.. _research:

Research
========

Citing mjlab
------------

If you use mjlab in your research, please cite:

.. code-block:: bibtex

    @article{Zakka_mjlab_A_Lightweight_2026,
        author = {Zakka, Kevin and Liao, Qiayuan and Yi, Brent and Le Lay, Louis and Sreenath, Koushil and Abbeel, Pieter},
        title = {{mjlab: A Lightweight Framework for GPU-Accelerated Robot Learning}},
        url = {https://arxiv.org/abs/2601.22074},
        year = {2026}
    }

Publications
------------

Papers that use mjlab. To add your work, open a pull request or post in
`Show and Tell <https://github.com/mujocolab/mjlab/discussions/categories/show-and-tell>`_.

.. list-table::
   :header-rows: 1
   :widths: 60 30 10

   * - Title
     - Authors
     - Year
   * - `HUSKY: Humanoid Skateboarding System via Physics-Aware Whole-Body Control
       <https://arxiv.org/abs/2602.03205>`_
     - Han, Wang, Zhang, Liu, Luo, Bai, Li
     - 2026
   * - `DynaRetarget: Dynamically-Feasible Retargeting using Sampling-Based
       Trajectory Optimization <https://arxiv.org/abs/2602.06827>`_
     - Dhedin, Taouil, Omar, Yu, Tao, Dai, Khadiv
     - 2026
   * - `CLOT: Closed-Loop Global Motion Tracking for Whole-Body Humanoid
       Teleoperation <https://arxiv.org/abs/2602.15060>`_
     - Zhu, Cai, Yang, Ren, Xie, Wang, Wu, et al.
     - 2026
   * - `PAC-MAN: Perception-Aware CBF-RL for Whole-Body Safety in Humanoid
       Dodgeball <https://arxiv.org/abs/2607.28623>`_
       (`code <https://github.com/lzyang2000/perceptive_cbf_rl>`_)
     - Yang, Li, Ames
     - 2026

Projects
--------

Projects built on mjlab. To add yours, open a pull request or post in
`Show and Tell <https://github.com/mujocolab/mjlab/discussions/categories/show-and-tell>`_.

.. list-table::
   :header-rows: 1
   :widths: 35 65

   * - Project
     - Description
   * - `menloresearch/asimov-mjlab <https://github.com/menloresearch/asimov-mjlab>`_
     - Locomotion fork for the Asimov bipedal robot.
   * - `Nagi-ovo/mjlab-homierl <https://github.com/Nagi-ovo/mjlab-homierl>`_
     - H1 locomotion across multiple tasks with robustness to upper body disturbances.
   * - `MyoHub/mjlab_myosuite <https://github.com/MyoHub/mjlab_myosuite>`_
     - Musculoskeletal simulation integration with MyoSuite.
   * - `MarcDcls/mjlab_upkie <https://github.com/MarcDcls/mjlab_upkie>`_
     - Velocity control for the Upkie wheeled biped.
   * - `unitreerobotics/unitree_rl_mjlab <https://github.com/unitreerobotics/unitree_rl_mjlab>`_
     - Official Unitree RL environments for Go2, G1, and H1\_2.
   * - `pal-robotics/pal_mjlab <https://github.com/pal-robotics/pal_mjlab>`_
     - PAL Robotics robots and tasks.
   * - `Msornerrrr/in-hand-rotation-mjlab <https://github.com/Msornerrrr/in-hand-rotation-mjlab>`_
     - Sim to real RL for in hand cube rotation with the LEAP Hand.
   * - `project-instinct/InstinctMJ <https://github.com/project-instinct/InstinctMJ>`_
     - mjlab version of Project-Instinct, a whole-body control toolchain to study Instinct-Level intelligence.
   * - `lzyang2000/twist2_mjlab <https://github.com/lzyang2000/twist2_mjlab>`_
     - mjlab port of `TWIST2 <https://arxiv.org/abs/2511.02832>`_.
   * - `wuji-technology/wuji-mjlab <https://github.com/wuji-technology/wuji-mjlab>`_
     - In-hand cube reorientation on the Wuji Hand with sim-to-real deployment.
   * - `wbc-mjlab/wbc-mjlab <https://github.com/wbc-mjlab/wbc-mjlab>`_
     - Configurable whole-body control — shared MDP with task configs, one policy for many skills.
   * - `pollen-robotics/microduck_rl <https://github.com/pollen-robotics/microduck_rl>`_
     - Sim-to-real locomotion for `Microduck <https://pollen-robotics.com/microduck/>`_, an
       open-source 25 cm bipedal robot from Pollen Robotics and Hugging Face.