Learning-based optimal control shows potential to handle large-scale and possibly unknown systems. How can we achieve these tasks with rigorous guarantees for optimal performance and safety?  We explore methods such as model predictive control and control barrier function in this direction.

 Examples of publications:

• S. Shi, A. Tsiamis, and B. De Schutter. "Suboptimality analysis of receding horizon quadratic control with unknown linear systems and its applications in learning-based control."IEEE Transactions on Automatic Control (early access), 2026. Link:  ArXiv  IEEE

• K. He, S. Shi, T. van den Boom, and B. De Schutter. "State-action control barrier functions: Imposing safety on learning-based control with low online computational costs." IEEE Transactions on Automatic Control (early access), 2025. Link:  ArXiv  IEEE

Complex dynamical systems are often interconnections of subsystems. In this direction, we consider data-driven modeling problems of large-scale networks, with applications to biological networks, e.g., brain imaging analysis (left figure).

Examples of publications:

• S. Shi, X. Cheng, and P. M. J. Van den Hof. “Generic identifiability of subnetworks in a linear dynamic network: the full measurement case,” Automatica, 2022. Link: Elsevier  ArXiv   

• R. J. C. van Esch, S. Shi*, A. Bernas, S. Zinger, A. P. Aldenkamp, and P. M. J. Van den Hof. “A Bayesian method for inference of effective connectivity in brain networks for detecting the Mozart effect,” Computers in Biology and Medicine, 2020. *Corresponding author. Link: Elsevier  ArXiv