Physics-Informed Neural Network Modeling of Vehicle Collision Dynamics in Precision Immobilization Technique Maneuvers

TL;DR

This work tackles the challenge of real-time, accurate prediction of vehicle collision dynamics in precision immobilization technique scenarios by integrating physics-informed neural networks with uncertainty modeling. It introduces a dual-PINN framework: a PINN-GMM for learning time-varying impact-force distributions under momentum and energy constraints, and an Adaptive-PINN for post-collision dynamics with dynamic constraint weighting and an Adaptive Physics Guard to balance physics and data learning. The approach delivers robust uncertainty quantification, achieves force prediction errors below $15.0 ext{ extpercent}$ on FE analysis data, and reduces average trajectory error by $63.6 ext{ extpercent}$ relative to traditional 4DOF models, while preserving millisecond-level computation for real-time use. Validated through FE simulations, Carsim integration, and scaled vehicle experiments, the framework demonstrates strong potential for PIT safety systems and general collision-dynamics prediction, with opportunities for broader scenario coverage and automated hyperparameter tuning.

Abstract

Accurate prediction of vehicle collision dynamics is crucial for advanced safety systems and post-impact control applications, yet existing methods face inherent trade-offs among computational efficiency, prediction accuracy, and data requirements. This paper proposes a dual Physics-Informed Neural Network framework addressing these challenges through two complementary networks. The first network integrates Gaussian Mixture Models with PINN architecture to learn impact force distributions from finite element analysis data while enforcing momentum conservation and energy consistency constraints. The second network employs an adaptive PINN with dynamic constraint weighting to predict post-collision vehicle dynamics, featuring an adaptive physics guard layer that prevents unrealistic predictions whil e preserving data-driven learning capabilities. The framework incorporates uncertainty quantification through time-varying parameters and enables rapid adaptation via fine-tuning strategies. Validation demonstrates significant improvements: the impact force model achieves relative errors below 15.0% for force prediction on finite element analysis (FEA) datasets, while the vehicle dynamics model reduces average trajectory prediction error by 63.6% compared to traditional four-degree-of-freedom models in scaled vehicle experiments. The integrated system maintains millisecond-level computational efficiency suitable for real-time applications while providing probabilistic confidence bounds essential for safety-critical control. Comprehensive validation through FEA simulation, dynamic modeling, and scaled vehicle experiments confirms the framework's effectiveness for Precision Immobilization Technique scenarios and general collision dynamics prediction.

Figures (13)

• Figure 1: Schematic diagrams of the 4DOF vehicle model with impact forces applied
• Figure 2: A view of two-vehicle collision model
• Figure 3: The PIT scenario setting in FEA simulation: (a) before and (b) after collision
• Figure 4: Twelve sample results of FEA force calculated via $a_x$ and $a_y$
• Figure 5: Scheme of PINN-GMM neural network framework