Model predictive control

Summary

Provide an introduction to the theory and practice of Model Predictive Control (MPC). Main benefits of MPC: flexible specification of time-domain objectives, performance optimization of highly complex multivariable systems and ability to explicitly enforce constraints on system behavior.

Content

Keywords

Multi-variable control, Constrained systems, Model-based Control, Optimization

Learning Prerequisites

Required courses

• Automatique or Control Systems

Recommended courses

• Multivariable systems or Dynamic coordination

Important concepts to start the course

• State-space modeling
• Basic concepts of stability
• Linear quadratic regulation

Learning Outcomes

By the end of the course, the student must be able to:

• Design an advanced controller for a dynamic system, A11
• Assess / Evaluate the stability, performance and robustness of a closed-loop system, A12
• Work out / Determine the performance (by simulations or experiments) of a mechatronic system, A21
• Assess / Evaluate Define (specifications) the control performance for mechatronic systems, A18

Transversal skills

• Write a scientific or technical report.

Teaching methods

Lectures, exercises and course project

Expected student activities

• Participate in lectures, exercises and course project
• Homework of about 2 hours per week

Assessment methods

• Reports on weekly exercises
• Report on simulation-based project
• Written final exam

Supervision

Resources

Bibliography

All material can be downloaded from the moodle site.

Ressources en bibliothèque

• Predictive Control with Constraints / Maciejowski
• Model Predictive Control: Theory and Design / Rawlings
• Convex Optimization / Boyd
• Predictive Control for linear and hybrid systems / Borrelli
• Numerical Optimization / Nocedal

Websites

• http://la.epfl.ch/teaching/mpc

Moodle Link

• https://go.epfl.ch/ME-425