Installation¶

Prerequisites¶

OS: Ubuntu 20.04/22.04
RAM: 32GB+
GPU: NVIDIA RTX 2070+ (must with RTX cores)
NVIDIA Driver: 535.216.01+

For complete requirements, please see Isaac Sim’s Requirements.

GRUtopia is built upon NVIDIA’s Omniverse and Isaac Sim platforms, inheriting their dependencies. GRUtopia 2.0 specifically requires Isaac Sim 4.2.0. To ensure optimal performance and avoid any potential issues, it is essential to use this version rather than any other releases, such as 4.5.0 or 4.1.0.

Installation¶

Three ways of installation are provided:

Install from source (Linux): recommended for users who wants to thoroughly explore GRUtopia with Isaac Sim as a GUI application on Linux workstation with a GPU.
Install from PyPI (Linux): recommended for users who wants to use GRUtopia as a handy toolbox with Isaac Sim as a GUI application on Linux workstation with a GPU.
Install with Docker (Linux): recommended for deployment on remote servers or the Cloud using a Docker container.

See more: Differences Between Workstation And Docker.

Windows support is in our roadmap. Contributions are welcome!

Install from source (Linux)¶

Before proceeding with the installation, ensure that you have Isaac Sim 4.2.0 and Conda installed.

Clone the GRUtopia repository with git.

$ git clone git@github.com:OpenRobotLab/GRUtopia.git

Navigate to GRUtopia root path and configure the conda environment.

$ cd PATH/TO/GRUTOPIA/ROOT

# Conda environment will be created and configured automatically with prompt.
$ ./setup_conda.sh

$ cd .. && conda activate grutopia  # or your conda env name

Install from PyPI (Linux)¶

Before proceeding with the installation, ensure that you have Isaac Sim 4.2.0 and Conda installed.

Create conda env with python=3.10 specified.
```
$ conda create -n <env> python=3.10
```
Install GRUtopia through pip.

NOTE: Ensure you have git installed.
```
$ conda activate <env>
$ pip install grutopia
```

Configure the conda environment.

$ python -m grutopia.setup_conda_pypi

$ conda deactivate && conda activate <env>

Install with Docker (Linux)¶

Make sure you have Docker and NVIDIA Container Toolkit installed. You can refer to the container installation doc of Isaac Sim for detailed instructions.

Clone the GRUtopia repository to any desired location.

$ git clone git@github.com:OpenRobotLab/GRUtopia.git

Pull the Isaac Sim image (docker login is required, please refer to NGC Documents).
```
$ docker pull nvcr.io/nvidia/isaac-sim:4.2.0
```

Build docker image, replacing with your desired tag:

$ cd PATH/TO/GRUTOPIA/ROOT

$ docker build -t grutopia:<your tag> .

Start docker container, replacing with the above tag:

$ xhost +local:root # Allow the container to access the display

$ cd PATH/TO/GRUTOPIA/ROOT

$ docker run --name grutopia -it --rm --gpus all --network host \
  -e "ACCEPT_EULA=Y" \
  -e "PRIVACY_CONSENT=Y" \
  -e "DISPLAY=${DISPLAY}" \
  -v /tmp/.X11-unix/:/tmp/.X11-unix \
  -v ${PWD}:/isaac-sim/GRUtopia \
  -v ${HOME}/docker/isaac-sim/cache/kit:/isaac-sim/kit/cache:rw \
  -v ${HOME}/docker/isaac-sim/cache/ov:/root/.cache/ov:rw \
  -v ${HOME}/docker/isaac-sim/cache/pip:/root/.cache/pip:rw \
  -v ${HOME}/docker/isaac-sim/cache/glcache:/root/.cache/nvidia/GLCache:rw \
  -v ${HOME}/docker/isaac-sim/cache/computecache:/root/.nv/ComputeCache:rw \
  -v ${HOME}/docker/isaac-sim/logs:/root/.nvidia-omniverse/logs:rw \
  -v ${HOME}/docker/isaac-sim/data:/root/.local/share/ov/data:rw \
  -v ${HOME}/docker/isaac-sim/documents:/root/Documents:rw \
  grutopia:<your tag>

You are now ready to use GRUtopia in this container.

NOTE: If you are using a remote server without display, you can use the Omniverse Streaming Client to stream the simulation UI.

Prepare Assets¶

📝First of all you MUST complete the User Agreement for GRScenes-100 Dataset Access.

Then you can one of the following methods to get the assets:

Download the assets automatically with GRUtopia installed:
```
$ python -m grutopia.download_assets
```
During the script execution, you can choose to download full assets (~80GB) or a minimum set (~500MB), and you will be asked to specify the local path to store the downloaded assets.

NOTE: If GRUtopia is installed with Docker, We recommend downloading the assets to a location under /isaac-sim/GRUtopia/ in container (which is mounted from a host path) so that it can be retained across container recreations.
Download the assets manually from HuggingFace/ModelScope/OpenXLab, and then use the following command to tell GRUtopia where the assets locate if you are meant to use it with GRUtopia:
```
$ python -m grutopia.set_assets_path
```

Verify Installation¶

$ python -m grutopia.demo.h1_locomotion  # start simulation

If properly installed, Isaac Sim GUI window would pop up and you can see a humanoid robot (Unitree H1) walking following a pre-defined trajectory in Isaac Sim.

NOTE: A slowdown is expected during first execution. Isaac sim requires some one-time startup setup the first time you start it. The process could take up to 5 minutes. This is expected behavior, and should only occur once!