Predicting head impact conditions in vehicle-to-pedestrian impacts through computational human modeling

Area of Science:

• Biomechanical Engineering
• Automotive Safety
• Computational Modeling

Background:

• Vehicle front-end design significantly impacts pedestrian head injury severity.
• Accurate prediction of head impact conditions is crucial for evaluating vehicle safety.

Purpose of the Study:

• To develop prediction models for pedestrian head impact conditions using finite element (FE) simulations.
• To create a virtual database of pedestrian impacts for diverse vehicle geometries and pedestrian sizes.

Main Methods:

• Morphed 20 generic vehicle models into 20 U.S. vehicle front-end geometries.
• Conducted 240 FE pedestrian impact simulations with four pedestrian sizes at three speeds.
• Utilized vehicle geometry, pedestrian size, impact speed, and wrap-around distance (WAD) as predictors for head impact time (HIT), velocity, and angle.

Main Results:

• Prediction models achieved high accuracy for HIT (R²=0.979), head impact velocity (R²=0.799), and head impact angle (R²=0.846).
• Impact speed, WAD, hood angle, and hood height were significant predictors.
• HIT was highly predictable, while velocity and angle showed greater variation.

Conclusions:

• A virtual database and prediction models for pedestrian head impacts were successfully generated.
• Models effectively use vehicle and impact parameters to predict head impact conditions.
• Findings suggest varying impact velocity and angle in future vehicle evaluations for enhanced pedestrian head protection.