EUPanda is a End-User Programming framework for the FRANKA Panda Robot Arm that tries to replicate the proprietary Desk environment in a open source ROS-based fashion.
The framework allows to create from kinestethic teaching robot programs made of 5 primitives:
- move to pose,
- move to contact at pose,
- move fingers,
- apply force with fingers, and
- user synchronization primitive.
Watch EUPanda in action!
EUPanda development is stale:
- The backend (C++/Python) supports creation of programs, creation through kinesthetic teaching, and their execution (and reversion).
- The
masterbranch has the GUI (PyQT) used for tuning of already created robot programs (see [1]). - The
pbd_gui_developmentbranch integrates in the GUI the automatic creation of robot programs from unconstrained robot demonstrations (see [2]).
For the Active Learning code behind our HRI'20 paper [1], please check the tuning_al_gui repository.
[1] Mattia Racca, Ville Kyrki, and Maya Cakmak, "Interactive Tuning of Robot Program Parameters via Expected Divergence Maximization", HRI'20
@inproceedings{racca20,
author = {Racca, Mattia and Kyrki, Ville and Cakmak, Maya},
title = {Interactive Tuning of Robot Program Parameters via Expected Divergence Maximization},
year = {2020},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
booktitle = {Proceedings of the 2020 ACM/IEEE International Conference on Human-Robot Interaction},
pages = {629–638},
keywords = {human-robot interaction, end-user programming, active learning},
series = {HRI '20}
}[2] Petri Kalima, "Segmentation of Kinesthetic Demonstrations into End-User Robot Programs"
@mastersthesis{kalima20,
title={Segmentation of Kinesthetic Demonstrations into End-User Robot Programs},
author={Kalima, Petri},
year={2020},
pages={63}
}Everything authored by me is released under the GNU GPL 3.0 license. Some files are authored by others, and are included as part of my own setup; they still belong to the original authors.