Learning Stable Nonlinear Reduced Dynamics for Real-Time Deformable Simulation

Real-time simulation of deformable objects is essential for XR, AR, and VR applications, yet existing methods remain too expensive for interactive use. We will develop a data-driven framework that learns both a nonlinear deformation manifold and its dynamics from simulation data. The resulting method will enable fast, stable, and physically faithful simulation of deformable objects in interactive XR environments.

Project Team

Prof. Stelian Coros

Stelian Coros leads the Computational Robotics Lab (CRL). The lab is focused on develop theoretical foundations to shape the way future generations of robots are made, how they operate in complex environments, and how they interact with us. In this quest, Computational Robotics is defined as the fusion of simulation, algorithms and data. Essentially, the lab works on formalizing advanced simulation models to equip robots with an inherent understanding of physical laws. Utilizing these models, practical algorithms are designed to address motion planning, motion control, and computational design challenges. Additionally, whenever feasible, data is leveraged to efficiently learn solution spaces, create accurate digital twins through real-to-simulation methodologies, and allow humans to teach robots new skills.

Prof. Bernd Bickel

Bernd Bickel is a Full Professor of Computational Design, leading the Computational Design Laboratory  in the Department of Architecture and equally affiliated with the Department of Civil, Environmental and Geomatic Engineering (D-​BAUG) at ETH Zurich. He conducts research into digital technologies with an emphasis on artificial intelligence and extended reality. He has a particular interest in computer graphics at the intersection between robotics, computer vision, machine learning, materials science and digital fabrication. The aim of his research is to create new and efficient ways of modelling, simulating and fabricating digital content. His work has won multiple prizes, including an Oscar (Academy of Motion Picture Arts and Sciences Technical Achievement Award) in 2019 and an ERC Starting Grant in 2016.

Dr. Bernhard Thomaszewski

Haoyang Zhou

Dr. Thabo Beeler

Thabo Beeler is currently a Research Director at Google, where he is heading the Syntec team within AR Perception. They work on digital humans in the context of virtual and augmented reality, focusing on capture, reconstruction, appearance acquisition, generative modeling, and synthesis. Prior to that, Thabo Beeler was a research scientist at Disney Research | Studios where he built up the Capture and Effects group, focusing on digital humans for film.

Dr. Hugo Bertiche Argila

Jesus Perez

Publications

No entries.