fixed READMR
Browse files
README.md
CHANGED
|
@@ -1,51 +1,28 @@
|
|
| 1 |
---
|
| 2 |
license: bsd-2-clause
|
| 3 |
---
|
| 4 |
-
#
|
| 5 |
-
Software that integrates various imitation learning methods and benchmark task environments to provide baselines for robot manipulation
|
| 6 |
|
| 7 |
-
|
| 8 |
-
[
|
| 9 |
-
[](https://github.com/isri-aist/RoboManipBaselines/blob/master/LICENSE)
|
| 10 |
|
| 11 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
|
| 13 |
-
##
|
| 14 |
-
|
| 15 |
-
Spatial attention recurrent neural network
|
| 16 |
-
|
| 17 |
-
### [ACT](./robo_manip_baselines/act)
|
| 18 |
-
Action Chunking with Transformers
|
| 19 |
-
|
| 20 |
-
### [DiffusionPolicy](./robo_manip_baselines/diffusion_policy)
|
| 21 |
-
Diffusion Policy
|
| 22 |
-
|
| 23 |
-
### [MT-ACT](./robo_manip_baselines/mt_act)
|
| 24 |
-
Multi-Task Action Chunking Transformer
|
| 25 |
-
|
| 26 |
-
## Data
|
| 27 |
-
### Publicly available datasets
|
| 28 |
-
|
| 29 |
-
### Data collection by teleoperation
|
| 30 |
-
|
| 31 |
-
## Environments for robot manipulation
|
| 32 |
-
|
| 33 |
-
## Utilities
|
| 34 |
-
|
| 35 |
-
## Evaluation results
|
| 36 |
-
|
| 37 |
-
## License
|
| 38 |
-
Files that originate from this repository are subject to the BSD 2-Clause License. If a file explicitly states a different license, or if there are different license files in a directory, those licenses will take precedence. For files in third-party directories, please follow the respective licenses.
|
| 39 |
-
|
| 40 |
-
## Citation
|
| 41 |
-
You can cite this work with:
|
| 42 |
```bib
|
| 43 |
-
@
|
| 44 |
-
author = {
|
| 45 |
-
title = {
|
| 46 |
-
|
| 47 |
-
|
| 48 |
-
|
| 49 |
-
|
| 50 |
}
|
| 51 |
-
```
|
|
|
|
| 1 |
---
|
| 2 |
license: bsd-2-clause
|
| 3 |
---
|
| 4 |
+
# Diffusion Policy
|
|
|
|
| 5 |
|
| 6 |
+
## Install
|
| 7 |
+
See [GitHub](https://github.com/isri-aist/RoboManipBaselines/tree/master/robo_manip_baselines/diffusion_policy) for installation.
|
|
|
|
| 8 |
|
| 9 |
+
## Policy rollout
|
| 10 |
+
Run a trained policy:
|
| 11 |
+
```console
|
| 12 |
+
$ python ./bin/rollout/RolloutDiffusionPolicyMujocoUR5eCable.py \
|
| 13 |
+
--checkpoint <checkpoint_path>/checkpoints/200.ckpt \
|
| 14 |
+
--skip 3 --world_idx 0
|
| 15 |
+
```
|
| 16 |
|
| 17 |
+
## Technical Details
|
| 18 |
+
For more information on the technical details, please see the following paper:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
```bib
|
| 20 |
+
@INPROCEEDINGS{DiffusionPolicy_RSS23,,
|
| 21 |
+
author = {Chi, Cheng and Feng, Siyuan and Du, Yilun and Xu, Zhenjia and Cousineau, Eric and Burchfiel, Benjamin and Song, Shuran},
|
| 22 |
+
title = {Diffusion Policy: Visuomotor Policy Learning via Action Diffusion},
|
| 23 |
+
booktitle = {Proceedings of Robotics: Science and Systems},
|
| 24 |
+
year = {2023},
|
| 25 |
+
month = {July},
|
| 26 |
+
doi = {10.15607/RSS.2023.XIX.026}
|
| 27 |
}
|
| 28 |
+
```
|