Dynamics Modeling of Soft Robots Based on Attention-enhanced Lagrangian Deep Neural Networks

Yeqi Wei, Xiangyu Shao*, Jingyue Liu, Shaojie Zhang, Linke Xu

*Corresponding author for this work

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

Abstract

This study explores a method for the dynamic modeling of soft robots, focusing on enhancing the deep learning-based Lagrangian modeling approach through the attention mechanism, which enriches the training process by allocating focused attention and analytical weighting to critical state features, thereby increasing the model's sensitivity to changes in the robot's state. We compared our method through simulation, demonstrating that the model is effective in long-term prediction and noise rejection.

Original languageEnglish
Title of host publicationProceedings of the 3rd Conference on Fully Actuated System Theory and Applications, FASTA 2024
PublisherIEEE
Pages200-207
Number of pages8
ISBN (Electronic)979-8-3503-7369-1
DOIs
Publication statusPublished - 2024
Event3rd Conference on Fully Actuated System Theory and Applications, FASTA 2024 - Shenzhen, China
Duration: 10 May 202412 May 2024

Conference

Conference3rd Conference on Fully Actuated System Theory and Applications, FASTA 2024
Country/TerritoryChina
CityShenzhen
Period10/05/2412/05/24

Bibliographical note

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

  • Attention-enhanced deep learning
  • Lagrangian neural networks
  • Machine learning in robotics
  • Predictive modeling
  • Soft robots

Fingerprint

Dive into the research topics of 'Dynamics Modeling of Soft Robots Based on Attention-enhanced Lagrangian Deep Neural Networks'. Together they form a unique fingerprint.

Cite this