• Journal of information and communication convergence engineering
• /
• 제11권2호
• /
• pp.95-102
• /
• 2013

This paper presents a stochastic driver behavior modeling framework which takes into account both individual and general driving characteristics as one aggregate model. Patterns of individual driving styles are modeled using a Dirichlet process mixture model, as a non-parametric Bayesian approach which automatically selects the optimal number of model components to fit sparse observations of each particular driver's behavior. In addition, general or background driving patterns are also captured with a Gaussian mixture model using a reasonably large amount of development data from several drivers. By combining both probability distributions, the aggregate driver-dependent model can better emphasize driving characteristics of each particular driver, while also backing off to exploit general driving behavior in cases of unseen/unmatched parameter spaces from individual training observations. The proposed driver behavior model was employed to anticipate pedal operation behavior during car-following maneuvers involving several drivers on the road. The experimental results showed advantages of the combined model over the model adaptation approach.

• International Journal of Computer Science & Network Security
• /
• 제24권3호
• /
• pp.142-150
• /
• 2024

A conceptual model represents an understanding of a system that is going to be developed, which in this research, a driving simulation software to study driver behavior at signalised intersections. Therefore, video observation was conducted to study driver compliance behaviour within the dilemma zone at signalised intersection, with regards to driver's distance from the stop line during yellow light interval. The video was analysed using Thematic Analysis and the data extracted from it was analysed using Chi-Square Independent Test. The Thematic Analysis revealed two major themes which were traffic situation and driver compliance behaviour. Traffic situation is defined as traffic surrounding the driver, such as no car in front and behind, car in front, and car behind. Meanwhile, the Chi-Square Test result indicates that within the dilemma zone, there was a significant relationship between driver compliance behaviour and driver's distance from the stop line during yellow light interval. The closer the drivers were to the stop line, the more likely they were going to comply. In contrast, drivers showed higher non-compliant behavior when further away from stop line. This finding could help in the development of conceptual model of driving simulation with purpose in studying driver behavior.

• 대한교통학회지
• /
• 제33권2호
• /
• pp.159-169
• /
• 2015

과속운전으로 인한 교통사고는 치사율이 높고 그에 따른 많은 사회적비용의 지출이 따른다. 본 연구는 운전자의 행동특성이 과속의도와 과속행동에 미치는 영향과 그에 따른 인과관계를 밝히는데 목적을 두었다. 본 연구에서는 운전행동설문지로 활용되는 DBQ(Driver Behavior Questionnaire)를 이용하여 과속운전 의도와 행동에 영향을 미치는 운전자의 행태와 인적특성을 분석하고, 구조방정식 모형을 통하여 행동특성과 과속의도, 과속행동들간의 인과관계에 대하여 검증하였다. 이에 따른 가설을 검증하기 위하여 구조방정식 모형에 의한 경로분석을 실시한 결과, 과속의도에 영향을 미치는 DBQ의 속성은 Violation으로 나타났고, 과속의도는 과속행동에 영향을 미치는 것으로 나타났다. 연구결과를 바탕으로 선행연구들과 비교하여 논의하면, DBQ의 속성은 Violation, Mistake, Lapse 순으로 과속행동에 영향을 미치는 것으로 나타났다. 운전행동 척도인 DBQ의 세가지 속성 Lapse, Mistake, Violation이 과속행동에 유의한 영향을 미친다는 선행연구를 지지하여 DBQ를 활용한 운전행동분석 및 위험운전행동의 예측수단으로 활용될 수 있을 것으로 기대된다.

• International Journal of Computer Science & Network Security
• /
• 제22권11호
• /
• pp.163-171
• /
• 2022

A conceptual model represents an understanding of a system that is going to be developed, which in this research, a driving simulation software to study driver behavior at signalised intersections. Therefore, video observation was conducted to examine driver compliance behaviour within the dilemma zone at signalised intersection, pertaining to driver's distance from the stop line during yellow light interval. The video was analysed using Thematic Analysis and the data extracted from it was analysed using Chi-Square Independent Test. The Thematic Analysis revealed two major themes which were traffic situation and driver compliance behaviour. Traffic situation is defined as traffic surrounding the driver, such as no car in front and behind, car in front, and car behind. Meanwhile, the Chi-Square Test result indicates that within the dilemma zone, there was a significant relationship between driver compliance behaviour and driver's distance from the stop line during yellow light interval. The closer the drivers were to the stop line, the more likely they were going to comply. In contrast, drivers showed higher noncompliant behavior when further away from stop line. This finding could help us in the development of conceptual model of driving simulation with purpose of studying driver behavior.

• 한국산업융합학회 논문집
• /
• 제5권3호
• /
• pp.239-246
• /
• 2002

The propose of this study is to analysis driver's behavior of traveler information. This research made an attempt to explore driver's route change behavior in the en-route stage. Model were developed for each analysis with LIMDEP software which was developed by Willams H Greene. Commuters' transportation change in before trip stage are affected by their income, travel time, and incident information and constant of this model showed their reluctance of change mode. This was resulted from the inappropriateness of traffic information to general commuters which is the main target of traffic information.

• 한국컴퓨터정보학회논문지
• /
• 제26권11호
• /
• pp.173-182
• /
• 2021

본 논문에서는 운전자의 주의산만을 유발하는 운전자, 탑승자의 동작을 분석하고 핸드폰과 관련된 운전자의 행동 10가지를 인식하였다. 먼저 주의산만을 유발하는 동작을 환경 및 요인으로 분류하고 관련 최근 논문을 분석하였다. 분석된 논문을 기반으로 주의산만을 유발하는 주요 원인인 핸드폰과 관련된 10가지 운전자의 행동을 인식하였다. 약 10만 개의 이미지 데이터를 기반으로 실험을 진행하였다. SURF를 통해 특징을 추출하고 3가지 모델(CNN, ResNet-101, 개선된 ResNet-101)로 실험하였다. 개선된 ResNet-101 모델은 CNN보다 학습 오류와 검증 오류가 8.2배, 44.6배가량 줄어들었으며 평균적인 정밀도와 f1-score는 0.98로 높은 수준을 유지하였다. 또한 CAM(class activation maps)을 활용하여 딥러닝 모델이 운전자의 주의 분산 행동을 판단할 때, 핸드폰 객체와 위치를 결정적 원인으로 활용했는지 검토하였다.

• International Journal of Automotive Technology
• /
• 제7권2호
• /
• pp.145-154
• /
• 2006

In the companion paper, the composition and structure of the MATDYMO (Multi-Agent for Traffic Simulation with Vehicle Dynamic Model) were proposed. MATDYMO consists of the road management system, the vehicle motion control system, the driver management system, and the integration control system. Among these systems, the road management system and the integration control system were discussed In the companion paper. In this paper, the vehicle motion control system and the driver management system are discussed. The driver management system constructs the driver agent capable of having different driving styles ranging from slow and careful driving to fast and aggressive driving through the yielding index and passing index. According to these indices, the agents pass or yield their lane for other vehicles; the driver management system constructs the vehicle agents capable of representing the physical vehicle itself. A vehicle agent shows its behavior according to its dynamic characteristics. The vehicle agent contains the nonlinear subcomponents of engine, torque converter, automatic transmission, and wheels. The simulation is conducted for an interrupted flow model and its results are verified by comparison with the results from a commercial software, TRANSYT-7F. The interrupted flow model simulation is implemented for three cases. The first case analyzes the agents' behaviors in the interrupted flow model and it confirms that the agent's behavior could characterize the diversity of human behavior and vehicle well through every rule and communication frameworks. The second case analyzes the traffic signals changed at different intervals and as the acceleration rate changed. The third case analyzes the effects of the traffic signals and traffic volume. The results of these analyses showed that the change of the traffic state was closely related with the vehicle acceleration rate, traffic volume, and the traffic signal interval between intersections. These simulations confirmed that MATDYMO can represent the real traffic condition of the interrupted flow model. At the current stage of development, MATDYMO shows great promise and has significant implications on future traffic state forecasting research.

• 한국자동차공학회논문집
• /
• 제12권5호
• /
• pp.136-145
• /
• 2004

In companion paper, the composition and structure of the traffic environment is derived. Rules to regulate agent behaviors and the frameworks to communicate between the agents are proposed. In this paper, the model of a driver agent which controls a vehicle agent is constructed. The driver agent is capable of having different driving styles. That is, each driver agent has individual behavior settings of the yielding index and the passing index. The yielding index can be defined as how often the agent yields in case of lane changes, and the passing index can be defined as how often the agent passes ahead. According to these indices, the agents overtake or make their lanes for other vehicles. Similarly, the vehicle agents can have various vehicle dynamic models. According to their dynamic characteristics, the vehicle agent shows its own behavior. The vehicle model of the vehicle agent contains the nonlinear subcomponents of engine, torque converter, automatic transmission, and wheels. The simulation has proceeded for an interrupted flow model. The result has shown that it is possible to express the characteristics of each vehicle and its driver in a traffic flow, and that the change of the traffic state is closely related with the distance and the signal delay between intersections. The system developed in this paper shows the effectiveness and the practical usefulness of the traffic simulation.

• International Journal of Internet, Broadcasting and Communication
• /
• 제14권2호
• /
• pp.119-128
• /
• 2022

Recently, due to the development of related technologies for autonomous vehicles, driving work is changing more safely. However, the development of support technologies for level 5 full autonomous driving is still insufficient. That is, even in the case of an autonomous vehicle, the driver needs to drive through forward attention while driving. In this paper, we propose a method to monitor driving tasks by recognizing driver behavior. The proposed method uses pre-trained deep convolutional neural network models to recognize whether the driver's face or body has unnecessary movement. The use of pre-trained Deep Convolitional Neural Network (DCNN) models enables high accuracy in relatively short time, and has the advantage of overcoming limitations in collecting a small number of driver behavior learning data. The proposed method can be applied to an intelligent vehicle safety driving support system, such as driver drowsy driving detection and abnormal driving detection.

• 한국자동차공학회논문집
• /
• 제5권3호
• /
• pp.209-219
• /
• 1997

A fuzzy driver model based on a preview-predictor and yaw rate is developed. The model is used to investigate the handling performance of two wheel steering system(2WS) and four wheel steering system(4WS) vehicles. The two degree-of- freedom model which has yaw and lateral motion predicts the path of the vehicles. Based upon the yaw rate and lateral deviations, the fuzzy engine describes the human driver's complicated control behavior which is adjusted for the driving environment. Both typical single lane change maneuver and double lane change maneuver are adopted to demonstrate the feasibility of fuzzy driver model.