Course Information

Course objectives

The aim of this course is to provide the basic theory required for solving complex control problems in the face of uncertainty. The course presents theory and methodology for modelling and analysis of signals and systems, and methods for design of feedback controllers. Special emphasis is placed on:

  • Control of systems with multiple inputs and outputs
  • Fundamental limitations for control performance
  • Sensitivity and robustness in feedback systems
  • Synthesis of controllers through optimization
  • Predictive control with constraints

Personnel

Elling W. Jacobsen, Course responsible and Lectures

Victor Nan Fernandez-Ayala, Teaching assistant

Jiabao He, Teaching assistant

EECS Student Service Center, Course Administration

Elling, Victor and Jiabao are all sitting on floor 6 and 7, Malvinas v. 10. 

 

Course litterature and software

The course follows the book:

Torkel Glad and Lennart Ljung; Control Theory - Multivariable and Nonlinear Methods, Taylor and Francis Ltd, ISBN 0748408789 (Swedish version: T. Glad and L. Ljung; Reglerteori, flervariabla och olinjära metoder, Studentliteratur, 2:a upplagan, ISBN 91-44-03003-7)

The course book is available at Kårbokhandeln (the student book store) and can also be bought in online bookstores like bokus and adlibris.

Lecture notes for Lectures 1-13 will be published, together with the slides used, under Lectures. Lecture notes will be available at least two days in advance and the slides will be available at latest the day before the lecture.

All exercises with solutions, computer exercises and the laboratory project can be downloaded from the homepage:

The following material contains high-quality supplementary reading. Note that the book by Skogestad and Postlethwaite can used instead of Glad and Ljung and reading assignments will be provided also for this book (the topics of the last two lectures on MPC and constraint handling are not covered in Skogestad and Postlethwaite, but the lecture notes should be sufficient for these parts):

  • Skogestad, S. och Postlethwaite, I.: Multivariable Feedback Control, Analysis and Design, John Wiley & Sons, 1996
    • reading assignments will be provided for Skogestad since many students prefer this book over Glad and Ljung.
  • Åström, K.J. och Wittenmark, B.: Computer Controlled Systems, Theory and Design, 3rd ed, Prentice Hall, 1997
  • Goodwin, G.C., Graebe, S.F. och Salgado M.E.: Control System Design, Prentice Hall 2001
  • Maciejowski, Multivariable Feedback Design, Addison-Wesley, 1989

If you need to catch up on the basic control course, see litterature list for that course. A recommended basic control course book in english is Feedback Control of Dynamic Systems by Franklin, Powell and Emami-Naeini, Pearson 2010, ISBN 978-0-13-500150-9.

Matlab is available in most of the computer rooms at KTH. A student version of Matlab is available for use at home via KTH Program Distribution.

Disposition

The course consists of 14 lectures and 10 exercise sessions. In addition, four compulsory computer exercises and a compulsory laboratory project form part of the course. Grading is based on a 5-hour written final exam. See grading criteria below.

The lectures will be given in class. Lecture notes and slides used in each lecture will be posted under Lectures before each lecture. The exercises will be given in class; see Excercises. We strongly recommend that you also try to solve exercise problems on your own or in a group; recommended problems to do at home are listed under Exercises.

Exam and Grading Criteria

The grading of the exam will be according to the intended learning outcomes as stated in the course plan; On completion of the course, the student should be able to
•    formulate basic theory and definitions of important concepts in multivariable control
•    apply analysis and design methods in multivariable control.

  • Grade E: on simple problems
  • Grade D: on advanced problems within some part of the course
  • Grade C: on advanced problems within several parts of the course
  • Grade B: on advanced problems within all parts of the course or solve problems combining several parts of the course
  • Grade A: on advanced problems within all parts of the course and solve problems combining several parts of the course

Grading criteria on the exam will be

  • Grade E: At least 23p and at least 18p on problems 1, 2 and 3
  • Grade D: At least 28p and at least 18p on problems 1, 2 and 3
  • Grade C: At least 33p and at least 18p on problems 1, 2 and 3
  • Grade B: At least 38p
  • Grade A: At least 43p 
  • For FX at least 20p and at least 18p on problems 1, 2 and 3

Course evaluation

Please help us to improve the course by answering the course evaluation which will be available at the end of the course. We listen to all constructive criticism and make our best to continually improve the course.