Repository Structure

Simulation & Core Code

• maze_project/: Core simulation, planning, control, and ML code
• maze_project/main.py: Entry point — CLI arguments and main simulation loop
• maze_project/simulation/: PyBullet simulation layer (robot, maze, sensors, orchestrator)
• maze_project/maze_generator/: Procedural maze generation and JSON serialization
• maze_project/planning/: Path planning algorithms (A*, BFS, DFS)
• maze_project/control/: Motion control layer (PD waypoint-tracking controller)
• maze_project/logger/: Rate-limited CSV data logger for collecting training data
• maze_project/tools/: Helper scripts for maze generation and batch data collection
• maze_project/data/mazes/: Saved maze JSON files used for simulation and data collection
• maze_project/README.md: Detailed documentation of the simulation architecture and usage

Machine Learning

• maze_project/learning/: ML experiment notebooks
• maze_project/learning/archive/supervised_linear_mlp.ipynb: Supervised heading regression (linear, ridge, MLP) with maze-grouped train/test splits
• maze_project/learning/archive/unsupervised_clustering.ipynb: Unsupervised clustering (K-Means, GMM) with rotation-invariant features and PCA visualization

Environment Setup

• conda/: Conda environment configuration and setup scripts
• conda/environment.yml: Dependency definitions (Python 3.11, PyBullet, scikit-learn, etc.)
• conda/setup.sh: Setup script for Linux / macOS
• conda/setup.bat: Setup script for Windows

Archive

• archive/docs/images/: Figures and visualizations referenced in the midterm and final report
• archive/index.html : Report for project proposal (includes general contributions from teammates)
• archive/midterm.html : Report for project midterm (includes general contributions from teammates)
• archive/final.html : Report for project final (includes general contributions from teammates)