Semantic Information Integration with Transformations for Stream Reasoning
Linköping University, Department of Computer and Information Science, Artificial Intelligence and Intergrated Computer systems
Linköping University, The Institute of Technology
Conference paper (Refereed)
16th International Conference on Information Fusion
16th International Conference on Information Fusion, 9-12 July 2013, Istanbul, Turkey
The automatic, on-demand, integration of informationfrom multiple diverse sources outside the control of theapplication itself is central to many fusion applications. Animportant problem is to handle situations when the requestedinformation is not directly available but has to be generatedor adapted through transformations. This paper extends thesemantic information integration approach used in the streambasedknowledge processing middleware DyKnow with supportfor finding and automatically applying transformations. Twotypes of transformations are considered. Automatic transformationbetween different units of measurements and betweenstreams of different types. DyKnow achieves semantic integrationby creating a common ontology, specifying the semantic contentof streams relative to the ontology and using semantic matchingto find relevant streams. By using semantic mappings betweenontologies it is also possible to do semantic matching overmultiple ontologies. The complete stream reasoning approach isintegrated in the Robot Operating System (ROS) and used incollaborative unmanned aircraft systems missions.