Feb 23, 2020   3:20 p.m. Roman, Romana
Academic information system

Final theses


Basic information

Basic information about a final thesis

Type of thesis: Dissertation thesis
Thesis title:Nonlinear model predictive control of underactuated mechatronic systems
Written by (author): doc. Ing. Martin Gulan, PhD.
Department: Institute of automation, measurement and applied informatics (FME)
Thesis supervisor: prof. Ing. Boris Rohaľ-Ilkiv, CSc.
Opponent 1:prof. Ing. Štefan Kozák, PhD.
Opponent 2:prof. RNDr. Ing. Tomáš Březina, CSc.
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:Nonlinear model predictive control of underactuated mechatronic systems
Summary:Nonlinear model predictive control (NMPC) is currently one of the most powerful tools that can be used to address nonlinear control problems, which in fact arise in nearly all engineering applications, often due to nonlinear dynamics. Many predictive control applications are, however, still based on the use of linear or linearized models, which facilitates the identification and optimization tasks, and leads to a reasonably good behaviour in the neighbourhood of the respective operating point. After decades of research, linear MPC is nowadays considered to be a mature technique for linear and rather slow systems like the ones usually encountered in the process industry. However, more complex systems, such as nonlinear or very fast processes have still called for the research in the field of nonlinear MPC. Solving a nonlinear optimal control problem comes usually with rapidly increased computational complexity, which thanks to the enormous progress achieved recently in nonlinear optimization is now possible to be addressed even in case of processes with very fast sampling times. Nevertheless, the number of NMPC applications is still limited, in particular for fast systems, which makes this topic very actual and relevant. The main aim of this work is to contribute with a new practical implementation of NMPC to an underactuated mechanical system (UMS) with fast dynamics, namely the so-called Pendubot. The thesis gradually introduces the NMPC problem formulation, briefly describes the associated nonlinear programming and reviews computationally efficient strategies to tackle the optimal control problem in view of fast real-time mechatronic applications. The practical part of this study starts off with providing a necessary background on UMSs with emphasis on the Pendubot system. An overview of existing and commonly used Pendubot control tasks and techniques is given as well. This is followed by deriving the nonlinear model of the system. Its physical parameters are based on a real-world laboratory model assembled for the purpose of practical tests. After describing the hardware setup, the software approach is presented. Central to the control scheme are the algorithms proposed in the theoretical part. These are appropriately integrated within the framework tailored for specific control tasks. Its functionality and performance are experimentally verified, with the objective of tracking the unstable equilibrium positions of the Pendubot system. The control sheme is moreover augmented by a nonlinear estimation scheme to obtain the unmeasured states, a nonlinear friction model, and parallelization of particular algorithmic routines to speed up the online execution. Finally, the thesis concludes with a summary of achieved results and its contribution, as well as an outline of potential objectives of further research.
Key words:nonlinear model predictive control, underactuated mechanical system, real-time nonlinear optimization, fast NMPC, Pendubot

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