01RKYQW / 01RKYOV - Estimation, Filtering and System Identification

(was: 01PDDQW / 01PDDOV - Identification and Control Methodologies)

Master Course in Mechatronic Engineering

Master Course in Computer Engineering

Course program and exam rules (academic year 2024/2025) NEW!!

LAIB exam rules NEW!!

Schedule of laboratory activities c/o LABINF / ACSLAB (academic year 2024/2025): NEW!!

19/03/2025: first computer practical
26/03/2025: second computer practical
02/04/2025: third computer practical
07/05/2025: fourth computer practical
28/05/2025: fifth computer practical
04/06/2025: sixth computer practical

Special announcements (academic year 2024/2025): NEW!!

First lecture on 24/02/2025, from 08:30 to 11:30, c/o Classroom 9T

Lecture Notes / Slides

1. Essentials of probability theory (PDF file)

Random experiment and random source of data; Probability distribution and density functions; Vector random variables; Correlation coefficient and correlation matrix; Normal random variables

2. Estimation theory (PDF file)

Estimation problem; Estimator probabilistic characteristics; Cramér-Rao inequality; Estimation methods: Least-Squares,Weighted Least-Squares, Maximum Likelihood, Bayesian, Recursive Bayesian

3. Set Membership estimation theory (PDF file)

Unknown but bounded errors; Estimate Uncertainty Set (EUS) and Intervals (EUI); Feasible Parameter Set (FPS) and Parameter Uncertainty Intervals (PUI); Optimal estimates

4. Kalman filter theory (PDF file)

Filtering problem; One-step and Multi-step Kalman predictor; Kalman filter; Steady-state Kalman predictor; Nonlinear Kalman filtering: Linearized and Extended Kalman predictor

5. System identification (PDF file)

Model classes; Prediction-Error method (PEM): Least-Squares and Recursive Least-Squares; Anderson's whiteness test; Model stucture selection and validation criteria: FPE, AIC, MDL

6. System identification MATLAB Toolbox (PDF file)

Introduction to System Identification Toolbox under MATLAB and the Graphical User Interface GUI

7. Nonlinear system identification (PDF file)

Parametric approach; Fized and tunable basis functions; Parametric models; Nonlinear regression systems

0. Lecture Notes, a.a. 2012/13 (PDF file)

Notes handwritten by the teacher during the lessons, to be carefully checked

Official formulary (PDF file, definitive version)

The only formulary allowed to be used during the exam (please, send an email message to the teacher for asking to add possible missed useful formulas or for warning for possible typos)

Laboratory Lecture Notes / Slides and Auxiliary Lectures

Lab #1: parametric estimation from data of a position transducer model (PDF file)

MATLAB data file sensor.mat
Main steps and proposed solution under MATLAB using the statistical approach (file Lab1.m , Lab1.html and Lab1P.pdf )

Lab #2: parametric estimation from data of a position transducer model (PDF file)

MATLAB data file sensor.mat
Main steps and proposed solution under MATLAB using the Set Membership approach (file Lab2.m , Lab2.html and Lab2P.pdf )

Lab #3.1: parametric estimation from data of a resistance (PDF file)

MATLAB data files resistor_data_1.mat and resistor_data_2.mat
Main steps and proposed solution under MATLAB for resistor_data_1.mat (file Lab3_E1A.m , Lab3_E1A.html and Lab3_E1AP.pdf )
Main steps and proposed solution under MATLAB for resistor_data_2.mat (file Lab3_E1B.m , Lab3_E1B.html and Lab3_E1BP.pdf )

Lab#3.2: parametric estimation from data of an auto-regressive model (PDF file)

MATLAB data file AR_data.mat
Main steps and proposed solution under MATLAB (file Lab3_E2.m , Lab3_E2.html and Lab3_E2P.pdf )

Lab #4: design and simulation of different Kalman predictors and filters (PDF file)

MATLAB data file data.mat
Main steps and proposed solution under MATLAB (file Lab4.m , Lab4.html and Lab4P.pdf )

Lab #5.1: parameter convergence in ARX model identification (PDF file)

Main steps and proposed solution under MATLAB (file Lab5_1.m , Lab5_E1.html and Lab5_E1P.pdf)

Lab #5.2: hair dryer model identification using the MATLAB GUI (PDF file)

MATLAB data file dryer2.mat
Main steps and proposed solution under MATLAB without using the GUI (file Lab5_2.m , Lab5_E2.html and Lab5_E2P.pdf )

Lab #6: model identification for a water heater (PDF file)

MATLAB data file heater.mat
Main steps and proposed solution under MATLAB using linear models (file Lab6_1.m , Lab6_1.html and Lab6_1.pdf ))
Main steps and proposed solution under MATLAB using nonlinear models based on neural networks (file Lab6_2.m , Lab6_2.html and Lab6_2.pdf )

System Identification Toolbox User's Guide (for use with MATLAB 5.3 and higher) (PDF file)

Neural Network Based System Identification Toolbox version 2 (for use with MATLAB 5.3 and higher) by M. Noergaard (ZIP file)

Support material on MATLAB

MATLAB Primer - III edition, by K. Sigmon (1993): older but simpler introduction to MATLAB, based on MATLAB 4.1
MATLAB Primer - VII edition, by T. A. Davis and K. Sigmon (2005): newer and much more complete introduction to MATLAB, based on MATLAB 7.0
Introduzione al programma MATLAB, by M. Canale: introductory slides on MATLAB (in Italian), based on MATLAB 5 and 6

Exams

Sample examination paper (PDF file)
- MATLAB data files data1.mat and data2.mat
- Introduction and proposed solution under MATLAB (file MICexam_es1.m , MICexam_es1.html and MICexam_es1.pdf ; file MICexam_es2.m )
- Report of the solution (file MICexam_report.pdf)

Academic Year 2023/24, Exam #1: June 27, 2024, at 14:00, c/o LAIB 3 - 4: results
Academic Year 2023/24, Exam #2: July 25, 2024, at 11:00, c/o LAIB 1 - 4: results
Academic Year 2023/24, Exam #3: September 18, 2024, at 14:00, c/o LAIB 1 - 4: results
Academic Year 2023/24, Exam #4: February 12, 2025, at 11:00, c/o LAIB 3: results NEW!!

Past exams

Academic Year 2022/23, Exam #1: June 29, 2023, at 08:30, c/o LAIB 1 - 4: results
Academic Year 2022/23, Exam #2: July 17, 2023, at 10:00, c/o LAIB 1 - 4: results
Academic Year 2022/23, Exam #3: September 18, 2023, at 11:00, c/o LAIB 3 - 4: results
Academic Year 2022/23, Exam #4: February 22, 2024, at 08:30, c/o LAIB 4: results

Academic Year 2021/22, Exam #1: June 22, 2022, at 15:00, c/o LAIB 3 - 4: results
Academic Year 2021/22, Exam #2: July 18, 2022, at 13:30, c/o LAIB 1 - 4: results
Academic Year 2021/22, Exam #3: September 12, 2022, at 10:00, c/o LAIB 3 - 4: results
Academic Year 2021/22, Exam #4: November 7, 2022, at 14:30, c/o LAIB 4: results
Academic Year 2021/22, Exam #5: February 15, 2023, at 08:30, c/o LAIB 4: results

Academic Year 2020/21, Exam #1: July 02, 2021, at 08:30, c/o VLAIB: results
Academic Year 2020/21, Exam #2: July 22, 2021, at 08:30, c/o VLAIB: results
Academic Year 2020/21, Exam #3: September 10, 2021, at 08:30, c/o VLAIB: results
Academic Year 2020/21, Exam #4: November 10, 2021, at 08:30, c/o VLAIB: results
Academic Year 2020/21, Exam #5: January 18, 2022, at 08:30, c/o AULA 5D: results

Academic Year 2019/20, Exam #1: July 02, 2020, at 14:00, c/o VLAIB: results
Academic Year 2019/20, Exam #2: July 23, 2020, at 14:00, c/o VLAIB: results
Academic Year 2019/20, Exam #3: September 04, 2020, at 08:00, c/o VLAIB: results
Academic Year 2019/20, Exam #4: February 12, 2021, at 08:30, c/o VLAIB: results

Academic Year 2018/19, Exam #1: July 05, 2019, at 13:30, c/o LAIB 3 - 4: results
Academic Year 2018/19, Exam #2: July 18, 2019, at 13:30, c/o LAIB 3 - 4: results
Academic Year 2018/19, Exam #3: September 20, 2019, at 08:30, c/o LAIB 4: results
Academic Year 2018/19, Exam #4: February 06, 2020, at 08:30, c/o LAIB 2B: results

Teacher: prof. Michele Taragna
email: michele.taragna [at] polito.it
office phone: (+39) 011-090-7063
office fax: (+39) 011-090-7099

Last update of this page: 20/02/2025, 16:00 (M.T.)