Semantic world models for object identification and localization in mobile robotics

Semantic world models are experiencing increasing attention as a relevant field of research in robotics. They extend classical world models by describing the state of an environment based on information that has a tangible meaning for humans. This includes, for example, the class membership of objects located within the environment and their spatial relationship. Thus, semantic models bridge the gap between humans and robots and open up new possibilities for the industrial application of mobile robotics. In this paper, a software architecture is presented that implements a semantic world model focusing on the localization and identification of object instances.