AI FINAL REVIEW

Name a type of Data Structure.

-Arrays -Link lists -Queues -Stacks -Heaps -Graphs -Scene Graphs -Octrees -Quad-trees -Binary-trees -Minimax trees -kd trees -Sphere trees -Bounding volume hierarchies -Hash tables -Portals and sectors

T/F Effective and efficient pathfinding is a seldom used task that nonplayer characters use in few games.

-FALSE

Name five ways to use flocking in a game situation.

1. In Assassin's Creed, NPC's follow the player and surrounds them 2. In GTA, cars follow after each other and cause traffic 3. In RTS games, units can be placed and grouped without colliding with each other. 4. In Battlefield, the enemies veer towards their groups and stay grouped. 5. In hunting games, animals stay together with their kind.

Select the scripting languages listed below: A. Lua B. Python

A Lua B Python

Define a finite state machine and tell what each part does.

A finite state machine is an abstract machine that can be in one of many different and predefined states, also it can define a set of conditions that determines when the state should change. The character will behave differently in each state and the player's actions determines their transitions.

Describe what Influence Mapping is.

A way to vary the cost of the A * nodes, depending on what is happening in the game. It is used for calculating the costs of different paths.

The _____ algorithm provides an effective solution to the problem of pathfinding.

A*

What are the two (and sometimes three) distinct problems in chasing and evading.

A. deciding to chase or evade C. effecting the chase or evasion D. obstacle avoidance

Define Heuristic

An estimate of the number of steps from the given tile to the destination file.

What are the advantages of a finite state machine compared to hard-coded AI?

Because finite state machines are easier to understand, to implement, and to debug, developers prefer finite state machines over hard-coded AI.

Describe why the shortest path between start and end points may not be the fastest.

Because it depends on the games environment and the terrain cost. A long walk along a road may be faster than taking a short walk through a swamp.

_____ can make computer-controlled characters seem intelligent.

Breadcumb Pathfinding

In a real game, the player never sees the _____ .

Breadcure trail

The modified _____ line-of-sight algorthim is used to calculate line segments.

Bresenham

Describe the most widely used AI technique in games.

Cheating is a mechanism that helps give the computer an edge against intelligent human players.

Define Defuzzification.

Defuzzification is the process of taking fuzzy output membership and producing a corresponding crisp numerical output.

_____ controlling an opponent's behavior is one of the most common uses of scripting in game AI.

Directly

What do we call it when ambiguous transitions exist? What are three ways of dealing with them?

Double covered and uncovered are the ambiguous FSM types. Check for these by unit testing or code review

T/F Finite State Machines do not lend themselves well to game AI development.

FALSE

T/F Fuzzification involves processing fuzzy data into crisp data.

FALSE

T/F Game AI is considered in the category of strong AI.

FALSE

T/F In Threat Assessment the idea is to evaluate the size of the threat and the level of threat and then respond with maximum defensive force to provide protection

FALSE

T/F Obstacle avoidance can be achieved by using the attractive nature of the Lenard-Jones function.

FALSE

T/F Seldom in video games do non-player characters move in cohesive groups.

FALSE

T/F Swarming is similar to flocking, however the resulting behavior is more consistent.

FALSE

T/F The Lenard-Jones potential function nears infinite repulsion as the units get farther apart.

FALSE

T/F The narrow-field-of-view model restricts each unit to seeing only what is directly next to it.

FALSE

T/F The scripting language to use in a game is Lua.

FALSE

T/F When using the A* algorithm in a tiled situation before you start a branching search technique, you need to to keep track of which tiles need to be searched by using a closed list.

FALSE

T/F While benevolent characters can provide helpful information to game players, it is unethical to have characters intentionally mislead players.

FALSE

T/F Writing logical rules given fuzzy input is not possible.

FALSE

T/F Visual interaction seems most intelligent and immersive when it relates to the current game situation.

FALSE, VERBAL INTERACTION

T/F In physically simulated environments, you specify an object's position explicitly during the simulation. Why or Why not?

FALSE, You cannot specify an object's position because the physics is constantly acting on the object so it is not able to follow a pattern.

T/F The techniques for learning and reacting to character behavior fall under the deterministic AI and its difficulties apply also.

FALSE, nondeterministic AI

T/F In traditional Fuzzy Logic applications, rules evaluate to absolutely True or False.

FALSE, traditional Boolean logic application

Pac Man ghosts are an example of _____ .

FSM Finite State Machine

The purpose of an Empty node table is to establish the path between nodes.

False, the purpose of an empty node table is to establish the CONNECTIONS between the nodes.

What is Fuzzy Logic?

Fuzzy Logic is used in fuzzy state machines to make resulting actions somewhat less predictable and to reduce the burden of having to incorporate huge numbers of if-then rules.

Describe "Fuzzy Logic"

Fuzzy logic is a way to present problems to computers in a way that is similar to how humans solve problems.

Explain the main basic difference between Game Physics and Game AI.

Game Physics determines how the object is going to move while Game AI supplies what the object is going to do and how it is going to react.

Describe Cohesion.

Having each unit steer towards the average position of a group.

Describe the difference between regular line-of-site chasing vs line-of-site chasing in Tiled Environments.

In a Continuous environment, changing positions is unrestricted and then the positions are mapped to the nearest screen pixel. However, in tile-based games changing positions is more restricted, not every tile is mapped to a screen pixel and there are eight possible directions of movement.

Describe the simpler approach to understand what is said in a game instead of using a sophisticated language parser.

Instead of checking for each literal text string, you can search for and respond to particular keywords.

Give a short and basic description of how a game reads and parses a script.

It searches a given file for a specific keyword and then returns the value associated with that keyword.

What are open lists used for?

Keeping track of which tiles need to be searched and used to determine if you have reached a dead end.

Who is Lofti Zadeh and what did he do?

Lofti Zadeh is a professor at the University of California Berkeley, that wrote a fuzzy set theory and became the father of fuzzy logic.

_____ factors contribute to how challenging a potential opponent will be.

MANY

How often do humans use Fuzzy Logic in their daily lives?

Most of the time

What do we mean by ambiguous transitions?

Multiple valid transitions from one state at the same time.

Which is more important for the game programmer, deterministic or nondeterministic AI.

Nondeterministic is more important because it creates pressure to produce stronger AI and to produce the next hit game.

_____ is simply the process of moving the position of a game character from its initial location to a desired destination.

Pathfinding

_____ is a simple way of to give the illusion of intelligent behavior.

Pattern movement

_____ AI systems are probably the most widely used AI systems for real-world applications.

Rule-Based

Name three rules of Classic Flocking

Separation Cohesion Alignment

T/F A more natural path to a destination can be created by using the Bresenham line algorithm.

TRUE

T/F Another name for a data structure is a class.

TRUE

T/F By combining the chase and obstacle avoidance algorithms with the swarming algorithm would allow swarms to swarm and chase prey and avoid obstacles along the way.

TRUE

T/F Game AI has to be able to act, has to be intelligent, and has to deal with changing conditions.

TRUE

T/F In the use of tile-based algorithms of pattern movement, you must consider certain caveats specific to such environments.

TRUE

T/F In tile based games the game domain is divided into discrete tiles - squares, hexagons, etc. - and the player's position is fixed to a discrete tile.

TRUE

T/F It is the pathfinding algorithm's job to find the best path between any two points, avoiding any obstacles.

TRUE

T/F Line-of-site chase algorithm is relentless, the predator will always head directly toward the prey.

TRUE

T/F Many times it is necessary to move characters in a realistic way even though they may not have a destination.

TRUE

T/F Steering forces from a variety of rules are accumulated in the same variable and then applied all at once to control unit motion, you can effectively layer any number of rules on top of previous ones.

TRUE

T/F Terrain cost is usually hardcoded into the game world.

TRUE

T/F The basic premise of the FSM is to link individual and unique states to different types of behavior.

TRUE

T/F The interesting thing about using the Lenard-Jones potential function for many tasks is that this one function enables the programmer to create many sorts of seemingly intelligent behavior.

TRUE

T/F To build a path simply refer to the completed path table.

TRUE

T/F To normalize a pattern, subtract the starting position from all the positions stored in the pattern arrays

TRUE

T/F Tracing around _____ is a relatively simple method of obstacle avoidance.

TRUE

T/F each unit in a flock must be aware of its neighbor.

TRUE

In your own words, describe a line-of-site approach.

The overall gist of line-of-sight approach is, to have the predator take a straight-line path to reach the prey.

Describe in your own words why the line-of-site chase algorithm is not always the shortest path to the target.

The reason why the line-of-sight chase algorithm is not the always the shortest path is, because of the range to the target or how long it would take to reach the target.

What are some indicators that finite state machine is inappropriate to use?

Trying to create different characters objects with completely different states.

The standard method for implementing pattern movement takes the desired pattern and encodes the control data into a(n) _____ .

array

To create a swarm, have the units _____ each other.

attract

T/F The simplest chase algorithm involves correcting the prey's coordinates based on the predator's coordinates so as to reduce the distance between their positions.

correct the predators coordinates based on the prey's coordinates

To give each unit its own "personality", you can give each unit _____ parameters settings.

different

Another word for a global coordinate system is _____ .

earth-fixed

The Bresenham algorithm is a(n) _____ method for tiled environments.

effective

You can add AI to flocking leaders to make their leading as intelligent as you want. Further, you can define and add other unit types to the mix, creating even greater variety. the possibilities are _____ .

endless

Wall tracing is more of an _____ technique.

exploration

The programmer's _____ is the only limit when using the L-J potential function.

imagination

The point of using scripting is to make gameplay more _____ .

immersive

Fuzzy logic can be used to control such things as _____ .

improving on basic chasing and evading

If units are close together, the repulsive force between them will be ____ .

large

Nondeterministic methods facilitate _____ .

learning & unpredictable gameplay

In physics, the Lenard-Jones potential represents the potential energy of attraction and repulsion between _____ .

molecules

The results of Threat Assessment will be harder for the player to predict because there are _____ defined cutoffs, or breakpoints, at which the defensive deployments change sharply.

no clearly

A simple way of finding a dead end is to monitor the _____ list.

open

Chapter 8 showed you how to implement basic scripting that enables you to alter game AI _____ the main game program.

outside

To find the point where the predator and prey will meet at the same time, you must consider their _____ .

relative velocities

Square pattern control inputs are ____ .

simple

Deterministic behavior or performance is _____ .

specified and predictable

In flocking situations, the programmer can use _____ _____ the same as used in chasing, evading, and pattern movement.

steering forces

Cheating in gameplay gives an advantage to _____ .

the game

In the Steering Model tuning requires _____ .

trial and error

Nothing can break the immersive effect of a game faster than seeing a game character _____ navigate a set of obstacles.

unable to

Nondeterministic behavior or performance is _____ .

unspecified & unpredictable

Like path-following, _____ does not calculate a path from a starting point to an ending point.

wall-tracing