| license: mit | |
| tags: | |
| - robotics | |
| - simulation | |
| - deformable-objects | |
| pretty_name: AdaptiGraph simulation data | |
| # AdaptiGraph simulation data | |
| Simulation data for [AdaptiGraph: Material-Adaptive Graph-Based Neural Dynamics for Robotic Manipulation](https://robopil.github.io/adaptigraph/) (RSS 2024). | |
| Code: [github.com/Boey-li/AdaptiGraph](https://github.com/Boey-li/AdaptiGraph) · Checkpoints: [boeyyyy/AdaptiGraph](https://huggingface.co/boeyyyy/AdaptiGraph) | |
| 100 episodes per material (98 train, 2 validation/visualization): `rope.zip`, `granular.zip`, `cloth.zip`. | |
| ``` | |
| hf download boeyyyy/AdaptiGraph-data --repo-type dataset --local-dir sim_data | |
| cd sim_data && for m in rope granular cloth; do unzip -q $m.zip; done | |
| ``` | |
| ## Citation | |
| ```bibtex | |
| @inproceedings{zhang2024adaptigraph, | |
| title={AdaptiGraph: Material-Adaptive Graph-Based Neural Dynamics for Robotic Manipulation}, | |
| author={Zhang, Kaifeng and Li, Baoyu and Hauser, Kris and Li, Yunzhu}, | |
| booktitle={Proceedings of Robotics: Science and Systems (RSS)}, | |
| year={2024} | |
| } | |
| ``` | |