Examples of how to use NVIDIA Omniverse Isaac Sim for to solve Reinforcement Learning Games (RL-Games)

Follow the Isaac Sim documentation to install the latest Isaac Sim release (2023.1.1)

To install omniisaacgymenvs, first clone this repository:

git clone https://github.com/KhaledSharif/omniverse-gym.git

Once cloned, locate the python executable in Isaac Sim. By default, this should be python.sh. We will refer to this path as PYTHON_PATH.

To set a PYTHON_PATH variable in the terminal that links to the python executable, we can run a command that resembles the following. Make sure to update the paths to your local path. For Linux:

alias PYTHON_PATH=~/.local/share/ov/pkg/isaac_sim-2023.1.1/python.sh

Install the repository and its dependencies:

PYTHON_PATH -m pip install -e .

To run a simple form of PPO from rl_games, use the single-threaded training script:

PYTHON_PATH run.py task=Cartpole

The result is saved to the current working directory in a new directory called runs.

You can now evaluate your model by running the same environment in test (inference) mode using the saved model checkpoint.

PYTHON_PATH run.py task=Cartpole test=True checkpoint=runs/Cartpole/nn/Cartpole.pth