Return to Dr. Jacob Schrum's Homepage

Companion Bots in Unreal Tournament 2004

This page presents videos associated with a human subject study focused on understanding what sorts of behaviors players want in companion bots in the First-Person Shooter video game Unreal Tournament 2004. Specifically, the study has each participant play the game with two different companion bots in team deathmatch mode against two native bots. The two bots are Jude, a hand-coded bot that purposefully follows its human teammate so it is ready to assist in combat, and Ethan, a skilled bot whose combat behavior is optimized using MM-NEAT. This research was conducted by Adina Friedman as part of Southwestern University's Summer research program SCOPE. All code is part of the MM-NEAT repository. The bots were written using the Java middleware Pogamut, which can communicate with Unreal Tournament 2004 via a mod called GameBots.

Human Subject Sessions With Jude

Videos of sessions from all 30 participants in the human subject study are included in the playlist below. These videos show the match that each participant played with Jude as a companion. (Note: Videos for subjects 25 and 26 were inadvertantly recorded at the wrong resolution, causing the view of the screen to be clipped)


Human Subject Sessions With Ethan

Videos of sessions from all 30 participants in the human subject study are included in the playlist below. These videos show the match that each participant played with Ethan as a companion. (Note: Videos for subjects 25 and 26 were inadvertantly recorded at the wrong resolution, causing the view of the screen to be clipped)


Associated Publications


Peer-Reviewed Conference Publications


Invited Book Chapters/Articles


Extended Abstracts


Undergraduate Poster Presentations Supervised


Associated Movies



Miscellaneous Content

  • Spring 2015: BotPrize 2012 Champion: A Human-like Bot for Unreal Tournament: A poster included as part of a demo presented at the AI Open House of the Twenty-Ninth Conference of the Association for the Advancement of Artificial Intelligence.
  • Fall 2013: A Retrospective Look at Botprize presented to students in the Computational Intelligence in Game Research track of UT's Freshman Research Initiative.
  • Fall 2012: Making AI that Plays like Humans (Winning the BotPrize). Interview with victors of BotPrize 2012 (Me and Mihai Polceanu) on AiGameDev.com
  • Fall 2012: New Scientist Article About Our Victory in Botprize 2012.
  • Fall 2012: BBC Article About Our Victory in Botprize 2012.
  • Fall 2012: Slashdot Article About Our Victory in Botprize 2012.
  • Summer 2012: The UT^2 Game Bot Judged More Human Than Humans. A UTCS news post about our victory in the WCCI 2012 Humanlike Bots Competition.
  • Summer 2012: UT^2: Human-like Behavior via Neuroevolution of Combat Behavior and Replay of Human Traces slides from the WCCI 2012 Humanlike Bots Competition.
  • Fall 2011: BEACON Researchers at Work Blog Post about my research.
  • Spring 2011: Botprize 2010 Competition: Revised/Improved Slides About Our 2nd Place UT^2 Bot presented to Student ACM Chapter at Southwestern University
  • Fall 2010: Botprize 2010 Competition: Slides About Our 2nd Place UT^2 Bot presented at DeVry University in Austin, TX

  • Last Updated: 5/28/2019