Crowd Dynamics Simulation Research

1
Crowd Dynamics Simulation Research

• Dr. D. J. Kaup
• Dr. T. L. Clarke
• Rex Oleson
• Dr. Linda C. Malone
• University of Central Florida
• 3100 Technology Parkway
• Orlando, FL 32826

2
Overview

• Generalize Lakoba-Kaup- Finkelstein crowd
  simulation model to incorporate personality
  factors
• Improved simulation of crowd behaviors in crowded
  environments.

3
HMFV Equations
4
Sample HMFV Crowd Simulation

• Mason Animation
• Standard HMFV Model

5
Density Dependent Panic Factors Added LKF Model
Snapshot of a non-panicking crowd. Larger
circles represent bodies, and smaller ones,
noses, of pedestrians in this way, the direction
of motion of a pedestrian becomes apparent.
6
Break Away Behavior
Snapshot of a non-panicking crowd a few seconds
after a group of pedestrians at the back of the
crowd breaks away".
7
Exit Angle Results

• Based on NSF-funded Lakoba-Kaup-Finkelstein model

Number of people exiting per 100 seconds
8
Verification Using Crowd VideoCitrus Bowl
Crowd UCF-Tulane 2005
9
Vector Field of Optical FlowLucas-Kanade
Algorithm
10
Add Personality

• Incorporate Personality Factors into HMFV/LKF
• Big Five
• Neuroticism
• Extroversion
• Openness
• Agreeableness
• Conscientiousness

11
Math Approach

• Methods
• Add personality vector to simulation running
  under MASON

12
Math Approach

• Quadratic form implements interpersonal force
  term

13
Interaction Matrix
14
Trial Matrix
Neurotic gt Blue Extrovert gt Cyan Openness gt
Green Agreeable gt Yellow Consci gt Red
Neur Extro Open Agree Consc
Neur 1 -0.1 -0.1 -0.1 -0.1
Extro -0.1 1 1 1 1
Open -0.1 1 1 1 1
Agree -0.1 1 1 1 1
Consc -0.1 1 1 1 1
15
HMFV? NEO Transition
16
More Distant NEO Transition

• Mason Animation

17
Less Distant NEO Transition

• Mason Animation

18
Incorporating Gender
19
Future Research

• Refine
• B Matrix
• HMFV? NEO Transition
• Products
• Model of crowds and third-party actor behavior
• Detect IEDs via behavior anomalies

20
Acknowledgements
This research has been supported in part by the
National Science Foundation under Grant No.
BCS-0527545 and by the U.S. Army Research
Development and Engineering Command, Simulation
Technology Center, Contract N61339-02-C-0107.