1. Bbm421 – Computer Games Technology
Fall ’12
Lecture 7
By Umut Rıza ERTÜRK

2. Outline AI
Network

3. Resource Also probably of the best resources for game development
AI Game Engine Programming – Brain Schwab
Dan Witzner Hansen - Lecture Notes

4. What is Artificial Intelligence? (John McCarthy, Stanford University)
It is the science and engineering of making intelligent machines, especially intelligent computer programs. It is related to the similar task of using computers to understand human intelligence, but AI does not have to confine itself to methods that are biologically observable.
Yes, but what is intelligence?
Intelligence is the computational part of the ability to achieve goals in the world. Varying kinds and degrees of intelligence occur in people, many animals and some machines.
Isn't there a solid definition of intelligence that doesn't depend on relating it to human intelligence?
Not yet. The problem is that we cannot yet characterize in general what kinds of computational procedures we want to call intelligent. We understand some of the mechanisms of intelligence and not others.

5. What’s involved in Intelligence?
Ability to interact with the real world
to perceive, understand, and act
e.g., speech recognition and understanding and synthesis
e.g., image understanding
e.g., ability to take actions, have an effect
Reasoning and Planning
modeling the external world, given input
solving new problems, planning, and making decisions
ability to deal with unexpected problems, uncertainties
Learning and Adaptation
we are continuously learning and adapting
our internal models are always being “updated”
e.g., a baby learning to categorize and recognize animals

6. What is AI?

7. What is AI in games?
AI is (traditionally) the control of every non-human entity in a game?
The other cars in a car game
The opponents and monsters in shooter games
Your units, your enemy’s units and your enemy in a RTS game
But, typically does not refer to passive things that just react to the player and never initiate action
That’s physics or game logic
For example, the blocks in Tetris are not AI, nor is a flag blowing in the wind
It’s a somewhat arbitrary distinction

8. Example AI Game Scenarios
Learning the habits and preferences of a human player in the context of game play
Game enemy finding the shortest route to the player
Learning and adapting avoidance algorithms during the game
Game enemy adaptive collaboration with other game enemies toward a common game objective
Generating authentic and adaptive game dialogue and commentary e.g sports
Generation of game environment and levels

9. Perception Management
Perceive (Sensor information)
Think (Evaluate perceived data)
Act (execute planned action)

10. AI Update Step The sensing phase determines the state of the world
May be very simple - state changes all come by message
Or complex - figure out what is visible, where your team is, etc
The thinking phase decides what to do given the world
The core of AI
The acting phase tells the animation what to do
Generally not that interesting (for this course)
AI Module
Sensing
Game Engine
Thinking
Acting

11. AI in the Game Loop
AI is updated as part of the game loop, after user input, and before rendering
There are issues here:
Which AI goes first?
Does the AI run on every frame?
Is the AI synchronized?

12. AI and Animation AI determines what to do and the animation does it
AI drives animation, deciding what action the animation system should be animating
Scenario 1: The AI issues orders like “move from A to B”, and it’s up to the animation system to do the rest
Scenario 2: The AI controls everything down to the animation clip to play
Which scenario is best depends on the nature of the AI system and the nature of the animation system
Is the animation system based on move trees (motion capture), or physics, or something else
Does the AI look after collision avoidance? Does it do detailed planning?

13. AI by Polling The AI gets called at a fixed rate
Senses: It looks to see what has changed in the world. For instance:
Queries what it can see
Checks to see if its animation has finished running
And then acts on it
Why is this generally inefficient?

14. Event Driven AI
Event driven AI does everything in response to events in the world
Events sent by message (basically, a function gets called when a message arrives, just like a user interface)
Example messages:
A certain amount of time has passed, so update yourself
You have heard a sound
Someone has entered your field of view
Note that messages can completely replace sensing, but typically do not. Why not?
Real system are a mix - something changes, so you do some sensing

15. Requirements for Games AI
They need to be fast (i.e contrary to some academic AI approaches which are ‘optimal’ – we will give you both)
Games do not need optimal solutions (they need to be fun)
Games may need to combine several techniques
The goal is to design agents that provide the illusion of intelligence
Illusion of being intelligent

16. Goals of Game AI
Goal driven - the AI decides what it should do, and then figures out how to do it:
Reactive - the AI responds immediately to changes in the world
Knowledge intensive - the AI knows a lot about the world and how it behaves, and embodies knowledge in its own behavior
Characteristic - Embodies a believable, consistent character
Fast and easy development
Games do not need optimal solutions (they need to be fun)
Games may need to combine several techniques
The goal is to design agents that provide the illusion of intelligenc
Low CPU and memory usage
These conflict in almost every way – how?

17. AI Techniques in Games
Basic problem: Given the state of the world, what should I do?
A wide range of solutions in games:
Finite state machines, Decision trees, Rule based systems, Neural networks, Fuzzy logic
A wider range of solutions in the academic world:
Complex planning systems, logic programming, genetic algorithms, Bayes-nets
Typically, too slow for games

18. Techniques used today Deterministic: Statistical:
Finite State Machines
Decision trees
Fuzzy State machines
Statistical:
Baysian Approaches
Neural Networks
Evolutionary algorithms/ Artificial life

19. Genetic Algorithm Generate a random population of potential solutions
Mate (crossover) selected individuals to create offspring
Replace some individuals in the population with new offspring
Apply mutation to population and evaluate against solution objective
if solution found stop
else go to step one
Use a Binary GA or Real GA to encode problem
See Haupt and Haupt (2004), Practical Genetic Algorithms for a detailed discussion and practical examples

20. Searching for Optimum Solutions
AI systems search to find an optimum solution to a given problem
The optimum solution may be known by the AI system as an objective function which is compared with a current potential solution
Human evaluation may take place to provide an interactive process where each potential solution is rated as being a worthy candidate

21. Example AI Game Scenarios
Learning the habits and preferences of a human player in the context of game play
Game enemy finding the shortest route to the player
Learning and adapting avoidance algorithms during the game
Game enemy adaptive collaboration with other game enemies toward a common game objective
Generating authentic and adaptive game dialogue and commentary e.g sports
Generation of game environment and levels

22. Simulated AI Random movement Tracking algorithms (chase / evade)
Swarming
Patterns of movement
Patterns linked to states
Random patterns of movement linked to random states
Probabilistic states and associated movement

23. Finite State Machine (FSM)
Notion of game being in one of a finite number of states
Behavior and direction of game logic dependant on current state
Ideal properties for a robust FSM
A reasonable number of states, each of which represents a different goal or motive within the game
Lots of input to the FSM, such as state of the game environment and the other objects within the game environment

24. Pathfinding Searching for the shortest path between two points
Search everywhere?
Search and record the best routes - store a heuristic (cost)
Widely used pathfinding algorithm is the A* algorithm
A* works by assigning a cost of moving from one position to another in terms of finding the required point
The A* algorithm is CPU intensive - but some versions have been optimized to reduce computational overheads
Good examples of A* implementation in Brackeen’s book ‘Developing Games in Java’

25. Fuzzy Logic True / False used in traditional logic
Fuzzy logic seeks to give an input to a fuzzy system a fuzzy truth as to its membership of vaguely defined collections of fuzzy states.
Fuzzy logic allows for set membership values to range (inclusively) between 0 and 1, and in its linguistic form, imprecise concepts like "slightly", "quite" and "very". Specifically, it allows partial membership in a set.

26. Discrete Value Ranges
Cold Cool Warm Hot 1
-10 10 20 30 0C

27. Fuzzy Sets Heat Value Ranges
Cold Cool Warm Hot 1
-10 10 20 30 0C

28. Fuzzy Associated Matrix
A fuzzy associated matrix FAM is a matrix that shows the membership grades of all the elements of a fuzzy relation. This is a convenient way to view fuzzy relations.

29. Fuzzy Sets Centriod
The final stage is defuzzification into ‘crisp’ output is value derived from a fuzzy centroid 1
-10 10 20 30 0C

30. Two Measures of Complexity
Complexity of Execution
How fast does it run as more knowledge is added?
How much memory is required as more knowledge is added?
Determines the run-time cost of the AI
Complexity of Specification
How hard is it to write the code?
As more “knowledge” is added, how much more code needs to be added?
Determines the development cost, and risk

31. Different games need different methods
Action Games (lead a character through a set of levels)
Fighting Games Physical combat
Sports Games (icehockey and football)
Racing Games (Car, bikes, speedboats)
Adventure games ( plots based on exploration and problem solving)
Role Playing games (D&D)
Board games (Chess, checkers)
Strategic games (i.e. military type battle scenarios)
Simulation Games (simcity)

32. Cases Chess: Large search trees
Packman: Random FSM (weighted randomness)
Smart environments: Information is not coded in the NPC but in their environment

33. Is AI in Games really only about the players in the game?
What about the surroundings?
Interaction methods?

34. AI Timeline