Abstract:

Robots that operate reliably in the real world must reason about forces, not just positions. I will present a robotics architecture designed around force intelligence—the ability to decide when, where, and how much force to apply. I will show how this perspective unifies whole-body control and dexterous manipulation, and argue that force-centric design is a key missing ingredient for scalable real-world autonomy.

Bio:

Pulkit Agrawal is an Associate Professor in the Department of Electrical Engineering and Computer Science at MIT. He earned his Ph.D. from UC Berkeley and co-founded Eka Robotics and SafelyYou. Pulkit completed his bachelor’s from IIT Kanpur and was awarded the Director’s Gold Medal. His work has received multiple Best Paper Awards, the IEEE Early Career Award in Robotics and Automation, the IROS Toshio Fukuda Young Professional Award, the IIT Kanpur Young Alumnus Award, the Sony Faculty Research Award, the Salesforce Research Award, the Amazon Research Award, the Signatures Fellow Award, and the Fulbright Science and Technology Award.

Speaker: Wolfram Burgard
Professor and Founding Chair, Department of Computer Science & Artificial Intelligence at University of Technology Nuremberg
Editor-in-Chief at IEEE Transactions on Robotics

Presentation Title: "The Challenges to Realize Embodied AI"

Date: Wednesday, April 22 at 11:00 AM

Location: Tech, M416 and Zoom
https://tinyurl.com/CRBSeminar

Abstract:
To ultimately achieve Embodied Artificial Intelligence, we need robots capable of robustly perceiving their environments and executing their actions. The key challenge is that no sensors and actuators are perfect, which means that robots need the ability to properly deal with the resulting uncertainty. Professor Burgard will discuss the opportunities of the probabilistic and deep-learning approaches to robotics and how they can be combined to get the best of both worlds. He will describe how we can utilize the potential of foundation models to even better deal with complex and changing real-world environments.

Bio:
Wolfram Burgard is a professor of Computer Science at the University of Technology Nuremberg, where he heads the research group on Robotics and Artificial Intelligence. Prior, he was a professor of autonomous intelligent systems at the University of Freiburg, Germany. From 2019 until 2021, he was VP for Automated Driving and Machine Learning at the Toyota Research Institute in Los Altos, USA. In his career, he published over 400 papers. He received multiple awards, including the IEEE Technical Field Award, the IEEE RAS Pioneer Award, and the Leibniz Prize of the German Research Foundation. He is a fellow of the IEEE, the AAAI, and the EurAI, and a member of the German Academy of Sciences Leopoldina and the Heidelberg Academy of Sciences and Humanities.

Related Info

Biophysical modeling of synaptic plasticity

Abstract

Neurotransmission occurs at different length scales. Experiments have shown that neuronal structure and function are tightly coupled across these scales. In this talk, I will discuss recent efforts to develop quantitative models to understand these structure-function relationships. One focus is how the shapes of dendritic spines and the presence of organelles within them affect calcium transients in spines and dendrites. I will also present an example at the scale of axons, focusing on how the mechanics of the neuronal membrane influence axonal morphology and how the organization of the membrane periodic skeleton affects tension propagation along axons.

Bio

Padmini Rangamani is a Professor in the Departments of Pharmacology and Mechanical and Aerospace Engineering at the University of California, San Diego. Prior to this, she was a UC Berkeley Chancellor’s Postdoctoral Fellow. She obtained her Ph.D. in Biological Sciences from the Icahn School of Medicine at Mount Sinai. She received her B.S. and M.S. in Chemical Engineering from Osmania University (Hyderabad, India) and the Georgia Institute of Technology, respectively. She is the recipient of the PECASE, ARO, AFOSR, and ONR Young Investigator Awards, as well as a Sloan Research Fellowship in Computational and Molecular Evolutionary Biology. She was recently elected a Fellow of the American Institute for Biological and Medical Engineers and the American Physical Society.

ABSTRACT

GScientific Machine Learning (SciML) integrates data-driven inference with physical modeling to solve complex problems in science and engineering. However, the design of SciML architectures, loss formulations, and training strategies remains an expert-driven research process, requiring extensive experimentation and problem-specific insights. We introduce AgenticSciML, a collaborative multi-agent system in which over 10 specialized AI agents collaborate to propose, critique, and refine SciML solutions through structured reasoning and iterative evolution. The framework integrates structured debate, retrieval-augmented method memory, and ensemble-guided evolutionary search, enabling the agents to generate and assess new hypotheses about architectures and optimization procedures. Across physics-informed learning and operator learning tasks, the framework discovers solution methods that outperform single-agent and human-designed baselines by up to four orders of magnitude in error reduction. The agents produce novel strategies—including adaptive mixture-of-expert architectures, decomposition-based PINNs, and physics-informed operator learning models—that do not appear explicitly in the curated knowledge base. These results show that collaborative reasoning among AI agents can yield emergent methodological innovation, suggesting a path toward scalable, transparent, and autonomous discovery in scientific computing.

BIO
George Karniadakis is from Crete. He is an elected member of the National Academy of Engineering, National Academy of Arts and Sciences, and a Vannevar Bush Faculty Fellow. He received his S.M. and Ph.D. from the Massachusetts Institute of Technology (1984/87). He was appointed Lecturer in the Department of Mechanical Engineering at MIT and subsequently joined the Center for Turbulence Research at Stanford/NASA Ames. He joined Princeton University as Assistant Professor in the Department of Mechanical and Aerospace Engineering and as Associate Faculty in the Program of Applied and Computational Mathematics. He was a Visiting Professor at Caltech in 1993 in the Aeronautics Department and joined Brown University as Associate Professor of Applied Mathematics in the Center for Fluid Mechanics in 1994. After becoming a full professor in 1996, he continued to be a Visiting Professor and Senior Lecturer of Ocean/Mechanical Engineering at MIT. He is an AAAS Fellow (2018–), Fellow of the Society for Industrial and Applied Mathematics (SIAM, 2010–), Fellow of the American Physical Society (APS, 2004–), Fellow of the American Society of Mechanical Engineers (ASME, 2003–), and Associate Fellow of the American Institute of Aeronautics and Astronautics (AIAA, 2006–). He received the William Benter Award (2026), the SES G.I. Taylor Medal (2014), the SIAM/ACM Prize on Computational Science & Engineering (2021), the Alexander von Humboldt Award (2017), the SIAM Ralf E. Kleinman Award (2015), the J. Tinsley Oden Medal (2013), and the CFD Award (2007) by the US Association for Computational Mechanics. His h-index is 160 (highest in Applied Mathematics) and he has been cited over 156,000 times.

BIO

Treasured member of the Northwestern faculty from 1977 until his death in 2014, Ted Belytschko was a central figure in the McCormick community and an internationally renowned researcher who made major contributions to the field of computational structural mechanics. One of the most cited researchers in engineering science, Belytschko developed explicit finite element methods that are widely used in crashworthiness analysis and virtual prototyping in the auto industry. He received numerous honors, including membership in the U.S. National Academy of Engineering, U.S. National Academy of Sciences, and the American Academy of Arts and Sciences. He was a founding director of the U.S. Association for Computational Mechanics, and in 2012, the association named a medal in his honor. The ASME Applied Mechanics Award was renamed the ASME Ted Belytschko Applied Mechanics Division Award in November 2007. Belytschko also served as editor-in-chief of the International Journal for Numerical Methods in Engineering, and he was co-author of the books “Nonlinear Finite Elements for Continua and Structures” and “A First Course in Finite Elements.”

ABOUT TED BELYTSCHKO
Treasured member of the Northwestern faculty from 1977 until his death in 2014, Ted Belytschko was a central figure in the McCormick community and an internationally renowned researcher who made major contributions to the field of computational structural mechanics. One of the most cited researchers in engineering science, Belytschko developed explicit finite element methods that are widely used in crashworthiness analysis and virtual prototyping in the auto industry. He received numerous honors, including membership in the U.S. National Academy of Engineering, U.S. National Academy of Sciences, and the American Academy of Arts and Sciences. He was a founding director of the U.S. Association for Computational Mechanics, and in 2012, the association named a medal in his honor. The ASME Applied Mechanics Award was renamed the ASME Ted Belytschko Applied Mechanics Division Award in November 2007. Belytschko also served as editor-in-chief of the International Journal for Numerical Methods in Engineering, and he was co-author of the books “Nonlinear Finite Elements for Continua and Structures” and “A First Course in Finite Elements.”

Co-sponsored by the Departments of Mechanical Engineering and Civil & Environmental Engineering

Mechanics of architected materials across length and time scales

Abstract

Architected materials (or mechanical metamaterials) across length scales—from nanometers to centimeters— have enabled previously unachievable mechanical properties through a variety of 3D material morphologies. Significant advances in our understanding of these materials have thus pointed to structure-property relations that lead to unique macroscopic mechanical properties. Despite this progress, several hurdles have precluded widespread application of these materials to solve engineering challenges. First, clear routes to scalably design and fabricate these architected materials have remained elusive; with most designs reported to date targeting high stiffness but low deformability. Second, since most of the studies to date have characterized architected materials under quasi-static deformation, their dynamic-property regime remains to be fully characterized and understood—essential to a variety of envisioned applications. In this talk, we present efforts towards addressing these long-standing challenges, specifically by proposing routes for designing architected materials with extreme compliance and by presenting two types of high-throughput characterization methods that enable exploration of architected materials under dynamic conditions. We discuss efforts implementing and understanding compliant architected materials by proposing metamaterial design paradigms inspired by polymer-network architectures. We also present efforts performing non-contact characterization of materials through laser-induced vibrational signatures, providing a new route to uncover dynamic elastic responses of materials and unparalleled throughput rates. Lastly, we discuss efforts characterizing architected materials under extreme dynamic conditions through use of microparticle impact experiments at the microscale, shedding light on energy dissipation mechanisms that emerge from the use of 3D microstructure.

Bio

Carlos Portela is the Robert N. Noyce Career Development Associate Professor in Mechanical Engineering at MIT. Portela received his Ph.D. and M.S. in mechanical engineering from the California Institute of Technology, where he was given the Centennial Award for the best thesis in Mechanical and Civil Engineering. His research lies at the intersection of mechanics, nano-to-macro fabrication, and materials science with the objective of designing and testing novel materials—with features spanning from nanometers to centimeters—that yield unprecedented mechanical and acoustic properties. Portela is the recipient of the 2026 ONR YIP Award, a 2024 ARO Early Career Program Award, was recognized as an MIT TR Innovator Under 35 in 2022, and was a recipient of the 2022 NSF CAREER Award. His teaching efforts have been recognized by MIT’s 2025 Jr. Bose Award and the 2023 Spira Award for Excellence in Teaching.