Dataset Card for HannesImitation

The HannesImitation dataset is a collection of vision-based grasping demonstrations with the Hannes prosthetic hand.
The dataset is intended for training control policies using Imitation Learning approaches such as Behavior Cloning.

• 3 grasping scenarios: Table Grasp, Shelf Grasp, Human-to-Hannes Handover.
• 15 objects from the YCB dataset.
• 10 demonstrations per scenario per object for a total of 450 demonstrations.

For additional description, please refer to the paper linked below.

Dataset Details

Dataset Description

• Curated by: Carlo Alessi (@carlo-alessi), Humanoid Sensing and Perception (@HSP-IIT)
• Funded by: Project Future Artificial Intelligence Research (FAIR)
• License: cc-by-4.0

Dataset Sources

• Repository: github.com/hsp-iit/HannesImitation
• Paper: HannesImitation: Grasping with the Hannes Prosthetic Hand via Imitation Learning (arXiv)

Dataset Structure

[More Information Needed]

Dataset Creation

Source Data

Data Collection and Processing

[More Information Needed]

Citation

BibTeX:

@inproceedings{alessi2025hannesimitation,
  title={HannesImitation: Grasping with the Hannes Prosthetic Hand via Imitation Learning},
  author={Alessi, Carlo and Vasile, Federico and Ceola, Federico and Pasquale, Giulia and Boccardo, Nicol{`o} and Natale, Lorenzo},
  booktitle={IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  year={2025},
  pages={},
  organization={}
}

APA:

Alessi, C., Vasile, F., Ceola, F., Pasquale, G., Boccardo, N., & Natale, L. (2025).
HannesImitation: Grasping with the Hannes Prosthetic Hand via Imitation Learning.
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).