EECI GSC course on LMIs for optimization and control
(Didier Henrion, LAAS-CNRS, Toulouse, France and FEL-CVUT, Prague,
Czech Republic)
Online course organized by the Department of Information Engineering, Computer Science and Mathematics of the University of L'Aquila, Italy from 26 to 30 April 2021
Venue and dates
The course is given online. It consists of 14 lectures of 90 minutes each, given
on Monday 26 (2pm-5:30pm), Tuesday 27 (9:30am-1pm and 2pm-5:30pm), Wednesday 28 (2pm-5:30pm), Thursday 29 (9:30am-1pm and 2pm-5:30pm) and Friday 30 April (9:30am-1pm) 2021, following this schedule.
Documents
The course outline and keywords are described in this document.
Video recordings of the lectures are available below:
Lecture 1 - Introduction to the course and the participants
Lecture 2 (sound only) - Background on convexity and cones
Lecture 3 (sound only) - Geometry of the semidefinite cone
Lecture 4 - Finite-dimensional conic duality
Lecture 5 - Semidefinite programming duality
Lecture 6 - Matlab demo with SeDuMi, MOSEK and YALMIP - scripts
Lecture 7 - Background on measure theory, infinite-dimensional conic duality
Lecture 8 - Positive polynomials and moments, the moment-SOS hierarchy
Lecture 9 - Solutions to exercises and the moment-SOS hierarchy
Lecture 10 - Polynomial optimization - slides with exercises and notes
Lecture 11 - Polynomial optimization continued - Matlab demo with GloptiPoly - scripts
Lecture 12 - Positively invariant sets
- slides with exercises and notes
Lecture 13 - Polynomial optimal control - slides with exercises and notes
Lecture 14 - Polynomial optimal control continued - Matlab demo with GloptiPoly - scripts
The course follows the lecture notes:
D. Henrion. Optimization on linear matrix inequalities
for polynomial systems control. International Summer School of Automatic Control, Grenoble, France, September 2014.
Homeworks and exam
Homeworks are given during the course.
A written examination is available on request.
Certificates of attendance can be provided on request.
Last updated on 30 April 2021.