PURPOSES : This study was purposed to investigate whether the current passing sight distance standard is appropriate. Therefore, to determine a safe passing sight distance, data collected from the observations of a two-lane road were implemented in reliability analysis to develop a realistic passing sight distance model. METHODS : First, passing sight distance data were collected for analysis from two-lane roads in Jeollabuk-do. Next, the speed profile was generated based on the collected data in order to analyze driver behavior. Finally, a model reflective of the driving behavior of motorists was developed and compared to the existing passing sight distance standard using reliability analysis. RESULTS : As compared to the existing model for safe passing sight distance, the actual speed of the overtaking vehicle indicated higher acceleration and a longer acceleration duration. The entire model, with exception of the d1 section, was modified to reflect the overtaking behavior of motorists d1. d2 was set to correspond to the time until the vehicles run side by side (t21); Time t22 was set to correspond to the time until subsequent recovery to the main lane, and d3 was set to range between 19 and 47 m following review of the actual passing sight distance data. In addition, d4 was computed to be two-thirds of the distance d2, and the oncoming car was found to drive at a constant speed that was similar to the speed associated with d2 . Analysis of the reliability of the existing Korean passing sight distance standard revealed that the current standard does not recommend a safe passing sight distance for drivers. CONCLUSIONS : As compared to the existing passing sight distance criterion, the actual speed and acceleration that driver apply to overtake a vehicle are higher, and the existing passing sight distance criterion is shorter than the actual passing sight distance. Additionally, the results of reliability analysis revealed that the proposed model developed to reflect driver overtaking behavior has more explanatory power than the existing one. Therefore, it is necessary to introduce new design standards that implement passing sight distance variables that are relevant to local drivers.

PURPOSES : Because elderly drivers are more prone to becoming confused when approaching an urban intersection and thus may yield prolong judgment and decision times than non-elderly drivers, to increase the comfort and safety of the intersection environment for elderly drivers, this study applied autonomous driving tests at an urban intersection to examine their driving characteristics. METHODS: To obtain a more comprehensive understanding of driving features, this study collected drive data of non-elderly drivers and elderly drivers via an autonomous experiment using OBD2 and an eye-tracker, in addition to performing a literature review on the measured visibility range of elderly drivers at intersections. This literature review was conducted considering the general knowledge of elderly drivers having relatively reduced visibility. Additionally, as they are commonly more vulnerable, this study analyzes characteristics of elderly drivers as compared to those of non-elderly drivers. CONCLUSIONS: The results of this study can be summarized as follows: 1) the peripheral visible distance of elderly drivers is reduced as compared to that of non-elderly drivers; 2) elderly drivers approach and proceed through intersections at slower speeds than non-elderly drivers; and 3) elderly drivers yield increased driving distances when performing a right or left turn as compared to non-elderly drivers as a result of their reduced speed and acceleration and larger turning radii relative to non-elderly drivers.

PURPOSES: According to accident statistics for road built in 2004, the ratio of accidents on frozen roads to normal roads is 0.9%, whereas the fatality ratio is 2.7%. The risk of accidents on frozen roads is very high. Measures taken every year to prevent traffic accidents of frozen roads in the winter season are still insufficient. Additionally, measures have been established mainly on rural roads. Therefore, for urban roads, analyses and measures to prevent accidents are lacking. In this study, data on accidents on frozen roads was used to search for the causes behind these accidents and measures to reduce accidents have been recommended. METHODS: In this study, collected data from the TAMS (Traffic Accident Management System), which were collected by the Seoul National Police Agency was used. The data were divided into vehicle, people, and condition of road. The analytical model used here was the Logistic Regression Model, which is frequently used for traffic safety and accident analysis. This study uses the odds ratio analysis to search for variables related to frozen road traffic accidents in each category. A total of 18 out of 47 variables were found to be the causes of accidents. RESULTS: From the results of the comparative analysis of 18 variables, the category of the condition of the road was found to be the most critical. Contrary to expectations, more accidents occurred in clear weather than in other conditions. Accidents on bridges occurred frequently, and its odds ratio was the highest compared with other road types. When BPT is operated, the probability of accidents on frozen roads is lower than in general conditions, and accidents occurred frequently on roads with less than four lanes. CONCLUSIONS : Based on the results of this study, suggestions for reducing the risk of future domestic road accidents in freezing conditions are indicated as follows. First, it is necessary to perform a technical review of the urban road traffic accidents caused by frozen roads. Second, it is necessary to establish criteria for the study of the road environment based on the major causes of road accidents on frozen roads. Third, improvements in urban road environmental factors should be made.

PURPOSES: This study is to conduct the research on the design principle for the driver's safety and comfort in installing consecutive curves of superhighway. METHODS: Superhighway does not currently exist in domestic area. Thus, this study is conducted by collecting driving behavior usage of 30 people who are involved in the members of the virtual driving simulation. By identifying the distribution characteristics of each scenario in ANOVA & Tukey Test, the distribution are categorized into three groups. RESULTS : In the case of Group A in Section 3 (R2 entry part), lane departure exceeds the safety standard, which means to be risky condition. And then in the case of Group B and C, the lane departure values applying theoretical formula was evenly distributed compared to the proven values. CONCLUSIONS : Based on the result, the continuous curve design principles at superhighway should follow three standards as follow. First, an additional linear part needs to be inserted between two curves. Second, what if inserting the linear part is difficult, it would be better to insert a curve more than 2,000m. Third, R1/R2 ratio should not be over two. This design primarily aims to the safety of the operator. Such road alignment also meets the expectations of drivers, thus, it may help drivers to be compatible and amenable while driving continuous curve in superhighway.