Game Development Reference
In-Depth Information
there is the necessity of having a “commander” who will decide when the strategy
should be changed.
The realm of strategy and planning can get incredibly complex, especially if it's
for a fully featured RTS such as
StarCraft
. For further information on this topic,
check the references at the end of this chapter.
Summary
The goal of a game AI programmer is to create a system that appears to be intelli-
gent. This may not be what a researcher considers “true” AI, but games typically
do not need such complex AIs. As we discussed, one big problem area in game AI
is finding the best path from point A to point B. The most common pathfinding al-
gorithm used in games is A*, and it can be applied on any search space that can be
represented by a graph (which includes a grid, path nodes, or navigation meshes).
Most games also have some sort of AI behavior that is controlled by a state ma-
chine, and the best way to implement a state machine is with the State design pat-
tern. Finally, strategy and planning may be utilized by genres such as RTS in order
to create more organic and believable opponents.
Review Questions
1
.
Given the following graph, what are the adjacency lists for each node?
2
.
Why are navigation meshes generally preferred over path nodes?
3
.
When is a heuristic considered admissible?
