Building Maps Based on a Learned Classification of Ultrasonic Range Data

This paper introduces an approach for learning environmental maps based on ultrasonic range data. A neural network concept (self-organizing feature map) is used to learn a classification of the range data which makes it possible to discern situations. As a consequence the free-apace is partitioned into situation areas which are defined as regions wherein a specific situation can be recognized. Using dead-reckoning such situation areas can be attached to graph nodes generating a map of the free-space in the form of a graph representation. In this context it is discussed how the dead-reckoning drift can be compensated.