Accident Reconstruction Using Data Retrieval from Crash-Test Video Images 2016-01-1464

The main source for the estimation of stiffness coefficients to be used in accident reconstruction calculations is a very large database of crash-test related information from NHTSA. However, that database includes only car models sold in the USA. Unfortunately, there is no such information for European-only cars besides the raw video recordings of EuroNCAP crash tests. In the present work a methodology is proposed to estimate the stiffness coefficients of European-only models from video images of EuroNCAP crash tests. However, these images are intricate to assess, because the car front is crushed into a deformable barrier at 40% of the front width and usually the bonnet (hood) hides most of the crash damage. Therefore, the top images could not be used straightforward, so a procedure was envisaged to circumvent this difficulty and still allow to calculate stiffness coefficients for European-only cars.

The methodology for stiffness coefficients estimation involves:

1. 1
   calculation of an average value for rebound based on NHTSA tests;
2. 2
   calculation of the deformation of the deformable barrier as a function of vehicle class;
3. 3
   measurement of the maximum deformation, using the superposition of video images of the crash-test;

Since the video images are subject to error, such as parallax, a method to reduce these errors was established and used. Then, from the video image it was possible to calculate the permanent deformation by subtracting the rebound (1) and the deformation of the barrier (2) from the maximum deformation (3) that was measured from the image.

This methodology allowed to extend the existing database of US cars to European models and to significantly improve the calculation of the dissipated energies associated to the permanent damage caused by accidents.