Recently, the number of elderly driver accidents has been steadily increasing. EDR(Event Data Recorder) helps a lot in understanding traffic accidents. In particular, as anxiety about SUA(Sudden Unintended Acceleration) increases, EDR data is playing an important role in accident analysis. In this study, EDR data of an accident vehicle suspected of SUA was analyzed to identify traffic accident circumstances and detailed accidents. Experimental results were derived and analyzed by simulating the situation of SUA while driving a car. As a result, it was found that normal braking is performed when the brake pedal is operated even in dangerous situations such as mechanical defects and driver malfunctions. Rather than finding the cause of an accident after a traffic accident, countermeasures are needed to prevent mechanical defects and driving malfunctions before a traffic accident.

The event recording devices such as EDR and DTG have recently been developed and installed in automobiles. The reliability of EDR application is being conducted in many previous studies, but with the development and development of autonomous vehicles, it is necessary to study the reliability of EDR and DTG application results for programs that can analyze autonomous driving. Therefore, in this study, the analysis was carried out in a way to secure the reliability of the application results of EDR and DTG of Carmaker, which can analyze the traffic accident analysis program PC-Crash and autonomous driving, and secure the reliability of the Carmaker program that applied EDR in Korea and abroad. As a result of the analysis, it was found that the speed error rate gradually increased from high speed to low speed, and the maximum error speed was less than 5 km/h through the average error rate for each speed. In the future, it is thought that it can be usefully used for analysis of traffic accidents in the event of autonomous vehicle accidents.