Lukes - 2
• Live simulation—real people operating with real equipment in the real world with appropriate simulation
interfaces;
• Virtual simulation—real people in simulators and simulations fighting in synthetic worlds;
• Constructive simulation—aggregated wargames and analytical models operating in synthetic worlds.
In the military context, “Everything short of combat is simulation” wherein advanced technology offers unique
advantages to support complex operations, safely, economically, with enhanced opportunities for analysis and
learning, and negligible environmental impact.
To meet these technical challenges, the ADS Synthetic Environments Program seeks to create robust virtual
worlds from operational terrain and environmental data sources of sufficient fidelity and currency to interact
with the real world. These Synthetic Environments must realizable, affordable and supportable within disparate
real-time simulators and simulations linked by local and long-haul networks. On-going efforts under the Synthetic
Environments Program support new developments to incorporate weather, dynamic phenomena, dynamic terrain
and objects within DIS. The program is now being expanded to address rapid construction of virtual worlds with
research initiatives in digital mapping, softcopy workstations and cartographic image understanding.
This objective of this paper is to highlight cartographic issues critical to Synthetic Environments and to
identify some of the scene analysis and machine vision research issues central to this growing problem domain. A
more general description of issues in the generation and use of synthetic environments in networked simulation
can be found in Mamaghani. 3
2 Cartographic Components of Synthetic Environments
The cartographic components of the synthetic environment are largely familiar and not particularly unique
or remarkable in themselves. The interesting aspect lies in the assessment of the available source data, efforts to
enhance and augment this baseline data, how the components are assembled into an unified representation that is
spatially coherent, and the generation of diverse run-time representations tailored to meet various needs within an
ADS exercise. The generation of unified representations having spatial coherency at a level of detail and fidelity
that balance suitable realism in distributed simulation with hardware constraints in computer image generation
is the subject of ongoing research. 5 The primary cartographic components from which synthetic environments
are constructed include the underlying ’bare earth’ terrain surface, man-made and natural features represented
as two or three-dimensional boundaries or objects, and textures that can be used to convey natural patterns of
vegetation, soils, water bodies, or synthetic man-made materials.
2.1 Terrain Surface
Generation of a synthetic environment typically starts with definition of the terrain surface geometry. Often,
this is a gridded digital elevation model comprised by a matrix of elevation posts with spacing ranging from
kilometers to meters depending on the application. Alternatively, Triangular Irregular Networks (TIN) provide the
flexibility to construct a continuous surface of irregular triangles adaptively shaped to represent the terrain surface.
Such gridded or TIN representations are used to characterize both the terrestrial landmass and underwater
bathymetry. Attribution of surface material composition ranges from null to multiple values including soil type,
land cover and/or land use. In many applications, terrain surfaces are characterized by a single elevation value
at a given location; more advanced representations support driving ground vehicles over and under bridges or
tunnels, conducting infantry operations within multi-story buildings or caves, and supporting naval operations
on and under the ocean surface. These applications require the capability to represent multiple elevation values
at a given location.