The creation and development of a synthetic environment for a simulation exercise can be a laborious, time consuming, and sometimes undefined process. This is especially disheartening because it involves re-inventing the wheel. A synthetic environment is common to most simulations, only the fidelity, extent, and interoperability of the environment varies. There are a multitude of decisions to be made and requirements to consider, many of them reaching beyond the intended goal of simulating an entity or exercise to collect valuable data. These decisions may involve large quantities of information pertaining to the integration of environmental models, simulation entities of differing fidelities and architectures, selection of models appropriate for the simulation, network communication architecture interfaces and data transmission mediums. There is currently no central repository for environmental or entity models or descriptive information about such models. The developer is faced with hard decisions that are typically out of their scope of expertise and redevelopment of a new environment to refine decisions can be costly and time consuming. There is currently no effective way to create a distributed synthetic environment, in a quick and adaptive manner.

This paper presents an approach to rapidly developing a synthetic environment using defined processes, architectures, and repositories of environmental software and data. The synthetic environment of interest in this paper is a family of models and related knowledge that provides for the simulation of the tactical and natural environments within which a simulation entity operates. The synthetic environment is not simply an exercise or scenario manager. The approach presented in this paper focuses on efficient information management with the goal of significantly reducing the time and effort required to create a synthetic environment and perform a simulation exercise. The approach involves architecture based selection and adaptation of environmental models, accessed through the world wide web and assembled by a knowledge based system. This approach allows the user to focus on the simulation at hand and not on the repetitive task of environment development.