Ontology-driven middleware for next-generation train backbones

The current evolution in the railway management software domain heads in the direction of highly modular systems with many different smaller components working together, tuned towards the operator's needs. In order to create an integrated management system for the train and railway domain in general, many different applications from both operators and manufacturers have to cooperate. To create a robust integrated system, a good supporting middleware infrastructure is needed. To the authors' knowledge, very few integrated or standardized techniques for creating higher-level intelligent middleware in the railway domain are publicly defined to date. The solution today towards establishing connectivity through the train makes use of a messaging bus. A service layer, providing mechanisms for e.g. life-cycle management, discovery and information aggregation, is however not provided in this current approach. Therefore, a distributed and modular architecture using ontologies, and widely used standards, such as Ethernet and IP, is detailed in this paper. The architecture provides the required intelligence needed for monitoring distributed applications in the train environment. The middleware allows information to be aggregated and analysed on different levels. Information querying, based on the ontology in the middleware is also discussed. By means of directory functionality, the ontology-driven middleware provides intelligent discovery as well. Finally, the ontologies used in the middleware to structure the domain and corresponding methods for creating such ontologies are presented.

[1]  Eduardo Tovar,et al.  Real-time fieldbus communications using Profibus networks , 1999, IEEE Trans. Ind. Electron..

[2]  Atanas Kiryakov,et al.  Ontology Middleware: Analysis and Design , 2002 .

[3]  Stefan Decker,et al.  Creating Semantic Web Contents with Protégé-2000 , 2001, IEEE Intell. Syst..

[4]  Douglas C. Schmidt,et al.  An overview of the Real-Time CORBA specification , 2000, Computer.

[5]  Volker Haarslev,et al.  Racer: An OWL Reasoning Agent for the Semantic Web , 2003 .

[6]  Frank van Harmelen,et al.  Sesame: A Generic Architecture for Storing and Querying RDF and RDF Schema , 2002, SEMWEB.

[7]  Filip De Turck,et al.  Modeling the performance of the Web service platform using Layered Queueing Networks , 2005, Software Engineering Research and Practice.

[8]  Bijan Parsia,et al.  Pellet: An OWL DL Reasoner , 2004, Description Logics.

[9]  Tao Gu,et al.  A service-oriented middleware for building context-aware services , 2005, J. Netw. Comput. Appl..

[10]  V. A. Minin,et al.  Development of the communications-based train control system for Moscow Metro , 1997, Proceedings of the 1997 IEEE/ASME Joint Railroad Conference.

[11]  Tao Gu,et al.  Toward an OSGi-based infrastructure for context-aware applications , 2004, IEEE Pervasive Computing.

[12]  Thomas R. Gruber,et al.  A translation approach to portable ontology specifications , 1993 .

[13]  Koen De Bosschere,et al.  Towards an Extensible Context Ontology for Ambient Intelligence , 2004, EUSAI.

[14]  Deborah L. McGuinness,et al.  OWL Web ontology language overview , 2004 .