Semantic Labeling of Places with Mobile Robots

Service robots that are able to operate autonomously and to interact with their users are one of the major research goals in robotics. In previous years there has been a tremendous progress in the area of robot navigation and perception. However, the question of how to bridge the gap between such subsymbolic and symbolic tasks like planning or interaction has remained an open research question.

In this book, O´ scarMart´inezMozos addresses the problem of assigning semantic labels to sensor data. He considers problems like labeling places in the environment, extracting topological representations from geometric maps, utilizing semantic labels for exploration and navigation as well as for human-robot interaction. All approaches have been validated on real robotic systems and on data gathered with real robots.

This book describes an approach that builds the basis of a roadmap towards getting from subsymbolic sensations to symbolic representations. It describes methods for the direct interpretation and also for the temporal and spatial integration.