It knows what you're going to do: adding anticipation to a Quakebot

  title={It knows what you're going to do: adding anticipation to a Quakebot},
  author={John E. Laird},
  booktitle={AGENTS '01},
  • J. Laird
  • Published in AGENTS '01 28 May 2001
  • Computer Science
The complexity of AI characters in computer games is continually improving; however they still fall short of human players. In this paper we describe an AI bot for the game Quake II that tries to incorporate some of the missing capabilities. This bot is distinguished by its ability to build its own map as it explores a level, use a wide variety of tactics based on its internal map, and in some cases, anticipate its opponents actions. The bot was developed in the Soar architecture and uses… 

Figures from this paper

Using a Cognitive Architecture for Opponent Target Prediction
This paper discusses how a cognitive architecture based on the simulation theory of mind to opponent prediction within a gaming scenario was adapted and the techniques used to improve the accuracy of predictions and reduce some of the computational cost of using this approach.
An Exploration into Computer Games and Computer Generated Forces
The similarities and differences between AIs for computer games and computer generated forces (CGFs) are studied, and it is argued that computer games have advantages for doing certain types of research on complex, human-level behavior.
Partial observability during predictions of the opponent's movements in an RTS game
  • S. Butler, Y. Demiris
  • Business
    Proceedings of the 2010 IEEE Conference on Computational Intelligence and Games
  • 2010
The effect of imposing partial observability onto an RTS game with regard to making predictions is shown, and two different mechanisms that decide where best to direct the attention of the observers to maximise the benefit of predictions are compared.
Investigating individual game-play patterns using a self-organzing map
This paper investigates the possibility of understanding decision making patterns of an individual player using play data from the 2D arcade game Pacman via an unsupervised learning approach.
Predictive Control for Artificial Intelligence in Computer Games
The authors' bot, Randomly Planning Fighter (RPF), defeats the bot earlier designed for Half-Life with the use of behavior-based techniques, proving that on-line planning can be feasible in rapidly changing environment of modern computer games.
Artificial Intelligence Design in a Multiplayer Online Role Playing Game
The result of the project is a design that allows to have diverse AIbehavior and flexibility to reuse code to create new behaviors, but may prove to be inefficient if implemented on systems with a large number of players or many AI controlled game entities.
Evolution of Reactive Rules in Multi Player Computer Games Based on Imitation
A system in which no direct programming of the behaviour of the opponents is necessary, rules are gained by observing human players and then evaluated and optimised by an evolutionary algorithm to optimise the behaviour.
Game Over: You have been beaten by a GRUE
Some preliminary results from an evaluation ofGRUE in a Tileworld environment suggest that the features of GRUE are useful in dynamic environments, and a brief description of future work is presented.
Learning Agents in Quake III
It is found it is possible to improve a Q ❥ bot by using Reinforcement Learning to train combat movement, and the trained bot outperforms its non-learning counterparts, but does not appear smarter in combat with human opponents.
Towards a game agent
The objective of this report is to give the reader a survey on state-of-the-art techniques and academic research in the field of artificial life where the simulation of complex and emergent behavior


Developing an Artificial Intelligence Engine
Recent efforts by a group at the University of Michigan Artificial Intelligence Lab to apply state of the art artificial intelligence techniques to computer games are discussed.
Creating Human-like Synthetic Characters with Multiple Skill Levels: A Case Study using the Soar Quakebot
This paper reports on a preliminary attempt to develop and evaluate synthetic characters with multiple skill levels and with human-like behavior. Our goal was to determine which aspects of behavior
RESC: An Approach for Real-time, Dynamic Agent Tracking
Together, RESC's situatedness, immediate commitment, and single-state backtracking conspire in providing RESC its real-time character.
Automated Intelligent Pilots for Combat Flight Simulation
The TacAir-Soar system is an intelligent, rule-based system that generates believable "human-like" behavior for military simulations that is capable of executing most of the airborne missions that the United States military flies in fixed-wing aircraft.
Intelligent Agents for Interactive Simulation Environments
An overview of the TacAir-Soar project, aimed at developing humanlike, intelligent agents that can interact with each other, as well as with humans, in virtual environments, and the challenges that automated pilots face in battlefield simulations is analyzed.
Integrating, Execution, Planning, and Learning in Soar for External Environments
How the Soar architecture supports planning, execution, and learning in unpredictable and dynamic environments is described and capabilities are demonstrated on two robotic systems controlled by Soar.
The Final Hours of Quake III Arena: Behind Closed Doors at id Software, GameSpot,
  • 1999
The Soar Tutorial, Part V. The Soar Quakebot
  • The Soar Tutorial, Part V. The Soar Quakebot
  • 2000
The Soar Tutorial, Part V. The Soar Quakebot,
  • 2000