---
license: bsd-2-clause
pipeline_tag: robotics
---

# Spatial attention recurrent neural network (SARNN)

This repository contains a SARNN (Spatial attention recurrent neural network) model trained with the [MujocoUR5eCable dataset](https://huggingface.co/datasets/RoboManipBaselines/MujocoUR5eCable).

The model is part of **RoboManipBaselines**, a unified framework for imitation learning in robotic manipulation across real and simulation environments.

[[Paper](https://huggingface.co/papers/2509.17057)] [[Project Page](https://isri-aist.github.io/RoboManipBaselines-ProjectPage/)] [[GitHub](https://github.com/isri-aist/RoboManipBaselines)]

## Install
See [GitHub](https://github.com/isri-aist/RoboManipBaselines/blob/master/doc/install.md#SARNN) for installation instructions.

## Policy rollout
To run a trained policy, navigate to the top directory of the `robo_manip_baselines` repository and run:

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

## Technical Details
For more information on the technical details of the SARNN architecture, please see the following paper:
```bibtex
@INPROCEEDINGS{SARNN_ICRA2022,
  author = {Ichiwara, Hideyuki and Ito, Hiroshi and Yamamoto, Kenjiro and Mori, Hiroki and Ogata, Tetsuya},
  title = {Contact-Rich Manipulation of a Flexible Object based on Deep Predictive Learning using Vision and Tactility},
  booktitle = {International Conference on Robotics and Automation},
  year = {2022},
  pages = {5375-5381},
  doi = {10.1109/ICRA46639.2022.9811940}
}
```

## Citation
If you use this framework or model in your work, please cite the RoboManipBaselines paper:

```bibtex
@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}
}
```