Institutional Repository of Vilnius University
Bluetooth švyturėlių išdėstymo optimizavimas navigacijai patalpose /
Abstract
dc:description”Bluetooth Low Energy” beacons are often used to help determine a device position indoors. One of the most frequent issues with this technology is how to position limited number of beacons in an optimal way so that these would cover the largest possible indoor area fit for navigation. To solve this problem, the aim of this work is to develop a genetic algorithm capable of finding fit solutions to beacon placement within indoors. To justify the algorithm’s solutions, additional research will be done with real world data in a simulated two-dimensional environment, which will be based on a real world location with existing radio signal strength data. To create a physical environment for the simulation, multiple possible software candidates will be reviewed with emphasis on functional expendability, performance. The chosen software - video game engine ”Godot”, will be used to create the simulation environment in which radio propogation models: ”Log-distance path loss”, ”ITU” and ”Motley-Keenan” will be analyzed and compared against real world data. The most suitable radio propogation model for the placement optimization task will be used in development of a genetic algorithm. In the development process of genetic algorithm, two different fitness functions will be created and their set of solutions are going to be compared with their correctness to fit in a real world situation. In addition to development of genetic algorithm, a survey will be done on how the genetic algorithm’s parameters influence the amount of generations needed to find near-optimal beacon placement solution. Following the research, it is concluded that even if most radio propagation models excel in different scenarios, ”Log-distance path loss” radio propogation model is the most fitting model for most of the use cases. In addition to previous point, it is seen that fitness function has to be carefully modelled to real world situation. To tackle this problem, it is recommended to develop fitness function in small iterations while observing all the changes. Furthermore, it is proven that a video game engines can be used for scientific research purposes as long as the physical scale is within observable space to humans, which is often used to create realistic virtual worlds.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2021
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Ivoškus, Aivaras,
- Contributors dc:contributor
-
- Raila, Tomas
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:ELABAETD88378770&prefLang=en_US