There are a number of methods to interact with your colleagues, the professors and the TAs during the course:

Exercise / TP sessions

Help on the TPs, exercises, lectures, or general control discussion is available in person on Tuesdays from 15h - 17h in MED 21120 and MED 22524.

TP and exercise sessions can both also be done remotely (in-person is prefered of course), and TAs will be available online to support you via Ed Discussion during the TP sessions.

Recorded videos

All the lecture material has been pre-recorded and will be made available on Moodle below.

Ed Discussion

Post questions and answers using the Ed Discussion link below. The profs and TAs will monitor and respond through this platform too.

The schedule is given below week-by-week.

Explanations of the various activities:

Videos: Covered in the lecture
This material will be covered during the in-person live lectures on Tuesday morning. You can use these videos if you can't attend the lectures, or to review the material later on.

Videos: Review material
These are topics that you should already know, but this material can help you review if you are uncertain. This material will not be covered during the in-person lecture.

Exercises
Exercise sets aligned with the flow of the lectures. 

TPs
TPs enabling you to try the various techniques developed on the experimental platforms in the teaching labs. These can be done in-person, or online.

If you're having trouble printing the slides with the grey font, then here's a copy with all the fonts set to black.

For the following three weeks, we'll cover the first of three main control approaches that we'll study this term: PID. This is the simplest form of control, but also by far the most common and is present in every device everywhere.

This week we continue on studying the theory of PID control by adding the integrator and the derivative terms.

This week we'll finish our study of the PID controller by looking a few methods of tuning the controller to achieve good performance.

This week we'll start our second approach to controller design: loop shaping. This is also a very commonly applied technique, and we'll study the basic, but commonly used, version of this. The advanced control course takes these ideas much further by looking at multi-dimensional optimal loop shaping techniques.

The idea of loopshaping is to modify the frequency response of our open-loop system to get good properties in our closed-loop system. Since everything here requires very good familiarity with the frequency domain, this first week we'll review the frequency response.

This week we'll consider the concept of stability more carefully by developing the Nyquist criterion. This will lead us to the ability to be able to measure how far we are from becoming unstable, and therefore how wrong our model can be, or how different the world can be from what we think it is, and yet our controller still works.

This week we'll be studying robustness margins, which are measures that tell us how far our system is from becoming unstable. We care about this because the models we make of the world are always highly approximate, and the real world can be very different from what we expect. 

Next week we'll move on to loopshaping techniques, where the goal will be to design frequency responses that trade robustness (i.e., being stable despite a changing world, or wrong model) against performance/aggressiveness of the conroller.

This week we'll put together the tools we've learned so far and develop a procedure to modify the frequency response of our open-loop system to achieve good closed-loop behaviours.

This week we'll start studying the third class of control techniques covered in the course: state-space design approaches.

This week we will first finish the segment on state-feedback control design, before moving onto the second topic in state-space design, that of state estimation.

In this last week, we'll have a brief look at another very common method of control linear quadratic regulation. From one perspective, this is just another method of pole placement, but it has a very different twist to it. As a form of "optimal" control, we specify the type of behaviour that we would like to see at a higher level, and then leave the optimizer to find us the best controller to achieve it. This is a philosophy that is common in many more advanced forms of control, and one that you will see in your future control classes.

This week we will make a brief introduction to controller deployment - how to go from paper to hardware.

Each week you have written / matlab exercises to do. These are interspersed with the lectures and TPs and are given on the MOOC platform. 

The full set of questions and their solutions are given here below. (Only the questions are on the MOOC platform).

• The exam is closed-book and closed-notes
• You may bring one sheet, double-sided, with anything you like written on it. Hand written, typed, printed anything that fits on two sides of an A4 sheet.
• You may bring a non-graphing calculator
• The information on the sheet below will be included in the exam: https://moodle.epfl.ch/mod/resource/view.php?id=1008003

The lectures are all pre-recorded and are available on the MOOC platform. Please follow the schedule indicated above to keep up with the course.

The slides used are all available below.