Institutional Repository of Vilnius University
TurtleBot Burger robotui skirtų aplinkos išžvalgymo algoritmų palyginimas realiose sąlygose /
Abstract
dc:descriptionFor autonomous mobile robots operating in unknown indoor environments, it is not suf ficient to build a map and plan a path to a given goal, because the robot also has to decide which unexplored part of the environment should be investigated next. This thesis addresses the comparison of environment exploration algorithms: methods based on different decisionmaking principles may perform differently in real indoor environments with different geometries. Five algorithms belonging to different groups of exploration strategies are compared using a common soſtware basis. The experiments were carried out in three real indoor environments: a room with objects, a multiple room layout with a narrow passage between the rooms, and a long corridor with alcoves and obstacles. The algorithms were evaluated using exploration time, map coverage, traveled path length, and navigation goal execution results. The results showed that all analyzed algorithms were able to explore real indoor environments, but their performance depended on the geometry of the environment. In the simpler room, the results of the algorithms were similar, while environments with narrow passages and a long corridor revealed clearer differences between the exploration strategies. The nearest frontier and information gain methods showed a good ba lance between exploration time, traveled path length, and map coverage in most experiments. The RRT and Voronoibased methods also achieved high coverage in some environments, but their performance was less stable.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Bačianskas, Gvidas,
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:ELABAETD308336008&prefLang=en_US