Mission with image processing based unmanned aerial vehicle moving platform landing-taking system design

Unmanned aerial vehicles (UAVs), with their ability to take off and land vertically and perform specified tasks, are used in civil and military areas such as carrying payloads, search and rescue activities, agricultural practices, and exploration in dangerous areas. UAVs have become an important research topic in recent years due to their ability to be used in environments dangerous to human life and their low cost. It is a major problem for unmanned aerial vehicles to land quickly and stable on a predetermined moving area without detouring from their path. In this thesis, an image processing-based landing algorithm is improved for the UAV to land on the moving platform. The enhanced landing algorithm consists of three steps. The first of these steps is to detect the platform moving on the path determined by the camera located at the bottom of the UAV. The second is that the UAV follows the moving platform. The third is landing on the platform detected by the camera under the UAV. In order to reduce the processing load in the thesis study, it is recommended to change the image processing frequency adaptively depending on the change in speed and height of the moving platform. The developed image processing-based landing algorithm uses the circular Hough transform, one of the image processing approaches, to detect the image consisting of two circles in an interlocking structure on the moving platform and performs the landing of the UAV on the moving platform. The effect of salt and pepper noise at different levels on the detection of the circular image on the moving platform of the UAV is also examined. The performance of the image processing algorithm, both the noise on the image and the speed change effect of the moving platform are investigated. By comparing the effect of adaptive and fixed image processing frequency performances on the processing load, the effect of the adaptive approach on real-time operating performance is evaluated.