A Scalable and Extensible Interactive Scenario Architecture for Distributed Command and Control Simulations

Interactive virtual environments are becoming increasingly popular for their utility in virtual training, distributed decision-making and collaborative environments. Some of these applications rely on a scenario that is revealed to the user as he/she interacts with synthetic objects and characters that inhabit virtual worlds. The development and authoring of interactive dynamic scenarios is often hard and difficult to accomplish using current techniques. Many interactive scenario developers use decision trees, which yield very limiting and unfulfilling training experiences, because they do not stimulate learning or thinking beyond the scripted paths. Some researchers proposed plan-based interactive narrative architectures, which, although superior to decision trees, do not scale and do not address user’s goals and intentions, yielding inflexible scenarios that do not adapt suitably to players’ goals or behaviors. In this paper, we propose a dynamic scenario architecture that aims at enhancing scalability and reuse by using a multi-agent layered problem solving technique. Additionally, the interactive scenario architecture will automatically adapt to users’ goals by integrating a user model and a user monitoring technique.