// the find
ShisatoYano/AutonomousVehicleControlBeginnersGuide
Python sample codes and documents about Autonomous vehicle control algorithm. This project can be used as a technical guide book to study the algorithms and the software architectures for beginners.
A Python simulation library covering the full autonomous vehicle stack: localization (EKF, UKF, particle filter), mapping (occupancy grid, NDT, potential field), path planning (A*, RRT variants, PRM, PSO), path tracking (Stanley, LQR, MPC, MPPI), and basic perception. It's a learning resource first, not a deployable system — everything runs in matplotlib with simulated sensors.
Unusually broad coverage for a single repo: most similar projects pick one domain (just path planning, or just SLAM), this one implements the whole pipeline end to end. Each algorithm gets its own isolated module with a matching simulation script, so you can run a specific algorithm in 30 seconds without untangling a monolith. CI runs on Linux, Windows, and macOS, which is rare for a Python robotics edu project and means the numpy/scipy dependencies are actually verified across platforms. The animated GIFs in the README are genuinely useful — you can see what each algorithm does before reading a line of code.
Perception coverage is thin: rectangle fitting from LiDAR point clouds and sensor calibration only, no camera pipeline, no fusion between sensors — someone wanting to study perception will hit a wall fast. The codebase implements its own KD-tree from scratch rather than using scipy.spatial.KDTree, which is a red flag for correctness and performance in a teaching context where you'd want learners comparing against known-good implementations. Documentation says 'still not completed' and the design documents cover only a fraction of the algorithms, so the code itself has to carry explanatory weight it wasn't fully written to carry. No ROS integration or real hardware path — this is strictly a matplotlib sandbox, which limits how far a learner can carry these implementations toward an actual vehicle.