Transcript Game AI

Artificial Intelligence in
Video Games
Jason Fuller
1
What is Game AI?
 Imitate intelligence in the actions of non-player
characters (NPCs).
 Make the game “feel” real.
 Obey laws of the game
 Show decision making
and planning
2
Goals of Game AI
 Be fun!
 Be challenging but not
overwhelming (unless the
hardest difficulty is selected)
 Do not hog all the resources!
(CPU time)
 Make sure the AI does not
cheat! (At least do not get
caught)
 AI often get bonuses when
difficulty increases
3
Types of Games
1. Action games
 Shooters (FPS and Third-Person)
 Racing, Sports
2. RPG games (Role Playing Game)
 Often include many
action game aspects of AI
3. RTS games (Real Time Strategy)
4
Game AI Types
 Action and RPG AI tend to work better with Finite State
Machine based AI
 RTS AI used Finite State Machines in the early years of AI
development.
 RTS AI work best with Artificial Neural Networks and Fuzzy
Logic
 Both contain path finding components
5
AI Path Finding
Dijkstra’s Algorithm
A* Algorithm
 Most commonly used
 Finds the shortest path
 The world or map of the game is represented by a grid of points
6
A* Algorithm
 Allows for high optimization
 Either by changing the search algorithm to better suit the game or by
changing the data structures.
 Very similar to how people move between locations in a city.
7
Finite State Machines (FSM)
 Simplest and most basic AI model.
 Consists of:
 States
 State Transitions
 Most common for Action games!
 Not many different actions for NPCs
8
Finite State Machines
 Among the States and State Transitions there are 4
components:
 States which define behavior
 State transitions which are the movement
from one state to another
 Conditions which must be met for state transition
 Events/Actions which are internally or externally
generated which may lead to a state transition
9
FSM Example
10
FSM Disadvantages
 Very predictable
 Too many states get tough to organize
 Since there are such crisp rules between states, NPC does not
feel natural
11
FSM within a State
 States have a FSM within them
12
Modern FSM Example
13
History of Finite State Machines
 In 1952, the game Nim used AI to play against an opponent.
 1960’s & 1970’s
 Spacewar!
 Pong
 Space Invaders
 1980’s
 Simcity
14
History Continued
 1990’s
 Dragon Quest IV
 Warcraft
 Half-Life
15
Artificial Neural Networks (ANN)
 No agreed definition, most common one is “a network of
simple processing elements, which can exhibit complex global
behavior, determined by the connections between the
processing elements and element parameters.”
 Mathematical model inspired by biological neural networks.
 An adaptive structure that can learn.
16
ANN Structure
 Very similar to the structure of our brain.
 Input layer, processing (hidden) layer, output layer
 Learns by example
17
ANN Structure
 The hidden layer is not just a straight line of nodes
 Each node in the hidden layer will contain just a small part of
the overall calculation
 The nodes have connections
between each other with
certain weights
18
ANN Structure
 The weight of the connections between the nodes determine the outcomes
of the calculations
 If a node is triggered by 2 different nodes it can then determine which one is
more important
19
ANN Learning
20
21
22
Black & White
 Came out in 2001
 First to effectively use
Artificial Neural Networks
23
Halo Wars
 Came out in 2009
 Featured a “Custom” difficulty level that used Artificial Neural
Networks
24
Fuzzy Logic
 Introduced in 1965 for use in Artificial Intelligence research
 Present problems to computers in a way similar to how humans
solve problems and that everything is a matter of degree (or
preference or context).
25
Example Problem
A store owner needs to decide how much produce to order.
Elements to take in to consideration:
 Time of year?
 What is the weather like?
 Is there a Holiday coming up?
26
Video
 Fuzzy Logic: An Introduction
27
Games
Now that the major types of AI have been covered, I will go into
more detail about what games they are used in.
28
Racing Game AI
 Large-scale cheating!
 AI already know the track and optimal path
 AI already has complete behavior determined before the start
of the race
29
Racing AI graphs
General Path
Optimal Path
30
Racing Game AI
 In its basic form, it is the most basic of game AI but in some of
the racing simulators, the AI are more complicated
 If the player is using the optimal path, the AI will actively try to
push them off of it.
 The AI will also use tricks such as spinning out opponents by
making their back tires lose grip.
31
FPS Game AI
 Implemented with a layered structure
 Bottom layers control the path finding tasks and animation
selection
 Higher layers control the tactical reasoning which is where the
Finite State Machine would be.
32
A* Graph of FPS or RPG World
General path
Playable Zone
Unplayable Zone
33
FPS Continued
 F.E.A.R. series has revolutionary AI
 AIs have knowledge of map elements and will flank the player
 AI will break through walls and windows to get to the player
 AI will rush when they heavily outnumber the player
34
RPG Game AI
 Many encounters with AI are unscripted
 GTA IV and Far Cry 2 made great leaps in “friendly” AI
35
Elder Scrolls IV: Oblivion
 Released in 2006
 During testing, a story important NPC kept being found dead.
 A mechanic of the AI was the cause.
36
Bioshock Infinite
 The player companion, Elizabeth (who is an AI), is almost
entirely unscripted.
37
RTS Game AI
 Started out using Finite State Machines to control AI
 Too many options to cover
 AI was “dumb”
 AI would build up in a strict way
 Once the player found a strategy that worked against the AI, it
would always work.
 RTS AI switched to a combination of Fuzzy Logic and Artificial
Neural Networks
38
RTS Game AI
 By changing to Fuzzy Logic and Artificial Neural Networks (ANN):
 Fuzzy Logic led to smarter responses to attacks
 ANN led to smarter development of base and better long term decisions
39
A* Graph of RTS Map
40
RTS Continued
 Maxis is again changing the simulation landscape with the new
Simcity
 Every “Sim” is a full AI
 Have there own agenda
 Have specific wants
and needs
41
Video
 SimCity: Economics AI
42
Future of Game AI
 Game AI have made great leaps forward since they were first
developed.
 An AI that can learn how you play a game would be a great
opponent
43
References

Champandard, Alex. "Top 10 Most Influential AI Games." Aigamedev.com. N.p., 12 Sept.
2007. Web. 25 Feb. 2013. <http://aigamedev.com/open/review/top-ai-games/>.

Grant, Eugene, and Rex Lardner. "The Talk of the Town." TheNewYorker.com. The New
Yorker, 02 Aug. 1952. Web. 25 Feb. 2013.
<http://www.newyorker.com/archive/1952/08/02/1952_08_02_018_TNY_CARDS_00023
6053>.

Grzyb, Janusz. "Artificial Intelligence in Games." - CodeProject. Software Developer's
Journal, n.d. Web. 25 Feb. 2013. <http://www.codeproject.com/Articles/14840/ArtificialIntelligence-in-Games>.

"Neural Networks: A Requirement for Intelligent Systems." N.p., 2007. Web. 25 Feb. 2013.
<http://www.learnartificialneuralnetworks.com/#training>.

"Short Term Decision Making with Fuzzy Logic And Long Term Decision Making with
Neural Networks In Real-Time Strategy Games." Hevi.info. N.p., n.d. Web. 25 Feb. 2013.
<http://www.hevi.info/tag/artificial-intelligence-in-real-time-strategy-games/>.
44
Videos
 Fuzzy Logic
http://www.youtube.com/watch?feature=player_detailpage&v=P8wY6mi1vV8#
t=117s
 SimCity
http://www.youtube.com/watch?feature=player_detailpage&list=UUnje_8ilXP7
KB2vdssyAWug&v=MxTcm1YFKcU#t=37s
45
QUESTIONS?
46