We are the Predictive Dynamics Laboratory in the Department of AI and Robotics at Sejong University. We develop computational and learning-based methods for modeling, predicting, and understanding complex dynamical systems. Our goal is to build AI that understands dynamics—models that go beyond fitting observed data by respecting the underlying physical structure and remaining reliable beyond the training regime.
Our current research focuses on three closely connected areas:
Computational Dynamics and Multibody Systems
Multibody dynamics, nonlinear vibration, constrained mechanical systems, and numerical time integration
Physics-Structured Scientific Machine Learning
Structure-preserving learning, Hamiltonian and Lagrangian neural networks, neural differential equations, and long-horizon prediction
Uncertainty-Aware Dynamics and Simulation
Stochastic dynamics, uncertainty quantification, robust prediction, and learning from sparse or noisy data
We apply these methods to a broad range of mechanical, robotic, vehicle, and manufacturing systems, working at the intersection of computational mechanics, scientific machine learning, and Physical AI.