TOWARDS A DISTRIBUTED INFORMATION ARCHITECTURE FOR AVIONICS DATA Chris Mattmann Jet Propulsion Laboratory California Institute of Technology 4800 Oak Grove Drive M/S 171-264 Chris.Mattmann@jpl.nasa.gov Dana Freeborn Jet Propulsion Laboratory California Institute of Technology 4800 Oak Grove Drive M/S 171-264 Dana.Freeborn@jpl.nasa.gov Dan Crichton Jet Propulsion Laboratory California Institute of Technology 4800 Oak Grove Drive M/S 306-438 Dan.Crichton@jpl.nasa.gov ABSTRACT Avionics data at the National Aeronautics and Space Administration’s (NASA) Jet Propulsion Laboratory (JPL) consists of distributed, unmanaged, and heterogeneous information that is hard for flight system design engineers to find and use on new NASA/JPL missions. The development of a systematic approach for capturing, accessing and sharing avionics data critical to the support of NASA/JPL missions and projects is required. We propose a general information architecture for managing the existing distributed avionics data sources and a method for querying and retrieving avionics data using the Object Oriented Data Technology (OODT) [3] framework. OODT uses an XML [4] messaging infrastructure that profiles data products and their locations using the ISO-11179 [2] data model for describing data products. Queries against a common data dictionary (which implements the ISO model) are translated to domain dependent source data models, and distributed data products are returned asynchronously through the OODT middleware. Further work will include the ability to plug in new manufacturer data sources, which are distributed at avionics component manufacturer locations throughout the United States. KEYWORDS Information Architecture, Distributed Data Management, XML 1. INTRODUCTION In October 2003, the Center for Advanced Avionics (CAA) at the Jet Propulsion Laboratory funded the State of the Art Survey Task that seeks to bring component manufacturers such as Boeing, Ball Aerospace, and Honeywell to JPL in order to survey state of the art products being developed in industry for possible use on future JPL missions. Manufacturers provide product specifications in the form of presentation slides, data sheets, and documents which describe their state of the art products. These product specification documents are stored in Docushare [6], a web-based content retrieval system. However, Docushare’s standard set of metadata for describing a document does not support the search and retrieval of avionics components required by flight system design engineers. An existing approach at JPL for managing this survey data, as well as other sources of avionics product information, is in the form of a set of HTML web pages. HTML data tables are used to store component metadata and links to the component data product sheets; however, these tables provide no underlying method for search and retrieval other than the primitive find on page methods of the internet browsers. Consequently, while the State of the Art Survey Task has been successful in gathering data specifications for new flight technology, this information is still not available to flight system design engineers through simple search and retrieval mechanisms. Furthermore, the ability to easily compare the survey data with other existing avionics data sources is not possible. Currently, these distributed data sources suffer from common data management problems such as domain dependent data models and heterogeneous data sources.