JournalTOCs

Facta Universitatis, Series : Automatic Control and Robotics
Number of Followers: 1

This is an Open Access Journal

Open Access journal
ISSN (Print) 1820-6417 - ISSN (Online) 1820-6425
Published by U of Niš

[11 journals]

CHATBOT APPLICATION AS SUPPORT TOOL FOR THE LEARNING PROCESS OF BASIC
CONCEPTS OF TELECOMMUNICATIONS AND WIRELESS NETWORKS
- Authors: Mihailo Jovanović, Kristijan Kuk, Vladica Stojanović, Edis Mehić
  Pages: 057 - 073
  Abstract: There are several applications for Chatbots in education, as well as their contributions to mentoring in the learning process. Bots can assist teachers with staying up to date on new standards and evaluation methodologies. Bots can assist students in understanding tough subjects in a way that makes it appear as if they are being taught by another person. Chatbots serve as virtual assistants in the educational setting, improving efficiency or answering frequently asked questions. In this case, we work on the premise of investigating the potential of Chatbots as analytical tools for analyzing preferred types of learning material in a mobile learning environment, which leads to the acquisition of a proper level of knowledge on the topics of telecommunication and wireless networks.
  PubDate: 2024-02-06
  DOI: 10.22190/FUACR230719005J
  Issue No: Vol. 22, No. 2 (2024)
INTELLIGENT COMPUTER VISION SYSTEM FOR SCORE DETECTION IN BASKETBALL
- Authors: Ivan Ćirić, Maša Milošević, Nikola Ivačko, Aleksandra Cvetković, Milan Pavlović, Dušan Krstić
  Pages: 075 - 085
  Abstract: Development of an intelligent computer vision system for Smart IoT basketball training and entertainment includes the development of a range of various subsystems, where score detection subsystem is playing a crucial role. This paper proposes the architecture of such a score detection subsystem to improve reliability and accuracy of the RFID technology used primarily for verification purposes. Challenges encompass both hardware-software interdependencies, optimal camera selection, and cost-effectiveness considerations. Leveraging machine learning algorithms, the vision-based subsystem aims not only to detect scores but also to facilitate online video streaming. Although the use of multiple cameras offers expanded field coverage and heightened precision, it concurrently introduces technical intricacies and increased costs due to image fusion and escalated processing requirements. This research navigates the intricate balance between achieving precise score detection and pragmatic system development. Through precise camera configuration optimization, the proposed system harmonizes hardware and software components.
  PubDate: 2024-02-06
  DOI: 10.22190/FUACR230822006C
  Issue No: Vol. 22, No. 2 (2024)
REDUCTION OF GIBBS PHENOMENON IN EOG SIGNAL MEASUREMENT USING THE MODIFIED
DIGITAL STOCHASTIC MEASUREMENT METHOD
- Authors: Jelena Đorđevic Kozarov, Milan Simić, Milica Stojanović, Dragan Živanović
  Pages: 087 - 101
  Abstract: The method of digital stochastic measurement is based on stochastic analog-to-digital conversion, with a low-resolution A/D converters and accumulation. This method has been mainly tested and used for the measurement of stationary signals. This paper presents, analyses and discusses a simulation model development for an example of electrooculography (EOG) signal measurement in the time domain. Tests were carried out without adding a noise, and with adding a noise with various level of signal-to-noise ratio. For these values of signal-to-noise ratio, the mean and maximal relative errors are calculated and the significant influence of Gibbs phenomenon is noticed. In order to eliminate Gibbs phenomenon and decrease measurement error, a modified stochastic digital measurement method with overlapping measurement intervals has been developed and applied. On the basis of obtained results, the possibility of design and realization of an instrument with sufficient accuracy benefiting from the hardware simplicity of the method has been formulated. Also, the idea for the future research for developing a simulation model with a lower sampling frequency and implementing the proposed method is outlined.
  PubDate: 2024-02-06
  DOI: 10.22190/FUACR231102007D
  Issue No: Vol. 22, No. 2 (2024)
NON-INTRUSIVE LOAD MONITORING USING CURRENT HARMONIC VECTORS AND ADAPTIVE
FEATURE SELECTION
- Authors: Srđan Đorđević, Milan Simić
  Pages: 103 - 113
  Abstract: The non-intrusive load monitoring method presented in this paper uses changes in current harmonic vectors to identify the operational state of appliances. The algorithm based on this feature has low complexity, but it may suffer from an information loss caused by a random fluctuation of the current harmonic vectors. In order to deal with this problem, we propose the algorithm which includes a stage which identifies and select a subset of relevant features in the set of available appliance features. The proposed load disaggregation algorithm is demonstrated through experiments on a representative set of household appliances.
  PubDate: 2024-02-06
  DOI: 10.22190/FUACR231123008D
  Issue No: Vol. 22, No. 2 (2024)
ON A DEFUZZIFICATION PROCESS OF FUZZY CONTROLLERS
- Authors: Dušan Simjanović, Marko Milojković
  Pages: 115 - 130
  Abstract: In this paper, innovations in the field of automatic control systems with fuzzy controllers have been considered. After a short introduction on fuzzy controllers, four different ways of a defuzzification process were introduced, and verified on the simulation of nuclear reactor fuzzy controller. The default Matlab fuzzy toolbox solution is timely most demanding, while two solutions based on the defuzzification on trapezoidal fuzzy numbers have the advantage in the process of crisp numbers calculation. Also, a solution based on the determination of the line dividing the obtained polygon into two parts of equal areas is presented.
  PubDate: 2024-02-06
  DOI: 10.22190/FUACR231202009S
  Issue No: Vol. 22, No. 2 (2024)

HOME > Browse the 73 Subjects covered by JournalTOCs
Subject	Total Journals