LAUSR

Abstract Finite element (FE) human body models, though providing effective means for vehicle design and assessment from the safety point of view, are still under development stage due to their limited validation. These models have been validated for cadaveric test data available in the literature for different parts. However, it has been identified that the reconstruction of the real-world accident using these FE models is required in order to evaluate their biofidelity and robustness for real crash scenarios. Hence, in the present study FE reconstruction of three real-world pedestrian accidents from RAIDS, UK database using the THUMS pedestrian model has been carried out to study the behaviour of the model during the real-world crash scenario and evaluate its biofidelity. Methodology of reconstruction has been divided into two steps: estimation of unknown kinematic parameters related to crash conditions by kinematic reconstruction using multibody simulation, and FE estimation of injuries using the THUMS pedestrian model. Because, lower extremities are most commonly injured body part in pedestrian accidents, the present work focuses on lower extremity injuries. From the reconstruction study, it has been observed that kinematic and injury response of the THUMS lower extremities model shows good conformance with the actual crash data, in terms of both kinematic and injury response.