This work presents a machine learning-based approach for proprioception in soft robots, leveraging recurrent neural networks trained on sensor and actuation data to accurately estimate the robot's position while maintaining its inherent compliance.
This article is published in IEEE Robotics and Automation Letters (RA-L): IEEE Xplore Link
If you found this useful, you could cite the article as follows:
@article{pagliarani2025softtex,
title={SoftTex: Soft robotic arm with learning-based textile proprioception},
author={Pagliarani, Niccol{\`o} and Alessi, Carlo and Arleo, Luca and Campinoti, Giulia and Maselli, Martina and Falotico, Egidio and Cianchetti, Matteo},
journal={IEEE Robotics and Automation Letters},
year={2025},
publisher={IEEE}
}For questions or collaborations regarding Machine Learning-based proprioception, contact @carlo-alessi.