Part-Time Programs in Engineering and Applied Science, Johns Hopkins University

Control System Design Methods
525.777

Course Description
This course examines recent multivariable control system design methodologies and how the available techniques are synthesized to produce practical system designs. Both the underlying theories and the use of computational tools are covered. Topics include review of classical control system design and linear system theory, eigenstructure assignment, the linear quadratic regulator, the multivariable Nyquist criterion, singular value analysis, stability and performance robustness measures, loop transfer recovery, H-infinity design, and mu-synthesis. An introduction to nonlinear techniques includes sliding mode control and feedback linearization. Recent papers from the literature are discussed. Each student will be assigned a design project using PC-based design and analysis software.

Syllabus

  1. Fundamental Design Principles, Nyquist Criterion, Single Loop Compensation
  2. Review of Matrix Theory, Linear Dynamic Systems, Multivariable System Descriptions
  3. Linear Quadratic Regulator, Riccati Equation
  4. Multivariable System Stability, Multivariable Nyquist Criterion, Singular Values
  5. LQR Stability Properties, LQR Loop Shaping, Linear Observers and LQG Regulator
  6. Loop Transfer Recovery, Loop Shaping, LQG/LTR Design Examples
  7. Normed Linear Spaces, H2 and H-infinity
  8. Performance Specifications
  9. H-infinity Design Examples
  10. Structured and Unstructured Model Uncertainty, Structured Singular Value, Performance and Stability Robustness, Mu-synthesis
  11. Feedback Linearization
  12. Sliding Mode Control
  13. Design Examples
  14. Design Project Presentations

Prerequisites
525.466 Linear System Theory and 525.409 Continuous Control Systems or the equivalent.

Instructor
Alan Pue is assistant supervisor of the Guidance and Navigation Systems Group in the Power Projection Systems Department at the Johns Hopkins University Applied Physics Laboratory. Dr. Pue holds B.S. and M.Eng. degrees in electrical engineering from Cornell University and a Ph.D. in electrical engineering from the University of Maryland.

Computer Lab Requirements
Matlab with Control Systems Toolbox will be used

Return to Electrical Engineering Courses | Electrical Engineering | Part-Time Engineering

January 1998