Open-Source Self-Driving Textbook on GitHub Builds an Autonomous System Chapter by Chapter
4310sy · x · 2026-11-14
A Japanese developer is writing a free, open textbook on autonomous driving algorithms, published incrementally on GitHub. The approach is "learn by building": starting from a bare world with a vehicle that moves by given commands, each chapter adds one stage to the system in a Python simulation — course following, sensing, localization, mapping, route planning, perception, and optimal control — before advancing to driving among other vehicles and obstacles.
The book has three parts: the base (simulator, frames/units, vehicle model), building the system (7 chapters from course following to optimal control), and driving among others (tracking moving objects, obstacle avoidance, lane changing, dynamic vehicle model). Every chapter ends with a runnable "system so far", and alternative algorithms for a stage can be swapped into the same system for comparison. The book is a work in progress with most chapters planned, and the author welcomes feedback in Discussions.
More from Research
- Recursive Model Collapse: How It Learned to Stop Worrying and Become the Assistant — sebkrier · 2026-11-14
- AI helps unlock hidden discoveries in math genius Paul Cohen's notebooks — irrelevantusername24 · 2026-11-14
- SGLang Summit talk details battle-tested NVFP4 4-bit RL across thousands of GPUs — ericzelikman · 2026-11-14
- Decision models tested: training buys speed and cost, not reasoning back — someone_12443 · 2026-11-14
- Ropedia open-sources Xperience 2: 1,589 3D scene scans and 131 hours of egocentric demos for physical AI — liuziwei7 · 2026-11-14
- RobotAgentBench: hybrid Codex-InternVLA-A1.5 system averages 48.5 on household agent tasks — jiqizhixin · 2026-11-14