Selected Chaps From Automatic Control
FEKT-DKA-AM1Acad. year: 2020/2021
The subject focuses at studies of the methods of design of advanced control algorithms including classical control structures as well as algorithms of robust, adaptive and predictive control. Attention is also paid to information processing algorithms and state observers for realization of so called virtual sensors and algorithms of sensorless control. Traditional methods for systems control and processing information are complemented by artificial intelligence-based approaches. In addition to the theoretical aspects of the given topic, sample algorithms for advanced drives, mechatronic systems and mobile robots are also solved.
Learning outcomes of the course unit
The graduate should be able to design and tune sophisticated control systems with elements of AI
Principle of continuous control theory. Principle of discrete control theory. State control.
Recommended optional programme components
Recommended or required reading
Skogestad S., Postletwaite I.: Multivariable Feedback Control, John Wiley & Sons, 2007. (EN)
Slotine J. J. E., Li W.: Applied Nonlinear Control, Prentice-Hall, 1991 (EN)
Gu D.-W., Petkov P. H.: Robust Control Design with MATLAB, Springer, 2013 (EN)
Russell S., Norvig P.: Artificial Intelligence a Modern Approach. Prentice Hall 2010 (EN)
Voseelman G., Mass H-G. Airborne and Terrestrial Laser Scanning, CRC Press, 2010 (EN)
Goodwin G.C., Seron M.M. , Doná J.A. Constrained Control and Estimation, Springer, 2005 (EN)
Hermann R. ,Krener A., Nonlinear controllability and observability, IEEE Transactions on Automatic Control, vol. 22, no. 5, pp. 728–740, 1977 (EN)
Planned learning activities and teaching methods
Techning methods include lectures and selfstudy. Course is taking advantage of e-learning (Moodle) system
Assesment methods and criteria linked to learning outcomes
Elaboration of 2-3 projects (30 points). Final exam (70 points).
Language of instruction
1. Modern approaches in automatic control
2. Robust control of dynamic systems with uncertainty considerations
3. Specific Adaptive Control Problems
4. State controller as the basic structure for model based predictive control
5. State observability theory of nonlinear dynamic systems
6. Principles of using virtual sensors for sensorless control, example of control applications for actuators with asynchronous and synchronous motors
7. Artificial neural networks (NS) and their learning methods.
8. Control theory and artificial intelligence, NS-based control algorithms.
9. Identification of systems using NS, adaptive optimal controller based on NS identification.
10. Modern methods of autonomous outdoor and indoor self-localisation.
11-12. Advanced 3D mapping - sensors, data fusion methods, data representation, practical use.
The objective of the course is to develop students' knowledge and competences in the field of design of advanced control algorithms and algorithms for information processing based on classical mathematical approaches as well as artificial intelligence, including possible advanced applications such as mobile robotics.
Classification of course in study plans
- Programme DKA-KAM Doctoral, any year of study, winter semester, 4 credits, compulsory
- Programme DKA-EKT Doctoral, any year of study, winter semester, 4 credits, compulsory-optional
- Programme DKA-MET Doctoral, any year of study, winter semester, 4 credits, compulsory-optional
- Programme DKA-SEE Doctoral, any year of study, winter semester, 4 credits, compulsory-optional
- Programme DKA-TLI Doctoral, any year of study, winter semester, 4 credits, compulsory-optional
- Programme DKA-TEE Doctoral, any year of study, winter semester, 4 credits, compulsory-optional
Type of course unit
39 hours, optionally
Teacher / Lecturer