Rocket League Bot

Notes from building a hard-coded Rocket League bot in Python — game-space vectors, ground and aerial kinematics, path planning and shot alignment.

Notes from building a Rocket League bot in Python using the RLBot framework. The bot is hard-coded rather than machine-learned — every behaviour is explicit maths and logic — which makes the project a nice excuse to work through the game’s physics properly. These pages document that maths so it’s useful to anyone building their own bot.

The bot’s source code is on GitHub.

Sections

1. Project Overview — what the RLBot framework provides, how the bot is structured, and the in-game debug rendering used during development.
2. 3D Vector System — the Vec3 class underpinning everything: distances, angles and dot products in game space.
3. Ground Kinematics — how the car accelerates, brakes, coasts and boosts on the ground, with the values that govern each.
4. Aerial Kinematics — solving for the orientation and boost needed to reach a point in the air at a given time.
5. Aligning Shots on Target — hitting the ball is easy; hitting it towards the goal needs an offset approach point.
6. Path Optimisation Analysis — turning radius vs speed, and estimating time-to-target along realistic paths.