PURPOSES : For autonomous vehicles, abnormal situations, such as sudden changes in driving speed and sudden stops, may occur when they leave the operational design domain. This may adversely affect the overall traffic flow by affecting not only autonomous vehicles but also the driving environment of manual vehicles. Therefore, to minimize the traffic problems and adverse effects that may occur in mixed traffic situations involving manual and autonomous vehicles, an autonomous vehicle driving support system based on traffic operation optimization is required. The main purpose of this study was to build a big-data-classification system by specifying data classification to support the self-driving of Lv.4 autonomous vehicles and matching it with spatio-temporal data. METHODS : The research methodology is explained through a review of related literature, and a traffic management index and big-dataclassification system were built. After collecting and mapping the ITS history traffic information data of an actual Living Lab city, the data were classified using the traffic management indexing method. An AI-based model was used to automatically classify traffic management indices for real-time driving support of Lv.4 autonomous vehicles. RESULTS : By evaluating the AI-based model performance using the test data from the Living Lab city, it was confirmed that the data indexing accuracy was more than 98% for the KNN, Random Forest, LightGBM, and CatBoost algorithms, but not for Logistics Regression. The data were severely unbalanced, and it was necessary to classify very low probability nonconformities; therefore, precision is also important. All four algorithms showed similarly good performances in terms of accuracy. CONCLUSIONS : This paper presents a method for efficient data classification by developing a traffic management index to easily fuse and analyze traffic data collected from various institutions and big data collected from autonomous vehicles. Additionally, EdgeRSU is presented to support the driving of Lv.4 autonomous vehicles in mixed autonomous and manual vehicles traffic situations. Finally, a database was established by classifying data automatically indexed through AI-based models to quickly collect and use data in real-time in large quantities.

일반차량과 자율주행차량이 혼재하는 상황에서 발생가능한 미래 재난상황에 대한 관리방안 준비가 필요하다. 특히 재난 상황 중 안 개 발생 시 시야 확보가 어려운 일반차량 운전자와 센서기반 자율주행차량의 주행 특성이 다를 수 있다. 해당 상황에서의 문제점을 도출하고 이를 극복하기 위해 혼합교통류 관리 방안을 제안하고자 한다. 본 연구에서는 다양한 재난 상황 중 안개를 연구 대상으로 설정하였다. 과거 기상 상황별 일반차량을 주행 특성을 이력자료로 분석한 후, 안전한 교통흐름을 유지하기 위하여 자율주행차에게 정 보를 제공하는 방안을 제안한다.

자율주행차가 보급되어 도로에서 사람 운전자와 함께 운영되는 미래가 다가오고 있다. 사람 중심으로 운영되는 도로 체계가 자율주 행차와 공존하는 형태로 변화하고 있으며, 도로 시스템도 사람 운전자와 자율주행차가 혼재된 혼합교통류를 대상으로 변화하고 있다. 현재 도로에서는 예상하지 못한 상황들이 다양하게 발생한다. 교통사고, 도로 낙하물 등 교통흐름에 영향을 주는 상황들이 발생하며, 대응을 위한 전략들이 각 지방자치단체에서 준비되어 있다. 미래 교통상황에는 도로상에 자율주행차가 혼재되어 있으며 이를 포함하 는 돌발 및 재난상황에 대한 제어전략은 아직 부재하다. 본 연구에서는 돌발 및 재난상황 발생 시 자율주행차 제어전략에 대한 설계 방안을 제안한다. 돌발 및 재난상황 범위에 대해 정의하며, 상황 구분을 위한 기준을 제시하여 각 상황에서 자율주행차가 안전하게 대 응할 수 있도록 제어전략을 제시한다.

PURPOSES : This study presents a general guideline for the initial management of traffic signal timings in response to traffic incidents, prior to the implementation of specific treatments in detail. The proposed solution includes a set of optimal reductions in the green time rates at three signalized intersections upstream. METHODS : To account for the various traffic and incident conditions that may be encountered, a total of 36 traffic-condition scenarios were prepared. These scenarios encompass a wide range of conditions, from unsaturated to near-saturated conditions, and were designed to provide a comprehensive understanding of the impact of traffic conditions on signal timing. For each of the traffic conditions, all 27 traffic signal timing combinations were subjected to testing. A total of 972 simulation analyses were conducted using the SUMO model. The results indicated that the scenario with the lowest control delay was the optimal choice. RESULTS : The results indicated that the most effective initial management for the traffic incident would be to reduce the green signal timings by 20% at the first two upstream intersections and by 40% at the third intersection. CONCLUSIONS : We propose reducing the green times by 20% at the first and second intersections and by 40% at the third intersection as the initial response of the traffic signal control center when a traffic incident occurs.

PURPOSES : Even when autonomous vehicles are commercialized, a situation in which autonomous vehicles and regular drivers are mixed will persist for a considerable period of time until the percentage of autonomous vehicles on the road reaches 100%. To prepare for various situations that may occur in mixed traffic, this study aimed to understand the changes in traffic flow according to the percentage of autonomous vehicles in unsignalized intersections. METHODS : We collected road information and constructed a network using the VISSIM traffic simulation program. We then configured various scenarios according to the percentage of autonomous vehicles and traffic volume to understand the changes in the traffic flow in the mixed traffic by scenario. RESULTS : The results of the analysis showed that in all scenarios, the traffic flow on major roads changed negatively with the mix of autonomous vehicles; however, the increase or decrease was small. By contrast, the traffic flow on minor roads changed positively with a mix of autonomous vehicles. CONCLUSIONS : This study is significant because it proactively examines and designs traffic flow changes in congested traffic that may occur when autonomous vehicles are introduced.

PURPOSES : Because a driving simulator typically focuses on analyzing a driver’s driving behavior, it is difficult to analyze the effect on the overall traffic flow. In contrast, traffic simulation can analyze traffic flow, that is, the interaction between vehicles; however, it has limitations in describing a driver’s driving behavior. Therefore, a method for integrating the simulator and traffic simulation was proposed. Information that could be controlled through driving experiments was used, and only the lane-change distance was considered so that a more natural driving behavior could be described in the traffic flow. METHODS : The simulated connection method proposed in this study was implemented under the assumption of specific traffic conditions. The driver’s lane-changing behavior (lane-changing distance, deceleration, and steering wheel) due to the occurrence of road debris was collected through a driving study. The lane-change distance was input as a parameter for the traffic simulation. Driving behavior and safety were compared between the basic traffic simulation setting, in which the driver's driving behavior information was not reflected, and the situation in which the driving simulator and traffic simulation were integrated. RESULTS : The number of conflicts between the traffic simulation default settings (Case 1) and the situation in which the driving simulator and traffic simulation were integrated (Case 2) was determined and compared for each analysis. The analysis revealed that the number of conflicts varied based on the level of service and road alignment of the analysis section. In addition, a statistical analysis was performed to verify the differences between the scenarios. There was a significant difference in the number of conflicts based on the level of service and road alignment. When analyzing a traffic simulation, it is necessary to replicate the driving behavior of the actual driver. CONCLUSIONS : We proposed an integration plan between the driving simulator and traffic simulation. This information can be used as fundamental data for the advancement of simulation integration methods.

PURPOSES : In this study, model-agnostic methods are applied for interpreting machine learning models, such as the feature global effect, the importance of a feature, the joint effects of features, and explaining individual predictions. METHODS : Model-agnostic global interpretation techniques, such as partial dependence plot (PDP), accumulated local effect (ALE), feature interaction (H-statistics), and permutation feature importance, were applied to describe the average behavior of a machine learning model. Moreover, local model-agnostic interpretation methods, individual conditional expectation curves (ICE), local surrogate models (LIME), and Shapley values were used to explain individual predictions. RESULTS : As global interpretations, PDP and ALE-Plot demonstrated the relationship between a feature and the prediction of a machine learning model, where the feature interaction estimated whether one feature depended on the other feature, and the permutation feature importance measured the importance of a feature. For local interpretations, ICE exhibited how changing a feature changes the interested instance’s prediction, LIME explained the relationship between a feature and the instance’s prediction by replacing the machine model with a locally interpretable model, and Shapley values presented how to fairly contribute to the instance’s prediction among the features. CONCLUSIONS : Model-agnostic methods contribute to understanding the general relationship between features and a prediction or debut a model from the global and/or local perspective, securing the reliability of the learning model.

해상교통관제 업무의 최적화를 위하여 요구되는 인적요소 분야 중 관제사의 상황인식(SA: Situation Awareness)와 관제 업무부하 (Workload)와의 관계성을 확인하는 것이 해상교통 분야에서는 중요한 실정이다. 이 연구에서는 관제사의 상황인식과 업무부하를 상황인식 평가기술(SART)과 다차원 작업부하 지표(NASA-TLX)를 실제적으로 측정하고, 측정 결과를 비교함으로서 개념들에 대한 이해와 시스템적 으로 관리할 수 있는 방법을 제시함으로써 해상교통관제사 전문성 제고방안에 기여하고자 한다.

철도안전법 상의 사고 분류 이외에도 철도운영기관의 실무 차원에서 인적 오류를 유발한 당사자에게 책임이 있다고 판정할 경우 처벌과 불이익을 부과하는 책임사고 판정제도가 있다. 책임사고경험이 있는 당사자는 심리적, 신체적 피로와 긴장감을 경험하는 한편, 사고 및 장애발생의 가능성을 늘 염려하면서 주어진 과업을 수행하게 된다. 본 연구는 그 동안 철도분야 인적오류 연구의 공식적 대상에서 제외되어 왔던 철도차량 검수직 종사자를 대상으로 책임사고경험이 이례상황 스트레스 및 건강의 한가지 척도인 신체적 우울감에 미치는 인과관계를 AMOS 통해 밝혀보았다. 연구 결과, 검수직 종사자의 책임사고 경험은 이례상황 스트레스를 부분 매개로 하여 신체적 우울감에 유의한 영향을 미치는 것으로 나타났다.

최근 고속도로 터널부 대형사고의 발생으로 사회적 관심이 급증하고 있다. 고속도로의 터널부는 일반부와는 다른 터널부만의 주행특성을 가지고 있으며, 이에 따라 차량 주행시 고속도로 일반부와 터널부의 속도차이가 발생하게 된다. 이와 같은 현상은 터널 구간이 운전자의 차량 간 거리 감지능력과 급격한 명암 변화에 의해 암순응과 명순응의 시간이 일반구간과 달리 길기 때문에 발생하는 운전자의 자기방어 성격이라 할 수 있다. 이처럼 고속도로 터널부의 경우 고속도로 시설 중 사고에 가장 취약한 구간이라 할 수 있으며 이러한 구간에서의 속도의 변화는 대형 사고로 이어질 가능성이 크다. 본 연구의 목적은 터널 내 설치된 VDS 데이터 자료(속도, 교통량, 점유율)를 활용하여 터널 내 사고 발생시의 기존 대처방안을 보완하고, 터널 이전 지점의 교통량, 점유율 패턴을 통해 터널 내 사고 발생 가능성을 예측하며 사전에 터널 내 소통상황을 향상시키는 방안을 제시하는것이다. 이를 위해 본 연구에서는 각 터널 내 속도패턴, 터널 이전 지점의 교통량 , 점유율 패턴 등을 분석하고, 해당 패턴에서의 사고 발생 가능성을 파악하는 등 터널 부 사고시 신속한 대처 방안 및 터널 내 소통향상 기법 등을 제시한다.

항만 및 연안의 관제구역 내에서는 VHF를 이용하여 24시간 해상교통관제가 이루어지고 있다. 그러므로 VHF 교신을 분석하면 관제구역 내의 선박 움직임이나 관제사의 관제패턴을 확인할 수 있다. 따라서 본 연구에서는 VHF 교신분석으로 관제사가 관제구역 내 위험상황을 관제하는 간격을 도출하고 관제 가이드라인 및 위험한 상황을 사전에 대비할 수 있는 기초 자료로 제공하고자 한다. 이를 위하여 부산항을 대상으로 7일간 VHF 교신을 청취하고, VTS가 직간접적으로 관제한 선박에 대하여 Park 모델을 이용하여 위험도를 도출하였다. 이를 이용하여 단위시간당 일정 위험도 이상의 선박을 관제하는 빈도확률이 푸아송 분포를 따르고 있음을 확인하였고, 그 결과 VTS가 선박을 직접 관제에 개입할 경우는 3.50시간마다, 특히 주간시간대의 경우 2.85시간마다 관제하는 것으로 분석되었다. 그리고 3.84시간마다 일정 위험도 이상의 선박간의 교신이 있음을 확인하였다.

PURPOSES: The purpose of this study is to develop a methodology for estimating additional carbon emissions due to freeway incidents. METHODS : As our country grows, our highway policy has mainly neglected the environmental and social sectors. However, with the formation of a national green growth keynote and an increase in the number of people interested in environmental and social issues, problems related to social issues, such as traffic accidents and congestion, and environmental issues, such as the impact of air pollution caused by exhaust gases that are emitted from highway vehicles, are beginning to be discussed. Accordingly, studies have been conducted on a variety of environmental aspects in the field of road transport, and for the quantitative calculation of greenhouse gas emissions, using various methods. However, in order to observe the effects of carbon emissions, microscopic simulations must use many difficult variables such as cost, analysis time, and ease of analysis process. In this study, additional greenhouse gas emissions that occur because of highway traffic accidents were classified by type (incident handling time, number of lanes blocked, freeway level of service), and the annual additional emissions based on incidents were calculated. According to the results, congestion length and emissions tend to increase with an increase in incident clearance time, number of occupied lanes, and worsening level of service. Using this data, we analyzed accident data on the Gyeong-bu Expressway (Yang-Jae IC - Osan IC) for a year. RESULTS : Additional greenhouse gas emissions that occur because of highway traffic accidents were classified by type (incident handling time, number of lanes blocked, freeway level of service) and annual additional emissions caused by accidents were calculated. CONCLUSIONS: In this study, a methodology for estimating carbon emissions due to freeway incidents was developed that incorporates macroscopic flow models. The results of the study are organized in the form of a look-Up table that calculates carbon emissions rather easily.

PURPOSES : This study aims to investigate the direct and indirect influence areas from incidents on urban interrupted roadways and to develop traffic management strategies for each influence area. METHODS : Based on a literature review, various traffic management strategies for certain incidents were collected. In addition, the relationship between the measure of effectiveness and the characteristics of incidents was explored using an extensive simulation study. RESULTS : From the simulation studies, traffic delays increased as the number of lane closures increased, and the impact of lane closures was reduced to the direction upstream from the incident site. However, the magnitude of the delay change depended on the degree of saturation. Using these characteristics, the direct and indirect influence areas resulting from incidents were defined, and traffic management strategies were established for each direct and indirect influence area and for each level of incident. CONCLUSIONS: The results of this study will contribute to the improvement of national traffic safety by preventing secondary incidents and by effective adaptation to incident events.