Smart Terrain for Autonomous Agent Applications

Current combat simulations use polygonal representations of the terrain, augmented with vector data, for terrain reasoning. While these data provide the basic terrain representation needed for vehicle dynamics and weapon system effects, they do not provide the semantic information needed for higher level reasoning, especially for modeling of human behaviors within a combat environment. Semantic information goes beyond the physical characteristics that most terrain databases provide, and includes relationships between terrain features and how they can be used in the performance of specific combat missions. Humans can interpret this additional semantic data and make decisions based on it, but human behavior algorithms need these data to be represented within the combat simulations. MÄK Technologies proposes to develop terrain representation enhancements to augment common terrain representations that support the incorporation of semantic data that will enhance the current polygon and vector representation. We will develop automated tools to extract semantic information where possible, and other tools to assist terrain developers in extracting semantic information where it is not possible to fully automate the process. We will demonstrate this terrain representation in MÄK VR-Forces, a commercial-off-the-shelf CGF system, and the IWARS dismounted infantry simulation.