A Computationally Efficient Mathematical Model of the Pedestrian Lower Extremity

In this dissertation, a computationally-efficient multi-body mathematical model of the 50th percentile male lower extremity capable of predicting the risk of injury in vehicle-pedestrian collisions was constructed and comprehensively validated at both the component and full-scale levels. The model was constructed for the MADYMO software platform using 50 th percentile male anthropometric and inertial properties, and structural response and injury tolerance properties from the literature. In the absence of available data in the literature, the structural response and injury tolerance of the human leg and thigh were experimentally characterized by subjecting isolated post-mortem leg and thigh specimens to dynamic three-point bending experiments designed to provide optimal data for model validation. The scaled response data were used to generate a series of response corridors and the scaled fracture tolerances were used to develop a series of parametric injury risk functions from Weibull survival models fit to the data. To optimize component-level response of the model, leg and thigh model parameters were varied in simulations replicating the experiments until model responses matched scaled response corridor averages. In order to characterize the collective response of the lower extremity in vehicle-pedestrian impacts and generate kinematics data for model validation, seven post-mortem human surrogates (PMHS) were instrumented with custom six-degree of freedom instrumentation packages and subjected to full-scale vehicle-pedestrian impact experiments. Three-dimensional lower extremity kinematics in both local (body) and global (inertial) reference frames were calculated via a customized application of rigid body kinematics theory. Then the model was connected to existing foot/ankle and upper body models and its response to full-scale vehicle impact was verified by comparing its local rotational and global translational kinematics to the experimental data. This dissertation presents the most comprehensively validated pedestrian lower extremity model developed to date. This study presents a framework for detailed mathematical model development and validation that has not been previously employed. Additionally, extensive characterization of the structural response and injury tolerance of the leg and thigh in dynamic bending are presented. Lastly, the kinematic response data captured in the vehicle-pedestrian impact experiments is the most detailed pedestrian lower extremity kinematics data available.