The Community for Technology Leaders
Green Image
Virtual environments that allow multiple participants to cooperatively interact present complex design problems. A joint Loral-US Army project, the Close Combat Tactical Trainer (CCTT) described in this article will, when completed, train ground combat tank and mechanized infantry forces on simulated equipment using a high-fidelity representation of actual terrain. With over 50 different human-computer interfaces and ultimately consisting of over half a million lines of Ada program code, together with extensive distribution requirements, CCTT is the most complex virtual environment attempted to date. To solve some of the design problems, CCTT relies successfully on concurrent engineering, spiral development, and usability engineering. The CCTT system consists of networked simulators and workstations based on a Fiber Distributed Data Interface (FDDI) LAN with up to 100 IBM RS/6000 processors attached. Each manned module has a single RS/6000 processor to perform vehicle computations, transmit updated vehicle information to the FDDI network, and accept information about other simulated entities from the FDDI network. An Evans and Sutherland Image Generator 3000 generates the visual scene for each module. Software elements start with the network layer, proceed up through the service layer, and end at the application level. Each layer is based on data-driven, object- oriented program code that meets real-time performance requirements. The authors describe the CCTT architecture, their development approach, and lessons learned. For example, they explain that applying concurrent engineering is a significant effort and may actually increase the time needed to organize and train teams and accustom engineers to the collaborative process. They also briefly discuss costs and testing trends for contracted system developments.

T. W. Mastaglio and R. Callahan, "A Large-Scale Complex Virtual Environment for Team Training," in Computer, vol. 28, no. , pp. 49-56, 1995.
89 ms
(Ver 3.3 (11022016))