nielsr's picture
nielsr HF Staff
Improve model card
0551ae0 verified
|
raw
history blame
2.37 kB
metadata
license: bsd-2-clause
pipeline_tag: robotics

Diffusion Policy - RoboManipBaselines

This repository contains a Diffusion Policy model trained using the RoboManipBaselines framework. The model was trained with the MujocoUR5eCable dataset.

Installation

For installation instructions, please refer to the Diffusion Policy installation guide in the official repository.

Policy Rollout

To run the trained policy, use the following command from the top directory of the robo_manip_baselines repository:

# Go to the top directory of this repository
$ cd robo_manip_baselines
$ python ./bin/Rollout.py DiffusionPolicy MujocoUR5eCable --checkpoint ./checkpoint/DiffusionPolicy/<checkpoint_name>/policy_last.ckpt

Technical Details

This model implements the Diffusion Policy as described in:

@INPROCEEDINGS{DiffusionPolicy_RSS23,
  author = {Chi, Cheng and Feng, Siyuan and Du, Yilun and Xu, Zhenjia and Cousineau, Eric and Burchfiel, Benjamin and Song, Shuran},
  title = {Diffusion Policy: Visuomotor Policy Learning via Action Diffusion},
  booktitle = {Proceedings of Robotics: Science and Systems},
  year = {2023},
  month = {July},
  doi = {10.15607/RSS.2023.XIX.026}
}

Citation

If you use this model or the RoboManipBaselines framework in your research, please cite:

@article{RoboManipBaselines_Murooka_2025,
  title={RoboManipBaselines: A Unified Framework for Imitation Learning in Robotic Manipulation across Real and Simulation Environments},
  author={Murooka, Masaki and Motoda, Tomohiro and Nakajo, Ryoichi and Oh, Hanbit and Makihara, Koshi and Shirai, Keisuke and Ogata, Tetsuya and Domae, Yukiyasu},
  journal={arXiv preprint arXiv:2509.17057},
  year={2025}
}