A hybrid localization approach for UAV in GPS denied areas

In this paper, the localization problem of an autonomous unmanned aerial vehicle in case of losing the GPS signal is handled. A vision-based solution approach is proposed consisting of two phases. In the first phase a hybrid map is constructed. Such map consists of a set of reduced features obtained by information-theoretic analysis. This enables faster UAV localization processing without degenerating the accuracy. The features are represented by local descriptors which are additionally tagged with their metric position. The second phase localizes the UAV using the map, which is performed on two scales. A fast and coarse topological location is identified based on matching features of images taken by the camera with the local descriptors information in the map. This guides the UAV in fast and safe emergency homing. A second precise metric position can be estimated in extension with respect to a previously identified topological location and with the aid of the features' metric position information. This can assist the UAV navigation in case the mission should be completed without interruption despite the GPS signal loss.