Optimalno upravljanje nelinearnim sustavima primjenom neuronskih mreža

Kasać, Josip (1998) Optimalno upravljanje nelinearnim sustavima primjenom neuronskih mreža. = Optimal Control of Nonlinear Systems Using Neural Networks. Scientific master's thesis , Sveučilište u Zagrebu, Fakultet strojarstva i brodogradnje, UNSPECIFIED. Mentor: Novaković, Branko.

[img]
Preview
Text
22_02_2006_Kasac_Magisterij.pdf - Published Version Jezik dokumenta:Croatian

Download (3MB) | Preview

Abstract (Croatian)

U radnji je prikazan izvod numeričkoga logoritma za optimalno upravljanjene linearnim multivarijabilnim sustavima sa ograničenjima varijabli stanja i upravljanja. Izvod algoritma zasnovan je na BPTT (backpropagation-through-time) algoritmu koji se primjenjuje kao algoritam učenja za dinamičke neuronske mreže. Izveden je također algoritam za vremenski optimalno upravljanje koji je zasnovan na svojstvima kaznenih funkcija za ograničenja varijabli stanja i upravljanja. Dobiveni algoritmi primijenjeni su na razne probleme optimalnog upravljanja robotom s dva stupnja slobode. Najprije je razmatran problem vremenski optimalnog upravljanja robotom s ograničenjima varijabli upravljanja. Zatim je razmatran isti problem s dodatnim uvjetom; zaobilaženjem prepreke, odnosno ograničenjima varijabli stanja. Nakon toga razmatra se problem optimalnog upravljanja kooperativnim radom dva robota. Problem se sastoji u zajedničkom prenošenju krutog tereta s jednog mjesta na drugo uz ograničenja varijabli upravljanja, te uz uvjet međusobnog izbjegavanja sudara i održavanja konstantne udaljenosti između prihvatnica manipulatora. Nakraju je prikazan izvod algoritma za optimalno upravljanje s povratnom vezom za nelinearne multivarijabilne sustave s ograničenjima varijabli upravljanja. Izvod algoritma također je zasnovan na principu BPTT algoritma.

Abstract

The master's thesis presents the derivation of the numerical algorithm for optimal control of nonlinear multivariable systems with limited state and control vectors. The algorithm derivation is based on the BPTT (backpropagation-through-time) algorithm which is used as a learning algorithm for dynamic neural networks. Presented is also the derivation of the algorithm for time optimal control which is based on the characteristics of penalty functions for constraints of state and control vectors. The derived algorithms are used for various problems concerning optimal robot control with two degrees of freedom. The thesis first deals with the problem of time optimal robot control with control vector constraints. It also considers the same problem with an additional condition - avoidance of obstacles, that is, constraint of state vectors. The thesis further deals with the problem of optimal control of cooperative work of two robots, that is, of joint transfer of solid material from one place to another including control vector constraint and conditions of mutual avoidance of clashes and maintaining of constant distance between the hands of the robot. Derivation of the optimal control algorithm with a feedback for nonlinear multivariable systems with control vector constraint is presented at the end. This derivation is also based on the BPTT algorithm.

Item Type: Thesis (Scientific master's thesis)
Uncontrolled Keywords: neuronske mreže; optimalno upravljanje; upravljanje robotom; nelinearni sustavi
Keywords (Croatian): neural networks; optimal control; robot control; nonlinear systems
Divisions: 900 Department of Robotics and Production System Automation > 910 Chair of Engineering Automation
Date Deposited: 22 Sep 2014 18:00
Last Modified: 16 Oct 2015 13:20
URI: http://repozitorij.fsb.hr/id/eprint/37

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

Nema podataka za dohvacanje citata