This paper deals with self-localization of a mobile robot on the condition that no a-priori knowledge about the environment is available. The applied method features to be accurate, robust, independent of any artificial landmarks and feasible with such a moderate computational effort that all necessary tasks can be executed in real-time on a standard PC. The perception system used is a panorama laser range finder (PLRF) which takes scans of its present environment. A modified Dynamic Programming (DP) algorithm provides pattern matching and pattern recognition on the preprocessed panorama scans and thereby renders a qualitative fusion of the sensory data. For an exact quantitative estimate of the robot's current position, a robust localization module is employed. The knowledge gained about the environment along that way is stored in a self-growing, graph based map which combines geometrical information and topological restrictions. Preliminary experiments in a common office environment ...

This paper presents our approach for laser-based local position tracking based on the data explored in a three-dimensional environmental model of an indoor environment. This algorithm is used to substitute the dead reckoning on a mobile robot to allow robust map generation and position dependent task triggering. The underlying concept of the local environmental model used for filtering of the sensor information allows an easy fusion of di#erent sources of the available information, like: a-priori knowledge, explored information and even fusion of the information from other sensor systems. This system is implemented and tested on our mobile robot.