One important issue that may come up in the implementation of many diverse applications is determining the absolute position of a machine or person on earth or on a planet. While often on earth this issue can be solved with the deployment of GPS receivers, in work underground, in the forest, in situations with poor visibility or for planetary exploration, entirely new approaches at reaching a solution must be developed. MMI has developed the new VisualGPS process to this end together with its collaboration partners. How it works: a mobile system with an autonomous sensor system like laser scanners or (stereo) cameras captures its surroundings from the ground, establishes an area map from this sensor data, then compares its observations with a global navigation map and estimates its current position therefrom. The main component of both the local area map and the global navigation map are clearly identifiable environmental characteristics in the sensor data, both from remote sensing and from the ground. These characteristics are described with their distinctive features beside their position and organized into different classes of “semantic landmarks”. An example of this can be found in the area of planetary exploration, craters and chunks of rock, also on hill or mountain tops or in the trees surrounding the forest.