Foundation Models for Estimation and Control of Dynamical Systems - 2026_IDR_DEIB_77

The research programme focuses on the development of Foundation Models for estimation and control of dynamical systems, with applications in automotive, mechatronics, and robotics. The activities will investigate In-Context Learning methodologies to develop estimators and controllers capable of adapting to different systems and operating conditions without retraining. Learning approaches based on simulated and experimental data will be explored, with particular emphasis on generalization, robustness, and zero-shot transfer to real-world systems. The proposed methodologies will be validated through simulations and experimental applications.