Chapter 10: Artificial Intelligence: Tic Tac Toe

Overview

Developers use the term Artificial Intelligence (AI) in a general way to refer to a number of things. Modern AI often has to do with systems that learn through experience how to make decisions. AI also refers to the part of a game's code that controls the computer opponents. AI strategies are a complex enough topic that they could easily fill books on their own. Even though we won't be able to cover the complex AI algorithms, some systems allow us to make a fairly "dumb" AI to compete against the human player.

You'll often need to develop a simple AI. Each game type requires a unique style of AI, so programming AI takes many forms. For example, in a real-time strategy game, you must develop short behaviors for all the units in the game as well as the overall control of computer-controlled opponent teams . A game like that is played in real-time, so the AI also must operate in real-time. On the other hand, if you are programming a chess game in which the player and the computer take turns, the AI is completely different.

If you sit back and ponder the different kinds of games you might need to develop AI for, you'll find that nearly every one of them needs a custom solution. There are distinct types of games , however, that share common solutions. For example, you can program a game to play checkers in a similar way that you program a game to play chess.

In this chapter, you'll look at one of the more difficult types of AI programming: building AI that plays turn -based games such as checkers, chess, and backgammon.