ISS0023 Intelligent Control Systems

Course description




  • Autumn



Awarded ECTS points: 



The course gives an overview of :

  • complex systems modeling and control methods and their applications in design of reliable control systems. 
  • artificial intelligence methods (artificial neural networks, fuzzy systems, genetic algorithms) based systems identification and control techniques and their applications in development of intelligent control systems. 
  • artificial intelligence methods based classification and recognition techniques and their applications.
Topics include:
  • Nonlinear systems, Principes of nonlinear systems identification and control; 
  • Adaptive control systems; 
  • Artificial neural networks. Structures of artificial neural networks and training algorithms; 
  • Artificial neural networks based identification of nonlinear systems; 
  • Artificial neural networks based control of nonlinear systems; 
  • Self-learning neural networks; 
  • Artificial neural networks based image recognition and pattern classification;
  • Fractional Order Modelling and Control; 
  • Dynamic feedback linearization based control of nonlinear systems; 
  • Genetic algorithms and their applications for identification and control of nonlinear systems.


Lab reports:

6 labs = 6 reports
Each report gives up to 1 point.
Each report has to be presented during 2 weeks after the lab!
Later presented reports (before December 18) – multiplied by coefficient 0.8
After December 18 – coefficient 0.6
5 best report will give up to 5 points.
Exam prerequisites:
Course ISS0023 is declared (included into Your semester plan),
Laboratory trainings are performed,
Reports are presented and accepted (written report on each laboratory training).
Estimation criterion:
Grades are based on the report of the final small practical project (design of a control system).
Exam  - up to 72 hours
Small practical project – design of a control system according to given control criteria;
Simulation of the control system;
Analysis of results and writing a report.
2 tasks – each one gives maximum 5 points.
Average of 2 exam tasks and labs = YOUR COURSE GRADE
„5“ - student's knowledge is excellent, his/her answers are clear and complete, accurate in details,
deliberative and individual. Independent thinking. Student is capable of analyzing the problem and possible solutions, proposing his own solution and prooving its efficiency.
„4“ - student's knowledge is very good, his/her answers are clear and complete, accurate in details,
but less individual views. Student is capable of analyzing the problem, finding a suitable solution proving its efficiency.
„3“ - student's knowledge is good, his/her answers are clear, but there are a few errors in the
discussion and lack of personality and individual point of view. Student is capable of analyzing the problem and applying standard methods for is solution.
„2“ - student's knowledge is satisfactory, but all the answers are not clear and student makes more
errors (but not errors in basic concepts) and lack of personality and individual point of view.
„1“ - Student responses are weak, there are major mistakes in arguments (not knowing the content,
mistakes in concept), and student needs constant help and guidance in formulating responses to

January 2020

Mon Tue Wed Thu Fri Sat Sun


Titlesort descending Published Short description Files
Adaptive control 08.09.2016 Slides - part 1 adaptive_control_1_2016.ppt adaptive_control_1_2016.pdf
Adaptive control 08.09.2016 Slides - part 2 adaptive_control_2_2016.ppt adaptive_control_2_2016.pdf
Fractional-order Modeling and Control 05.12.2016 Slides for lecture: Fractional-order Modeling and Control ISS0023_FracCalc.pdf
Fuzzy Logic based Modeling and Control (S. Astapov) 28.11.2016 Lecture slides on fuzzy logic based modeling and control Sergei_Astapov_Fuzzy_Control_lecture_slides.pdf
Genetic algorithms in control system design 10.11.2016 Lecture slides on genetic algorithms in control system design Genetic2016.pdf Genetic2016.pptx
Introduction to artificial neural networks 23.09.2016 Introduction to neural networks, neural network structure and mathematical model - lecture slides NN2016_part1.ppt NN2016_part1.pdf
Lecture 1 12.09.2016 08.09.2016 Introductory lecture ISS0023_introduction_2016.pptx ISS0023_introduction_2016.pdf
NN supervised learning 23.09.2016 Training of NNs - lecture slides NN2016_part2.ppt NN2016_part2.pdf
NN-based identification and control - lecture slides 10.10.2016 Neural Networks based identification and control of nonlinear systems, supevised and unsupervised learning NN2016_part3.ppt NN2016_part3.pdf
NN-structures in control 10.11.2016 Lecture slides on NN-based control and custom network sttructures NN_structures2016.pdf NN_structures2016.pptx


Titlesort descending Published Short description Files
Adaptive control 14.09.2016 Nonlinear Adaptive Control adaptju1.mdl
Adaptive control 14.09.2016 Model Reference Control aw2d.m aw_dsim2.mdl aw_dsim44.mdl

Laboratory works

Titlesort descending Published Short description Files
Dynamic linearization based control 22.11.2017 Example of NN-ANARX model based control - liquid level control of a laboratory multi-tank system. Tank3_NN_control2013a.mdl
Fractional-order Modeling and Control 15.12.2016 Slides and materials for laboratory work: Fractional-order Modeling and Control iss0023_fraccalc_lab.pdf refgen_example.m
Fuzzy Logic Control 29.11.2016 Files and instructions for the fuzz logic laboratory work Sergei_Astapov_Fuzzy_Control_Lab.pdf fuzzy
Image recognition lab 01.11.2016 Example: Character recognition
Introduction to Neural Networks 02.10.2016 NN training data - two inputs, one output + function used to generate the data (you may use this file to validate the result) nn_data.mat answer.m
Neural Networks based control 17.10.2016 Example of NN-based control system design in MATLAB/Simulink lab_NN2_2016.pdf


Titlesort descending Published Short description Files
E. Rüstern Adaptiivjuhtimine (In Estonian, Eesti keeles) 08.09.2016 Prof. Ennu Rüsterni materjal - ülevaade adaptiivjuhtimidest (Eesti keeles) ISS0022_ülevaade_adaptiivsüsteemidest.pdf ISS0022_ülevaade_adaptiivsüsteemidest_2.pdf
Fuzzy control 08.09.2016 Book by K. Passino FCbook.pdf
Training Feedforward Networks with the Marquardt Algorithm 08.09.2016 Paper by Martin T. Hagan and Mohammad B. Menhaj in IEEE TRANSACTIONS ON NEURAL NETWORKS 00329697.pdf
Titlesort descending Published Short description Files
Exam, January 2017 16.01.2017 You are given 4 tasks. Please solve 2 tasks. On your choice you have 4 options to choose from. Option 1: solve Tasks 1 and 4; Option 2: Tasks 2 and 3; Option 3: Tasks 1 and 3; Option 4: Tasks 3 and 4. You have 3 days to solve the tasks . Exam task.pdf
Home work 1 (NN training) 02.10.2016 By training an Artificial Neural Network (ANN), identify relation between 3 inputs and one output. homework1.mat homework1_task.pdf
Home work 2 (NN based control) 01.11.2016 Design Neural Network based nonadaptive and adaptive control systems for a black box system given in the file. Present a report containing all design steps, simulation results (with disturbances) and analysis. NB! REPORT FORMAT *.DOC(X) or *.PDF nonlinear_system.slx
Home work 3 (pattern recognition) 01.11.2016 Design a supervised and an unsupervised NN based systems for recognition of numbers given in file all_numbers.m. Proof experimentally that the system recognizes numbers correctly. all_numbers.m
Home work 4 (NN-ANARX model based control) 15.11.2016 Design a NN-ANARX model based control system for a Jacketed CSTR (Continuous Stirred Tank Reactor) from the third laboratory work (Neural Network based control) j_CSTR_data.mdl
Home work 5 (Fuzzy Logic) 01.12.2016 Design Fuzzy Logic Controllers and present a report according to the given instructions Sergei_Astapov_Fuzzy_Control_Lab.pdf