Extended Semantic Annotations for Generating Translators in the Arrowhead Framework

In order to create distributed automation systems, it is required to ensure their interoperability; however, ensuring interoperability between heterogeneous systems (using different communication protocols, data formats, and semantics) is a challenging task. Among many interoperability challenges, this paper addresses issues concerning semantic and data interoperability, namely, it provides a contribution to support the semantic compatibility verification and the generation of translators for XML messages. Translators are generated based on XML-Schemas that are annotated with a reference ontology. We base our annotations on an extension of an existing declarative annotation method. In particular our extension explicitly addresses ambiguities of annotations, schema instance mismatches and coverage mismatches that frequently occur on IoT message schemas. We have evaluated our approach based on a set of interaction scenarios from the domain of the Arrowhead Project. A tool prototype that supports the semantic compatibility verification and the generation of XML translators is available at http://gres.uninova.pt/tag/.