Navigation methods using airlandscape data for small UAVs

The purpose of the research. UAVs are widely used in such fields as military, intelligence, and research. As a result, the issue of their use in electronic warfare conditions and the presence of various interference, both man-made and natural, is acute. Thus, the need for UAV navigation using visual data increases. The purpose is study of the main negative factors affecting the quality of aerial photography images and methods for eliminating them to construct a correct orthomosaic for UAV visual navigation. Methods. Mathematical methods are proposed to eliminate negative distortions in UAV camera images using various image transformation approaches. After applying these methods to the original images, corrected versions are obtained, which are used to construct the tile covering. Tiled coverage created from processed images provides continuous and uniform coverage of the area collected during the UAV’s flight. This allows you to obtain exact coordinates of images and objects on them. Results. An analysis of the main methods for eliminating negative factors that distort images during aerial photography for the visual navigation of UAVs, as well as a brief overview of the visual navigation methods themselves. Conclusion. To successfully implement visual navigation of a UAV, it is necessary to apply a number of methods for converting aerial photography images, as well as the use of certain algorithms for visual navigation. It is concluded that in addition to the use of mathematical and software algorithms, it will also be necessary to analyze and study the necessary computing power for the use of the entire hardware and software complex on board the UAV, taking into account its weight-dimensional properties.