Installation

Installation#

robosuite officially supports macOS and Linux on Python 3. It can be run with an on-screen display for visualization or in a headless mode for model training, with or without a GPU.

The base installation requires the MuJoCo physics engine (with mujoco, refer to link for troubleshooting the installation and further instructions) and numpy. To avoid interfering with system packages, it is recommended to install it under a virtual environment by first running virtualenv -p python3 . && source bin/activate or setting up a Conda environment by installing Anaconda and running conda create -n robosuite python=3.10.

Install from pip#

Note: For users looking to use the most up-to-date code and develop customized features, it is recommended to install from source.

  1. After setting up mujoco, robosuite can be installed with

    $ pip install robosuite

  2. Test your installation with

    $ python -m robosuite.demos.demo_random_action

    Attention Mac users!

    Mac users who wish to use the default mjviewer renderer need to prepend the “python” command with “mj”: mjpython ...

Install from source#

  1. Clone the robosuite repository

    $ git clone https://github.com/ARISE-Initiative/robosuite.git
$ cd robosuite

  2. Install the base requirements with

    $ pip3 install -r requirements.txt

    This will also install our library as an editable package, such that local changes will be reflected elsewhere without having to reinstall the package.

  3. (Optional) We also provide add-on functionalities, such as OpenAI Gym interfaces, inverse kinematics controllers powered by PyBullet, and teleoperation with SpaceMouse devices. To enable these additional features, please install the extra dependencies by running

    $ pip3 install -r requirements-extra.txt

  4. Test your installation with

    $ python robosuite/demos/demo_random_action.py

    Attention Mac users!

    Mac users who wish to use the default mjviewer renderer need to prepend the “python” command with “mj”: mjpython ...

Installing on Windows#

It is common to run into issues when installing robosuite on a Windows machine. robosuite can be installed on Windows using the following steps.

  1. Either follow step 1 from the section Install from pip or steps 1 and 2 in the section Install from source. During this process, you may run into some errors. Please refer to the steps below on how to fix these.

  2. If you run into the error FileNotFoundError: [Errno 2] No such file or directory: 'C:\\tmp\\robosuite.log', create a directory called tmp under C:\.

  3. You will also likely face the issue of mujoco.dll not found. If you are running in a conda environment (highly recommended), go to the location where your packages are installed (i.e. site-packages). If you are unsure where the MuJoCo package is located, open a new python shell and run the following.

    import mujoco
print(mujoco.__path__)

    If the MuJoCo package does not already exists, install it by running

    $ pip install mujoco

    Within the MuJoCo package, there should be a file called mujoco.dll. If you installed robosuite using pip, copy and paste this file into anaconda3\envs\{your env name}\Lib\site-packages\robosuite\utils . If you installed robosuite from source, copy and paste this file directly into robosuite\utils.

  4. You may also get an EGL issue. If this happens, please go into robosuite\utils\binding_utils.py (either in site-packages or in cloned repository depending on whether you installed from pip or source) and change "egl" to "wgl" at line 43. It should look like this:

     if _SYSTEM == "Darwin":
     os.environ["MUJOCO_GL"] = "cgl"
 else:
     os.environ["MUJOCO_GL"] = "wgl"

  5. Test your robosuite installation by running

    $ python robosuite/demos/demo_random_action.py