Template for Isaac Lab Projects

Overview

Isaac Lab extension for performing a transparent liquid pouring task using RL and image processing through PourIt. Based on the Isaac Lab extension template.

Work In Progress...

Keywords: extension, template, isaaclab, liquid, pouring, pourit

Installation

• Install Isaac Lab by following the installation guide. We recommend using the conda installation as it simplifies calling Python scripts from the terminal.

• Clone the repository separately from the Isaac Lab installation (i.e. outside the IsaacLab directory) and use he access token (this is a private repository for now):

# Option 1: HTTPS
git clone https://github.com/robegi/IsaacLabPouringExtension.git

• To rename from pouring_ext to a custom name:

# Enter the repository
cd IsaacLabExtensionTemplate
# Rename all occurrences of pouring_ext (in files/directories) to your_fancy_extension_name
python scripts/rename_template.py your_fancy_extension_name

• Using a python interpreter that has Isaac Lab installed, install the library

python -m pip install -e exts/pouring_ext

Now the extension can be executed from any Isaac Lab script by importing it:

import pouring_ext # Custom extension

Example: to execute skrl to train the model, use:

# Enter Isaac Lab folder
cd IsaacLab
# Execute the training script, after importing the extensions
./isaaclab.sh -p source/standalone/workflows/skrl/train.py --task Isaac-Franka-Pouring-Direct-v0 --num_envs 1 --device cpu --disable_fabric --enable_cameras 

This uses the default python interpreter. To use a custom one, simply replace ./isaaclab.sh -p with python inside a conda environment.