ME493 - Methods of Data-Driven Control

Preface

Documentation of the independent study ME493: Methods of Data-Driven Control, taken during Spring 2024.

Instructor: Dirk M. Luchtenburg
Co-conspirators: Sohaib Bhatti, Khushant Khurana, Eunkyu (Q) Kim, Sonam Okuda
Email: dirk.luchtenburg@cooper.edu

Description:
This independent study provides an introduction to state-of-the-art methods employed in the field of data-driven engineering. Data-driven methods such as the proper orthogonal decomposition and the dynamic mode decomposition are used in a variety of fields including fluid dynamics, climate analysis, mixing problems, the study of infectious diseases, etc. Data-driven methods rely heavily on concepts from linear algebra, calculus, probability, and statistics. We will review these essential elements and develop a hands-on understanding of a selection of data-driven methods. The study culminates with an application of some method(s) to a practical problem which is to be carefully detailed in a technical report.

Referenced texts include the following:

• Strang (2019)
  
• Brunton and Kutz (2022)
  
• Boyd and Vandenberghe (2018)
  
• Géron (2022)
  
• Sutton and Barto (2018)

Boyd, Stephen, and Lieven Vandenberghe. 2018. Introduction to Applied Linear Algebra: Vectors, Matrices, and Least Squares. Cambridge University Press.

Brunton, Steven L, and J Nathan Kutz. 2022. Data-Driven Science and Engineering: Machine Learning, Dynamical Systems, and Control. Vol. 2. Cambridge University Press.

Géron, Aurélien. 2022. Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow. O’Reilly Media, Inc.

Strang, Gilbert. 2019. Linear Algebra and Learning from Data. SIAM.

Sutton, Richard S, and Andrew G Barto. 2018. Reinforcement Learning: An Introduction. MIT Press.