Issue No. 05 - September/October (2006 vol. 21)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MIS.2006.92
Maria Cutumisu , University of Alberta
Duane Szafron , University of Alberta
Jonathan Schaeffer , University of Alberta
Matthew McNaughton , University of Alberta
Thomas Roy , University of Alberta
Curtis Onuczko , University of Alberta
Mike Carbonaro , University of Alberta
Many computer games use custom scripts to control the ambient behaviors of nonplayer characters. As a result, story authors must write computer code for the game world's hundreds or thousands of NPCs. Creating entertaining, nonrepetitive NPC behaviors without investing substantial programming effort to write customized, nontrivial scripts is challenging. Current computer games have simplistic ambient behaviors for NPCs; it's rare for such characters to interact with each other. To address this, we developed a model that uses generative behavior patterns, which let authors quickly and reliably generate realistic and entertaining NPC scripts. The approach has proven successful, even in the challenging area of creating interactions between NPCs. This article is part of a special issue on interactive entertainment.
ambient behavior, nonplayer character, intelligent agents, scripting language, generative pattern, collaborative behavior, computer games
J. Schaeffer et al., "Generating Ambient Behaviors in Computer Role-Playing Games," in IEEE Intelligent Systems, vol. 21, no. , pp. 19-27, 2006.