CSCE 624 Aim Trajectory User Identification Project

This project collects mouse aiming trajectories from a Godot 3D target acquisition task and uses them for user identification experiments inspired by sketch-recognition stroke analysis methods.

Each trial produces a trajectory:

TargetSpawned → MouseMove events → ShotFired

These trajectories are converted into feature datasets for classification experiments.

---

Requirements

• Godot Engine (standard version, not .NET)
• Python 3

Python dependencies: pip install pandas numpy scikit-learn matplotlib

---

Running the Experiment (Godot)

1. Install Godot

2. Open Godot

3. Import this repository using project.godot

4. Open the scene: 3d_test_arena.tscn

5. Run the scene

The experiment presents aiming targets and automatically logs mouse trajectory data.

Press Escape to save and exit early if needed.

After completion, a CSV telemetry file is generated automatically.

---

Running Without Godot (Executable Export)

Godot can export a standalone executable so participants do not need the engine installed.

To export:

1. Open project in Godot
2. Go to Project → Export
3. Add a target platform
4. Export executable

Participants can run the executable and send back generated CSV files.

---

Locating Generated CSV Files

Godot stores telemetry in its user data directory.

On Windows:

Press Win + R and run: %appdata%\Godot\app_userdata\

Open the project folder and retrieve: trajectory_data_*.csv

Move these into: analysis/data/raw/

---

Data Pipeline

The analysis workflow is: raw telemetry CSV ↓ segment_trials.py ↓ segmented trajectory dataset

---

Segmenting Trajectories

Place raw CSV files in: analysis/data/raw/

Run: python analysis/segment_trials.py

This converts event-level telemetry into one row per trajectory.

Example output features: target_index duration_ms path_length mean_velocity num_points misses

Saved to: analysis/data/segmented_trials.csv

---

Project Structure

• CSCE624Project/
  • analysis/
    • data/
      • raw/
    • segment_trials.py
  • resultsDisplay/
  • 3d_test_arena.tscn
  • character_body_3d.gd
  • main.gd
  • main.tscn
  • project.godot

---

Notes

• Generated CSV telemetry files should not be committed
• Raw data belongs in analysis/data/raw/
• The main experiment logic is implemented in: character_body_3d.gd