Case Studies: Brtal Legend tara@doublefine.com ~50 unique unit - - PowerPoint PPT Presentation

▶
case studies br tal legend
SMART_READER_LITE
LIVE PREVIEW

Case Studies: Brtal Legend tara@doublefine.com ~50 unique unit - - PowerPoint PPT Presentation

Sep 09, 2022 696 likes •1.73k views

Case Studies: Brtal Legend tara@doublefine.com ~50 unique unit types ~50 unique unit types ~50 unique unit types Navigation Navigation Brawler vs Strategic AI Tactics Brawler vs Strategic Perspective The Team Anna Kipnis Chad

slide-1
SLIDE 1

Case Studies: Brütal Legend

tara@doublefine.com

slide-2
SLIDE 2

~50 unique unit types

slide-3
SLIDE 3

~50 unique unit types

slide-4
SLIDE 4

~50 unique unit types

slide-5
SLIDE 5

Navigation

slide-6
SLIDE 6

Navigation

slide-7
SLIDE 7

Brawler vs Strategic AI Tactics

slide-8
SLIDE 8

Brawler vs Strategic Perspective

slide-9
SLIDE 9

The Team

  • Anna Kipnis
  • Chad Dawson
  • Joe Virskus
  • Kee Chi
  • Nathan Martz
slide-10
SLIDE 10

~50 Units in Detail

  • Component Architecture
  • Unique Double Teams
  • Avatar
slide-11
SLIDE 11

AI Avatar

  • Strategic decisions
  • Combat decisions
slide-12
SLIDE 12

AI Avatar – Strategic

  • 1. Retreat if low health
slide-13
SLIDE 13

AI Avatar – Strategic

  • 1. Retreat if low health
  • 2. Capture uncapped geysers
slide-14
SLIDE 14

AI Avatar – Strategic

  • 1. Retreat if low health
  • 2. Capture uncapped geysers
  • 3. Defend base
slide-15
SLIDE 15

AI Avatar – Strategic

  • 1. Retreat if low health
  • 2. Capture uncapped geysers
  • 3. Defend base
  • 4. Join battles in progress
slide-16
SLIDE 16

AI Avatar – Strategic

  • 1. Retreat if low health
  • 2. Capture uncapped geysers
  • 3. Defend base
  • 4. Join battles in progress
  • 5. Help units under attack
slide-17
SLIDE 17

AI Avatar – Strategic

  • 1. Retreat if low health
  • 2. Capture uncapped geysers
  • 3. Defend base
  • 4. Join battles in progress
  • 5. Help units under attack
  • 6. Hang out with largest army
slide-18
SLIDE 18

AI Avatar – Combat

  • 1. Rock Solo
  • 2. Double Team
  • 3. Melee and guitar attacks
slide-19
SLIDE 19

Double Teams

slide-20
SLIDE 20

Double Teams

slide-21
SLIDE 21

Double Teams

slide-22
SLIDE 22

Double Teams

slide-23
SLIDE 23

Double Teams

slide-24
SLIDE 24

Double Teams

  • How to use them
  • When to use them
slide-25
SLIDE 25

Double Teams

slide-26
SLIDE 26

Double Teams

Beast: CoopEval=CoopEvalManyTargets { AITargetSearchRadius=35; Priority=0.7; };

slide-27
SLIDE 27

Double Teams

HairHead: CoopEval=CoopEvalHighValueTarget { BestValueRatio=1.5; };

slide-28
SLIDE 28

AI Avatar

  • CoopEvalDistance
  • CoopEvalMegastage
  • SoloEvalHelper
slide-29
SLIDE 29

Questions?

slide-30
SLIDE 30

On the AI Strategy for KILLZONE 2’s Bots

Alex Champandard – AiGameDev.com Remco Straatman – Guerrilla Tim Verweij – Guerrilla

slide-31
SLIDE 31

Gameplay

(video)

slide-32
SLIDE 32

THE CHALLENGE

slide-33
SLIDE 33

Maps

slide-34
SLIDE 34

Game Modes

  • Search and Retrieve
  • Capture and Hold
  • Assassination
  • Body Count
  • Seach and Destroy
slide-35
SLIDE 35

Badges

slide-36
SLIDE 36

Weapons

slide-37
SLIDE 37

THE SOLUTION

slide-38
SLIDE 38

Take Away Strategy is more than the sum of its parts.

slide-39
SLIDE 39

Architecture Strategy AI

Squad AI Squad AI Squad AI Squad AI

Individual AI Individual AI Individual AI Individual AI Individual AI Individual AI Individual AI Individual AI Individual AI

slide-40
SLIDE 40

Architecture

Killzone 2 Multiplayer Bots

  • R. Straatman, T. Verweij.

Paris Game AI Conference, 2009.

slide-41
SLIDE 41

Internal Architecture

Capture & Hold Search & Destroy Assassination Body Count Search & Retrieve

Base Strategy

Objectives: Squads: Bots:

slide-42
SLIDE 42

Waypoint Network

slide-43
SLIDE 43

Strategic Graph

slide-44
SLIDE 44

Area Clustering Algorithms

Automated Terrain Analysis William van der Sterren

AiGameDev.com Masterclass, 2009.

slide-45
SLIDE 45

Manual Annotations

slide-46
SLIDE 46

Regroup Locations

slide-47
SLIDE 47

Mission-Specific Defense

slide-48
SLIDE 48

Sniping Locations

slide-49
SLIDE 49

Terrain Analysis

Terrain Analysis & Reasoning

Kevin Dill, Sergio Garces, William van der Sterren, Paul Tozour. AiGameDev.com Special Report, 2008.

Procedural Combat Tactics

Remco Straatman, William van der Sterren, Arjen Beij. Game Developers Conference, 2005.

slide-50
SLIDE 50

Influence Map

slide-51
SLIDE 51

Strategic Pathfinding

slide-52
SLIDE 52

Strategic Pathfinding

slide-53
SLIDE 53

Strategic Pathfinding

slide-54
SLIDE 54

Strategic Pathfinding

slide-55
SLIDE 55

Strategic Pathfinding

Realistic Autonomous Navigation in Dynamic Environments Alex J. Champandard

Masters Research Thesis, University of Edinburgh, 2002.

slide-56
SLIDE 56

Squad Assignment

slide-57
SLIDE 57

Objective Assignment

slide-58
SLIDE 58

Squad / Objective Assignment

Building a Better Battle: The Halo 3 AI Objectives System Damian Isla

Game Developers Conference, 2008.

slide-59
SLIDE 59

THE RESULTS

slide-60
SLIDE 60

Reception

Don‘t listen to anything he says; he‘s biased.

slide-61
SLIDE 61

Analysis

slide-62
SLIDE 62

Embracing Special Cases

slide-63
SLIDE 63

Explosive Ambush

(video)

slide-64
SLIDE 64

Turret Wall

(video)

slide-65
SLIDE 65

Modern Combat AI Less "Expert System" and more “Creative AI."

  • Gameplay data-mining.
  • Optimization algorithms.
slide-66
SLIDE 66

Embracing Special Cases

slide-67
SLIDE 67

What‘s Next?

slide-68
SLIDE 68

On the AI Strategy for Killzone 2’s Bots

Alex Champandard – AiGameDev.com Remco Straatman – Guerrilla Tim Verweij – Guerrilla

slide-69
SLIDE 69

Dawn of War 2 AI Postmortem

Chris Jurney Senior Programmer Double Fine Productions

slide-70
SLIDE 70

Overview

  • What Went Wrong
  • What Went Right
  • Assorted Technical and Production Nuggets
slide-71
SLIDE 71

Goals for Dawn of War 2

  • Bringing Dawn of War into the

Company of Heroes engine

slide-72
SLIDE 72

What Went Wrong

slide-73
SLIDE 73

What Went Wrong

  • No Team Overlap

1.5 Programmers 25 Programmers 20 Programmers Company

  • f Heroes

Dawn of War 2

slide-74
SLIDE 74

What Went Wrong

  • No Team Overlap
  • Results: Lots of rework and lost effort
slide-75
SLIDE 75

What Went Wrong

  • Designer Bandwidth
slide-76
SLIDE 76

What Went Wrong

  • Designer Bandwidth

– AI design is a deep specialty – AI changes cascade (balance)

slide-77
SLIDE 77

What Went Right

slide-78
SLIDE 78

What Went Right

  • AI Programmer/AI Designer pair
  • Short iteration cycles
slide-79
SLIDE 79

Search Based Melee

slide-80
SLIDE 80

Touching

  • Company of Heroes – no touching
  • Dawn of War 2 – lots of touching
  • No touching example
slide-81
SLIDE 81

No Touching

slide-82
SLIDE 82

Touching

slide-83
SLIDE 83

Search Based Melee

  • Tabletop solution
  • Consider ring base

in A* search

  • Only accept goal

cells where rings exactly touch

slide-84
SLIDE 84
slide-85
SLIDE 85
slide-86
SLIDE 86

Search Based Melee

  • Lots of benefits…
  • Exact distance for animators
  • Extra room between units to increase

mobility of other unit

slide-87
SLIDE 87

Problem: Group Overlap

  • Multiple units attacking a single target bump

and overlap

  • Fix: Enforce circular bases between units

with the same target

slide-88
SLIDE 88

Group overlap example screen showing melee rings

slide-89
SLIDE 89
slide-90
SLIDE 90

Problem: Group Overlap

  • Decide whether to re-move based on current
  • verlap and priority
  • In A* search, only accept goal cells with no
  • verlap
  • Only perform fancier steps when near

enough melee target to matter

slide-91
SLIDE 91

Search Based Melee

Ork Rugby

slide-92
SLIDE 92
slide-93
SLIDE 93

Search Based Melee

slide-94
SLIDE 94

What Went Right

Defining Personality with AI

slide-95
SLIDE 95

Defining Personality

  • Dawn of War 2 units has fewer animations

than Company of Heroes units

  • Compensate for animation with AI
  • Create new plan functionality for new unit

capabilities

slide-96
SLIDE 96

Defining Personality

Un Under der Attack! k! Al All Squa uad: : Seek ek Cover er Nearby y (5m) Al All Squa uad: : Wait Few w Seconds

  • nds

Core: re: Advance ce 15m To Bet etter er Cover Flanks nks: : Advance ce 15m To Bet etter er Cover er

Company of Heroes Squad Plan

slide-97
SLIDE 97

Defining Personality

Move out!

slide-98
SLIDE 98
slide-99
SLIDE 99
slide-100
SLIDE 100
slide-101
SLIDE 101

Defining Personality

slide-102
SLIDE 102

Dawn of War 2 AI Postmortem

Chris Jurney Senior Programmer Double Fine Productions http://gdc.chrisjurney.com jurney@gmail.com