Institutional Repository of Vilnius University
Sprendimų priėmimo algoritmo realizavimas IoT pagrindu sukurtų robotų judėjime /
Abstract
dc:descriptionThis paper analyzes implemantation of IoT robotics decision-making alghoritm. It is argued that, the potential of IoT technology has not been fully exploited. However, the synergy of the two technologies (IoT and robotics) could bring a lot of significant changes into our everyday life. Nowadays, many systems are hard coded to perform certain actions, they operate in accordance with a strictly defined algorithm and hence lack flexibility and robustness. For this reason, this paper presents the machine learning concept that gives systems the ability to learn. The first chapter describes the IoT concept and where IoT is being used. Next, the synergy of IoT and robotics is discussed. Then the new term – IoRC (Internet Of Robotics) is introduced. The second chapter explains machine learning concept. Author presents three types of machine learning: learning with supervisor, learning without supervisor and reinforcement learning. After that some machine learning methods are presented: artificial neural network, deep learning, q – learning, genetic alghoritms, Bayes networks, decision trees, similiarity learning and support vector machines. Author gives pros and cons of all presented methods. In the third chapter author describes the architecture of the IoT robot prototype – parts it consists of and the way it performs. After that, in order to implement IoT robotics decision–making alghoritm, author chooses an artifical neural network method. Author describes neural network structure and training dataset. Then author describes system‘s architecture and functionality.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2017
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Filon, Sergej,
- Contributors dc:contributor
-
- Uosis, Karolis
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD23166363&prefLang=en_US