Risky Driving Behavior Detection Using In-Vehicle WiFi Signals

When any distracted driving of the driver is concerned, use of smartphones and navigation systems while driving are major causes for traffic accidents. However, in a situation where navigation systems are commonly installed and smartphones are commonly used within vehicles, it is better to implement technologies that control the use of smartphones in high-risk situations and allow brief periods of use otherwise, rather than ban the use of smartphones altogether while driving. This study introduces the DWMS (Driver Workload Management System) and proposes service that controls the use of smartphones according to driver's workloads. The DWMS periodically monitors the driver's behaviors and smartphone usages in order to calculate driving workload. When the calculated workload exceeds the threshold value, the driver is deemed to be overburdened and all incoming calls and SMS (Short Message Service) is blocked. The service will then make a notification when this status is relieved. In addition, at the end of driving, the DWMS will provide a summary of the information on how the driver drives, workload, and behaviors taken as a text message to the driver's smartphone. The result of this study can increase both driving safety and convenience for drivers.

In this paper we investigate driver behavior variables proposed in literature from an objective approach; we consider studies which employed numeric data that can be objectively measured by standardized tools and electronic devices, such as speed, acceleration and position. Our analysis develops in three phases; first we identify recent fields that focused on modeling driver behavior, such as Usage based insurance and governement agencies of transport. Next we extract quantitative variables from literature and classify them according to their importance; to achieve this, we identify their priority in each paper depending on their results. Finally we order them by their rate use in scientific literature, in order to have a comprehensive synthesis about the most significant and prevalent variables in literature. This paper therefore serves as reference for future studies on driver behavior; the collaboration we provide is a set of crucial variables for modeling, understanding and analysis driver behavior in quantitative research. Results show that “Speed”, “Acceleration/deceleration” and “Braking” are the most used in recent studies as well as considered priority variables.

The use of driver behavior has been considered a complex way to solve road safety complications. Car drivers are usually involved in various risky driving factors which lead to accidents where people are fatally or seriously injured. The present study aims to dissect and rank the significant driver behavior factors related to road safety by applying an integrated multi-criteria decision-making (MCDM) model, which is structured as a hierarchy with at least one 5 × 5 (or bigger) pairwise comparison matrix (PCM). A real-world, complex decision-making problem was selected to evaluate the possible application of the proposed model (driver behavior preferences related to road safety problems). The application of the analytic hierarchy process (AHP) alone, by precluding layman participants, might cause a loss of reliable information in the case of the decision-making systems with big PCMs. Evading this tricky issue, we used the Best Worst Method (BWM) to make the layman’s evaluator task easier and timesaving. Therefore, the AHP-BWM model was found to be a suitable integration to evaluate risky driver behavior factors within a designed three-level hierarchical structure. The model results found the most significant driver behavior factors that influence road safety for each level, based on evaluator responses on the driver behavior questionnaire (DBQ). Moreover, the output vector of weights in the integrated model is more consistent, with results for 5 × 5 PCMs or bigger. The proposed AHP-BWM model can be used for PCMs with scientific data organized by traditional means.

Driver profiling – Data-based identification of driver behavior dimensions and affecting driver characteristics for multi-agent traffic simulation

운전자 사고와 교통 위반의 원인은 행동심리학적 요소에 의해 영향을 받고 있기 때문에, 운전자와 보행자의 행태 및 심리학적 연계성을 파악하고 세부적인 분석을 통해 개선점을 탐색하는 서비스디자인의 접근방법이 점차 중요해지고 있다. 서비스디자인은 눈에 보이지 않는 무형의 요소까지 통합하여 가시화하고 고객이 보다 높은 가치로 느낄 수 있게 하는 서비스 중심의 디자인 개발 방법론을 의미한다. 이에 따라 본 연구에서는 스쿨존 도로환경에 대한 근본적인 문제점들을 발견하여 서비스디자인 관점에서 해결하고, 디자인 사고를 통한 행동유도 시설물 및 쾌적한 도로환경 구축을 위한 연구를 시행하였다. 먼저 국내외 도로 현황조사 및 행동심리 분석을 통한 문제점을 도출하고 교통사고 발생원인 및 대책 등의 해외개선 사례를 조사 분석하여 스쿨존 도로환경을 개선하기 위한 아이디어를 도출하였다. 이중 차량의 속도를 줄이기 위한 행동 유도 방법이 가장 중요한 것으로 파악되었으며, 이에 대한 해결방법으로 로봇 신호등을 활용하여 차선의 폭을 가변적으로 줄일 수 있도록 함으로써 차량의 속도를 줄이는 것과 동시에 학생들이 횡단보도를 쉽게 건너갈 수 있도록 하였다. 이러한 스쿨존 환경을 조성하기 위해서는 차량의 흐름을 유도하기 위한 로봇 신호등, 횡단보도의 그래픽과 공간디자인에 대한 종합적인 접근방법이 유효하였고, 펜스와 도로 환경의 공공디자인을 통합하여 스쿨존의 새로운 서비스디자인을 제안할 수 있었다. The cause of traffic accidents and driver behavior is affected by psychological factors, because the behavior of drivers and pedestrians, and to identify the psychological linkage analysis to explore improvements in service design approaches are becoming increasingly important. Service design is a service-oriented development methodology that make invisible and intangible elements into visible and higher-value. Accordingly, through design thinking to action for development of products and a pleasant road environment were studied for school zone by discovering underlying issues to resolve in terms of service design. First, survey of domestic road psychological and behavioral problems were derived through the analysis of the causes of traffic accidents and researched the concept of school zones by analyzing the ideas. To reduced the width of the lane manipulate the speed of the vehicle and to cross the crosswalk safely. In order to encourage the flow of vehicle traffic, new solutions are proposed by offering robot traffic lights, graphic design of crosswalk, space, and the fence and convergence public design and services.

This paper describes a practicing method for enhancing driver performance, which it bases on cognitive driver behavior model of ACT-R--a dominant theory of human cognitive architecture. First, the paper elaborates the idea of driver behavior modeling in cognitive architecture. And then, it dedicates the implementation of driver's practice courses based on the cognitive architecture model. Finally, by comparing the experiments back and forth of the practice, the results indicate that the practice based on driver cognitive behavior model can improve the driver performances positively. The experiences of the practice also implied that both practice method is indeed an approach worthy of pursuit for enhancing driver performance at driving emergencies and driver behavior modeling in cognitive architecture is a substantial tool for practical applications in predicting and instructing driver behaviors.

Understanding of mechanisms and factors responsible for the driver behavior on the road is the subject of ongoing interest to transportation psychologists, occupational doctors and engineers. Models of driver behavior are a key point for the understanding the mechanisms and factors which may cause limitations to the optimal functioning on the road. They also systematize knowledge about the factors responsible for the behavior of the driver and thus constitute a starting point for formulating empirical or diagnostic hypotheses. The aim of this study is to present models of driver behavior from the descriptive and functional perspectives. Med Pr 2017;68(3):401-411.

Multi-scale driver behavior modeling based on deep spatial-temporal representation for intelligent vehicles

Inverse Reinforcement Learning Based Stochastic Driver Behavior Learning

Two factor authentication offers significant benefits but has had limited uptake as users find the extra step required at each authentication attempt (e.g., to enter a verification code) to be onerous. In this paper, we present a novel twofactor authentication system called Wi-Access that requires no additional effort from the user beyond typing her credentials. The first factor is established by the knowledge of the password. Wi-Access uses the unique perturbations in the ambient WiFi signals caused by the hand and finger motions of the user during password entry as a way to confirm inherence (i.e., something that the user is) as the second factor for authentication. We show that the way in which Alice moves her hands and fingers while typing her password is uniquely manifested in the Channel State Information (CSI) of ubiquitous WiFi signals. We also show that if Bob types Alice's password, then the corresponding CSI signals are significantly distinct to Alice. Wi-Access uses Principle Component Analysis to reduce the dimensionality of the CSI data. A variance-based thresholding algorithm is used to extract the CSI signals that precisely correspond to the typing action. A one-class classifier based on an autoencoder that uses specific thresholds tailored to each user's training data is used to make the authentication decision. We implement Wi-Access using offthe-shelf 802.11n devices and evaluate its performance with 10 subjects. Our tests reveal that Wi-Access can on average achieve 92% authentication accuracy with 8% false positives. Moreover, Wi-Access is robust in preventing attacks with a 91.8% attack detection accuracy.

In order to enhance the traffic road safety, applying the artificial intelligence theory and the technology in road traffic control system, that is intelligent driving vehicle, may effectively decrease traffic accidents, which are caused by drivers' neglect, fatigue and so on. This article is based on cognitive science to analyze driving behaviors. First, we estimated the Information Processing Model of Intelligent Vehicle. Second, based on prototype pattern recognition theory, we also built Pattern Recognition Model of Intelligent Driving Behavior, which provided the advantageous basis for the further research on intelligent vehicle. Third, virtual scene simulation system was established by using Creator and Vega Prime. Finally, according to test data, we investigated the features of speed and latitudinal distance in both artificial and intelligent U-Turn driving behavior. Our results describe that the intelligent vehicle does have the ability of cognizance, and it can make correct reactions corresponding to the driving environment, but its intelligence still needs improving.

The high impact of traffic accidents makes it imperative to formulate public policies to reduce their occurrence. In this task, knowing the cause of accidents is of paramount importance. The use of data mining and big data adapts to the complexity of the phenomenon under study. In order to classify some possible causes of traffic accidents, we built a data model to describe the behavior and dynamic of the participant agents in the traffic accident event in Colombia. This paper presents the application of MLP and Naive Bayes algorithms to identify the possible immediate cause and the rules decision algorithm PART for the root cause of traffic accidents. Models have been tested aiming to obtain the goodness of fit by increasing metrics like Recall, Precision, ROC and Kappa index, and minimize the RMSE.

Objective Before market introduction, the safety of highly automated driving systems needs to be assessed prospectively. BMW has developed a holistic approach for the assessment of the traffic safety impact by these systems in which stochastic traffic simulations play a significant role. A driver behavior model which represents realistic driver behavior ranging from performance in non-critical everyday driving toward performance in critical situations is key for this approach. To ensure trustworthy results, validation of the driver model is needed. The paper aims at demonstrating that the presented driver model acts realistically in different critical real-world traffic scenarios. Methods BMW has been developing the Stochastic Cognitive Model (SCM) which models cognitive processes in traffic situations. These processes range from information acquisition by gaze behavior, mental representation of the environment, recognition of situations from the visual information and reaction to the situation. The driver model combines these cognitive processes with stochastic driver parameters to obtain a variation in driver behavior in simulations. Especially visual attention modeling is key to realistic traffic interactions in simulations as this is the input for the sequential cognitive processes, i.e., the recognition of situations and the reaction to the situation. Modeling of driver’s gaze behavior with SCM is thus shown in this paper. Results SCM is applied in three critical real-world traffic scenarios in which gaze behavior, brake reaction times and time-to-collisions are evaluated and compared to the real-world data. Due to the stochastic approach not only a single SCM agent but a collective of virtual SCM test drivers is assessed. Results show that SCM is capable to simulate the influence of visual inattention on collision risk. Conclusion Realistic driver behavior in simulations can be achieved by using SCM. Especially in the presented critical scenarios SCM is able to represent real-world driving behavior which is determined particularly by its gaze behavior and subsequent reaction. Driving performance varies over different SCM agents which mean that different driving behavior can be simulated with SCM as well. However, the investigation in this paper included only three real-world cases. Therefore, further critical, and additionally non-critical scenarios need to be investigated in the future.

Adapting ISA system warnings to enhance user acceptance

Fatigue driving is one of the primary causes of traffic accidents, posing a serious threat not only to the driver’s life and property but also to the safety of surrounding vehicles and pedestrians. Therefore, accurate and efficient detection of driver fatigue is essential for preventing traffic incidents. This study proposes a novel driver fatigue detection framework that integrates AlphaPose* with a Long Short-Term Memory (LSTM) network. To enhance the performance of AlphaPose, the original human detector is replaced with an optimized YOLOv11n model, which incorporates a Hybrid Pooling Fusion Block (HPFB) to improve feature representation and meet the requirements of keypoint estimation. Furthermore, a multi-view data processing strategy based on the Driving Monitoring Dataset (DMD) is introduced to capture driver behavior from frontal, lateral, and hand views. To effectively model the temporal dynamics of driver behavior, a fatigue behavior monitoring network is designed using LSTM. Comparative experiments demonstrate that the proposed AlphaPose*-LSTM-based system achieves superior performance in fatigue detection tasks compared to existing state-of-the-art approaches.