Eunbi Jeong

Hazardous Driving Event Detection and Analysis System in Vehicular Networks: Methodology and Field Implementation

This study proposes a novel highway traffic surveillance system that is capable of detecting hazardous driving maneuvers through the use of an in-vehicle sensor and transmission of the event data detected to a traffic management center through vehicle-to-vehicle and vehicle-to-infrastructure wireless communication; this system is referred to as the “hazardous driving event detection and analysis system in vehicular networks.” Compared with existing surveillance systems, the main novel feature of the proposed system lies in its ability to detect hazardous driving maneuvers that have the potential to lead to crashes. Three major components of the system are introduced in this study: an algorithm for detecting hazardous driving events, a method for deriving a new index incorporating expert judgment for evaluation of the risk level of the traffic stream on the basis of analyses of hazardous events that are detected, and field implementation of the proposed system in a test bed with real-time and historical data. Extensive field tests were conducted in the test bed to fine-tune the prototypical system. The methodology and field implementation presented in this study have potential value to help highway traffic agencies monitor and evaluate traffic streams with a focus on traffic safety. The proposed system is expected to be effective as support for the development of various traffic information control strategies to enhance traffic safety on highways.

Safety Impacts of Intervehicle Warning Information Systems for Moving Hazards in Connected Vehicle Environments

Driver inattentiveness is one of the critical factors that contribute to vehicle crashes. The intervehicle safety warning information system (ISWS) is a technology to enhance driver attentiveness by providing warning messages about upcoming hazards under the connected vehicle environments. A novel feature of the proposed ISWS is its capability to detect hazardous driving events, which are defined as moving hazards with a high potential to cause crashes. The study presented in this paper evaluated the potential effectiveness of the ISWS to reduce crashes and to mitigate traffic congestion. The study included a field experiment that documented actual vehicle maneuvering patterns of accelerations and lane changes, which were used to enhance the realism of simulation evaluations. Probe vehicles equipped with customized onboard units, which consisted of a GPS device, accelerometer, and gyro sensor, were used. A microscopic simulator, VISSIM, was used to simulate a drivers responsive behavior after warning messages were delivered. A surrogate safety assessment model was used to derive surrogate safety measures to evaluate the effectiveness of ISWS in terms of traffic safety. The results showed a reduced number of rear-end conflicts when the ISWSs market penetration rate (MPR) and the congestion level of the traffic conditions increased. The reduced number of rear-end conflicts was approximately 84.3%, with a 100% MPR under Level of Service D traffic conditions. Analysis of the standard deviation of speed showed that a reduction of 39.9% was achieved. The outcomes of this study could be valuable to derive smarter operational strategies for ISWS.

Methodology for Estimating Safety Benefits of Advanced Driver Assistant Systems

최근 교통사고 및 교통사고로 인한 사상자수의 감소를 위해 기존의 자동차에 각종센서나 통신기술 등의 첨단 ICT기술을 융합한 첨단안전자동차에 대한 연구가 활발히 진행 중에 있다. 이러한 첨단안전자동차 기술의 시장진입 및 관련 기술 도입을 위해서는 자동차에 도입되는 첨단안전 기술에 대한 효과 분석에 따른 도입 타당성 평가가 필요하다. 본 연구에서는 첨단안전자동차 기술 중 운전자지원시스템(ADAS: Advanced Driver Assistant System)을 대상으로 효과분석 방법론을 개발하고, 개발한 방법론을 차선이탈경고장치(LDWS: Lane Departure Warning System)와 자동비상제동장치(AEBS: Automatic Emergency Braking System)에 적용한 사례분석을 통해 사고감소효과를 추정하였다. 분석결과, 차선이탈경고장치는 관련 사고유형(정면충돌, 도로외이탈, 공작물추돌, 전도전복)에 대해 약 10~14%의 사고감소 효과가 있는 것으로 나타났으며, 자동비상제동장치는 추돌사고에 대하여 약 50%의 사고감소 효과를 보이는 것으로 나타났다. 본 연구의 결과는 추후 첨단안전자동차 기술 개발에 따른 효과분석시 기초자료로 활용이 가능할 것이며, 운전자, 탑승자, 보행자를 위한 자동차 기술의 발전 및 기술도입 타당성을 제시하기 위한 근거 자료로 활용이 가능할 것으로 기대된다. 【Recent advanced sensors and communication technologies have been widely applied to advanced safety vehicle (ASV) for reducing traffic accident and injury severity. To apply the advanced safety vehicle technologies, it is important to quantify the safety benefits, which is a fundamental for justifying application. This study proposed a methodology for quantifying the effectiveness of the advanced driver assistant system (ADAS), and applied the methodology to lane departure warning system (LDWS) and automatic emergency braking system (AEBS) which are typical advanced driver assistant systems. When the proposed methodology is applied to 2008-2010 gyeonggi-province crash data, LDWS would reduce about 10~14% of relevant crashes such as head-on, run-off-the road, rollover and fixed-object collisions on the road. In addition, AEBS could potentially prevent about 50% of total rear-end crashes. The outcomes of this study support decision making for developing not only vehicular technology but also relevant safety policies.】

Evaluating the effectiveness of active vehicle safety systems

Advanced vehicle safety systems have been widely introduced in transportation systems and are expected to enhance traffic safety. However, these technologies mainly focus on assisting individual vehicles that are equipped with them, and less effort has been made to identify the effect of vehicular technologies on the traffic stream. This study proposed a methodology to assess the effectiveness of active vehicle safety systems (AVSSs), which represent a promising technology to prevent traffic crashes and mitigate injury severity. The proposed AVSS consists of longitudinal and lateral vehicle control systems, which corresponds to the Level 2 vehicle automation presented by the National Highway Safety Administration (NHTSA). The effectiveness evaluation for the proposed technology was conducted in terms of crash potential reduction and congestion mitigation. A microscopic traffic simulator, VISSIM, was used to simulate freeway traffic stream and collect vehicle-maneuvering data. In addition, an external application program interface, VISSIMs COM-interface, was used to implement the AVSS. A surrogate safety assessment model (SSAM) was used to derive indirect safety measures to evaluate the effectiveness of the AVSS. A 16.7-km freeway stretch between the Nakdong and Seonsan interchanges on Korean freeway 45 was selected for the simulation experiments to evaluate the effectiveness of AVSS. A total of five simulation runs for each evaluation scenario were conducted. For the non-incident conditions, the rear-end and lane-change conflicts were reduced by 78.8% and 17.3%, respectively, under the level of service (LOS) D traffic conditions. In addition, the average delay was reduced by 55.5%. However, the systems effectiveness was weakened in the LOS A-C categories. Under incident traffic conditions, the number of rear-end conflicts was reduced by approximately 9.7%. Vehicle delays were reduced by approximately 43.9% with 100% of market penetration rate (MPR). These results imply that from the perspective of traffic operations and control to address the safety and congestion issues of a traffic stream, smarter management strategies that consider both traffic conditions and MPR are required to fully exploit the effectiveness of the AVSS in the field.

Algorithm for Identifying Highway Horizontal Alignment using GPS/INS Sensor Data

Geometric information is a key element for evaluating traffic safety and road maintenance. This study developed an algorithm to identify horizontal alignment using global positioning system(GPS) and inertial navigation system(INS) data. Roll and heading information extracted from GPS/INS were utilized to classify horizontal alignment into tangent, circular curve, and transition curve. The proposed algorithm consists of two components including smoothing for eliminating outlier and a heuristic classification algorithm. A genetic algorithm(GA) was adopted to calibrate parameters associated with the algorithm. Both freeway and rural highway data were used to evaluate the performance of the proposed algorithm. Promising results, which 90.48% and 88.24% of classification accuracy were obtainable for freeway and rural highway respectively, demonstrated the technical feasibility of the algorithm for the implementation.

Multiple-Step Traffic Speed Forecasting Strategy for Winter Freeway Operations

Accurate and timely predictions of traffic conditions are required for congestion avoidance and route guidance in real-time freeway traffic operations. Special attention to winter operations is needed because prediction error could be amplified under severe weather conditions involving snow. This study employed a vehicle detection system to propose a speed prediction methodology that used the k–nearest neighbors algorithm. The speed prediction was further evaluated under different weather conditions with a road weather information system. Cross-comparisons of the mean absolute percentage error (MAPE) between three weather conditions (normal, light snow, and heavy snow) revealed that the MAPE tended to increase with increases in the forecasting time step (T) and snow intensity. The marginal MAPE over the time step was larger during heavy snow conditions than under normal and light snow conditions. These findings indicate that for winter freeway operations, the time step should be selected dynamically, depending on the weather conditions rather than with a static strategy for all conditions. To this end, this study proposes a framework to determine a dynamic forecasting T that is associated with weather conditions.