Machine Learning, Dynamical Systems and Control

This module gives an introduction to artificial intelligence approaches in dynamical systems and control, and shows connections between systems theory and machine learning, enabling the students to design modelling and control approaches that incorporate learning for dynamical systems approximation and controller design. The lectures are complemented with examples and a project in Matlab / Python.

Learning objectives and acquired competences:

Upon completion of the module, the student will be able to

• understand the core approaches of learning theory for dynamical systems and control,
• analyse machine learning algorithms for their usability to solve dynamical problems, and
• design classifiers for dynamical data and controllers incorporating artificial intelligence.

Content:

• Introduction to machine learning
• Static approaches in artificial intelligence
• Dynamical approaches in artificial intelligence, reinforcement learning
• Connecting systems theory with machine learning (Willems’ Fundamental Lemma, Koopman Theory, adaptive control, system identification, dynamic programming)