| Abstract [eng] |
Unmanned aerial vehicles are becoming very popular these days. They have a variety of applications: search and rescue missions, crop inspection, 3D mapping, surveillance and other military applications. Problem is that most UAV‘s do not have abject avoidance systems installed, and that can lead to broken equipment or people may get injured. To avoid that, it is needed to design a system capable of avoiding obstacles. In this paper, popular methods of object detection are reviewed and ones best suited for project are selected for further analysis. „Block match“ and „Semi global match“ methods are analysed and selected for application in object avoidance. Experiment is performed by building custom UAV platform and testing selected stereo algorithms using drone on a real life objects. |
