Multi-camera visual obstacle avoidance  for micro aerial vehicles

Morhunenko, Mykola

Домівка
→
Students Research & Project Works | Роботи студентів
→
Факультет прикладних наук
→
Бакалаврська програма "Комп'ютерні науки"
→
2022
→
Перегляд матеріалів

dc.contributor.author	Morhunenko, Mykola
dc.date.accessioned	2024-02-14T09:45:09Z
dc.date.available	2024-02-14T09:45:09Z
dc.date.issued	2022
dc.identifier.citation	Morhunenko, Mykola. Multi-camera visual obstacle avoidance for micro aerial vehicles / Morhunenko, Mykola; Supervisor: Ing. Matouš Vrba; Ukrainian Catholic University, Department of Computer Sciences. – Lviv: 2022. –44 p.	uk
dc.identifier.uri	https://er.ucu.edu.ua/handle/1/4420
dc.language.iso	en	uk
dc.title	Multi-camera visual obstacle avoidance for micro aerial vehicles	uk
dc.type	Preprint	uk
dc.status	Публікується вперше	uk
dc.description.abstracten	The 21st century is a time of innovation and exploration in the fields of applied science such as physics, medicine, biology, programming and robotics, as well as their intersections and fusions. One of the research topics, that have recently gained much popularity, are Micro unmanned Aerial Vehicles (MAVs). MAVs became smaller, cheaper and more readily available. However, due to the rising popularity and utility of MAVs, some of their problems and limitations are highlighted. MAVs often rely on the Global Navigation Satellite System (GNSS), but due to GNSS inaccuracy in closed environments, the MAV requires an obstacle avoidance system that is compact and reliable. In this thesis, a compact and reliable visual multi-camera obstacle avoidance system for MAVs is developed. Calibration of a non-planar stereo camera setup and extraction of obstacle positions in an unknown environment from pairs of 2D images are tackled in this work. The proposed solution is designed to run onboard an MAV with a limited computational power considering size, weight and payload limitations. Performance of a prototype of the proposed solution was measured in laboratory experiments. The result proved that the system is ready for on-drone deployment and real-life tests.	uk