Sep 19, 2019   8:20 a.m. Konštantín, pamätný deň - Deň prvého verejného vystúpenia Slovenskej národnej rady
Academic information system

Persons at STU


This page displays all publicly accessible information about the desired person. Some information about the person's occupation and offices may be hidden.

Ing. Martin Klaučo, PhD.
Identification number: 50920
University e-mail: martin.klauco [at] stuba.sk
 
Výskumný pracovník s VŠ vzdelaním - Department of Information Engineering and Process Control (IIEAM FCFT)
Vedecký pracovník KS II.b. CSc.,PhD. - Department of Information Engineering and Process Control (IIEAM FCFT)

Contacts     Graduate     Final thesis     Projects     
Publications     Placements     Supervised theses     

Basic information

Basic information about a final thesis

Type of thesis: Diploma thesis
Thesis title:Machine Learning Approaches Applied to Generation of Explicit Control Laws
Written by (author): Ing. Karol Kiš
Department: Department of Information Engineering and Process Control (IIEAM FCFT)
Thesis supervisor: Ing. Martin Klaučo, PhD.
Opponent:Ing. Juraj Oravec, PhD.
Final thesis progress:Final thesis was successfully defended.


Additional information

Additional information about the final thesis follows. Click on the language link to display the information in the desired language.

Language of final thesis:English

Slovak        English

Title of the thesis:Machine Learning Approaches Applied to Generation of Explicit Control Laws
Summary:The aim of this master thesis is to design a control law in the form of a neural network, whose behaviour is similar to an explicit model predictive control (MPC). The main advantages of explicit MPC are fast and efficient implementation, but on the other hand, the main disadvantage of explicit MPC is fixed structure of the original MPC problem, i.e. linear or quadratic cost function functions and linear constraints. However, the neural network control law must be trained on the basis of input data, which in this case are obtained by sequentially solving the MPC problem for different initial conditions. The advantage of this approach is the possibility to get control action not only for different initial states but also for different settings of prediction horizon or weight matrices in the cost function since not only states but also weight matrices can be input parameters of the neural network control law. This, in fact, creates a suboptimal tunable controller in explicit form.
Key words:machine learning, neural networks, approximation of model predictive control

Display and download files

To display the final thesis assignment form click on the Display the final thesis assignment form icon. The following icons - Final thesis, Thesis appendices, Supervisor's review, Opponent's review - relate to the final thesis and can be downloaded. They could be displayed on condition they have been inserted and are available publicly.

Display the assignment form

Parts of thesis with postponed release:

Final thesis (final thesis appendices) unlimited
Reviews for final thesis unlimited