Physics-informed Neural Networks to Model and Control Robots: a Theoretical and Experimental Investigation
Lists
Only you see this
Abstract
This work concerns the application of physics-informed neural networks to the modeling and control of complex robotic systems. Achieving this goal required extending Physics Informed Neural Networks to handle non-conservative effects. We propose to combine these learned models with model-based controllers originally developed with first-principle models in mind. By combining standard and new techniques, we can achieve precise control performance while proving theoretical stability bounds. These validations include real-world experiments of motion prediction with a soft robot and of trajectory tracking with a Franka Emika manipulator.
BibTeX
@article{liu2024physics,
author = {Jingyue Liu and Pablo Borja and Cosimo Della Santina},
title = {{Physics-informed Neural Networks to Model and Control Robots: a Theoretical and Experimental Investigation}},
journal = {Adv. Intell. Syst},
year = {2024},
doi = {10.1002/aisy.202300385},
}Loading related content...