New Mexico Supercomputing Challenge

AI in videogames

Team: 73

School: Los Alamos Mid

Area of Science: artificial intelligence


Interim: Problem Definition:
The term artificial intelligence is used every day, but `not everyone knows what it is. Artificial intelligence is a computer science that deals with writing computer programs that can solve problems creatively. The goal of this project is to create an artificial intelligence system that enables a computer to plan out a long term strategy and short term tactics in battlefield environments. It will need to weigh the long term consequences of its actions and uncertainties in the opponent’s response. It will also have to take into account the environmental factors around its units. It will also adapt and learn from past strategies and attempts, both failed and successful.

Problem solution:
To create such an artificial intelligence and environment, we will be using the programming language C++ with the IDE Dev C++ from Bloodshed.net and also the Allegro Graphics Library. We will set the artificial intelligence to use three different levels of strategy, being ranged to the artificial intelligence not making smart decisions to doing almost the best moves possible. The reason for not doing the very best moves possible is because the game would be impossible unless the player also makes the best moves possible. We will make it so that AI controlled players can fight each other while also fighting the human controlled players.

Progress:
At this time, although we have done very little and started very late, we are making quick progress. We currently have a working shell and most of our unit models have been finished. Also, we have a logic and frame rate timer so that it runs at the same speed for all computers. For the land, air, and sea units, we have all the different team colors and we have the different teams’ buildings and bases. Currently, the programming part can already do some basic functions, such as draw and process the different units and the playing field.
Expected Results
After testing, using, and playing the game multiple times, we will add more sophisticated artificial intelligences. We will also put a randomized battlefield into the simulation to create a more diverse playing fields and strategies. As technology advances, more and more sophisticated artificial intelligences will be created, making our simulation obsolete. We expect that later in the simulation for the artificial intelligence to build on its knowledge of the player’s strategies and counter the strategies before they can be fully put into effect.


Team Members:

  George Barnum
  Ben Liu
  Mohit Dubey

Sponsoring Teacher: Clara Vigil

Mail the entire Team

For questions about the Supercomputing Challenge, a 501(c)3 organization, contact us at: consult @ challenge.nm.org

New Mexico Supercomputing Challenge, Inc.
Post Office Box 30102
Albuquerque, New Mexico 87190
(505) 667-2864

Supercomputing Challenge Board of Directors
Board page listing meetings and agendas