Reconstructing Vehicle and Occupant Motion from EDR Data in High Yaw Velocity Crashes

Among the several data recorded by a vehicle’s onboard event data recorder (EDR) prior to, during and after a crash event, are time histories of longitudinal and lateral components of delta-v. The delta-v components are not measured directly but are calculated by numerically integrating the outputs of orthogonal accelerometers contained within the EDR box. As currently designed and implemented, automobile event data recorders typically measure longitudinal and lateral accelerations, but do not measure yaw velocity during the impulse phase of a crash event. A vehicle’s onboard EDR is rigidly fixed to the vehicle and therefore translates and rotates with the vehicle. Even though the EDR accurately measures the components of acceleration along the axes to which its accelerometers are mounted (typically aligned with the longitudinal and lateral axes of the vehicle), and correctly integrates the measured acceleration values to calculate the corresponding longitudinal and lateral delta-v components, these components of delta-v are calculated relative to the EDR accelerometer locations in the rotating vehicle, and should not be interpreted as representing absolute changes in vehicle velocity. Especially in cases where there is high-speed yaw, the delta-v values calculated by the EDR through direct integration of the measured components of acceleration must be adjusted to account for the yawing of the vehicle that occurred during acquisition of the data, i.e., the absolute values of delta-v must be derived from the EDR-calculated relative values. Requiring adjustment if yaw rates are high, the EDR-calculated delta-v components provide the basis, along with reasonable estimates of vehicle acceleration and yaw velocity, for determining the absolute change in the vehicle’s velocity and its resulting displacement during and immediately following the collision event that triggered the EDR to make a record. In turn, the motion of occupants relative to the interior of the impacted vehicle is determined by the rapid changes in vehicle position. A hypothetical crash is analyzed to demonstrate the important principles involved in using EDR data to reconstruct crash vehicle motion. Two full scale crash tests illustrate the results of utilizing EDR-calculated delta-v components to reconstruct vehicle motion without adjusting for vehicle yaw. Two equivalent methods of adjusting EDR delta-v output to account for vehicle rotation and to derive the absolute delta-v components are presented. These methods are shown to increase the accuracy of velocity and displacement calculations when reconstructing vehicle and occupant motion from EDR data in high yaw rotation crashes.