Thespian: using multi-agent fitting to craft interactive drama

  • Authors:

  • Mei Si;Stacy C. Marsella;David V. Pynadath

  • Affiliations:

  • University of Southern California, Los Angeles, CA;University of Southern California, Los Angeles, CA;University of Southern California, Los Angeles, CA

  • Venue:
  • Proceedings of the fourth international joint conference on Autonomous agents and multiagent systems
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

There has been a growing interest in designing multi-agent based interactive dramas. A key research challenge faced in the design of these systems is to support open-ended user interaction while ensuring dramatic user experiences and consistent character personalities. Autonomous agents with reactive and planning abilities are well suited for realizing characters that both adapt to user interactions and are consistent with their own goals. However, agents are often created manually and with extensive programming effort, that excludes authoring by non-technical authors. Thespian is a framework for realizing interactive drama that seeks to reduce programming effort. To start, an author provides linear scripts of the drama. An automated fitting algorithm then configures agents to behave according to the scripts via automated tuning of goal parameters. This capability allows authors to design in a familiar way by writing scripts. Thespian also supports reuse of characters and story elements. Given these advantages, new scenarios can be developed with less programming effort. We discuss the use of Thespian in fitting characters in the Tactical Language Training System and in a Grimms' fairy tale. We also present preliminary experiments on migrating characters between these stories.