🧭 VR Navigation Analysis

---

Cluster Analysis on Subject Data in Immersive Virtual Environments

A framework for analyzing navigation patterns in virtual reality environments
using k-means clustering to identify behavioral strategies in maze exploration

Ruida Zeng, Richard Paris, Bobby Bodenheimer

Learning in Virtual Environments (LIVE) Lab, Vanderbilt University

Overview • Team • Quick Start • Results • Structure • Requirements • Documentation

---

Overview

Analysis of navigation patterns from ~200 subjects in an immersive virtual maze environment. This research was conducted during the 2019 VUSE Undergraduate Summer Research Program.

The project applies k-means clustering to identify distinct navigation strategies and behavioral patterns as subjects explore and learn a virtual maze over 10-minute sessions.

---

Team

Role	Name	Contribution
Primary Investigator	Dr. Bobby Bodenheimer	Project oversight, EECS Department
Graduate Research Assistant	Richard Paris	VR environment creation (Unity), experiment design
Undergraduate Researcher	Ruida Zeng	Data analysis, clustering algorithms

📧 Contact: ruida dot zeng at vanderbilt dot edu

---

Quick Start

Installation

# Clone the repository
git clone https://github.com/ruidazeng/vr-navigation.git
cd vr-navigation

# Install dependencies
pip install -r requirements.txt

Run Examples

# Visualize a subject's navigation path
python visualization/visual.py

# Run k-means clustering on all subjects
python clustering/km_vector.py

# View 3D temporal clustering
python clustering/km_3d.py

---

Results

K-Means Clustering	3D Temporal View
Temporal Color Coding	Path Visualization

---

Project Structure

vr-navigation/
├── clustering/          # K-means clustering algorithms
├── visualization/       # Data visualization scripts  
├── preprocessing/       # Legacy data parsing scripts
├── Parsed Data/         # 143 subject CSV files
└── Resources/           # Images and documentation

Directory	Description	Docs
clustering/	K-means clustering for navigation pattern analysis	📖 README
visualization/	Matplotlib and turtle graphics visualizations	📖 README
preprocessing/	Legacy scripts for parsing raw experimental data	📖 README
Parsed Data/	143 parsed CSV files with subject navigation data	📖 README

---

Requirements

Package	Version	Purpose
numpy	1.19+	Numerical computing
pandas	1.0+	Data manipulation
scikit-learn	0.24+	K-means clustering
matplotlib	3.3+	Visualization

pip install -r requirements.txt

---

Documentation

Document	Description
Clustering Scripts	K-means algorithms and usage
Visualization Scripts	Path visualization tools
Preprocessing Scripts	Legacy data parsing documentation
Parsed Data	Data format and subject status
LICENSE	License information

---

_{Made with ❤️ at Vanderbilt University • 2019 Summer Research}