Embedded and Convex Optimization for Control
Stephen Boyd and Moritz Diehl
Leuven, 15-19 March 2010
COURSE INFORMATION
The course Embedded and Convex Optimization for Control is an interactive course to provide the participants with a strong working knowledge about the methods and applications of dynamic optimization in engineering applications. The course will consist of lectures, interactive sessions and guided computer exercises and is organized in Leuven as part of the ATHENS program 15 to 19 March 2010. It is aimed at 5th year master students with very strong skills in mathematics and programming in the languages C++ and MATLAB.
There will be applications from several fields that are treated in self-chosen tutorial projects by the participants in the last two days of the course. Particular emphasis is put on convex and embedded optimization.
The goal is that towards the end of the course, every participant will be working on formulating and solving a dynamic optimization problem of her/his own choice, so it is encouraged to think about interesting applications of dynamic optimization even before the course starts. The lectures and exercises will be given by S. Boyd, M. Diehl, B. Houska, H.J. Ferreau, J. Mattingley, and Y. Wang.
The language of the course is English.
Prerequisites
- Strong background in analysis and linear algebra corresponding to the first two years of a science or engineering degree with good grades. Knowledge of numerical mathematics is very helpful, but not manditory.
- Programming skills in the languages Matlab and C++ are needed.
Topics covered in the course
- Convex optimization problems and interior point algorithms
- Theory of nonlinear programming and optimal control
- Dynamic system modeling for optimization
- Model Predictive Control
- Convex optimization with the open-source package CVX - a matlab software for disciplined convex programming.
- Dynamic optimization using the open-source software package ACADO Toolkit - a toolkit for automatic control and dynamic optimization developed by OPTEC
Examination
- A short written exam for self-assessment and rehearsal will be held on Friday morning
Location
K.U. Leuven, Celestijnenlaan 200A (google maps), PC Room 00.124 and Seminar Room 00.144. The course will take place Mo-Fr 9:00 - 17:30
Accomodation
See Visitor Information.
The Program at a Glance:
| Monday 15 | Tuesday 16 | Wednesday 17 | Thursday 18 | Friday 19 | |
|---|---|---|---|---|---|
| 9:00 - 10:30 |
Introduction to Nonlinear Optimization M.Diehl |
Lecture on Dynamic Processes with Computer Exercise M. Diehl, B. Houska, H.J. Ferreau |
Demonstration at the Department of Mechanical Engineering |
Team-Work on Self Chosen Projects: Convex Optimal Control |
WRITTEN EXAM
|
| 10:45-12:15 |
Recent Advances in Convex Optimization S. Boyd |
Convex Optimization Methods for Stochastic Control
Y. Wang |
Computer Exercises: Solving Optimal Control Problems with ACADO B. Houska and H.J. Ferreau |
Team-Work on Self Chosen Projects: Convex Optimal Control |
Student Presentations of the Self Chosen Projects |
| Lunch | Lunch | Lunch | Lunch | Lunch together | |
| 13:30-15:00 |
Convex Sets, Functions,and Problems Y. Wang |
Nonlinear Optimal Control M. Diehl |
Automatic code generation for real-time convex optimization
J. Mattingley |
Team-Work on Self Chosen Projects: Nonlinear Optimal Control |
Student Presentations of Self Chosen Projects 13:00-14:00 |
| 15:15-16:15 |
Computer Exercise: CVX Tutorial J. Mattingley |
Nonlinear Programming M. Diehl |
Computer Exercise on Nonlinear Model Predictive Control H.J. Ferreau and B. Houska |
Team-Work on Self Chosen Projects: Nonlinear Optimal Control |
Reception at Arenberg Castle |
| 16:30-17:30 |
Office Hours: Time for Questions |
Office Hours: Preparation of the Projects for Thursday |
Nonlinear Model Predictive Control M. Diehl |
free | Reception at Arenberg Castle |
DETAILED PROGRAM:
Program on Monday, 15/3/2010
| Time | Topic | Room |
|---|---|---|
| 09.00 - 09.15 |
Welcome by the organizers Moritz Diehl |
00.124 |
| 09.15 - 9.45 |
Introduction to Embedded Optimization for Control Moritz Diehl |
00.124 |
| 9.45 - 10:30 |
Optimization: An Overview Moritz Diehl |
00.124 |
| 10.30 - 10.45 | Coffee break | close to 00.124 |
| 10.45 - 12.15 |
Recent Advances in Convex Optimization Stephen Boyd |
00.124 |
| 12.15 - 13.30 | Lunch break | ALMA 3 |
| 13.30 - 15.00 |
Convex Sets, Functions, and Problems Yang Wang |
00.124 |
| 15.00 - 15.15 | Coffee break | close to 00.124 |
| 15:15 - 15:45 |
Convex Functions and Sets Yang Wang |
00.124 |
| 15.45 - 16.45 |
Disciplined Convex Programming and CVX Interactive Lecture and Demonstration Jacob Mattingley and Yang Wang |
00.124 |
| 16:45 - 17:00 | short break | |
| 17.00 - 17:30 | office hours, time for questions | 00.124 |
Program on Tuesday, 16/3/2010
| Time | Topic | Room |
|---|---|---|
| 09.00 - 09.25 |
Dynamic Process Models Moritz Diehl |
00.124 |
| 09.25 - 9.45 |
Numerical Simulation of ODE models Boris Houska |
00.124 |
| 9.45 - 10:30 |
ACADO exercise: Simulation of a Crane Model Boris Houska and Hans Joachim Ferreau |
00.124 |
| 10.30 - 10.45 | Coffee break | close to 00.124 |
| 10.45 - 11.45 |
Convex Optimization Methods for Constrained Linear Stochastic Control Yang Wang |
00.124 |
| 11.45 - 12.15 |
Exercise using CVX for MPC Jacob Mattingley |
00.124 |
| 12.15 - 13.30 | Lunch break | ALMA 3 |
| 13.30 - 15.00 |
Numerical Optimal Control Moritz Diehl |
00.144 |
| 15.00 - 15.15 | Coffee break | close to 00.144 |
| 15.15 - 16.15 |
Newton Type Constrained Optimization in a Nutshell Moritz Diehl |
00.144 |
| 16:15 - 16:30 | short break | |
| 16.30 - 17:30 |
Preparation of the Projects for Thursday. Individual discussion with one of the teachers. |
00.144 |
Program on WedNesday, 17/3/2010
| Time | Topic | Room |
|---|---|---|
| 09.00 - 09.30 |
Introduction to Mechanical Engineering Applications Moritz Diehl |
00.124 |
| 09.30 - 10.30 |
Visit at the Department of Mechanical Engineering Lieboud Van den Broek, Jan Swevers, Moritz Diehl |
PMA |
| 10.30 - 10.45 | Coffee break | close to 00.124 |
| 10.45 - 11.45 |
Solving Optimal Control Problems with ACADO With Computer Exercise Boris Houska and Joachim Ferreau |
00.124 |
| 11.45 - 13.00 | Lunch break | ALMA 3 |
| 13.00 - 14.45 |
Embedded Convex Optimization Lecture and Computer Exercise Jacob Mattingley |
00.124 |
| 14.45 - 15:15 |
Computer Exercise on Nonlinear (Part 1) Model Predictive Control Boris Houska and Hans Joachim Ferreau |
00.124 |
| 15.15 - 15.30 | Coffee break | close to 00.124 |
| 15.30 - 16.15 |
Computer Exercise on Nonlinear (Part 2) Model Predictive Control Boris Houska and Hans Joachim Ferreau |
00.124 |
| 16:15 - 16:30 | short break | |
| 16.30 - 18:00 |
Nonlinear Model Predictive Control Moritz Diehl |
00.144 |
Program on THUrsday, 18/3/2010
| Time | Topic | Room |
|---|---|---|
| 09.00 - 10.30 |
Team Work on Self Chosen Projects: Convex Optimization for Control Stephen Boyd, Jacob Mattingley, and Yang Wang |
00.026 |
| 10.30 - 10.45 | Coffee break | close to 00.026 |
| 10.45 - 12.15 |
Team Work on Self Chosen Projects: Convex Optimization for Control Stephen Boyd, Jacob Mattingley, and Yang Wang |
00.026 |
| 12.15 - 13.30 | Lunch break | ALMA 3 |
| 13.30 - 15.00 |
Team Work on Self Chosen Projects: Nonlinear Optimal Control Moritz Diehl, Boris Houska, and Joachim Ferreau |
00.026 |
| 15.00 - 15.15 | Coffee break | close to 00.026 |
| 15.15 - 16.15 |
Team Work on Self Chosen Projects: Nonlinear Optimal Control Moritz Diehl, Boris Houska, and Joachim Ferreau |
00.026 |
| 16:15 - 16:30 | short break | |
| 16.30 - 17:30 | free, learing for exam, or finishing project | 00.026 |
Program on Friday, 19/3/2010
| Time | Topic | Room |
|---|---|---|
| 09.00 - 10.30 |
WRITTEN EXAM |
00.124 |
| 10.30 - 10.45 | Coffee break | close to 00.124 |
| 10.45 - 12.00 | Presentation of the Student Projects | 00.124 |
| 12.00 - 13.00 | Lunch break | ALMA 3 |
| 13.00 - 14.00 | Presentation of the Student Projects | 00.124 |
| 15.00 - 18.00 | Reception | Arenberg Castle |
Note that the final programme might be subject to minor changes.
LECTURES / SLIDE MATERIAL:
http://www.stanford.edu/class/ee364a/lectures.html - Lectures by Stephen Boyd
http://www.stanford.edu/class/ee364b/lectures.html - Lectures by Stephen Boyd
MATERIAL FOR THE EXERCISES
CVX - A matlab software for disciplined convex programming
ACADO Toolkit - An open source toolkit for automatic control and dynamic optimization.
CONTACT
Boris Houska
Optimization in Engineering Center (OPTEC), K.U.Leuven,
Electrical Engineering Department (ESAT),
Kasteelpark Arenberg 10 - bus 2446, BE-3001 Leuven-Heverlee,
Belgium
Office : 02.05
Phone: +32 (0)16 32 03 64
E-mail:
This e-mail address is being protected from spambots. You need JavaScript enabled to view it
LINKS
ATHENS Program
OPTEC Homepage
ATHENS Course 2007
ATHENS Course 2009
TEACHERS

Prof. Stephen P. Boyd
Information Systems Laboratory,
Department of Electrical Engineering,
Stanford University
Phone: (650) 723-0002
E-mail:
This e-mail address is being protected from spambots. You need JavaScript enabled to view it
Prof. Moritz Diehl
Optimization in Engineering Center (OPTEC) and
Electrical Engineering Departments (ESAT)
K.U. Leuven
Kasteelpark Arenberg 10, 3001 Leuven-Heverlee,
Belgium
E-mail:
This e-mail address is being protected from spambots. You need JavaScript enabled to view it
Phone: +32-16-321884
.jpg)
Boris Houska
E-mail:
This e-mail address is being protected from spambots. You need JavaScript enabled to view it
Phone: +32 (0)16 32 03 64

Hans Joachim Ferreau
E-mail:
This e-mail address is being protected from spambots. You need JavaScript enabled to view it
Phone: +32 (0)16 32 03 63

Jacob Mattingley (
)
Department of Electrical Engineering
Stanford University
E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Yang Wang
Department of Electrical Engineering
Stanford University
E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it



Athens 


