Tuesday, 16 January 2007: 9:30 AM
Making Space-based Sensors Discoverable on the Internet Using A Service Oriented Architecture and Open Geospatial Consortium Standards
214C (Henry B. Gonzalez Convention Center)
A number of experiments have been run at the NASA Goddard Space Flight Center to demonstrate interoperability between space and ground sensors in a variety of sensor webs experiments. These experiments involved the use of Earth Observing-1 (EO-1), Cosmic Hot Interstellar Plasma Spectrometer (CHIPS) and Space Technology 5 (ST-5)and have been reported in previous AMS conferences by the author of this paper. This past year, we have established a partnership with the Open Geospatial Consortium (OGC) in their OGC Webservices testbed which will run from June 2006 to December 2006 and whose purpose is to demonstrate a variety of new OGC interoperability standards of which Sensor Web Enablement(SWE)is one thread. Furthermore, we recieved a NASA Earth Science Technology Office (ESTO) award to further extend this work over three years. The OGC standards support a service oriented architecture which allows sensors, whether space-based on in-situ, to have a schema describing their capability and then to register these capabilities in an online searchable registry. For our demonstrations, we plan to demonstrate a few of the related OGC services with some of our satellites which include Sensor Planning Service (SPS), Sensor Alert Service (SAS) and Sensor Observation Services (SOS). A user will be able to search and "discover" one of the sensors on our satellites via generic Internet searches and then, given the proper permissions, be able to task our satellite with mouse clicks. Furthermore, we are working to demonstrate linking these various service requests into a service chain, thus making the effort to find and task a sensor much like searching "Google map". As an example, a user may search for sensors that can take a cloud-free image of a fire in Southern California. EO-1 would respond that it could take the image with its hyperspectral imager, using a subset of its available spectral bands and then provide a list of conflict-free dates and times that imaging services were available for the location specified. The user might also select automatic cloud screening to prevent images on cloudy days. Furthermore, the user may request other services to perform various science processing on the images and request delivery to their desktop, thus forming an automatic service chain tailored for that user. The plans for these efforts and some of the initial accomplishments will be presented in this paper. The implications of successfully infusing these capabilities into the world of available sensors would be to greatly increase sensor interoperability and in some cases reduce the cost of producing new science products by an order of magnitude.