Research and implementation of positioning algorithms for mobile robotic systems indoors
Students Name: Holovanchuk Danylo Ihorovych
Qualification Level: magister
Speciality: Information Technology Design
Institute: Institute of Computer Science and Information Technologies
Mode of Study: full
Academic Year: 2025-2026 н.р.
Language of Defence: ukrainian
Abstract: Holovanchuk D.I., Belej O.I. Research and implementation of positioning algorithms for mobile robotic systems indoors. Master’s thesis. - Lviv Polytechnic National University, Lviv, 2025. Extended abstract Location information is used not only to solve complex scientific and industrial problems, but also in everyday life. It helps people navigate, find interesting places, and plan routes. For geographers, the relationship between the object of study and its spatiotemporal location is crucial. Coordinated data forms the basis of spatial modeling in geographic information systems (GIS) and is used to index data in search engines. Many positioning methods and technologies have been developed over the millennia. However, the advent of the Global Navigation Satellite System (GNSS) revolutionized positioning. The global coverage, high accuracy, and speed of satellite positioning made this possible. Today, with the expansion of the mobile device market, the use of GNSS technology is becoming increasingly widespread. The spread of portable receivers in smartphones has not only contributed to the widespread use of location-based services (LBS) but also given rise to numerous derivative technologies: the phenomenon of "crowdsourcing" has opened up new opportunities for cartography. Although geodetic satellites cover the entire Earth’s surface, GNSS receivers are not ubiquitous. Indoor positioning technology has limited applications because signals are attenuated by passing through soil, walls, water, and other obstacles. Although positioning methods based on satellite signal repeaters exist, their high cost and relatively low accuracy make them less than ideal. Therefore, finding suitable alternatives for indoor use is of paramount importance. Indoors are usually enclosed. People spend most of their time indoors, making positioning systems critically important. Such systems can be used for traditional tasks such as navigation, mapping, environmental information dissemination, and advertising, as well as for more common applications such as building security, emergency response, and remote control of robotic systems. Currently, there is no universal indoor positioning method similar to the Global Navigation Satellite System (GNSS). Instead, many different systems are being developed, and the choice of a particular system often depends on the spatial structure, the positioning goal, and the required accuracy. The ranking of existing methods is primarily based on the implementation cost, which often plays a decisive role. This leads to a large number of unresolved issues in the field of positioning. The object of the study is the image preprocessing algorithms used for object detection and parameter modification using a television measurement system. The subject of the study is the development and analysis of a television measurement system based on digital image processing algorithms to create an indoor positioning system. The purpose of this work is to develop an indoor positioning algorithm that uses a monocular television system to process digital television images through color beacons under conditions of non-uniform background and variable illumination. The research methods include modern digital image processing tools, machine vision, image recognition, television measurements, mathematical analysis, and mathematical statistics. For practical application of the algorithm, modern numerical methods and programming languages, Matlab and C++ were used. This paper proposes an indoor television positioning algorithm based on a perspective projection model, which achieves a positioning accuracy of 2.8 ± 0.6 mm. This accuracy is comparable to the accuracy of laser and infrared reference beacons, while significantly reducing implementation time and material costs. The beacon recognition algorithm was improved to minimize the impact of low- light conditions and uneven background color. The following filtering parameters were defined for HSV saturation: S-curve curvature k = 0.3; S-curve threshold offset = 30; for HSV brightness, k = 0.1; offset = 50; for hue, the variance was defined as D = 10. For beacon recognition, the following parameters were defined: a threshold for the difference in the lengths of beacon vectors of 10% and a maximum angle between vectors of 7°. The results show that noise affects the positioning system at a level of 16.6 ± 0.1 dB. Keywords – positioning, mobile, robotic systems, beacon, traffic, recognition, digital processing, mathematical analysis. References. 1. Belej O., Kolesnyk K., Polotai O. (2021) Application of Neural Networks in Intrusion Monitoring Systems for Wireless Sensor Networks. In: Shakhovska N., Medykovskyy M.O. (eds) Advances in Intelligent Systems and Computing V. CSIT 2020. Advances in Intelligent Systems and Computing, vol 1293. Springer, Cham. 2. O. Belej, N. Nestor, and O. Polotai, "Developing a Local Positioning Algorithm Based on the Identification of Objects in a Wi-Fi Network of the Mall," 2019 IEEE XVth International Conference on the Perspective Technologies and Methods in MEMS Design (MEMSTECH), Polyana, Ukraine, 2019, pp. 32-36. 3. Beley O.I., Svatyuk O.R., Svatyuk D.R. Zastosuvannya z·hortkovykh neyronnykh merezh dlya bezpeky rozpiznavannya ob’yektiv u videopototsi, Kiberbezpeka: osvita, nauka, tekhnika, vol 4, no 8, pp 97-112, 2020.. 4. Dzhus O. P., Lobur M. V., Holovanchuk D. I., Belej O. I. Udoskonalennya systemy pozytsiyuvannya robotyzovanykh prystroyiv na osnovi alhorytmu kol?orovykh mayakiv // Komp’yuterni systemy proektuvannya. Teoriya i praktyka. – 2025. – Vyp. 7, No 2. – S. 14–29.