Feel free to join our Discord for help and discussions about Godot RL Agents.
Godot RL Agents is a fully Open Source package that allows video game creators, AI researchers and hobbyists the opportunity to learn complex behaviors for their Non Player Characters or agents. This repository provides:
- An interface between games created in the Godot Engine and Machine Learning algorithms running in Python
- Wrappers for four well known rl frameworks: StableBaselines3, Sample Factory, Ray RLLib and CleanRL.
- Support for memory-based agents, with LSTM or attention based interfaces
- Support for 2D and 3D games
- A suite of AI sensors to augment your agent's capacity to observe the game world
- Godot and Godot RL Agents are completely free and open source under the very permissive MIT license. No strings attached, no royalties, nothing.
You can find out more about Godot RL agents in our AAAI-2022 Workshop paper.
godot_rl_agents_trailer_v01_20211008.mp4
This quickstart guide will get you up and running using the Godot RL Agents library with the StableBaselines3 backend, as this supports Windows, Mac and Linux. We suggest starting here and then trying out our Advanced tutorials when learning more complex agent behaviors. There is also a video tutorial which shows how to install the library and create a custom env.
Install the Godot RL Agents library. If you are new to Python or not using a virtual environment, it's highly recommended to create one using venv or Conda to isolate your project dependencies.
Once you have set up your virtual environment, proceed with the installation:
pip install godot-rl
Download one, or more of examples, such as BallChase, JumperHard, FlyBy.
gdrl.env_from_hub -r edbeeching/godot_rl_JumperHard
You may need to add run permissions on the game executable.
chmod +x examples/godot_rl_JumperHard/bin/JumperHard.x86_64
Train and visualize, first download the Stable Baselines 3 example script in the github repo:
python examples/stable_baselines3_example.py --env_path=examples/godot_rl_JumperHard/bin/JumperHard.x86_64 --experiment_name=Experiment_01 --viz
Deprecated usage with an entrypoint (to be removed in version 1.0):
gdrl --env=gdrl --env_path=examples/godot_rl_JumperHard/bin/JumperHard.x86_64 --experiment_name=Experiment_01 --viz
You can also train an agent in the Godot editor, without the need to export the game executable.
- Download the Godot 4 Game Engine (.NET version) from https://godotengine.org/
- Open the engine and import the JumperHard example in
examples/godot_rl_JumperHard
- Start in editor training with:
python examples/stable_baselines3_example.py
There is a dedicated tutorial on creating custom environments here. We recommend following this tutorial before trying to create your own environment.
If you face any issues getting started, please reach out on our Discord or raise a GitHub issue.
The latest version of the library provides experimental support for onnx models with the Stable Baselines 3, rllib, and CleanRL training frameworks.
- First run train you agent using the sb3 example (instructions for using the script), enabling the option
--onnx_export_path=GameModel.onnx
- Then, using the mono version of the Godot Editor, add the onnx model path to the sync node. If you do not seen this option you may need to download the plugin from source
- The game should now load and run using the onnx model. If you are having issues building the project, ensure that the contents of the
.csproj
and.sln
files in you project match that those of the plugin source.
README_FPS.mp4
Please ensure you have successfully completed the quickstart guide before following this section.
Godot RL Agents supports 4 different RL training frameworks, the links below detail a more in depth guide of how to use a particular backend:
- StableBaselines3 (Windows, Mac, Linux)
- SampleFactory (Mac, Linux)
- CleanRL (Windows, Mac, Linux)
- Ray rllib (Windows, Mac, Linux)
We welcome new contributions to the library, such as:
- New environments made in Godot
- Improvements to the readme files
- Additions to the python codebase
Start by forking the repo and then cloning it to your machine, creating a venv and performing an editable installation.
# If you want to PR, you should fork the lib or ask to be a contibutor
git clone git@github.com:YOUR_USERNAME/godot_rl_agents.git
cd godot_rl_agents
python -m venv venv
pip install -e ".[dev]"
In order to run the tests, you'll need to make sure that you first have git-lfs installed. Once this is installed, you can download the examples which are used in the tests:
make download_examples
Note that the examples are only available for Linux and Windows. Once you've installed the examples, you can run the tests:
# check tests run
make test
Then add your features. Format your code with:
make style
make quality
Then make a PR against main on the original repo.
The objectives of the framework are to:
- Provide a free and open source tool for Deep RL research and game development.
- Enable game creators to imbue their non-player characters with unique behaviors.
- Allow for automated gameplay testing through interaction with an RL agent.
Please try it out, find bugs and either raise an issue or if you fix them yourself, submit a pull request.
This should now be working, let us know if you have any issues.
If the README and docs here not provide enough information, reach out to us on Discord or GitHub and we may be able to provide some advice.
We are inspired by the the Unity ML agents toolkit and we aim to provide a more compact, concise and hackable codebase, with little abstraction.
Godot RL Agents is MIT licensed. See the LICENSE file for details.
"Cartoon Plane" (https://skfb.ly/UOLT) by antonmoek is licensed under Creative Commons Attribution (http://creativecommons.org/licenses/by/4.0/).
@article{beeching2021godotrlagents,
author={Beeching, Edward and Dibangoye, Jilles and
Simonin, Olivier and Wolf, Christian},
title = {Godot Reinforcement Learning Agents},
journal = {{arXiv preprint arXiv:2112.03636.},
year = {2021},
}
We thank the authors of the Godot Engine for providing such a powerful and flexible game engine for AI agent development. We thank the developers at Sample Factory, Clean RL, Ray and Stable Baselines for creating easy to use and powerful RL training frameworks. We thank the creators of the Unity ML Agents Toolkit, which inspired us to create this work.