Authors: Berkan Guleyupoglu, Bharath Koya and F. Scott Gayzik—Wake Forest University School of Medicine, Virginia Tech – Wake Forest Center for Injury Biomechanics
Abstract
The objective of this study was to reduce the computational cost of human body model (HBM) simulations by developing a method to switch between models of varying detail during an event. One application is simulations involving active safety systems where the total simulation time can greatly exceed crash-only HBM simulations. We introduce an iterative morphing method to capture large changes in occupant position. A simplified HBM is used in the pre-crash portion of an event. Then, a detailed version of the same HBM is morphed to the final position of the simplified counterpart at the instant prior to the initiation of the crash. Each morphing step utilizes a second order regression and Gaussian correlation function. Simulation times for the pre-crash portion of the event were reduced by a factor of 29 times. The average coordinate differences for the head (2.4 mm), pelvis (4.1 mm) and thoracic (3.7 mm) CG’s matched closely between the simulated target and the morphed model. For robustness, the repositioned, detailed HBM was simulated in a NCAP level pulse and completed the simulation without error.
Pages: 3
Event: 61st Stapp Car Crash Conference
Type: Short Communication