Buchtitel: Proceedings of the 19th International Semantic Web Conference (ISWC'20)
In this paper, we introduce Aware, a knowledge-enabled framework for robots' situational awareness. It is designed to support autonomous logistics vehicles operating in automobile manufacturing plants. Aware comprises an ontology grounding robots' observations, a knowledge reasoner, and a set of behavioral rules: The Aware ontology models data streams of proprioceptive and exteroceptive sensors into high-level semantic representations. The knowledge reasoner infers adequate policy by reasoning over a sliding window of observations, presumably depicting the robot's perceptions and actual state of knowledge. The behavioral rules, in analogy to road traffic rules and common sense, regulate the operation of autonomous robots in a manufacturing environment despite their obvious peculiarity. Our rules are the first ones facilitating the orderly and timely flow of vehicles. We show the applicability of Aware in an industrial set up. Overall, we posit that situational awareness is a fundamental element towards functional autonomy and argue that it can provide a reliable basis for organizing and controlling robots in a smart factory in the near future.