This is the repository for training the Spill-Free Classifier (SFC) model presented in the Clutter-Aware Spill-Free Liquid Transport via Learned Dynamics paper (link).
- Mocap: Data collected from individuals handling different container scenarios captured using a motion capture system
- Panda: Data collected from the panda robotic arm handling different container scenarios. (Some files are currently missing)
git clone git@github.com:HIRO-group/trajectory-model.git
conda create -n "sfrrt"
conda activate sfrrt
cd trajectory-model
pip install -e .
mkdir data/
Copy the downloaded data in the data/
dir to execute the following commands.
python scripts/train_SFC.py
python scripts/predict_SFC.py
python scripts/calculate_tilt_angle.py
python scripts/plot_mocap_trajectory.py
python scripts/plot_panda_trajectory.py
python scripts/consume_data_from_mocap.py
python scripts/convert_panda_joints_to_cartesian.py
@inproceedings{abderezaei2024clutterawarespillfreeliquidtransport,
title={Clutter-Aware Spill-Free Liquid Transport via Learned Dynamics},
author={Abderezaei, Ava and Pasricha, Anuj and Klausenstock, Alex and Roncone, Alessandro},
year={2024},
booktitle={2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
}