Syllabus 2011
Syllabus 2011
Instructor: Professor Erik I. Verriest Oce: E-492 Phone (404) 894-2949 FAX: (404) 894-5953 e-mail: [email protected]. Be sure to put ECE6551 rst in the subject line, or your mail disappears with other SPAM. Secretary: Sharon Fennell, E-492, (404) 894-3782 Class Hours: TTh 4:00 - 5:45 pm, Van Leer C 457 Oce Hours: W 3:00-5:00 pm (tentative) Prerequisite: Graduate standing, and a good working knowledge of matrix algebra, elementary complex arithmetic and LTI-ODEs. A little signal and systems introduction: including wiring diagrams, the Laplace transform, and the Routh-Hurwitz criterion. Grading: Homework: 10 % Two tests: 25% each. Tentative dates: Tuesday June 14 and July 19. Final: 40% (Thursday, Aug 4, 2:50-5:40 pm). The nal exam MUST be taken at the scheduled time. Permitted only if either you have a legitimate excuse and I am notied in advance, or you are physically incapable of being in class due to an emergency. At the discretion of the instructor, make-up tests may either be oral or written, and may be given on the week before nals. In order to maintain academic honesty, all instances of academic misconduct will be reported to the Dean of Students.
Final: Make-ups:
Honesty:
Homework:
There will be an assignment approximately every week (except on the test weeks), due a week later. Carefully prepared and detailed solutions will be posted on T-Square. Expect to work at least another 3 hours outside the class for each class hour. The test problems will be similar in nature, but not identical to homework problems. Late homework will not be accepted.
Course Objective: To introduce the basic underlying principles in the study of digital control of physical systems. Mathematical rigor will be strived for, therefore providing a good training ground for the prospective graduate student. It is this instructors rm belief and experience that a good theoretical background provides a better mobility than training in a specic art or technique. Nothing is more practical than a good theory! However, let this not intimidate you. No prior knowledge of advanced or higher mathematics besides the cited prerequisites will be assumed. New concepts will be introduced as the course evolves, all that is required is a healthy curiosity, a creative mind and a willingness to learn. Emphasis: Will be on the basic underlying methods in the study of signals and systems. All derivations will start from rst principles.
Computer Usage: Individual computer experimentation (Matlab, Mathematica and/or Maple) is strongly recommended, but not required. Text book: K. Ogata, Discrete Time Control Systems, Second Edition, PrenticeHall 1995. (Required.) Selected classnotes will be made available on T-Square in due time.
Attention: The scheduled nal exam date is strict. No exceptions will be made for any reasons. If you need to schedule your ights home for the holidays, please do so accordingly.
Research: For students who are contemplating research under my guidance: obtaining an A in this course is a prerequisite. Students interested in my current research (mathematical system theory) may inspect my latest publications posted by my oce. Please do not remove these.
1. Introductory Material (a) Overview of the Course (b) Discrete Signals, Sequence Spaces, Behavioral Theory of Systems 2. Review Classical Control (a) Laplace Transform (b) Complex Integration, Residue Theorem (c) Input-Output Models and Transfer Functions (d) Stability and Feedback (e) Root Locus (f) Principle of the Argument, Nyquist Criteria 3. Z-Transform (a) Denition and Properties (b) Inverse Z-Transform (c) Dierence Equations 4. State Space Representation of Discrete Systems (a) Discrete and Continuous State Space Realizations (b) Similarity and Canonical Forms (c) Solutions of State Equations (d) Long Term Behavior, Stability 5. Z-Plane Analysis of Discrete Time Control Systems (a) Sample and Hold, Spectrum (b) Convolution Integral Method (c) Reconstruction, Nyquist Sampling Theorem (d) Emulation, Pulse Transfer Function (e) Realization of FIR and IIR Filters 6. Classical Discrete-Time Control (a) Conformal Mapping (b) Stability of Closed Loop Systems, Jury Test 4
(c) System Response (d) Control Design Methods 7. Properties of Realizations (a) Observability Problems (b) Reachability Problems (c) Minimality, PBH-Tests 8. Quadratic Forms and Applications (a) Quadratic Forms (b) Lyapunov Theory 9. Gramians and Energy Principles (a) Gramians and Energy Principles (b) Balancing (c) Model Reduction 10. Linear State Variable Feedback (a) Modal Control Problem, Feedback (b) Formulas for the Gains (c) The Algebraic Riccati Equation (d) Discrete-Time Scattering Theory (e) Bounded Real and Positive Real Lemmas 11. Linear State Observers (a) Modal Observer Problem (b) Combined Observer-Controller (c) Reduced Order Observers (d) Direct Transfer Function Design, Diophantine Equations 12. Multivariable System Theory (a) Matrix Fraction Descriptions (b) Polynomial Matrices (c) Rational Matrices and Smith-McMillan Form
13. Supplementary Topics, (time permitting) to be selected from: (a) Quantization Eects (b) Saturation Eects (c) Linear Matrix Inequalities (d) Time-Delay Systems (e) Periodic and General Time Varying Systems (f) Invariants and Parameterizations (g) Nonlinear Discrete-Time Systems (h) Subspace Identication, Least Squares and Total Least Squares (i) Minimum Sensitivity Design and Maximum Accuracy Control