Model Predictive Control
Description
Model predictive control is a flexible paradigm that defines the control law as an optimization problem, enabling the specification of time-domain objectives, high performance control of complex multivariable systems and the ability to explicitly enforce constraints on system behavior. This course provides an introduction to the theory and practice of MPC and covers advanced topics.
Tentative content:
- Review of required optimal control theory
- Basics on optimization
- Receding-horizon control (MPC) for constrained linear systems
- Theoretical properties of MPC: Constraint satisfaction and stability
- Practical issues: Tracking and offset-free control of constrained systems, soft constraints
- Robust MPC: Robust constraint satisfaction
- Simulation-based project providing practical experience with MPC
Requirements
One semester course on automatic control, Matlab, linear algebra.
Important concepts to start the course: State-space modeling, basic concepts of stability, linear quadratic regulation / unconstrained optimal control.
Literature
Class notes (will be available online on the Moodle class page).
Exam
Final written exam during the examination session, covers all material. Students are permitted to use two A4 cheat sheets (two sheets = four pages).
Grading
The final grade is based on an exam and an optional take-home project. The exam takes place during the examination session. The project is a continuous performance assessment (learning task) and requires the student to understand and apply the lecture material.
The grade of the project may contribute 0.25 grade points to the final grade, but only if it helps improving the final grade.
Repetition
The final exam is only offered in the session after the course unit. Repetition is only possible after re-enrolling.
Lectures
All lecture slides will be available on the Moodle class page.
Recitations
Weekly recitations start February 20, 2025. The teaching assistants discuss problem sets and/or illustrate with examples topics from the previous week's lecture.
Office Hours
Office hours are offered weekly after the recitation.
Problem Sets
Nongraded, optional problem sets are handed out weekly.
Project
During the semester, there will be a graded take-home project, which can be used to improve the final grade for the course (see "grading"). The project will require the student to apply the lecture material.
Plagiarism
When handing in any piece of work, the student (or, in case of a group work, each individual student) listed as author confirms that the work is original, has been done by the author(s) independently and that s/he has read and understood the ETH Citation etiquette. Each work submitted will be tested for plagiarism.
James B. Rawlings, David Q. Mayne, and Moritz M. Diehl. Model predictive control: Theory, Computation, and Design. Nob Hill Pub., 2020. external page Link