A study of intersection collision avoidance system in v2x environment based on collision probability calculation

Abstract

In 2012, 223,656 police-reported motor vehicle traffic crashes occurred in Korea. Traffic accidents with an estimated economic cost of 38 billion won represent of 31.5% of GDP in Korea. Accidents are severe problem worldwide, not only in South Korea. In the United States, 33800 of people were killed and 2.2 million injured by traffic crashes in 2009 causing $230 billion in estimated economic costs. With an increasing number of traffic accidents, more research is focused on movements in collision avoidance systems. Collision avoidance systems require analysis of target accident scenario and evaluation proves in hypothetical environments before implementation in reality. The number of traffic accidents tends to decrease with the use of collision avoidance systems with certain targets and implementations after hypothetical evaluation. Statistical analysis shows that intersection-related traffic represent 44% of the total number of car crashes; collision avoidance systems seem to reduce the total number of these types of traffic accidents. The target accident, side-impact crashes between two vehicles at an intersection, becomes more detailed especially with a 60km/h speed limit. Moreover, the accident is caused by a driver’s violation of the traffic signal. To describe the occurrence of the target accident in a hypothetical environment, the collision probability is calculated using the sample data. Using the multiple regression analysis model, the control parameter in the collision avoidance system was obtained based on the time lapse between vehicles at an intersection. Furthermore, the proposed system with the obtained control parameter is evaluated based on the calculation of the average accident rate after it is equipped with the proposed system. This research shows hypothetical analysis prior to implementation of collision avoidance system. It can be used as an early stage to propose guidelines for collision avoidance system with the target accident sce...