| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 6.08 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
We contributed a method for avoiding obstacles using monocular vision as the only sensor in UAV (Unmaned Aerial vehicle). The vision based ROS (Robotic operating system) node detects the known obstacles in front of the UAV. Unknown obstacles can be taken care of by adding he information of all the obstacles seen in the scene to a map. The distance to obstacle in this research is calculated by just increasing size of the obstacle in front of the UAV. The image processing libraries were used from OpenCV to do thresholding, noise removal and contours detection. This research also tests and evaluate the path planning of UAV using MoveIt architecture, and evaluates the different results obtained.Hence we show the effectiveness of the monocular vision and size as a constraint algorithm in UAVs to detect and avoid frontal obstacles.
Descrição
Palavras-chave
ROS UAV OpenCV MoveIt Thresholding Contours Monocular vision