Skip to main content

HEXFELLOW Mujoco Driver

Project description

HEXFELLOW MUJOCO DRIVER

DOI       


📖 Overview

What is hex_driver_mujoco

hex_driver_mujoco is a Python library for MuJoCo robot simulation with a unified interface. It supports the E3 Desktop (dual‑arm) and Archer Y6 (6‑DOF arm) robot models. The library runs each simulation in a separate process, streams robot states and camera images via shared‑memory ring buffers, and provides simple get_*_motor/get_*_color_img/get_*_depth_img APIs.

What problem it solves

  • Unified simulation interface: Use the same API for different robot models (E3 Desktop, Archer Y6).
  • Background simulation: Each simulation runs in its own process; the main thread only reads the latest state when needed.
  • Low‑latency sharing: States and images are passed through shared‑memory ring buffers, avoiding serialization overhead.
  • Real‑time control: Supports position and MIT (motor‑impedance‑torque) control modes with configurable gains.

Target users

  • Robotics researchers prototyping manipulation pipelines with MuJoCo simulations.
  • Developers who need a simple, consistent way to interface with simulated robots.
  • Anyone who wants to switch between different robot models without changing client code.

📦 Installation

Requirements

  • Python ≥ 3.10
  • OS: Linux (macOS/Windows may work but are less tested)
  • Core dependencies:
    • hex_util_runtime ≥ 0.0.0, < 0.1.0
    • opencv-python ≥ 4.10.0
    • mujoco ≥ 3.3.0

Install from PyPI

pip install hex_driver_mujoco

Install from Source

For those who need examples, you can install the package from source code with examples.

Note: We use uv to manage the Python environment. Please install it first.

  1. Clone and install in editable mode. The venv.sh script expects uv.

    git clone https://github.com/hexfellow/hex_driver_mujoco.git
    cd hex_driver_mujoco
    ./venv.sh
    
    • ./venv.sh — creates .venv, installs hex_driver_mujoco with example dependencies.
    • ./venv.sh --min — installs the core package only (without example dependencies). Some examples will not run.
    • ./venv.sh --pkg-only — installs the package only, skips example-related dependencies.
  2. Activate before running examples:

    source .venv/bin/activate
    

📑 API

See API for details of all Mujoco driver classes and parameters.

💡 Example

See Example for usage examples of each robot model.

📚 Citation

If you want to cite this project in your work, you can use the following BibTeX entry:

@software{hex_zmq_servers,
  author    = {Dong, Zhaorui and Chen, Zejun},
  title     = {Hex ZMQ Servers: A ZeroMQ-Based Embodied AI Communication Framework},
  year      = {2025},
  publisher = {Zenodo},
  version   = {v1.0.0},
  doi       = {10.5281/zenodo.18309960},
  url       = {https://doi.org/10.5281/zenodo.18309960}
}

📄 License

Apache License 2.0. See LICENSE.

🌟 Star History

Star History Chart

👥 Contributors

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

hex_driver_mujoco-0.0.1a5.tar.gz (4.9 MB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

hex_driver_mujoco-0.0.1a5-py3-none-any.whl (4.9 MB view details)

Uploaded Python 3

File details

Details for the file hex_driver_mujoco-0.0.1a5.tar.gz.

File metadata

  • Download URL: hex_driver_mujoco-0.0.1a5.tar.gz
  • Upload date:
  • Size: 4.9 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.10.12

File hashes

Hashes for hex_driver_mujoco-0.0.1a5.tar.gz
Algorithm Hash digest
SHA256 46162714b2b31ac9bc4d3a3b53099cdb43037b36bd7d8416d7c51da7189204f3
MD5 fd21d00982c56fa31890dcc2196ff288
BLAKE2b-256 e43edd9b30b9dd797751c578635a5ddc2737e4dd424500a4eb6f40d76ad13eef

See more details on using hashes here.

File details

Details for the file hex_driver_mujoco-0.0.1a5-py3-none-any.whl.

File metadata

File hashes

Hashes for hex_driver_mujoco-0.0.1a5-py3-none-any.whl
Algorithm Hash digest
SHA256 97afaafc0f2f76d2c75f6c1e4e5aa960f865ff89c0939a0aa6d41a12c321ac9b
MD5 c5bb53b700589d839e495dbc216d4524
BLAKE2b-256 88a431eae4d85f80a1a815cbe728c86174aa8cd5156951a624f94a4cba93eae6

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page