In this study, we investigate a new method of testing the occupant kinematics in a rollover crash situation. Much of this work is based on previous full scale vehicle studies by Orlowski and Bahling ( 1 , 2 ). Their work concentrated on FMVSS 208 dolly rollover tests of vehicles equipped with production and reinforced vehicle roofs. They found that the occupant's kinematics, as opposed to roof crush, were responsible for potentially injurious neck injuries as a result of diving type accident kinematics of the head and torso.

This led us to examine seat system, belt restraint system and belt restraint anchorage designs that could potentially improve the occupants head to roof clearance.

A simulated vehicle environment with representative seat and belt restraint systems was chosen as the baseline system. These quasistatic tests applied a rigid roof I seat and belt restraint geometry. Kinematics of a 50th percentile Hybrid Ill dummy were analyzed in the quasistatic test procedure. Modifications of the seat, belt restraint system, and its anchorages changed the trajectory and kinematics of the dummy. This paper describes the laboratory test fixture, test method for simulating rollover, and results of some of over thirty rollover tests. Based on these results, a seat, belt restraint, and belt restraint anchorage design is described for this vehicle environment that reduced the excursion of the dummies from their seats in these simulations.