Crowd Scattering A Good Starting Point

1
Crowd Scattering? A Good Starting Point

• Bradley C. Schricker
• Sonia R. von der Lippe
• ATT Government Solutions, Inc.
• Orlando, FL
• 2004 Spring SIW

2
Overview

• Introduction
• Military Operations Other Than War (MOOTW)
• What is Crowd Scattering?
• Similar Scattering Work
• Crowd Scattering Development
• Crowd Scattering and Beyond
• Summary and Conclusion

3
Introduction
4
Introduction, cont
5
Introduction, cont
Hiroshima mushroom cloud, August 6, 1945
6
Introduction, cont
Nagasaki mushroom cloud, August 9, 1945
7
Introduction, cont

• SO WHAT???

8
Introduction, cont

• Could be considered the introduction of mass
  counter-terrorism a major aspect of MOOTW
• Why?
• Up until then, military operations typically
  pitted equal forces against each other
• By dropping A-bombs, a small squadron was able to
  decimate entire cities with little or no warning
  (sound familiar?)

9
MOOTW

• Military Operations Other Than War
• Military operations fit into two categories
• War
• MOOTW

FM 100-11 Force Integration,
www.adtdl.army.mil/cgi-bin/atdl.dll/fm/10011/fm100
-11.pdf, January 5, 1998.
10
MOOTW, cont

• Characteristics of War
• Direct and usually declared combat
• Use of overwhelming force
• Infliction of as few military and particularly
  civilian casualties as possible
• Goal is to cease combat on terms favorable to
  United States and allies

11
MOOTW, cont

• Characteristics of MOOTW
• Mostly non-combat operations
• Can include armed conflict
• Counter-drug activities
• Disaster relief
• Civil support tasks
• National assistance

12
MOOTW, cont

• Strikes and raids
• Enforcement of the peace
• Support against violent insurgence
• Counter-terrorism
• Non-combat related evacuations

13
MOOTW, cont

• Specifically, counter-terrorism is the focal
  point of this paper
• It is also the tie-in with WWII A-bombs
• Furthermore, terrorism-related operations are
  witnessed on some scale by Americans every single
  day
• February 26, 1993 (World Trade Center Parking
  Garage)
• April 19, 1995 (Oklahoma City bombing)
• October 12, 2000 (Attack against the U.S.S. Cole)
• September 11, 2001 (needs no explanation)
• Daily attacks against U.S. soldiers in Iraq

14
MOOTW, cont

• What does all of this mean?

15
MOOTW, cont

• Two things
• Terrorism has become a daily part of our world
• Counter-terrorism must improve

Smith, Michael M. and Melinda Hoffstetter,
Conduit or Cul-de-sac? Information Flow in
Civil-Military Operations, Joint Force
Quarterly, Spring 1999, pages 100-5.
16
What is Crowd Scattering?

• The name for collective movements of a group of
  individuals fleeing in a panic state from a
  perceived danger
• An individual behavior
• At first, each individual will behave in a unique
  way
• Also a group behavior
• As individuals find successful avenues of escape,
  more will follow

17
Crowd Scattering, cont

• Could be applied to
• Counter-terrorism
• Hostage rescue
• Disaster relief
• Urban warfare

18
Crowd Scattering, cont

• Our scenario
• In an airport
• Populated with hundreds of civilians
• Pedestrians
• Airport employees
• With no warning, a single man begins to shoot a
  firearm, potentially striking some number of the
  civilians

19
Other Crowd Scattering Work

• Crowd movement
• Movies
• Disneys Hunchback of Notre Dame
• Jurassic Park
• Lord of the Rings trilogy
• Franco Tecchia Urban Behavior Representations
• Daniel Thalman Individual and group interactions

Christensen, Per H., Non-Photorealistic
Rendering, ACM SIGGRAPH, Vol. 32, No. 1,
February, 1999. 6 Hodgins, Jessica K. and James
F. OBrien, Computer Animation, Encyclopedia of
Computer Science, Fourth Edition, MacMillan
Publishers Ltd., August, 2000. 7 Tecchia,
Franco, Céline Loscos, Ruth Conroy, and Yiorgos
Chrysanthou, Agent Behaviour Simulator (ABS) A
Platform for Urban Behaviour Development,
Proceedings of the 2001 Games Technology
Conference, Hong Kong, China, January 17-20,
2001. 8 Thalmann, Daniel, Soraia Raupp Musse,
and Marcelo Kallmann, From Individual Human
Agents to Crowds, Informatik / Informatique,
No1, 2000.
20
Other Work, cont

• Crowd Fleeing
• Sounds similar
• Involves planned movement
• Typically, also involves modeling a crowd as a
  single unit

Ulicny, B., Thalmann, D., Crowd simulation for
interactive virtual environments and VR training
systems, Proc. Eurographics Workshop on Animation
and Simulation '01, pp. 163-170, Springer-Verlag,
2001.
21
Crowd Scattering Development

• Consists of two separate stages
• Random motion
• Freeze
• Duck
• Run in a random direction
• Exit immediately
• Possibly more extreme reactions such as heart
  attacks
• Flock
• Based on algorithm credited to Thomas G. Grubb
• Group behavior emerges at this point

22
Development, cont

• Demonstration application of Grubbs algorithm
  can be downloaded from
• http//www.riversoftavg.com/flocking.htm
• Individual entities can be controlled by
• Separation Strength
• Alignment Strength
• Cohesion Strength
• Avoidance Strength

23
Development, cont
24
Crowd Scattering and Beyond

• Already have the foundation for testing our
  scattering algorithm
• We have demonstrated a distributed simulation
  between a COTS game engine and our own simulation
  engine
• Game engine handles extensive graphics
  applications better than simulation engine
• Simulation engine handles communications and
  overhead operations better than game engine
• Used both to get best of both worlds
• Simulation was demonstrated at 2002 I/ITSEC in
  Orlando, FL

Kayne, L. and Chludzinski, J., Game
Engine-Simulation Engine Hybrid IRD Final
Report, ATT Government Solutions, Inc. December
31, 2002.
25
Beyond, cont
26
Beyond, cont
27
Summary and Conclusion

• Summary
• MOOTW
• Crowd scattering in general
• Previous crowd behavior work
• Our algorithm
• Applying the algorithm

28
Summary and Conclusion, cont

• Conclusion
• This behavior stands out as interesting because
  of its hybrid nature between individual and group
  modeling
• Could lead to improved training environments for
  specific MOOTW operations
• This work has clear benefits for MOOTW training
  research and development

29
(No Transcript)