Time history of movement (stroke) is crucial information for crash test analysis. The stroke is often calculated from double integration of linear acceleration data when direct measurement by a potentiometer is impossible. But this method may not be accurate for the cases with large rotations. Newer crash tests like IIHS small overlap and NHTSA oblique involve large rotation, creating 3-dimensional (3D) occupant motions compared to front rigid barrier tests which are primarily 2D events. To help overcome some of the challenges posed by newer crash tests, the authors developed a method to calculate accurate 3D motion with 6-DOF (Degree-of-Freedom) instrumentation including angular rate sensors (ARS).
The calculation of accurate 3D position and orientation for a rigid body requires data collection of 6-DOF: linear acceleration (ax, ay, az) and angular velocity (ωx, ωy, ωz). The mathematical calculation to account for rotation of a rigid body was done by using the screw-axis method. The quaternion was calculated using numerical integration via the 4th-order Runge-Kutta method.
A dynamic component test was designed and conducted with a linear impactor to validate the 3D stroke calculator. The test set up included a 6-DOF sensor pack mounted on a polyethylene stick with an offset. The stick was mounted on a base, which was pushed by a linear impactor with controlled speed. The sensor experienced 3D motion when the stick was decelerated by the base impacting a honeycomb backstop.
This method of 3D rigid body tracking has various crash testing applications. The authors compared occupant head kinematics among three different frontal offset crash modes. A finding of the study was that one test mode resulted in more driver head stroke (relative to vehicle interior) compared to the other two crash test modes. The maximum head stroke, compared to the least, was more by 64% (longitudinal) and 49% (lateral).