AI-driven engineering workflows face unique challenges in crash safety design: unlike aerodynamics, crash events involve highly nonlinear contact dynamics, material nonlinearity, and discrete state transitions that are difficult to capture with data-driven surrogate models. We present the first foundation model-orchestrated workflow for crash safety design, enabling surrogate-assisted exploration for pedestrian protection and reducing evaluation time from hours per CAE simulation to seconds.
The workflow integrates four components:
- A surrogate trained on CAE crash simulations to predict pedestrian leg injury metrics from design parameters, achieving an average $R^2=0.87$ and providing distribution-free conformal prediction intervals;
- Multiobjective evolutionary search (NSGA-II) to discover diverse feasible parameter sets under user-specified constraints;
- A morphing-based geometry generator that maps parameters to topology-preserving 3D shapes;
- A natural-language interface where an LLM orchestrates the workflow and a vision-language model supports semantic comparison of generated designs.
In an automotive front-bumper case study, the workflow produces 35 distinct safety-compliant alternatives from a single exploration, a process that would take weeks with conventional CAE iteration. These results suggest that foundation models can serve as integration layers between ML surrogates and physics-based simulation, bringing AI capabilities to safety-critical engineering domains.
Blogger's Review: This article demonstrates how foundation models can enhance the efficiency of crash safety design, particularly in pedestrian protection. By reducing simulation time and increasing design diversity, the future of automotive safety design is set to increasingly rely on intelligent workflows.