Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/Farama-Foundation/Miniworld
Simple and easily configurable 3D FPS-game-like environments for reinforcement learning
https://github.com/Farama-Foundation/Miniworld
Last synced: 3 months ago
JSON representation
Simple and easily configurable 3D FPS-game-like environments for reinforcement learning
- Host: GitHub
- URL: https://github.com/Farama-Foundation/Miniworld
- Owner: Farama-Foundation
- License: apache-2.0
- Created: 2018-08-28T03:12:59.000Z (about 6 years ago)
- Default Branch: master
- Last Pushed: 2024-04-11T18:07:11.000Z (7 months ago)
- Last Synced: 2024-05-01T10:44:00.587Z (6 months ago)
- Language: Python
- Homepage: http://miniworld.farama.org/
- Size: 146 MB
- Stars: 671
- Watchers: 18
- Forks: 126
- Open Issues: 6
-
Metadata Files:
- Readme: README.md
- Funding: .github/FUNDING.yml
- License: license
- Code of conduct: CODE_OF_CONDUCT.rst
Awesome Lists containing this project
- awesome-AI-books - gym-miniworld
- awesome-production-machine-learning - MiniWorld - Foundation/Miniworld.svg?style=social) - MiniWorld is a minimalistic 3D interior environment simulator for reinforcement learning & robotics research. (Industry Strength RL)
README
Miniworld is being maintained by the Farama Foundation (https://farama.org/project_standards). See the [Project Roadmap](https://github.com/Farama-Foundation/Miniworld/issues/103) for details regarding the long-term plans.
[](https://travis-ci.org/maximecb/gym-miniworld)
Contents:
- [Introduction](#introduction)
- [Installation](#installation)
- [Usage](#usage)
- [Environments](docs/environments.md)
- [Design and Customization](docs/design.md)
- [Troubleshooting](docs/troubleshooting.md)
## Introduction
MiniWorld is a minimalistic 3D interior environment simulator for reinforcement
learning & robotics research. It can be used to simulate environments with
rooms, doors, hallways and various objects (eg: office and home environments, mazes).
MiniWorld can be seen as a simpler alternative to VizDoom or DMLab. It is written
100% in Python and designed to be easily modified or extended by students.
Features:
- Few dependencies, less likely to break, easy to install
- Easy to create your own levels, or modify existing ones
- Good performance, high frame rate, support for multiple processes
- Lightweight, small download, low memory requirements
- Provided under a permissive MIT license
- Comes with a variety of free 3D models and textures
- Fully observable [top-down/overhead view](images/maze_top_view.jpg) available
- [Domain randomization](https://blog.openai.com/generalizing-from-simulation/) support, for sim-to-real transfer
- Ability to [display alphanumeric strings](images/textframe.jpg) on walls
- Ability to produce depth maps matching camera images (RGB-D)
Limitations:
- Graphics are basic, nowhere near photorealism
- Physics are very basic, not sufficient for robot arms or manipulation
List of publications & submissions using MiniWorld (please open a pull request to add missing entries):
- [Decoupling Exploration and Exploitation for Meta-Reinforcement Learning without Sacrifices](https://arxiv.org/abs/2008.02790) (Stanford University, ICML 2021)
- [Rank the Episodes: A Simple Approach for Exploration in Procedurally-Generated Environments](https://openreview.net/forum?id=MtEE0CktZht) (Texas A&M University, Kuai Inc., ICLR 2021)
- [DeepAveragers: Offline Reinforcement Learning by Solving Derived Non-Parametric MDPs](https://arxiv.org/abs/2010.08891) (NeurIPS Offline RL Workshop, Oct 2020)
- [Pre-trained Word Embeddings for Goal-conditional Transfer Learning in Reinforcement Learning](https://arxiv.org/abs/2007.05196) (University of Antwerp, Jul 2020, ICML 2020 LaReL Workshop)
- [Temporal Abstraction with Interest Functions](https://arxiv.org/abs/2001.00271) (Mila, Feb 2020, AAAI 2020)
- [Addressing Sample Complexity in Visual Tasks Using Hindsight Experience Replay and Hallucinatory GANs](https://openreview.net/forum?id=H1xSXdV0i4) (Offworld Inc, Georgia Tech, UC Berkeley, ICML 2019 Workshop RL4RealLife)
- [Avoidance Learning Using Observational Reinforcement Learning](https://arxiv.org/abs/1909.11228) (Mila, McGill, Sept 2019)
- [Visual Hindsight Experience Replay](https://arxiv.org/pdf/1901.11529.pdf) (Georgia Tech, UC Berkeley, Jan 2019)
This simulator was created as part of work done at [Mila](https://mila.quebec/).
## Installation
Requirements:
- Python 3.7+
- Gymnasium
- NumPy
- Pyglet (OpenGL 3D graphics)
- GPU for 3D graphics acceleration (optional)
You can install it from `PyPI` using:
```console
python3 -m pip install miniworld
```
You can also install from source:
```console
git clone https://github.com/Farama-Foundation/Miniworld.git
cd Miniworld
python3 -m pip install -e .
```
If you run into any problems, please take a look at the [troubleshooting guide](docs/troubleshooting.md).
## Usage
There is a simple UI application which allows you to control the simulation or real robot manually.
The `manual_control.py` application will launch the Gym environment, display camera images and send actions
(keyboard commands) back to the simulator or robot. The `--env-name` argument specifies which environment to load.
See the list of [available environments](docs/environments.md) for more information.
```
./manual_control.py --env-name MiniWorld-Hallway-v0
# Display an overhead view of the environment
./manual_control.py --env-name MiniWorld-Hallway-v0 --top_view
```
There is also a script to run automated tests (`run_tests.py`) and a script to gather performance metrics (`benchmark.py`).
### Offscreen Rendering (Clusters and Colab)
When running MiniWorld on a cluster or in a Colab environment, you need to render to an offscreen display. You can
run `gym-miniworld` offscreen by setting the environment variable `PYOPENGL_PLATFORM` to `egl` before running MiniWorld, e.g.
```
PYOPENGL_PLATFORM=egl python3 your_script.py
```
Alternatively, if this doesn't work, you can also try running MiniWorld with `xvfb`, e.g.
```
xvfb-run -a -s "-screen 0 1024x768x24 -ac +extension GLX +render -noreset" python3 your_script.py
```
# Citation
To cite this project please use:
```bibtex
@article{MinigridMiniworld23,
author = {Maxime Chevalier-Boisvert and Bolun Dai and Mark Towers and Rodrigo de Lazcano and Lucas Willems and Salem Lahlou and Suman Pal and Pablo Samuel Castro and Jordan Terry},
title = {Minigrid \& Miniworld: Modular \& Customizable Reinforcement Learning Environments for Goal-Oriented Tasks},
journal = {CoRR},
volume = {abs/2306.13831},
year = {2023},
}
```