The MMI Device Ontology: Enabling Sensor Integration

The MMI Device Ontology: Enabling Sensor Integration

Carlos Rueda1, Nancy Galbraith2, Robert A. Morris3, 
Luis E. Bermudez4, Robert A. Arko5, John Graybeal6

1 Monterey Bay Aquarium Research Institute, Moss Landing, CA, USA

2 Woods Hole Oceanographic Institution, Woods Hole, MA, USA

3 Department of Computer Science, University of Massachusetts at Boston, Boston, MA, USA

4 Open Geospatial Consortium, Herndon, VA, USA

5 Lamont-Doherty Earth Observatory, Columbia University, Palisades, NY, USA

6 California Institute for Telecommunications and Information Technology, University of California San Diego, La Jolla, CA, USA

The Marine Metadata Interoperability (MMI) project has developed an ontology for devices to describe sensors and sensor networks. This ontology is implemented in the W3C Web Ontology Language (OWL) and provides an extensible conceptual model and controlled vocabularies for describing heterogeneous instrument types, with different data characteristics, and their attributes. It can help users populate metadata records for sensors; associate devices with their platforms, deployments, measurement capabilities and restrictions; aid in discovery of sensor data, both historic and real-time; and improve the interoperability of observational oceanographic data sets.

We developed the MMI Device Ontology following a community-based approach. By building on and integrating other models and ontologies from related disciplines, we sought to facilitate semantic interoperability while avoiding duplication. Key concepts and insights from various communities, including the Open Geospatial Consortium (eg., SensorML and Observations and Measurements specifications), Semantic Web for Earth and Environmental Terminology (SWEET), and W3C Semantic Sensor Network Incubator Group, have significantly enriched the development of the ontology. Individuals ranging from instrument designers, science data producers and consumers to ontology specialists and other technologists contributed to the work.

Applications of the MMI Device Ontology are underway for several community use cases. These include vessel-mounted multibeam mapping sonars for the Rolling Deck to Repository (R2R) program and description of diverse instruments on deepwater Ocean Reference Stations for the OceanSITES program. These trials involve creation of records completely describing instruments, either by individual instances or by manufacturer and model.

Individual terms in the MMI Device Ontology can be referenced with their corresponding Uniform Resource Identifiers (URIs) in sensor-related metadata specifications (e.g., SensorML, NetCDF). These identifiers can be resolved through a web browser, or other client applications via HTTP against the MMI Ontology Registry and Repository (ORR), where the ontology is maintained. SPARQL-based query capabilities, which are enhanced with reasoning, along with several supported output formats, allow the effective interaction of diverse client applications with the semantic information associated with the device ontology. In this presentation we describe the process for the development of the MMI Device Ontology and illustrate extensions and applications that demonstrate the benefits of adopting this semantic approach, including example queries involving inference. We also highlight the issues encountered and future work.

American Geophysical Union Fall Meeting 2010 – 
Dec 16, 2010 – IN44B-08