Crowd Flow Statistical Modeling in the Unreal Development Kit Mid-quarter Update

Crowd Flow Statistical Modeling in the Unreal Development Kit

Brian May and Eric Baker

Computer Graphics II; 4005-762

Prof. Reynold Bailey

http://bam1286.blogspot.com/

Project Objectives

To generate a statistical model that will project actor behavior in a given space based on heatmap data from previously generated data.

• Crowd behavior implemented, still needs tuning

Heatmap Generation

• Parser written, needs tuning to get the appropriate data for our crowd agents
  
• Heatmap not generated yet

Project Responsibilities

Brian: Parser and heatmap generation
Eric: Crowd behavior and UDK Environment implementation

Project Timeline -- Currently on track to complete the project in time

• Week 5 - setup environment
• Week 6 - Initial Algorithm implementation / Heatmap generation
• Week 7 - Refine and identify problems with algorithm and statistics gathering
• Week 8 - Finalize implementation and begin to gather data
• Week 9 - Present Findings

What has been done so far

• Crowd behaviors implemented and working as we will want them to.
  
• UnrealScript is supporting our data collection nicely with its native logging functions
  
• Logging crowd agent positions effectively
  
• The log is parsed pretty easily in a simple C program
  

An assessment of your (team's) progress
We're in good shape to have the project done before the end of the quarter. We only have a couple more steps to achieve our goal of producing a heatmap based on the data gathered. After that we will need to tune / refine / revise our implementation to actually make it into a useful tool.