fabio-amadio/dmp_lib

Dynamic Movement Primitive (DMP) Library

dmp_lib

The dmp_lib library offers several Dynamic Movement Primitive (DMP) implementations for robot learning from demonstrations:

• Standard DMPs
• Bimanual DMPs
• Target-Referred DMPs

Install

The package can be installed by running

pip install -e .

Repository Contents

The repository contains two folders: dmp_lib and example.

The dmp_lib folder contains all the modules implementing DMPs.

The example folder contains example scripts (run them from inside it).

The complete library documentation is available at: https://fabio-amadio.github.io/dmp_lib/.

Citing

If you use this package for any academic work, please cite our original paper.

@inproceedings{amadio2022target,
  author={Amadio, Fabio and Laghi, Marco and Raiano, Luigi and Rollo, Federico and Zunino, Andrea and Raiola, Gennaro and Ajoudani, Arash},
  booktitle={2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids)}, 
  title={Target-Referred DMPs for Learning Bimanual Tasks from Shared-Autonomy Telemanipulation}, 
  year={2022},
  volume={},
  number={},
  pages={496-503},
  doi={10.1109/Humanoids53995.2022.10000233}
}

dmp_lib's Issues

use scipy.spatial.transform.Rotation instead of custom geometry module

improve documentation (Readme, document classes and methods, etc.)