It is known that some Electronic Control Units (ECUs) that are installed in a vehicle can record precrash and/or post-crash information in an accident. The aim of this study is to understand the availability and usefulness of the ECU data and to develop various analysis methods enhancing the accident investigation.

With respect to ABS-ECU, engine-ECU, and Event Data Recorder (EDR), two types of crash test data are analyzed in this study. The first type is the JNCAP crash tests, for understanding the EDR characteristics under standardized crash test conditions. The second type is the real-world accident reconstructions for evaluating the performance of those ECUs under highly complex and/or severe crash conditions, including multiple rear-end collisions, car-to-car side impacts, and frontal and side pole impacts. The data obtained from ECUs are compared with the results from the instrumented sensors.

The study concludes that, the pre-crash velocities recorded by the EDR were highly accurate and reliable when cars proceeded without braking prior to the collision. The accuracy and reliability of the EDR impact velocity could be affected by the braking conditions and the EDR time zero information. The accuracy and reliability of the maximum delta-V recorded by the EDR decreased under highly complex or severe crash conditions, especially in the pole impacts. The EDRs underestimated the maximum delta-V in almost all the J-NCAP tests. The difference between the EDR maximum delta-V and the reference value was greater than 10 % in 4 of 14 tests. One of the factors responsible for this result might be attributable to the characteristics of the accelerometers used in EDR.

Diagnosis freeze data recorded in ABS-ECU and engine-ECU have a potential to be utilized for the accident investigation by providing additional precrash vehicle information. However, further study is needed for understanding the reliability and accuracy of the diagnosis freeze data.