Geographic Subdivision and Top Level Data Structures, Columbus, Magellan, and Expanding CIG (Computer Image Generation) Horizons,

Recent advances in CIG technology have made it possible to accommodate the production of very large map correlate visual data bases of homogeneous feature density. Inherent in the top level design of a visual data base is a strategy for geographic subdivision. This strategy is implemented in a top level data structure which is processed by the CIG hardware to determine which portions of a visual data base to display. The top level data structure also dictates how the visual data base must be modeled. Since a visual data base is typically constructed in a 3D Cartesian coordinate system, the problems historically associated with round to flat earth map projections are also encountered in top level data structures. There are also constraints imposed by the CIG system used. We will review the standard methods of geographic subdivision and discuss their limitations when applied to training requirements. We will explore a new strategy for geographic subdivision of a visual data base utilizing the military UTM and UPS grid systems and the leverage derived from innovative modeling techniques i.e., semi-automatic generation of terrain DLMS data, and strategies for reusing common scene elements. We will discuss how these concepts are currently being implemented into visual data base design and will show how an extrapolation of these trends can be applied to the problem of whole earth navigation training.