站内导读
定位:MuJoCo(Multi-Joint dynamics with Contact)是 Google DeepMind 维护的通用物理引擎,用于机器人、生物力学、图形动画与机器学习中对关节结构及其环境接触的快速精确仿真。
核心能力
- 运行时仿真模块针对性能调优,运行在内置 XML 编译器预先分配的低层数据结构上,减少运行期内存开销。
- 以 C API 为核心,另提供 Python 绑定与 Unity 引擎插件,便于接入不同技术栈。
- 支持原生桌面窗口与浏览器内的交互式可视化,并提供 WebAssembly 版 MuJoCo Studio。
- 公开大量物理量计算工具函数,附有 Colab 入门教程与模型编辑教程。
适用人群:做机器人控制、生物力学、图形动画或强化学习仿真的研究与开发人员;需要会写 C 或 Python,本地跑查看器要按文档安装预编译版本。
位置づけ:MuJoCo(Multi-Joint dynamics with Contact)は Google DeepMind が保守する汎用物理エンジンで、ロボット工学、バイオメカニクス、グラフィックスアニメーション、機械学習において、関節構造と環境との接触を高速かつ正確にシミュレーションするために使います。
主な機能
- 実行時シミュレーションモジュールは性能に最適化され、内蔵 XML コンパイラが事前に割り当てた低レベルのデータ構造上で動作するため、実行時のメモリ負荷が小さく抑えられます。
- C API を中核とし、Python バインディングと Unity エンジン向けプラグインも提供され、異なる技術スタックへ組み込めます。
- ネイティブのデスクトップウィンドウとブラウザ内の双方でインタラクティブな可視化に対応し、WebAssembly 版の MuJoCo Studio もあります。
- 多数の物理量計算ユーティリティ関数を公開し、Colab の入門チュートリアルとモデル編集チュートリアルも付いています。
対象ユーザー:ロボット制御、バイオメカニクス、グラフィックスアニメーション、強化学習シミュレーションに取り組む研究者と開発者。C または Python が書けることが必要で、ローカルでビューアを使う場合はドキュメントに従ってビルド済み版をインストールします。
정의: MuJoCo(Multi-Joint dynamics with Contact)는 Google DeepMind가 유지보수하는 범용 물리 엔진으로, 로봇공학, 생체역학, 그래픽 애니메이션, 머신러닝에서 관절 구조와 환경 접촉을 빠르고 정확하게 시뮬레이션하는 데 사용합니다.
핵심 기능
- 런타임 시뮬레이션 모듈은 성능에 맞춰 최적화되어 있고, 내장 XML 컴파일러가 미리 할당한 저수준 데이터 구조 위에서 동작해 실행 시 메모리 부담이 적습니다.
- C API를 핵심으로 하며 Python 바인딩과 Unity 엔진 플러그인도 제공해 서로 다른 기술 스택에 붙일 수 있습니다.
- 네이티브 데스크톱 창과 브라우저 안 모두에서 인터랙티브 시각화를 지원하고, WebAssembly 버전의 MuJoCo Studio도 있습니다.
- 다양한 물리량 계산 유틸리티 함수를 공개하고, Colab 입문 튜토리얼과 모델 편집 튜토리얼도 제공합니다.
대상 사용자: 로봇 제어, 생체역학, 그래픽 애니메이션, 강화학습 시뮬레이션을 하는 연구자와 개발자에게 적합합니다. C나 Python을 쓸 수 있어야 하고, 로컬에서 뷰어를 실행하려면 문서에 따라 미리 빌드된 버전을 설치해야 합니다.
What it is: MuJoCo (Multi-Joint dynamics with Contact) is a general-purpose physics engine maintained by Google DeepMind, built for fast and accurate simulation of articulated structures interacting with their environment in robotics, biomechanics, graphics and machine learning.
Key capabilities
- The runtime simulation module is tuned for performance and works on low-level data structures preallocated by the built-in XML compiler.
- Core is a C API, with Python bindings and a plug-in for the Unity engine available for other stacks.
- Interactive visualization runs in native desktop windows and in browsers, including a WebAssembly build of the MuJoCo Studio app.
README · 详细文档
来源 · 查看原文
MuJoCo stands for Multi-Joint dynamics with Contact. It is a general purpose physics engine that aims to facilitate research and development in robotics, biomechanics, graphics and animation, machine learning, and other areas which demand fast and accurate simulation of articulated structures interacting with their environment.
This repository is maintained by Google DeepMind.
MuJoCo has a C API and is intended for researchers and developers. The runtime simulation module is tuned to maximize performance and operates on low-level data structures that are preallocated by the built-in XML compiler. The library includes interactive visualization in native desktop windows and web browsers. MuJoCo further exposes a large number of utility functions for computing physics-related quantities.
We also provide [Python bindings] and a plug-in for the [Unity] game engine.
Documentation
MuJoCo's documentation can be found at [mujoco.readthedocs.io]. Upcoming features due for the next release can be found in the [changelog] in the "latest" branch.
Getting Started
- Try MuJoCo in your browser.
Visit mujoco.org to explore MuJoCo on desktop or mobile with nothing to install, or open live.mujoco.org to inspect and drag and drop your own models in the WebAssembly version of the [MuJoCo Studio] app (slower than native C, since it runs in WebAssembly).
- Run a viewer on your machine.
Follow the [Getting Started] guide in the documentation to download a prebuilt release and run the [MuJoCo Studio] app or the classic simulate viewer (shown in this video) on your machine.
- Explore our online IPython notebooks.
If you are a Python user, you might want to start with our tutorial notebooks running on Google Colab:
- The introductory tutorial teaches MuJoCo basics: 1 - The Model Editing tutorial shows how to create and edit models procedurally: 2 - The rollout tutorial shows how to use the multithreaded rollout module: 3 - The LQR tutorial synthesizes a linear-quadratic controller, balancing a humanoid on one leg: 4 - The least-squares tutorial explains how to use the Python-based nonlinear least-squares solver: 5 - The MJX tutorial provides usage examples of MuJoCo XLA, a branch of MuJoCo written in JAX: 6 - The differentiable physics tutorial trains locomotion policies with analytical gradients automatically derived from MuJoCo's physics step: 7
Installation
Prebuilt binaries
Versioned releases are available as precompiled binaries from the GitHub [releases page], built for Linux (x86-64 and AArch64), Windows (x86-64 only), and macOS (universal). This is the recommended way to use the software.
Building from source
Users who wish to build MuJoCo from source should consult the [build from source] section of the documentation. However, note that the commit at the tip of the main branch may be unstable.
Python (>= 3.11)
The native Python bindings, which come pre-packaged with a copy of MuJoCo, can be installed from [PyPI] via:
pip install mujoco
Note that Pre-built Linux wheels target manylinux2014, see here for compatible distributions. For more information such as building the bindings from source, see the [Python bindings] section of the documentation.
Versioning
We aim to release MuJoCo in the first week of each month. Our versioning standards changed to modified Semantic Versioning in 3.5.0, see versioning for details.
Contributing
We welcome community engagement: questions, requests for help, bug reports and feature requests. To read more about bug reports, feature requests and more ambitious contributions, please see our contributors guide and style guide.
Asking Questions
Questions and requests for help are welcome as a GitHub "Asking for Help" Discussion and should focus on a specific problem or question.
Bug reports and feature requests
GitHub Issues are reserved for bug reports, feature requests and other development-related subjects.
Related software
MuJoCo is the backbone for numerous environment packages. Below we list several bindings and converters.
Bindings
These packages give users of various languages access to MuJoCo functionality:
First-party bindings:
- dm_control, Google DeepMind's related environment stack, includes PyMJCF, a module for procedural manipulation of MuJoCo models.
- JavaScript bindings and WebAssembly support (inspired stillonearth and zalo's community projects; mjswan extends these with real-time policy control, interactive force
application, and more).
Third-party bindings:
- MATLAB Simulink: Simulink Blockset for MuJoCo Simulator
by Manoj Velmurugan.
- Swift: swift-mujoco
- Java: mujoco-java
- Julia: MuJoCo.jl
- Rust: MuJoCo-rs
Converters
- OpenSim: MyoConverter converts
OpenSim models to MJCF.
- SDFormat: gz-mujoco is a
two-way SDFormat <-> MJCF conversion tool.
- OBJ: obj2mjcf
a script for converting composite OBJ files into a loadable MJCF model.
- onshape: Onshape to Robot
Converts onshape CAD assemblies to MJCF.
Citation
If you use MuJoCo for published research, please cite:
@inproceedings{todorov2012mujoco,
title={MuJoCo: A physics engine for model-based control},
author={Todorov, Emanuel and Erez, Tom and Tassa, Yuval},
booktitle={2012 IEEE/RSJ International Conference on Intelligent Robots and Systems},
pages={5026--5033},
year={2012},
organization={IEEE},
doi={10.1109/IROS.2012.6386109}
}
License and Disclaimer
Copyright 2021 DeepMind Technologies Limited.
Box collision code (0) is Copyright 2016 Svetoslav Kolev.
ReStructuredText documents, images, and videos in the doc directory are made available under the terms of the Creative Commons Attribution 4.0 (CC BY 4.0) license. You may obtain a copy of the License at https://creativecommons.org/licenses/by/4.0/legalcode.
Source code is licensed under the Apache License, Version 2.0. You may obtain a copy of the License at https://www.apache.org/licenses/LICENSE-2.0.
This is not an officially supported Google product.
[build from source]: https://mujoco.readthedocs.io/en/latest/programming#building-from-source [Getting Started]: https://mujoco.readthedocs.io/en/latest/programming#getting-started [MuJoCo Studio]: https://mujoco.readthedocs.io/en/latest/studio/ [Unity]: https://unity.com/ [releases page]: https://github.com/google-deepmind/mujoco/releases [mujoco.readthedocs.io]: https://mujoco.readthedocs.io [changelog]: https://mujoco.readthedocs.io/en/latest/changelog.html [Python bindings]: https://mujoco.readthedocs.io/en/stable/python.html#python-bindings [PyPI]: https://pypi.org/project/mujoco/
