Artificial Intelligence & Computer Games Research

I'm no longer actively doing research on computer games and AI. Here is some of my previous research.

General AI and Computer Games

  • A general, over-the-top paper attempting to rally people to work on artificial intelligence and computer games: Interactive Computer Games: Human-level AI's Killer Application. This was presented as an invited talk at the National Conference on Artificial Intelligence (AAAI) on August 2, 2000 with Michael van Lent.

  • A revised version of the above paper that emphasizes the different genres of AI in computer games: The Role of AI in Computer Game Genres. This will appear in a book someday. Written with Michael van Lent.

  • An article on using AI and games for research with details of the anticipation in Quakebot (see related paper below). Using a Computer Game to Develop Advanced AI, Computer, 34 (7), July 2001, pp. 70-75.

  • A more recent and briefer paper that gives an overview of our Research in Human-level AI using Computer Games. This was published in Communications of the ACM in January 2002.

  • I've written a short paper on Design Goals for Autonomous Synthetic Characters.

  • Laird, J. E. and van Lent, Michael, Developing an Artificial Intelligence Engine, Proceedings of the Game Developers Conference, March 16-18, 1999, San Jose, CA, pp. 577-588.

  • The first paper we wrote that describes interface Soar to Quake and Descent. van Lent, Laird, J. E., Buckman, J., Hartford, J., Houchard, S., Steinkraus, K., and Tedrake, R. Intelligent Agents in Computer Games, Proceedings of the National Conference on Artificial Intelligence, July 1999, Orlando, FL, pp. 929-930.

Haunt 2

We are now devoting our efforts to developing an adventure-style game where human-level AI characters really make a difference. We are doing it as a mod to Unreal Tournament. The working title is Haunt 2 (see this link for details of the original Haunt).

  • Brian Magerko, John E. Laird, Mazin Assanie, Alex Kerfoot, Devvan Stokes, AI Characters and Directors for Interactive Computer Games, Proceedings of the 2004 Innovative Applications of Artificial Intelligence Conference, San Jose, CA, July 2004. AAAI Press.

Directors for Interactive Drama

Brian Magerko is doing his thesis on directors for interactive drama.

  • Magerko, B. "A Proposal for an Interactive Drama Architecture", AAAI 2002 Spring Symposium Series: Artificial Intelligence and Interactive Entertainment, March 2002.

  • Magerko, B. & Laird, J. E. Towards Building an Interactive, Scenario-based Training Simulator, 10th Computer Generated Forces and Behavior Representation Conference, 2002, Orlando, FL.

  • Magerko, B. & Laird, J. E., Building an Interactive Drama Architecture, 1st International Conference on Technologies for Interactive Digital Storytelling and Entertainment, Darmstadt, Germany, March 24 - 26, 2003

Quakebot

  • A description of some experiments we have done with the Quakebot to test how human it behaves and whether we can modify some simple parameters to change skill levels: Creating Human-like Synthetic Characters with Multiple Skill Levels: A Case Study using the Soar Quakebot. This appeared in the AAAI 2000 Fall Symposium Series: Simulating Human Agents, November 2000.

  • A technical examination of the Soar Quakebot with emphasis on its ability to use anticipation: It Knows What You're Going to Do: Adding Anticipation to a Quakebot. This appeared in the Agents 2001 conference. An earlier version first appeared in the AAAI 2000 Spring Symposium Series: Artificial Intelligence and Interactive Entertainment, March 2000: AAAI Technical Report SS-00-02.

  • A comparison of AI in computer games to research on computer generated forces (CGF) for distributed training. This paper includes a shortened and updated version of some of the material in the following paper. It is targeted to the CGF community. An Exploration into Computer Games and Computer Generated Forces, The Eighth Conference on Computer Generated Forces and Behavior Representation, May 2000, Orlando FL.

  • The games research uses the same technology we've used for simulating military pilots. This paper gives an overview of our military pilot simulation work and relates it to issues in computer game AI development. Sorry but not all of the figures came through to PDF. Laird, J. E., and Jones, R. M. Building Advanced Autonomous AI Systems for Large Scale Real Time Simulations, Proceedings of the Computer Game Developers Conference, May 1998, Long Beach, CA.

AI and Games Tutorial

Mike van Lent and I gave a tutorial on AI Tactical Decision-Making Techniques at the 2001 and 2002 Game Developers Conference.

Game Lecture Series at UM in 2000 and 2001

  • Fall 2001 Distinguish Lecture Series on Computer Games
    • Doug Church's Talk from 2001 Lecture Series
    • Lars Liden's Talk from 2001 Lecture Series

  • Fall 2000 Distinguish Lecture Series on Computer Games

  • 2000 Poster

Contributors

Here are the people who have contributed to our research on AI and computer games over the years (in alphabetical order): Ben Bachelor (Unreal Interface), Nathan Benninghoff (Unreal 3D models), Josh Buckman (Descent 3 interface), Karen Coulter (Soar support), Syed Enam-Ur-Rehman (Haunt Bot), Joe Hartford (Quake II interface), Ben Houchard (speech and teamwork), Steve Houchard (Quake II interface), Brad Jones (SGIO, Unreal and Quake interfaces), Alex Kerfoot (Unreal script), John Laird (project management and Quake bot), Colin Lauber (Haunt Level Design), Brad Muir (Half-Life interface), Damien Neff (Descent 3 interface), Michael Parent (Haunt II), Jeff Sheiman (Unreal 3D models), Kurt Steinkraus (Socket-IO), Devvan Stokes (Haunt Level Design), Russ Tedrake (Descent bot), Amy Unger (Quake bot), Michael van Lent (project management and Quake bot), Mike Zielinski (Descent 3 interface).

AI Game Websites

http://ai.eecs.umich.edu/people/laird/Game-AI-Resources.htm

Pictures

Here are a bunch of pictures from AAAI '99 and our preparation for AAAI '99.