Online Unsupervised Arm Posture Adaptation for sEMG-based Gesture Recognition on a Parallel Ultra-Low-Power Microcontroller

Introduction

This repository contains the code developed for our paper M. Zanghieri et al., “Online unsupervised arm posture adaptation for sEMG-based gesture recognition on a parallel ultra-low-power microcontroller,” IEEE BioCAS, 2023 [1].

We also release our UniBo-INAIL dataset for research on multi-subject, multi-day, and multi-posture sEMG.

Usage

1. Run experiment_multiposture.ipynb (or equivalently experiment_multiposture.py) to perform the baseline experiments with multi-posture training.
2. Run experiment_calibmodes.ipynb (or equivalently experiment_calibmodes.py) for the online PCA posture adaptation experiments.
3. Run read_results.ipynb to get the results statistics.

Authors

This work was realized mainly at the Energy-Efficient Embedded Systems Laboratory (EEES Lab) of University of Bologna (Italy) by:

• Marcello Zanghieri - University of Bologna
• Mattia Orlandi - University of Bologna
• Dr. Elisa Donati - Institute of Neuroinformatics (INI) of University of Zürich and ETH Zürich
• Prof. Emanuele Gruppioni - University of Bologna, INAIL Prosthesis Centre in Vigorso di Budrio (Bologna)
• Prof. Luca Benini - University of Bologna, ETH Zürich
• Prof. Simone Benatti - University of Modena & Reggio Emilia, University of Bologna

Citation

When referring to our paper or using our UniBo-INAIL dataset, please cite our work [1]:

@INPROCEEDINGS{zanghieri2023online,
  author={Zanghieri, Marcello and Orlandi, Mattia and Donati, Elisa and Gruppioni, Emanuele and Benini, Luca and Benatti, Simone},
  booktitle={2023 IEEE Biomedical Circuits and Systems Conference (BioCAS)}, 
  title={Online Unsupervised Arm Posture Adaptation for {sEMG}-based Gesture Recognition on a Parallel Ultra-Low-Power Microcontroller}, 
  year={2023},
  volume={},
  number={},
  pages={1-5},
  doi={10.1109/BioCAS58349.2023.10388902}}

References

[1] M. Zanghieri, M. Orlandi, E. Donati, E. Gruppioni, L. Benini, S. Benatti, “Online unsupervised arm posture adaptation for sEMG-based gesture recognition on a parallel ultra-low-power microcontroller,” in 2023 IEEE International Conference on Biomedical Circuits and Systems (BioCAS), 2023, pp. 1-5. DOI: 10.1109/BioCAS58349.2023.10388902.

License

All files are released under the LGPL-2.1 license (LGPL-2.1) (see LICENSE).