Visual feature-based absolute localization and mapping in gps-denied environments for mobile robots

The problem addressed in this thesis is the global location of a mobile vehicle in a structured environment. To locate itself, the robot needs a map of the environment. If a map is not available, the robot can map the environment. The problem of explore the environment given map to determine the pose of the robot, if this pose is a-priori unknown, is called active localization. A method for the active localization using a given map (e.g. floor plan) has to be implemented. In this context, localization means that the robot finds its pose by exploring the environment