Search Results

This study concerns the generation of response surfaces for kinematics and injury prediction in pedestrian impact simulations using human body model. A 1000-case DOE (Design of Experiments) study with a Latin Hypercube sampling scheme is conducted using a finite element pedestrian human body model and a simplified parametric vehicle front-end model. The Kriging method is taken as the approach to construct global approximations to system behavior based on results calculated at various points in the design space. Using the response surface models, human lower limb kinematics and injuries, including impact posture, lateral bending angle, ligament elongation and bone fractures, can be quickly assessed when either the structural dimensions or the structural behavior of the vehicle front-end design change. This will aid in vehicle front-end design to enhance protection of pedestrian lower limbs.

The complexity of vehicle-pedestrian collisions necessitates extensive validation of pedestrian computational models. While body components can be individually simulated, overall validation of human pedestrian models requires full-scale testing with post mortem human surrogates (PMHS). This paper presents the development of a full-scale pedestrian impact test plan and experimental design that will be used to perform PMHS tests to validate human pedestrian models. The test plan and experimental design is developed based on the analysis of a combination of literature review, multi-body modeling, and epidemiologic studies. The proposed system has proven effective in testing an anthropometrically correct rescue dummy in multiple instances. The success of these tests suggests the potential for success in a full-scale pedestrian impact test using a PMHS.

The validity of the full human body model in PAM-CRASH, called the H-Model, was assessed using human volunteers subject to either frontal or rear impact conditions. The frontal impact condition was simulated using a HYGE sled with a belted volunteer under a pulse with a peak acceleration of 9.8 g. The rear impact condition was also simulated using a HYGE sled, but with unbelted volunteers under a relatively soft pulse with peak acceleration about 3.8 g. The sled tests were modeled using PAM-CRASH. Occupant kinematics of the H-Model in the mid-sagittal plane were obtained and compared to volunteer responses.