Tree Detection for UAV Localization
Detekce stromů pro lokalizaci UAV
Authors
Supervisors
Reviewers
Editors
Other contributors
Journal Title
Journal ISSN
Volume Title
Publisher
České vysoké učení technické v Praze
Czech Technical University in Prague
Czech Technical University in Prague
Date
Abstract
Detekce stromů z mračna bodů je zásadní pro autonomní pohybování dronů v lese, stejně tak jako pro určení jeho polohy ve známém lese, nebo pro přibližnou lokalizaci vůči místu vzletu v neznámem lese. Pro správnou lokalizaci je nutná správná klasifikace a rozpoznání stromů stejně tak jako vhodný algoritmus pro následné lokalizování. Zároveň použití informace o všech jednotlivých stromech vede ve srovnání s mřížkou obsazenosti k méně nepřesnostem při dlouhodobé lokalizaci a v možnosti rozpoznání míst, kde již dron letěl.
Tree detection from Point cloud (PCL) is crucial for the autonomous movement of drones in the forest as well as for determining its location in a known forest or for approximate localization relative to a takeoff point in an unknown forest. Correct classification and recognition of trees, as well as a suitable algorithm for subsequent localization, is necessary for proper localization. At the same time, using information about all individual trees leads to fewer inaccuracies in long-term localization and the ability to recognize places where the drone has already flown compared to the occupancy grid.
Tree detection from Point cloud (PCL) is crucial for the autonomous movement of drones in the forest as well as for determining its location in a known forest or for approximate localization relative to a takeoff point in an unknown forest. Correct classification and recognition of trees, as well as a suitable algorithm for subsequent localization, is necessary for proper localization. At the same time, using information about all individual trees leads to fewer inaccuracies in long-term localization and the ability to recognize places where the drone has already flown compared to the occupancy grid.