NOTE: This code is for the ICRA 2025 paper entitled "Bootstrapping Object-Level Planning with Large Language Models" (Authors: David Paulius, Alejandro Agostini, Benedict Quartey, and George Konidaris).
You will need to clone this repository and recursively clone and download its submodules using the following commands:
git clone https://github.com/davidpaulius/olp_llm --recursive
git submodule update --remote --merge --recursive
You will need to download the CoppeliaSim simulator and set up its Python API using the following pip command:
python3 -m pip install coppeliasim-zmqremoteapi-client
pip3 install -r requirements.txt
Simply run the Jupyter Notebook script named ICRA25_pipeline.py once you have set everything up as above. You can change the task setting by modifying the code as well as set the OpenAI GPT model of your choice.
To deeply understand what's happening in the code, I would recommend going through the tutorials (yes, I know...) that are available online. There may be specific things you require for your own planning problem, so you should definitely consult the forums.
Feel free to reach out to me via email!