Networking for Distributed Virtual Simulation

The data must be exchanged in real time, with level of fidelity sufficient to permit their acceptable representation by all other participating simulators. The question "what is sufficient?" requires a detailed understanding of the purpose for which DVS is being used. For training applications it turns out that a three-dimensional rendering with visual fidelity similar to that of an arcade or home computer game is accepted well by typical users, so long as the simulator's controls are virtually the same as those of the actual tank, aircraft, etc. being simulated. The artificiality of the visual interface is quickly forgotten as the user becomes psychologically absorbed, interacting with other participants that are behaving just as they would on a real battlefield. Distributed Virtual Simulation (DVS) is a new technology that extends Virtual Reality to the networked environment. It was developed for military training and is being used for civilian education and entertainment. Participants can interact over great distances in the virtual world if supported by a network that provides real-time, multicast packet communications with resource reservation for adequate performance, and satisfactory security. Requirements and techniques for real-time networking, internetted multicast, and resource reservations in support of DVS are described, and are used to define the set of missing Internet technologies that will be needed for DVS. An important class of simulators produce semi-automated forces (SAFORs) where a human commander directs the performance of a group of artificial tanks, aircraft etc. that exist only as computer models which emit real-time information equivalent to that produced by the "real" simulators. SAFORs allow DVS to represent larger forces in order to improve the richness of the training experience for small numbers of human participants. Their quality has not reached a level such that expert humans remain convinced that they are "real" during protracted interactions, and some believe this will not be possible. Nevertheless they add to the scope of the DVS environment, particularly as many actual combat encounters are brief enough that there is little time to form such evaluations. Higher-level automated forces also are being investigated. Those interested in commercial DVS see automated participants as a powerful way to enrich the virtual environment for entertainment.