AI and Computer Games at Lehigh

Over the past 20 years computer games have gone through a remarkable evolution in areas such as graphics and multi-player gaming. Another area where some evolution can be seen is in the quality of the computer opponent. Despite this evolution, there is a big mismatch between what people in the game industry refer to as the "AI" (roughly, the program controlling the computer opponent) and what Artificial Intelligence is ("...technology can be controlled especially if it is saturated with intelligence to watch over how it goes, to keep accounts, to prevent errors, and to provide wisdom to each decision"- Allen Newell, 1992). In our research, we are looking into contemporary computer games and studying opportunities to use Artifiical Intelligence to enhance the capabilities of the computer opponent.

Here are a few activities on this topic:

• Co-authored two chapters at the 2013 Dagstuhl workshop on Artificial and Computational Intelligence in Games.

• InSyTe Lab Lab at Lehigh University that is conducting research on intelligent decision systems including decision systems for games.

• We built the RETALIATE system which uses Reinforcement Learning to learn team strategies in a FPS game. Check the web site of this project here

• We wrote a Chapter for the upcomming AI Game Programing Wisdom 3: Coordinating Teams of Bots with Hierarchical Task Network Planning

• Want to know more about AI? Here is a good place to start

• We received a donation from the Poker Academy! Please see details Here

• Care for a game of poker? Here is a program developed by Frank Cremen, Ben Mautner and Chris Olsisnki. You'll need the latest version of JavaTM 2 Platform, Standard Edition (J2SETM). When accesing the poker program, please make the window a "full screen" (F11 on Internet Explorer) and then select the button "refresh". Here is the document describing their implementation

• We are also building bots for the first-person-shooter game Unreal Tournament. Aaron Batalion and Todd Fisher started this project.
  • Hai Hoang developed the first HTN planner controlling teams of UT Bots and conducted some experiments. Link to the HTNBots web page, which contains the source code and documentation.

  • Hai Hoang's master thesis on the topic.

  • Our AIIDE-05 paper on the topic.

  • A chapter on this topic also appears in AI Game Programming Wisdom 3

• We combined evolutionary algorithms with reinforcement learning to create Adaptive Game AI for a real-time strategy game. Here on the topic.

• We are looking into the Age of Kings game and experimenting with a rule-based system modifying the behaviour of the computer oponent. Kit Ming Chang is leading this effort. This opponent, code-named Hawk, will be made available soon.

• We are teaching a course on AI and Game Programing.

• We are participating in the Working Group on Goal-oriented Action Planning, Special Interest Group on Artificial Intelligence of the International Game Developers Association.

• We are also interested in developing techniques for building explanations of actions taken by an AI-controlled player in the context of RTS games. Here is a paper on the topic.

Last updated: Fri dec 14 14:49:51 EST 2007