Battle Swarm: An Evolutionary Approach to Complex Swarm Intelligence - PowerPoint PPT Presentation

About This Presentation
Title:

Battle Swarm: An Evolutionary Approach to Complex Swarm Intelligence

Description:

Battle Swarm: An Evolutionary Approach to Complex Swarm Intelligence Russel Ahmed Apu Marina Gavrilova Brief Outline Battle Swarms Tactics and battle efficiency Swarm ... – PowerPoint PPT presentation

Number of Views:367
Avg rating:3.0/5.0
Slides: 37
Provided by: RusselA2
Category:

less

Transcript and Presenter's Notes

Title: Battle Swarm: An Evolutionary Approach to Complex Swarm Intelligence


1
Battle Swarm An Evolutionary Approach to Complex
Swarm Intelligence
  • Russel Ahmed Apu
  • Marina Gavrilova

2
Brief Outline
  • Battle Swarms
  • Tactics and battle efficiency
  • Swarm Intelligence Missile Genotype Encoding
  • Evolutionary strategies for battle swarms
  • Experimental results and analysis

3
Objective
  • To utilize swarm based tactics evolutionary
    swarm strategies to increase tactical efficiency
    for offensive and defensive agents

4
Battle Swarms Agents
  • MISSILES
  • Autonomous
  • Limited sensory capabilities
  • Limited intelligence
  • Single objective Hit ship
  • Complex dynamic system
  • Behavior of one missile effect other missiles in
    the swarm
  • Evolutionary Strategy
  • DEFENSE TURRETS
  • Point Defense system
  • Only Visual/radar capabilities
  • Limited coverage
  • Single Objective Destroy missiles
  • Simple rule, complex outcome Select and fire
  • Behavior and efficiency cruicial to survival
  • Fixed Strategy

5
Defense Mechanism
Reaction
Missile
Reaction Radius
Missile
Missile
Point Defense
6
Actions of a Missile
Action Encoding Set LRUDNMFAXYZ
XRand YConverge L Roll Left UPitch Up
NNOP FFollow
ZDiverge RRoll Right DPitch Down
MMemory AAvoid
PITCH UP
Up
ROLL RIGHT
Constant Thrust
Heading
PITCH DOWN
ROLL LEFT
Right
Discussed in the next few slides
7
Basic Sensory Encoding and Actions
Follow Target
COG
Proj(u)
(2) Pitch up
Proj(v)
(1) Roll to match proj(v)proj(u)
Projection Plane
8
Basic Sensory Encoding and Actions
(2) Pitch up
Avoid Target
COG
(1) Roll to match proj(v).proj(u)0
Projection Plane
9
Target Relative Coordinate(Range, Heading,
Bearing)
10
Target relative Heading
11
Target Relative Bearing
COG
u
Rproj
b
PRproj
F
Projection Plane
12
Swarm Relative Encoding
  • Regulates the probability of Flocking Tendency
  • Y Flock and increase tendency (probability ? of
    Boids flocking)
  • Z Diverge from flock and decrease tendency
  • If an agents decides to flock (prob ?), the
    direction is determined using modified BOIDS

13
Decision Making
  • Event related decision are made by the swarm
    implicitly
  • Avoiding Incoming fire Ionization trail gives
    negative pheromone to allow flocking out of a
    region
  • Finding Weakness in Defense Combined usage of
    flocking tendency, gas and ionization pheromone
    trail

14
Basic Encoding of Missile genotype
  • String of Possible Action (I.e LYUXLY)
  • Action string is circular (iterative)
  • Missile DNAGene_String
  • Continuous execution of the string
  • Each action executed for an infinite time
  • Regulates Swarm Behavior/Tendency

LYUXLY
15
Encoding Basic Maneuvers
  • Maintain Current Heading N
  • Homing the Target F
  • Ring Motion U
  • Cork Screw LU, LUMMM
  • Evasive Approach XF, XMMMF
  • Basic Evasive Action A, AMMMX
  • Fall Back XU, XMMMU, AU
  • Scramble X

16
Basic Maneuvers
XF
F
U
N
LU
A
17
Different Complex Maneuvering Tactics
  • Retaliation frontal attack
  • Evasive avoid fire at all costs
  • Convergent approach approach target from a
    particular direction
  • Divergent approach surround and approach from
    different directions
  • Trail wind flocking one missile leads others
  • Distract and draw fire

18
Different Complex Maneuvering Tactical strategies
  • Diversion (b) Trail Wind Flocking
  • (c) Retaliation (d) Divergence

19
Mutating and Evolving the Missile Genotype
  • Fitness Define a fitness function for the
    desired action
  • Crossover Augment/concatenate Genes
  • LUMU AMD ?LUAMDMU LUMD AMDUMU
    LMDMU
  • Randomization Replace arbitrary symbols with
    X run the simulation and convert meta genes to
    real genes
  • FFLLU ? FXLXU ? FULLU, FFLFU, FNLNU,
    FMLNU ? BestFULLU, FNLNU

20
Induced Evolution
  • We can introduce certain desired behavior in
    addition to natural evolution
  • Step 1 Train Missiles separately to obtain
    certain desired behavior without any other
    consideration. Obtain Viral strain W
  • Step 2 Infect All current Genotype with viral
    Strain W (crossover)

21
The Game Co-evolution
  • Implement basic missile F and basic Turret
    Select X, Fire_at_trajectory
  • Adjust physical property to match
  • Fitness50 (50 missiles hit the target)
  • Evolve Missiles and turrets against previous
    strain
  • Repeat step 3 for several Games cycles
  • If fitness falls or rises dramatically increase
    physical strength of opposing swarm (Missile
    Acceleration, velocity, turning. Turrets Speed
    of fire, number of turrets, firing frequency)

22
The Fitness Function Hetero-Sexual Mating
  • Use a two dimensional Fitness Function
  • Every missile has a masculine and a feminine
    fitness
  • Masculine Ability to Attack
  • Feminine Ability to Survive

23
Results
  • - Strategies evolved, Runtime and other aspects

24
Fitness Function
50
45
40
35
30
Feminine
25
20
15
10
5
0
0
20
40
60
80
100
Masculine
No Randomization
Randomized
25
Complex Tactics Convergent Approach
  • High Efficiency
  • Low evasion
  • Highly Masculine
  • Strength in numbers
  • Less exposure to incoming fire
  • Increase of spatial threat
  • Decrease of temporal threat

26
Complex Tactics Divergent Approach
  • Cause more distraction and confuse the defense
    system
  • Less likelihood for a missile to draw fire
  • Decrease of spatial threat
  • Increase of temporal threat
  • Lower Efficiency
  • Highly Evasive
  • Highly Feminine

27
Convergent VS Divergent Approach
  • DIVERGENT
  • More defense turrets
  • Draw more fire
  • Distracting and hard to shoot down
  • CONVERGENT
  • Less defense turrets
  • Draw less fire
  • Easy to shoot down

28
Complex Tactics Trail Wind Flocking
  • Better than Convergent Approach
  • Least exposure to incoming fire
  • Lot of opportunity for diversion/distraction
  • Decrease of spatial threat
  • Decrease of average temporal threat

29
More Results
30
More Results
31
More Results
  • See Animation Demos

32
Rendering and Physical Engine
  • Regular physics engine will not suffice
  • Approximation aggravates trajectory computation
  • Construct everything from scratch
  • Advanced look-ahead estimation based physics
    engine
  • Robust Rendering engine
  • Anisotropic Texture filtering
  • Multiple LOD based geometry rendering
  • Particle engine
  • Highly optimized exclusive API for performance
  • Flexibility

33
The Simulation Engine
  • Robust design Separation of Rendering modules
    from the simulation
  • Implements Command Console
  • Runtime performance is highly efficient
  • For 50 missiles
  • Full quality rendering_at_ 50FPS !!!
  • Simulation runs upto 50 times faster (FPS2200)
    is rendering is turned off (for evolutionary
    algorithm)
  • Excellent Rendering quality (anisotropic texture
    mapping, particle engine)

34
Runtime Performance
35
Summary
  • Using Swarm Intelligence to evolve battle tactics
    for
  • Missiles
  • Point Defense Turrets
  • Evolutionary strategies
  • Gene_String evolution
  • The novel Induced Evolution strategy
  • Co-evolutionary strategy
  • Implementation
  • Rendering and physical Engine
  • Genotype encoding
  • Basic maneuvers
  • Complex maneuvers
  • Integration

36
Thank you
Write a Comment
User Comments (0)
About PowerShow.com