Safe Robot Learning in Assistive Devices Through Neural Network Repair
Abstract
Assistive robotic devices are a particularly promising field of application for neural networks (NN) due to the need for personalization and hard-to-model human-machine interaction dynamics. However, NN based estimators and controllers may produce potentially unsafe outputs over previously unseen data points. In this paper, we introduce an algorithm for updating NN control policies to satisfy a given set of formal safety constraints, while also optimizing the original loss function. Given a set of mixed-integer linear constraints, we define the NN repair problem as a Mixed Integer Quadratic Program (MIQP). In extensive experiments, we demonstrate the efficacy of our repair method in generating safe policies for a lower-leg prosthesis.
Cite
Text
Majd et al. "Safe Robot Learning in Assistive Devices Through Neural Network Repair." Conference on Robot Learning, 2022.Markdown
[Majd et al. "Safe Robot Learning in Assistive Devices Through Neural Network Repair." Conference on Robot Learning, 2022.](https://mlanthology.org/corl/2022/majd2022corl-safe/)BibTeX
@inproceedings{majd2022corl-safe,
title = {{Safe Robot Learning in Assistive Devices Through Neural Network Repair}},
author = {Majd, Keyvan and Clark, Geoffrey Mitchell and Khandait, Tanmay and Zhou, Siyu and Sankaranarayanan, Sriram and Fainekos, Georgios and Amor, Heni},
booktitle = {Conference on Robot Learning},
year = {2022},
pages = {2148-2158},
volume = {205},
url = {https://mlanthology.org/corl/2022/majd2022corl-safe/}
}