Successful, seamless interchange of simulation databases has long proved surprisingly difficult to achieve. Numerous technical difficulties, arising from the different environmental representations used by different simulation systems, have proved to be only one facet of this difficulty. Often such problems are in fact the result of more fundamental underlying issues, such as the mathematical relationships between different co-ordinate systems. Logistical issues, and collaborative aspects of database interchange between different groups or companies, also contribute to the problems.
Thales has encountered many of these issues over the years in generating a range of databases for its simulation systems. These databases are required to correlate closely with other sensor systems, in particular the visual, but it is often the case that these other systems are third party products, using databases modelled by companies other than Thales. In these circumstances, the strategy used by Thales to generate its databases has typically been to derive them directly from the visual database. This has involved directly processing the visual database, extracting relevant geometry and attribution and formatting it for use by the Thales simulation systems. Historically, such visual databases have been provided by third parties using the SIF/HDI interchange format and imported directly into the Thales database generation toolset. While generating such derived databases in this way has been achieved successfully, many interchange issues referred to above were encountered and needed to be addressed.
When the need arose to replace SIF/HDI, the opportunity was taken to seek a replacement that would not only provide better representational capabilities but also address many of the wider, non-technical issues as well. Analysis of a variety of formats was undertaken and SEDRIS emerged as by far the strongest contender. Not only did it provide the best all round support for existing data representation requirements, it also gave good support for addressing wider interchange issues and offered a variety of opportunities to enhance the database generation toolset, both during initial development and over time.
This paper will discuss experiences using SEDRIS in this context. It will examine the basic representational requirements that needed to be met and the interchange problems that were to be overcome. The ways in which SEDRIS was seen to address these problems will be considered, along with the other advantages SEDRIS offered. Experiences developing SEDRIS software and interchanging databases using SEDRIS will also be described, including some lessons learned concerning both the use of SEDRIS and database interchange in general.