High Biofidelity Modification of Thoracolumbar Structure in THOR model for Reclined Occupant

With the rapid development of automated driving and the increasing adoption of zero-gravity seats, the crash safety of highly reclined occupants has become a critical issue. The current THOR model, designed for frontal impacts in the standard upright posture, shows limitations when applied directly for reclined occupants, including excessive thoracolumbar flexion stiffness and large backward pelvis rotation. In this study, we analyzed the thoracolumbar spine kinematics of the THUMS human body model, which had been reconstructed against post-mortem human subject (PMHS) tests. A two-segment fitting was used to “dummy-ize” the spinal flexion response, yielding a virtual rotation hinge located close to the thoracic hinge of the original THOR model. The hinge rotation angle exhibited a strong linear correlation with the flexion moment of the L1–T12 segment. Based on this relationship, angular stiffness derived from the fitting was applied to the thoracolumbar hinge of the dummy. Additional modifications were also implemented in the hip region to improve applicability. Comparative simulations demonstrated that our modified THOR model achieved closer agreement with PMHS responses than both the HIII model and the baseline THOR model (educational version). In particular, the pelvic rotation problem was alleviated, reducing the posterior tilt angle from about 20° in the baseline THOR to approximately 10°. These results indicate that introducing PMHS-based thoracolumbar flexion characteristics and targeted hip modifications substantially improves the biofidelity of THOR model for reclined occupant crash scenarios, providing a promising foundation for future dummy development and safety assessment.