Automated run-time mission and dialog generation

Abstract

Current mission driven systems, be they games or training simulations, are generally restricted to using a set of training missions that are hard coded into the system. This has the unfortunate effect of limiting the number of times a person or team can be sent through a simulator before it begins to lose its training value or the number of times a person can replay a game without it becoming predictable and somewhat boring. The fact that all of the mission parameters must be hard coded also increases the time required for scenario development. This study defines an architecture for automating the creation of missions at run time allowing much larger variety in the number and content of missions in a given system. Our architecture also allows for the creation of varied and more believable dialog with minimal scenario creation time required. We also explore an alternate method for determining agents attitudes towards the users avatar which is more robust than the more commonly used system and which can be used as an input to dialog generation further improving the realism of the dialog. A commercial game, Neverwinter Nights by Bioware, has been used to produce a proof of concept.