This project connects reinforcement learning (RL) in Python with interactive simulations in the Godot game engine. It uses Stable Baselines 3 for RL and supports custom environments built in Godot.
The project is under heavy refactoring to support Godot4 (which is totally incompatible with the current codebase). Development is done on the godot4-meson branch until things start getting usable.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results