Roland Schwan

Roland Schwan is a researcher at the Automatic Control Laboratory (LA) within the School of Engineering at the École Polytechnique Fédérale de Lausanne (EPFL). His work focuses on control systems, optimization algorithms, and the integration of machine learning with physical models. He is actively involved in projects involving model predictive control (MPC), stability verification of neural network controllers, and real-time optimization-based control for robotics applications.

His research interests span physics-informed machine learning, quadratic programming solvers, and embedded systems. Key contributions include the development of the PIQP solver for convex quadratic programming and stability analysis frameworks for neural network controllers. He has also explored applications in hovercraft dynamics identification and rocket trajectory control using advanced MPC techniques.

Schwan collaborates closely with institutions like the Swiss National Science Foundation and has published extensively in venues such as IEEE Transactions on Automatic Control and IEEE Conference on Decision and Control. His work bridges theoretical control methodologies with practical embedded system implementations.