Shadow-based matching for robust absolute localization during lunar landings

Autor: Kaufmann, Hannah
Jazyk: angličtina
Rok vydání: 2014
Předmět:
Popis: The exploration of the moon can give insights to the evolution and history of the solar system. Future lunar exploration missions require autonomous, safe and precise landings close to scientific spots of interest. Currently the German Aerospace Center (DLR, ger.:Deutsches Zentrum für Luft- und Raumfahrt) investigates several optical localization methods and the fusion of their results within the project ATON (Autonomous Terrain based Optical Navigation system). This shall allow precise landing in a region of a couple of hundred meters around a designated landing site. This thesis investigates an image-based, monocular method for precise position estimation during lunar landings. It is especially designed for the last phase until touchdown, which is still challenging. The image taken by the spacecraft during the descent is matched with an image of the same scene, which is preprocessed prior on earth. The position can be estimated from the correlation of both images. To correlate both images, the shadows seen in the two images, are matched. The proposed method is evaluated in software simulation. It is shown, that the method has the potential to precisely estimate the position during lunar landings, especially during the last landing phase until touchdown. The method is mostly invariant to sun angle variations and relatively large perspective differences between the images.
Databáze: OpenAIRE