436-405 Advanced Control Systems



 Handbook Details


 Lecture and Tutorial Notes

 Project Assignments

 Past Exam Papers

 MATLAB/SIMULINK information

 Contact Information

Schedule (Semester 2, 2004)


Tuesday, 12:00 am – 1:00 pm, IDTC Theatre

Thursday, 12:00 am – 1:00 pm, Theatre C2


For detailed program click here.

Back to top


Franklin GF, Powell DJ & Workman ML (1990, 1997) Digital Control of Dynamic Systems (2nd and 3rd edns), Addison Wesley [629.8 FRAN (2nd ed), 629.89 FRAN (3rd ed)]

Anderson BDO & Moore JB (1990) Optimal Control: Linear Quadratic Methods, Prentice Hall [629.831 ANDE]

Santina MS, Stubberud AR & Hostetter GH (1994) Digital Control System Design (2nd edn), Saunders College Publishing [629.89 SANT]

Brogan WL (1991) Modern Control Theory (3rd edn), Prentice Hall [629.831 BROG]

Skogestad S & Postlethwaite I (1996) Multivariable Feedback Control: Analysis and Design, Wiley [629.83 SKOG]

Kailath, T (1980) Linear Systems, Prentice-Hall [003 KAIL]

Glad, T & Ljung, L (2000) Control Theory, Taylor and Francis [629.8312 GLAD]

Dutton, K, Thompson, S & Barraclough, B (1997) The Art of Control Engineering, Addison-Wesley [629.8 DUTT]

Back to top

Lecture and Tutorial Presentations

A. Nonlinear systems (Dr Chris Manzie)

Dr Manzie’s notes.

B. Multivariable linear systems (Prof Good)

Click on the lecture title to see a slide-by-slide presentation (without the lecture-room animations). If you wish, you may download the PDF file to view in Acrobat Reader. M-files for Matlab v6 used to illustrate key concepts are also offered for download and exploration.

·         Lecture 1: Examples of MIMO systems, representation, controllability and observability. Print file: lecture01_4.pdf

·         Lecture 2:  State-space realisations of transfer function matrices. Print file: lecture02_4.pdf

·         Lecture 3: Review of SISO pole-placement design, digital implementation, MIMO pole placement, eigenstructure assignment, optimisation.  Print file: lecture03_4.pdf

flex.m: M-file illustrating SISO digital pole-placement design for flexible structure.

·         Lecture 4: Time-varying and infinite-horizon approaches, linear quadratic regulators, output weighting.  Print file: lecture04_4.pdf

lqr_des.m: Interactive M-file for trial-and-error design of optimal regulator for hydraulic servo example.

·         Lecture 5: Symmetric root locus, LQR design example, prediction and current estimators.  Print file: lecture05_4.pdf

flex_srl.m: M-file to produce symmetric root locus for compliant structure, which uses the function srl.m. (I would be grateful for bug reports, etc. for the latter function, which will plot an SRL for a continuous or discrete system.)
flex_lqr.m: Interactive M-file for trial-and-error design of discrete optimal regulator for compliant structure.
flex_predict.m: M-file for design of LQR for flexible structure, with multi-input prediction estimator.
flex_current.m:M-file for design of LQR for flexible structure, with multi-input current estimator.

·         Lecture 6: Effect of number of sensed outputs, comparison with full state-feedback, reduced-order estimators, optimal estimation, Kalman filter.  Print file: lecture06_4.pdf

flex_current1.m:M-file for design of LQR for flexible structure, with single-input current and prediction estimators.
flex_reduced.m:M-file for design of LQR for flexible structure, with multi-input reduced-order estimator.

·         Lecture 7: LQG regulator, Control-estimation duality, SRL for optimal estimator, example of LQG design for SIMO plant, robustness.

flex_srlc.m: M-file for SRL pole-placement design of controller and estimator for flexible structure.
flex_lqg.m: M-file for LQG design of regulator for flexible structure, with multi-input Kalman estimator.
sim_flex_lqg.mdl: Simulink model for evaluating LQG design.

·         Tutorial 1: Multivariable systems

prob1.m: M-file solution for Matlab problem 1.

·         Lecture 8: Tracking systems, integral control, disturbance estimation.

flex_track.m: M-file for tracking system, with pole-placement design of full state-feedback controller, for flexible structure
flex_int_full.m: M-file for tracking system, with integral control and full state-feedback controller, for flexible structure
sim_flex_int_full.mdl: Simulink model to evaluate previous design
flex_int_est.m: M-file for tracking system, with integral control and full-order estimator, for flexible structure
sim_flex_int_est.mdl: Simulink model to evaluate previous design
flex_disturb.m: M-file for tracking system with full-order estimator and rejection of constant disturbance, for flexible structure
sim_flex_disturb.mdl: Simulink model to evaluate previous design
sat_sin_load.m: M-file for tracking system with full-order estimator and rejection of sinusoidal disturbance, for satellite attitude control
sim_sat_sin_load.mdl: Simulink model to evaluate previous design

Back to top


There will be three assignments, all concerned with a precision X-Y table in the Mechatronics Lab. A schematic of the lab equipment is shown below.

To introduce control design challenges typical of those encountered in motion control practice, the normal couplings between the brushless AC servo motors and the lead screws can be replaced by couplings which introduce compliance and backlash. Variable amounts of Coulomb friction can also be introduced. These non-ideal characteristics will compromise the contouring performance of the X-Y table.

Assignment 1 will be performed in the laboratory. It will involve an experimental investigation of contouring performance with a conventional CNC control architecture: current and velocity loops closed by the motor drive units, position loop closed by the digital controller (a DSP in this case). Click here for Release Notes, and templates for MATLAB scripts. 

Downloads for Assignments 1 and 2: system_model.mdl, XYtableLibrary.mdl, system_param.m, coulomb.m

Assignment 2 will be performed using a detailed nonlinear simulation model of the equipment, and will involve system identification experiments on the system studied in Assignment 1, and investigation of alternative control strategies to improve the contouring performance.

Assignment 3 will be performed using direct access to the laboratory equipment. The challenge here will be to implement and test a controller which will provide superior contouring performance.

Back to top

Past Exam Papers

436-405, 1999

436-405, 2000

436-405, 2001

436-405, 2002

436-405, 2003

Back to top


Matlab resources from The MathWorks, Inc.

 Getting started with Matlab. A tutorial introduction to Matlab basics. View this document with your Web browser while you have a Matlab session going. It invites you to try various Matlab commands and to think about and learn from the system's response.

  An easy guide to Control System Toolbox. A tutorial introduction to Version 4 of the toolbox, by Finn Haugen of Telemark College in Norway.

  Animated SIMULIMK demonstrations from the MathWorks.

 Control Tutorials for Matlab. A very comprehensive set of 'tutorials' from The University of Michigan and Carnegie-Mellon University, on Matlab basics, modelling of seven different physical systems, and the design of both analog and digital controllers for these systems using classical and state-space techniques. Note that these are based on earlier versions of Matlab, and do not exploit the power of the LTI objects introduced in more recent versions. The 'matrix' of tutorials is as follows:

 Related Products

Free scientific portal for MATLAB/MIDEVA m-files and toolboxes, and Excel/Java/Fortran/C++ resources and links: http://www.mathtools.net

Complementary products for MATLAB, like MIDEVA (fast MATLAB replacement), MATCOM (Compiler for MATLAB), Visual MATCOM (integrate m-files into Visual C++) and others, all available for download: http://www.mathtools.com

Contact Information

Lecturers: Professor Malcolm C Good, Dr Chris Manzie

Electronic mail addresses: mcgood@unimelb.edu.au, manziec@unimelb.edu.au

Back to top

Last Revised: 12/11/04 9:16 AM