1. Introduction

When a traffic accident is reported to the police station in the vicinity, the on-site officer and the traffic accident investigator identify the cause of the accident based on various types of evidence (e.g., skid marks, a vehicle’s damaged parts, simulation etc.) and identify the perpetrator and the victim, with many refusing to acknowledge that they themselves are the perpetrator. Recent traffic accident investigation involves the use of video records and driving records (EDR, DTG) to identify the cause of the accident and distinguish the perpetrator from the victim. When there are no video records or driving records, however, there are difficulties in determining the exact cause of the accident when the investigation is based on statements from both parties.

Self-driving cars should be designed and developed considering a variety of cases, including vehicle structure, various sensors, actuators, communication failures, potential software errors, inadequate control, potential collision possibilities, off-road, loss of stability, violations of traffic laws, and abnormal driving. Among the matters considered in the design of self-driving cars, the Ministry of Land, Infrastructure, and Transport recommends that mode data should be traceable in the final report of the 2018 self-driving car convergence future forum. In addition, the early self-driving vehicle was equipped with sensors in the vehicle to recognize the surrounding environment and drive autonomously based on the collected information. However, to overcome various constraints such as the price burden of sensors and limitations of perception by weather or road environment, we are developing them by sharing various information through communication with roads and infrastructure. In advanced countries such as the U.S., Germany, and the U.K., Cooperative ITS (C-ITS), which focuses on the exchange of information between combined vehicles and road infrastructure, is actively promoting in advanced countries such as the U.S. and Europe ^[1].

Referring to Figure 1, Digital data collected by sensors for autonomous driving is stored within a vehicle, but numerous IT companies and motor companies such as Samsung, Harman, Microsoft, and Baidu are developing ways to store data in a cloud because of the large amount of data. South Korea, in particular, defined the method of transmitting and storing vehicle operation data to the center in real time, as in the figure below, based on location (track) information for safe driving and operation status information based on the location of the RSU by way of wireless communication. When a vehicle equipped with telematics service transmits CAN or ECU information, vehicle status information (direction, speed etc.) obtained through OBD-II, location (track) information calibrated by correcting the GPS received information within the vehicle, and information on vehicle events to a nearby roadside device (RSU), the location-based vehicle data for each vehicle collected by the roadside device is automatically sent to a center at regular intervals or when it occurs and stored in a database (DB) within ^[2].

Figure 1. Location-Based Vehicle Data Collection Service.

Countries leading the development of self-driving cars such as the U.S., the U.K., Germany, and Japan are pushing for laws to identify the attribution of responsibility for accidents in Level 3 self-driving vehicles from drivers to manufacturers or software developers ^[1]. However, there is no way to prove the cause of the accident. Even if the law is enacted, the law is useless. Therefore, in order to prove the cause and responsibility of the accident, the integrity of the accident data must be ensured, among other things, for accuracy and fairness. Therefore, using blockchain technology, a technology that has proven the accuracy and integrity of current data, it is possible to identify the exact cause of the accident.