NODE IK: Neural ODE Inverse Kinematics

Requirements

python >= 3.7
pytorch 1.0.1
matplotlib
sklearn
torchdiffeq
urdfpy
warp-lang
usd-core
pytorch_lightning
pyquaternion
tqdm

Version Numbers

As of 2023, some breaking changes have happened.

Install:

pip3 install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu113
numpy==1.22.3
pytorch_lightning==1.5.10
warp-lang==0.2.3

Git LFS

We use git lfs to reduce the dae file size on clone. Please install git lfs before cloning the repo. Otherwise install after and use git lfs fetch.

How to train

The example training code can be found in examples directory.

cd examples
python train_panda_urdf.py

How to visualize the reuslt

The example visualization code can be found in examples directory.

cd examples
python visualize_panda_urdf.py

This script creates usd files. NVIDA Omniverse can open these files and you can see the generated visual results.

How to evaluate the trained model

The example evaluation code can be found in examples directory.

cd examples
python evaluation_panda_urdf.py

The model checkpoint can be designated by modifying args.

Credits

If you find this repository useful, please cite:

@inproceedings{park2022nodeik,
  title={NODEIK: Solving Inverse Kinematics with Neural Ordinary Differential Equations for Path Planning},
  author={Park, Suhan and Schwartz, Mathew and Park, Jaeheung},
  booktitle={2022 22nd International Conference on Control, Automation and Systems (ICCAS)},
  pages={944--949},
  year={2022},
  organization={IEEE}
}

References

nodeik/layers are orignated from FFJORD and SoftFlow repos. Thank the authors for these codes.

FFJORD: https://github.com/rtqichen/ffjord
SoftFlow: https://github.com/ANLGBOY/SoftFlow

cadop/nodeik