DOI: 10.1016/j.cma.2025.118413

Authors: Vijay Dubey, Collin E. Haese, Osman Gültekin, David Dalton, Manuel K. Rausch, Jan Fuhg

Abstract: Surrogate models for the rapid inference of nonlinear boundary value problems in mechanics are helpful in a broad range of engineering applications. However, effective surrogate modeling of applications involving the contact of deformable bodies, especially in the context of varying geometries, is still an open issue. In particular, existing methods are confined to rigid body contact or, at best, contact between rigid and soft objects with well-defined contact planes. Furthermore, they employ contact or collision detection filters that serve as a rapid test but use only the necessary and not sufficient conditions for detection. In this work, we present a graph neural network architecture that utilizes continuous collision detection and, for the first time, incorporates sufficient conditions designed for contact between soft deformable bodies. We test its performance on two benchmarks, including a problem in soft tissue mechanics of predicting the closed state of a bioprosthetic aortic valve. We find a regularizing effect on adding additional contact terms to the loss function, leading to better generalization of the network. These benefits hold for simple contact at similar planes and element normal angles, and complex contact at differing planes and element normal angles. We also demonstrate that the framework can handle varying reference geometries. However, such benefits come with high computational costs during training, resulting in a trade-off that may not always be favorable. We quantify the training cost and the resulting inference speedups on various hardware architectures. Importantly, our graph neural network implementation results in up to a hundred- to thousand-fold speedup on GPU, and twenty- to two hundred-fold speedup on CPU for our benchmark problems at inference.

Keywords: Graph neural network; Contact mechanics; Soft tissue mechanics; Surrogate model; Computational mechanics

How to Cite

@article{dubey_graph_2026,
  title = {Graph neural network surrogates for contacting deformable bodies with necessary and sufficient contact detection},
  volume = {448},
  issn = {0045-7825},
  url = {https://doi.org/10.1016/j.cma.2025.118413},
  doi = {10.1016/j.cma.2025.118413},
  urldate = {2025-09-26},
  journal = {Computer Methods in Applied Mechanics and Engineering},
  author = {Dubey, Vijay K. and Haese, Collin E. and Gültekin, Osman and Dalton, David and Rausch, Manuel K. and Fuhg, Jan},
  month = jan,
  year = {2026},
  keywords = {Contact mechanics, Computational mechanics, Graph neural network, Soft tissue mechanics, Surrogate model},
  pages = {118413},
}

Can a graph neural network be used as a surrogate for nonlinear mechanics, especially in the presence of contact?

Summary

Surrogate models, including those based on machine learning, struggle with contact modeling. A common simplification assumes rigid body contact, thereby excluding soft material interactions. Prior works often rely on fast but insufficient necessary conditions, which can miss some contact pairs.

In our study, we present a graph neural network that utilizes continuous collision detection and, for the first time, incorporates sufficient conditions for contact detection within this framework. The robust contact detection computes a penalty that is minimized during the training.

What are our findings?

• Including contact loss regularizes training and improves generalization of contact behavior.
• Applies to complex contact: i.e., contact planes aren’t fixed, element normal angles are non-uniform, and reference configuration varies.
• Incurs significant computational cost at training time (10x) but leads to rapid inferences (1000x).

How to Cite

For more details read our paper:

Dubey VK, Haese CE, Gültekin O, Dalton D, Rausch MK, Fuhg J. Graph neural network surrogates for contacting deformable bodies with necessary and sufficient contact detection. Computer Methods in Applied Mechanics and Engineering. 2026;448:118413.

Access to complete code on Soft Tissue Biomechanics Lab github here.

Access to all the dataset used in the work on the Soft Tissue Biomechancs Lab’s Texas DataVerse here.

Earlier arxiv preprint.

Citation details:

@article{dubey_graph_2026,
  title = {Graph neural network surrogates for contacting deformable bodies with necessary and sufficient contact detection},
  volume = {448},
  issn = {0045-7825},
  url = {https://doi.org/10.1016/j.cma.2025.118413},
  doi = {10.1016/j.cma.2025.118413},
  urldate = {2025-09-26},
  journal = {Computer Methods in Applied Mechanics and Engineering},
  author = {Dubey, Vijay K. and Haese, Collin E. and Gültekin, Osman and Dalton, David and Rausch, Manuel K. and Fuhg, Jan},
  month = jan,
  year = {2026},
  keywords = {Contact mechanics, Computational mechanics, Graph neural network, Soft tissue mechanics, Surrogate model},
  pages = {118413},
}

LinkedIn Announcement

ELI5

View this post on Instagram

A post shared by Los Angeles Public Library (@lapubliclibrary)

Launch Window

Possibly between 16 July to 14 August 2025 and 1130-1530 (UTC) as per NOTAM.

A1877/25 (Issued for VOMF PART 1 OF 2) - GSLV-F16 ROCKET LAUNCH FM SHAR RANGE, SRIHARIKOTA WILL TAKE 
PLACE AS PER FLW DETAILS.THE LAUNCH WILL BE ON ANY ONE
OF THE DAY DRG THIS PERIOD.ACTUAL DATE OF LAUNCH WILL BE
INTIMATED ATLEAST 24 HR IN ADVANCE THROUGH A SEPARATE NOTAM.
LAUNCH PAD COORD: 134312N0801348E
NO FLT IS PERMITTED OVER THE DNG ZONES.
A)DANGER ZONE -1:IS A CIRCLE OF 10 NAUTICAL MILES AROUND THE 
LAUNCHER.
B)DANGER ZONE -2:IS AN AREA BOUNDED BY FOLLOWING COORD:
I.103000N 0824500E
II.105000N 0830500E
III.085533N 0844109E
IV.091743N 0834543E
V.103000N 0824500E
RTE AFFECTED IN CHENNAI FIR:
W20, L896, N563, N564, Q11, Q23, Q24, V4, V9, T3
CLOSURES/ALTN RTE FOR OVERFLYING:
1. W20 NOT AVBL BTN MMV-KAMGU
   ALTN: MMV-DCT-DOHIA-DCT-RAMDO-DCT-KAMGU
2. Q24 NOT AVBL BTN MMV-KAMGU
   ALTN: MMV-DCT-DOHIA-DCT-RAMDO-DCT-KAMGU (UNI DIRECTIONAL)
PART 1 OF 2. 1130-1530, 16 JUL 11:30 2025 UNTIL 14 AUG 15:30 2025. CREATED:
19 JUN 10:22 2025

More Details

NISAR

Authors: Vijay Dubey, Collin E. Haese, Osman Gültekin, Jan Fuhg, Manuel K. Rausch.

How to Cite

@inproceedings{dubey_graph_2025_sb3c,
	address = {Austin TX USA},
	title = {Graph Neural Networks for Soft-Body Contact Mechanics},
	booktitle = {Proceedings of the ASME Summer Bioengineering Conference 2025},
	author = {Dubey, Vijay and Haese, Collin E. and Gültekin, Osman and Fuhg, Jan and Rausch, Manuel K.},
	year = {2024},
}