• Aviation, Semester 3, position 4
• Automatic Control, Semester 3, position 4
• Weapon Systems, Semester 3, position 4
• Mechanics, Semester 3, position 4

• Introduction to methods for the analysis and design of intelligent control systems. • Gaining practical knowledge of several of the main techniques of intelligent control and an introduction to some promising research directions. • Use of the computer for simulation and evaluation intelligent control systems.

On successful completion of the course the students should be able to: • Understand of the functional operation of a variety of intelligent control techniques. • Understand of control-theoretic foundations. • Carry out synthesiis and analysis of intelligent control systems based on combinations of various theories: simulation, neural networks, fuzzy systems, genetic algorithms, biologically inspired algorithms, etc. • Use of the computer for simulation and evaluation intelligent control systems through Matlab/Simulink software, as and practical realization of control algorithms on various control plants using programming software Matlab.

Introduction of intelligent control. Conventional and intelligent control. Fundamentals of artificial neural networks: architecture, classification, basic properties. Neural network learning rules: principles, Hebbian learning law, Widrow-Hoff learning, delta rule. Single layer feedforward neural networks, perceptron, linear network. Multilayer feedforward networks with backpropagation error. Radial basis function neural networks. Support vector machines. Dynamical neural networks. Neural networks for nonlinear dynamic system modelling and identification. Neural networks for control: direct control and indirect control, direct inverse control,model predictive control. Deep learning and convolutional neural network. Biologically-inspired and evolutive algorithms.

PA: Practical work includes computational exercises that follow the content of course. PL: Practice and experiments: computer applications in simulation and evaluation of intelligent control systems, as well as their practical realization using Matlab/Simulink for control different plants within a modular educational real-time control system (double inverted pendulum, coupled tanks experiment, DC servo motor, heat flow exepriment).

Defined by curriculum of the study programme.

• Modular educational real time control system with various control plants (DC servo motor, inverted pendulum, double inverted pendulum, heat flow experiment, coupled water tanks experiment), with acquisition hardware and software. • PC and PC Embedded controllers, Siemens Simatic PLC, National Instruments controllers. • Installation for control system testing and acquisition of electrical variables. • Intelligent Control Systems Laboratory, Control Systems Laboratory.

Total assigned hours: 75

New material: 20
Elaboration and examples (recapitulation): 10

Auditory exercises: 15
Laboratory exercises: 10
Calculation tasks: 0
Seminar paper: 5
Project: 0
Consultations: 0
Discussion/workshop: 0
Research study work: 0

Review and grading of calculation tasks: 0
Review and grading of lab reports: 0
Review and grading of seminar papers: 6
Review and grading of the project: 0
Test: 4
Test: 0
Final exam: 5

Activity during lectures: 5
Test/test: 50
Laboratory practice: 5
Calculation tasks: 0
Seminar paper: 0
Project: 0
Final exam: 40
Requirement for taking the exam (required number of points): 30

Radiša Jovanović, Introduction to Neural Networks and Fuzzy Systems, Lecture notes, Faculty of Mechanical Engineering, 2023.; Simon Haykin, Neural Networks and Learning Machines, Vol. 3. Upper Saddle River, NJ, USA:: Pearson, 2009.; Radiša Jovanović, Mаtlab and Simulink in Automatic Control, Faculty of Mechanical Engineering, Belgrade, ISBN 978-86-7083-896-3, 2021 (in Serbian).