CarCrashNet: A Large-Scale Dataset and Hierarchical Neural Solver for Data-Driven Structural Crash Simulation

2026-05-07

AI Summary

This work addresses the high computational cost and modeling challenges posed by complex physical phenomena—such as nonlinear contact, large deformations, and material failure—in structural crash simulations. To this end, the authors construct a large-scale, open-source, high-fidelity crash dataset comprising over 14,000 component-level and 825 full-vehicle simulations, and introduce the first public benchmark covering multiple vehicle models and crash scenarios. Building upon this foundation, they propose CrashSolver, a hierarchical neural solver tailored for high-resolution finite element data that integrates geometric deep learning with a Transformer architecture to enable efficient and accurate full-vehicle crash prediction. Experiments demonstrate that the open-source pipeline based on OpenRadioss aligns closely with LS-DYNA simulations and physical test data, and that CrashSolver outperforms existing neural solvers across multiple benchmark tasks, establishing a reproducible foundation for AI-driven virtual crash testing.

Abstract

Crash simulation is a cornerstone of modern vehicle development because it reduces the need for costly physical prototypes, accelerates safety-driven design iteration, and increasingly supports virtual testing workflows. At the same time, modeling structural crash mechanics remains exceptionally challenging: the response is governed by nonlinear contact, large deformation, material plasticity, failure, and complex multi-body interactions evolving over space and time on high-resolution finite-element meshes. In this work, we introduce \textsc{CarCrashNet}, a public high-fidelity open-source benchmark for data-driven structural crash simulation. \textsc{CarCrashNet} combines component-scale and full-vehicle simulations in a multi-modal format, including more than 14{,}000 bumper-beam pole-impact simulations with varying geometry, materials, and boundary conditions, together with 825 full-vehicle crash simulations built from three industry-standard vehicle models of increasing structural complexity: Dodge Neon, Toyota Yaris, and Chevrolet Silverado. To establish the reliability of the benchmark, we validate our open-source finite-element workflow based on OpenRadioss against both experimental crash data and the commercial solver Ansys LS-DYNA. We also introduce \textsc{CrashSolver}, a machine-learning model designed for full-vehicle crash prediction from high-resolution finite-element crash data. We further perform extensive benchmarking across the released datasets and evaluate \textsc{CrashSolver} against state-of-the-art geometric deep learning and transformer-based neural solvers. Our results position \textsc{CarCrashNet} as a foundation for reproducible research in structural simulation, crashworthiness modeling, and AI-driven virtual crash testing. The dataset is available at https://github.com/Mohamedelrefaie/CarCrashNet.

Problem

• crash simulation
• structural crash mechanics
• nonlinear contact
• large deformation
• material plasticity

Innovation

• CarCrashNet
• data-driven crash simulation
• CrashSolver
• geometric deep learning
• finite-element modeling