CMPE243: System Identification
Spring 2019
MonWedFri 10:40-11:45AM                                                            J Baskin Engr 165
Lecture/Exam Dates   Your Grade   Homework   Useful Info  Course Material  Instructor    Cheating  

Course provides an introduction to the construction of linear dynamical models from experimental data using parametric and non-parametric identification techniques. Theoretical and practical aspects of these techniques will be addressed. Students will learn through hands-on experience; therefore, the course will be homework intensive.


Literature: My lecture notes will be used as the basis for the course and the exams. Textbook: Lennart Ljung, System Identification, 2nd Edition, Prentice Hall PTR (strongly suggested). Audio/video recording of the class is not permitted.



Lectures (subject to change)

         Introduction

         Time Invariant Linear Systems

         Simulation and Prediction 

         Models of Linear Time-Invariant (LTI) Systems 

         Models of Systems and Parameter Estimation (Fri, May 3-Midterm)

         Selected Topics

         Nonparametric Time- and Frequeny-Domain Methods 

         Recursive Estimation Methods

         Experiment Design  (Note: Mon, May 27-Memorial Day)

         Selected Topics




YOUR GRADE

Your final grade will be based on your homework (25%), one midterm exam (30%), the final exam (40%) and attendance above 70% (5%). Please bear in mind that your grade will be heavily based on the quality and completeness of problem solutions, and not only on their correctness.

Please note that your exam scores are not 'curved' and I do not assign letter grades to exams. I use the raw scores to compute your overall score in the class (which is also not 'curved'), and only then do I assign letter grades. Your letter grade is determined by your overall score according to the following (approximate) ranges:

A- to A+ 90 - 100
B- to B+ 78 - 89
C to C+ 65 - 77
D- to D+ and C- 50 - 64
F 0 - 49

There are small variations in these ranges from time to time (up to 1% in either direction), and intangible factors like improvement throughout the quarter can help in borderline cases.



Instructor: Dr.-Ing. Dejan Milutinović
Office: Engineering 2 319
Office Hours: Mon 4:10-5:15pm, Thu 12-1:05pm 
Email: dmilutin@ucsc.edu