Courses

Data-Driven Control Systems

This course on Data-Driven Control Systems provides a focused study on advanced methodologies in control system design. Part A covers topics such as Unfalsified Adaptive Switching Supervisory Control, Virtual Reference Feedback Tuning (VRFT), and Koopman Theory. Part B delves into philosophical discussions on the paradigm shift towards data-driven control. Through a blend of theory and practical applications, students will gain insights into cutting-edge approaches shaping the future of control systems.

Multivariable Control Systems

The Multivariable Control Systems course offers a comprehensive study of Multiple-Input Multiple-Output (MIMO) systems, covering essential topics such as system representation, zeros and poles analysis, state-space control design, stability considerations, and performance limitations. Students will explore the intricacies of uncertain MIMO systems, classic design methodologies, PI controllers, and the challenges posed by nonsquare MIMO configurations. By delving into both theoretical foundations and practical applications, participants will develop a deep understanding of designing and analyzing MIMO control systems, equipping them with the knowledge to tackle complex control challenges in diverse engineering domains.

System Identification

The System Identification course offers a comprehensive exploration of methodologies for modeling and understanding complex systems. Covering topics from static linear system identification to dynamic nonlinear system modeling, students will delve into parameter estimation, prediction error minimization, and closed-loop system identification. The course also delves into nonlinear parameter optimization, neural networks, fuzzy models, and their applications in system identification. By examining real-world applications and dimension reduction techniques, participants will gain a deep understanding of how to effectively identify and model diverse systems for prediction, simulation, control, and fault detection purposes.

Machine Learning

This course provides a foundational understanding of machine learning concepts, algorithms, and applications, empowering participants to extract valuable insights from vast datasets, build predictive models, and automate decision-making processes. Through theoretical lectures, practical exercises, and hands-on projects, students will explore various machine-learning techniques, including supervised, unsupervised, and reinforcement learning.

Model Predictive Control

The Model Predictive Control (MPC) System course offers a comprehensive exploration of this advanced control strategy, spanning linear MPC design to advanced applications. Students will delve into MPC’s industrial structure, historical evolution, and algorithmic architecture. The course covers linear MPC design principles, encompassing steady-state and dynamic optimization, numerical challenges, and code implementation for linear time-invariant discrete-time systems. Additionally, topics include stability analysis, feasibility, invariance, and practical MPC considerations. Advanced discussions on nonlinear systems, moving horizon estimators, explicit MPC, economic MPC, robust MPC, distributed MPC, stochastic MPC, and data-driven MPC enhance understanding of MPC’s diverse applications and capabilities in modern control systems.

Game Theory