affiliations include The Alan Turing Institute and NASA’s Jet Propulsion Laboratory.

Abstract

When we think of engineering materials, we often picture solid blocks such as steel or plastic with fixed properties—soft, lightweight, or strong. In contrast, granular materials such as sand or rice flow and shear. What if a material could do both? Polycatenated Architected Materials (PAMs) are a new class of structures that bridge the gap between solids and fluids. Made of interlocked particles forming intricate 3D networks—akin to modern-day chainmail—PAMs can switch from flowing like granular matter to behaving as solid elastic materials, depending on the applied forces. This unique duality defies conventional theories and enables applications ranging from safer sports gear, reconfigurable robotics, and smart devices for extreme environments. 

Bio

Chiara Daraio is the G. Bradford Jones Professor of Mechanical Engineering and Applied Physics at Caltech. Her work is focused on developing new materials with advanced mechanical and sensing properties, for application in robotics, wearable medical devices, and vibration absorption. She received her undergraduate degree in Mechanical Engineering from the Universita’ Politecnica delle Marche, Italy (2001) and her M.S. (2003) and Ph.D. degrees (2006) in Materials Science and Engineering from the University of California, San Diego. She joined Caltech in fall of 2006 and was promoted full professor in 2010. From 2013 to 2016, she served as a Professor of Mechanics and Materials at ETH Zürich. She received a Presidential Early Career Award (PECASE) from President Obama, a US Office of Naval Research Young Investigator Award and a National Science Foundation CAREER award. She was elected as a Sloan Research Fellow and selected by Popular Science magazine among the “Brilliant 10." Chiara also serves as a Director of Research Science at Meta Reality Lab Research.

Abstract

This talk will focus on new and developing simulation methods for problems involving flowing fluids laden with particles.  The particles may have complex shapes (such as gravel) or complex mechanical behaviors (such as soft biological media).  As an example of a relevant system, we will discuss a new large-scale model for riverbed dynamics which can be studied and validated in simulation, using a procedure to create digital twins of actual riverbed gravel and then solving for the combined fluid-grain motion by combining a non-spherical Discrete Element Method for the grains and Lattice-Boltzmann Method for the water.  For soft grains that undergo high deformations during flow, we have developed a fully-Eulerian-frame method called the Reference Map Technique, which models the entire process one one fixed grid, allowing the solid bodies to obey any chosen finite-deformation mechanical model.  Systems of many submerged and interacting solids can be simulated, and, by activating the solids internally, we can simulate systems of soft active swimmers. 

Bio

Ken Kamrin received a BS in Engineering Physics with a minor in Mathematics at UC Berkeley in 2003, and a PhD in Applied Mathematics at MIT in 2008. Kamrin was an NSF Postdoctoral Research Fellow at Harvard University in the School of Engineering and Applied Sciences before joining the Mechanical Engineering faculty at MIT in 2011, where he was appointed the Class of 1956 Career Development Chair and later received a second faculty appointment in Applied Mathematics. After 13 years as a professor at MIT, Ken joined the UC Berkeley Mechanical Engineering faculty in 2024. Kamrin’s research focuses on constitutive modeling and computational mechanics for large deformation processes, with interests spanning elastic and plastic solid modeling, granular mechanics, amorphous solid mechanics, and fluid-structure interaction. Kamrin’s honors include the 2010 Nicholas Metropolis Award from APS, the NSF CAREER Award in 2013, the 2015 Eshelby Mechanics Award for Young Faculty, the 2016 ASME Journal of Applied Mechanics Award, and the 2022 MacVicar Faculty Fellowship from MIT. He sat for three years on the Board of Directors of the Society of Engineering Science and serves as an associate editor for the International Journal of Solids and Structures, Granular Matter, and Computational Particle Mechanics. He is co-author of the recent undergraduate textbook Introduction to Mechanics of Solid Materials (Oxford).

Abstract

The coordinated flight of a bird flock, the relentless propulsion of a bacterium, and the flickering of red blood cells all share a common physical origin: they are hallmarks of active matter. Unlike inert materials that dominate traditional engineering, active matter consists of units that continuously consume energy to generate mechanical forces. In the realm of soft matter, and particularly in biological systems, fluctuations are not mere background noise; they are central to function, enabling sensitivity, adaptability, and motion. While decades of research in equilibrium statistical mechanics have provided phenomenal insights into “passive” systems that exhibit thermal fluctuations, living systems are rather “alive” with their own energy source capable of circumventing equilibrium considerations. Life is powered by continuous energy transduction, most notably through ATP hydrolysis, driving novel steady states far from equilibrium. 

In this talk, I will present a unified approach that combines continuum theory with non-equilibrium statistical mechanics to study soft active biological matter, specifically biological membranes and explore its application to address some important questions in biophysics. In the first example, we show that active forces can endow biological vesicles with the ability to attain size distributions that are improbable for passive vesicles governed only by thermal fluctuations. In another example, we demonstrate how active matter serves as an underpinning agency for the extraordinary sensitivity of some biological membranes to electric fields.

Bio

Yashashree Kulkarni is the Bill D. Cook Professor in the Department of Mechanical and Aerospace Engineering at University of Houston. She received her bachelor's degree from the Indian Institute of Technology Bombay in India in 2001 and her Ph.D. in Applied Mechanics from Caltech in 2007. After spending two years as a postdoctoral scholar at University of California San Diego, she joined the University of Houston as an Assistant Professor in 2009. Her research focuses on understanding physical phenomena in materials science and biology using continuum mechanics, statistical mechanics, and multi-scale modeling. She served as the President of the Society of Engineering Science (SES) in 2024 and is currently on the Executive Committee of the Applied Mechanics Division of the American Society of Mechanical Engineers (ASME). She currently serves as an associate editor for ASME’s Applied Mechanics Reviews. She is the recipient of the 2010 DARPA Young Faculty Award, the 2017 Sia Nemat-Nasser Early Career Award and the 2024 Mid-Career Centennial Award by the ASME Materials Division. She became an ASME Fellow in 2022. 

Bio

Ali Mani is a professor of Mechanical Engineering at Stanford University. He is also a faculty affiliate of the Institute for Computational and Mathematical Engineering at Stanford. He received his PhD in Mechanical Engineering from Stanford in 2009. Prior to joining the faculty in 2011, he was a senior postdoctoral associate at Massachusetts Institute of Technology. His research group builds and utilizes large-scale high-fidelity numerical simulations, as well as methods of applied mathematics, to enable quantitative understanding and reduced-order modeling of transport processes that involve strong coupling with fluid flow and commonly involve turbulence or chaos. His current research includes turbulence modeling, data-driven methods for closure analysis, and computational simulation of multiphase flows.

Abstract

The emergence of machine intelligence poses a singular moment in the evolution of research approaches. The interactions in natural language and the multifaceted nature of machine intelligence, for instance, analyzing visual data, provide a unique pathway for planning scientific exploration. However, it is the ability to reason and even to abstract that makes machine intelligence a formidable partner in human-driven advances in science and engineering. In this talk, we lay out the essential components of human-guided machine exploration. In particular, we focus on automating non-computational components of computation, unlocking a new frontier in engineering and scientific research. We envision a new paradigm, termed persistent computing machines, that can autonomously explore research domains. Initial results on these topics will be presented. This work is in collaboration with Vansh Sharma, Tim Welch, Anthony Carreon at UM, and Shivam Barwey at Notre Dame University.

Bio

Venkat Raman is the James Arthur Nicholls Collegiate Professor of Engineering and the Director of the Center for Prediction, Reasoning and Intelligence for Multiphysics Exploration (C-PRIME) at the University of Michigan. Venkat Raman received his PhD from Iowa State University in the Department of Chemical Engineering and was a NASA/Center for Turbulence Research Postdoctoral Fellow at Stanford University. Before joining the University of Michigan, he was on the faculty at The University of Texas at Austin. He is now a tenured professor at the University of Michigan in the Department of Aerospace Engineering. Raman received an NSF CAREER award in 2008, a distinguished paper award at the International Combustion Symposium in 2013, and he held the Eli. H and Ramona Thornton Centennial Fellow in Engineering at UT Austin from 2013-2014. He is a recipient of the George J. Huebener, Jr. Research Excellence Award from the University of Michigan. He was elected Fellow of the Combustion Institute in 2022 and serves as an Associate Editor of Combustion and Flame and the AIAA Journal of Propulsion and Power. As a faculty member, Raman has advised/is currently advising 40 PhD/MS students and has published more than 175 peer-reviewed articles in archival journals and conferences. Raman’s research interests lie in the broad area of computational propulsion but have more recently focused on detonation engines and scramjet-based hypersonic propulsion.

Abstract

Whereas human tissues and organs are mostly soft, wet, and bioactive, machines are commonly hard, dry, and abiotic. Merging humans and machines is of critical importance in addressing grand societal challenges in AI, health, environment, security, education, and happiness in life. However, merging humans and machines is extremely challenging due to their fundamentally contradictory properties. At the MIT Zhao Lab, we invent, study, and translate soft materials and systems to form long-term, robust, non-fibrotic, and high-bandwidth interfaces between humans and machines. In this talk, I will discuss the fundamental mechanics of soft materials for merging humans and machines. I will conclude the talk with a vision for the future convergence of humans and machines—especially between humans and AI.

Bio

Xuanhe Zhao is the Uncas and Helen Whitaker Professor at MIT. The mission of the MIT Zhao Lab is to advance science and technology at the intersection of humans and machines. Dr. Zhao is a Humboldt Research Awardee, Clarivate Highly Cited Researcher, and American Institute for Medical and Biological Engineering (AIMBE) fellow. He received early career awards from NSF, ONR, ASME, SES, AVS, Adhesion Society, and Materials Today. To translate technologies into societal impacts, he co-founded and/or advised multiple startup companies, including SanaHeal, Magnendo, Sonologi, Orbit, and Pelva. More than 15 patents from the Zhao Lab have been licensed by companies and contributed to FDA-approved and widely used medical devices.