Automotive Crash Dynamics Modeling Accelerated with Machine Learning

2025-10-16

AI Summary

To address the high computational cost and slow iteration inherent in finite element analysis (FEA) for automotive crashworthiness assessment, this paper proposes a hybrid machine learning surrogate model integrating graph neural networks (GNNs) and Transformer architectures for body-in-white (BIW) collision dynamics modeling. Leveraging NVIDIA’s PhysicsNeMo framework, the method incorporates time-conditional inputs and a stability-enhanced autoregressive strategy to perform end-to-end learning on 150 high-fidelity FEA time-series datasets. This work presents the first empirical validation of MeshGraphNet and Transolver for transient structural crash simulation, successfully capturing realistic deformation trends. Compared to conventional FEA, the surrogate achieves 3–4 orders-of-magnitude speedup in inference time, enabling rapid design space exploration and crashworthiness optimization. The approach demonstrates strong potential for industrial deployment in automotive development workflows.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Crashworthiness assessment is a critical aspect of automotive design, traditionally relying on high-fidelity finite element (FE) simulations that are computationally expensive and time-consuming. This work presents an exploratory comparative study on developing machine learning-based surrogate models for efficient prediction of structural deformation in crash scenarios using the NVIDIA PhysicsNeMo framework. Given the limited prior work applying machine learning to structural crash dynamics, the primary contribution lies in demonstrating the feasibility and engineering utility of the various modeling approaches explored in this work. We investigate two state-of-the-art neural network architectures for modeling crash dynamics: MeshGraphNet, and Transolver. Additionally, we examine three strategies for modeling transient dynamics: time-conditional, the standard Autoregressive approach, and a stability-enhanced Autoregressive scheme incorporating rollout-based training. The models are evaluated on a comprehensive Body-in-White (BIW) crash dataset comprising 150 detailed FE simulations using LS-DYNA. The dataset represents a structurally rich vehicle assembly with over 200 components, including 38 key components featuring variable thickness distributions to capture realistic manufacturing variability. Each model utilizes the undeformed mesh geometry and component characteristics as inputs to predict the spatiotemporal evolution of the deformed mesh during the crash sequence. Evaluation results show that the models capture the overall deformation trends with reasonable fidelity, demonstrating the feasibility of applying machine learning to structural crash dynamics. Although not yet matching full FE accuracy, the models achieve orders-of-magnitude reductions in computational cost, enabling rapid design exploration and early-stage optimization in crashworthiness evaluation.

Problem

• Developing machine learning surrogates for crash deformation prediction
• Evaluating neural network architectures for structural crash dynamics
• Reducing computational cost in automotive crashworthiness assessment

Innovation

• Machine learning surrogate models accelerate crash dynamics prediction
• MeshGraphNet and Transolver architectures model structural deformation
• Autoregressive schemes with rollout training enhance transient dynamics modeling