Driving Behavior Data Research Articles

Development and Comparative Analysis of Advanced Deep Learning Techniques for Crash Prediction in Advanced Driver Support Systems

Motor vehicle crashes claimed 38,800 lives and caused 4.4 million injuries in 2019 alone. Studies have shown that 94% of these crashes are because of driver errors. Such a huge contribution of driver errors to crashes indicates that efforts at improving safety should be directed toward both vehicles and drivers through advanced driver assistance systems (ADAS) and vehicular technologies. This study investigates the potential that real-time driver behavior data collected through vehicular technologies offer to predict crashes, as the first line of defense to avoid them. Three deep learning models were developed including multilayer perceptron neural networks (MLP-NN), long-short-term memory networks (LSTMN), and convolutional neural networks (CNN) using vehicle kinematics time series data extracted from the Second Strategic Highway Research Program Naturalistic Driving Study (SHRP2 NDS) dataset. The study builds on the hypothesis that crashes are preceded by turbulences that take place over time (turbulence horizon). If these turbulences are detected promptly they can help predict and avoid crashes. Several values were tested for the turbulence horizon and the prediction horizon (how long before the crash impact it can be predicted) to identify the optimal values. The results showed that the CNN model can predict all crashes with a 100% accuracy and zero false alarms 3 s before the crash impact time when a 6-s turbulence horizon is used. This outstanding performance demonstrates the developed model is a promising tool for implementation in ADAS.

Sound Effects on Physiological State and Behavior of Drivers in a Highway Tunnel.

Driving behavior in a highway tunnel could be affected by external environmental factors like light, traffic flow, and acoustic environments, significantly when these factors suddenly change at the moment before and after entering a tunnel. It will cause tremendous physiological pressure on drivers because of the reduction of information and the narrow environment. The risks in driving behavior will increase, making drivers more vulnerable than driving on the regular highways. This research focuses on the usually neglected acoustic environment and its effect on drivers' physiological state and driving behavior. Based on the SIMLAB driving simulation platform of a highway tunnel, 45 drivers participated in the experiment. Five different sound scenarios were tested: original highway tunnel sound and a mix of it with four other sounds (slow music, fast music, voice prompt, and siren, respectively). The subjects' physiological state and driving behavior data were collected through heart rate variability (HRV) and electroencephalography (EEG). Also, vehicle operational data, including vehicle speed, steering wheel angle, brake pedal depth, and accelerator pedal depth, were collected. The results indicated that different sound scenarios in the highway tunnel showed significant differences in vehicle speed (p = 0.000, η2 = 0.167) and steering wheel angle (p = 0.007, η2 = 0.126). At the same time, they had no significant difference in HRV and EEG indicators. According to the results, slow music was the best kind of sound related to driving comfort, while the siren sound produced the strongest driver reaction in terms of mental alertness and stress level. The voice-prompt sound most likely caused driver fatigue and overload, but it was the most effective sound affecting safety. The subjective opinion of the drivers indicated that the best sound scenario for the overall experience was slow music (63%), followed by fast music (21%), original highway tunnel sound environment (13%), and voice-prompt sound (3%). The findings of this study will be valuable in improving acoustic environment quality and driving safety in highway tunnels.

Individual Driver Crash Risk Classification Based on IoV Data and Offline Consumer Behavior Data

With the development of big data technologies, usage-based insurance (UBI) has received considerable attention from insurance companies. UBI products focus on identifying the relationship between the individual driver’s risk and online channel behavior variables from Internet of Vehicles (IoV) data. Although omnichannel information integration has promoted the development of many industries, it has not been used to improve the accuracy of driver risk classification models in insurance industries. This paper investigates the role of combining different channel variables in improving the classification of driver’s risk. Specifically, several models, including logistic regression and three different data mining techniques (neural networks, random forests, and support vector machines), augmented with driving behavior data based on the IoV and offline consumer behavior data collected from 4S (Sale, Spare part, Service, Survey) dealers, are applied to the classification model of risk. The empirical results show that the inclusion of online and offline channel data improves the different risk assessments; results also demonstrate the importance of offline consumer behavior variables in different models. These insights have important implications for insurance companies on UBI pricing strategy and cost management.

Influence of prompt timing and messages of an audio navigation system on driver behavior on an urban expressway with five exits

An audio navigation system is a very useful tool for driving path guidance with less distraction. However, the influence of the audio navigation system on driver behavior, especially in complicated road environments, is still not entirely clear. This study aims to investigate navigation prompt timing (NPT), navigation prompt message (NPM), and their combination in an audio navigation system on driving behavior on an urban expressway with five exits. Driving simulator technology was used to reproduce the experimental environment and obtain driver behavior data. Four indicators-speed, standard deviation of speed, absolute values of acceleration, and depth of accelerator-were selected to examine the main and interactive effects of NPT, NPM, and their combination based on repeated measures analysis of variance. The results show that the driver’s psychological state and operation of the vehicle on the urban expressway were affected by the prompt timing and messages of the audio navigation system. An interaction effect existed between prompt timing and prompt messages of the system, and this effect adjusted the effect on the driver’s psychological state and vehicle operation caused by these two important factors. This study contributes to our understanding of the influence of audio navigation systems on driver behavior in complex road environments and thus lays a foundation for developing standard audio navigation broadcast guidelines to improve drivers’ acceptance of navigation systems, reduce drivers’ cognitive workload, and improve the level of vehicle operation safety.

Deep driver behavior detection model based on human brain consolidated learning for shared autonomy systems

To achieve efficient shared autonomy, driver behavior detection (DBD) is undoubtedly required. This paper investigates a deep driver behavior detection (DDBD) model. To overcome the low accuracy of DBD due to a lack of driver behavior data, the similarity of some driver behavior characteristics, and the ignorance of multi-scale structure and texture information, a DDBD model based on human brain consolidated learning (HBCL) is proposed. First, multiple DBD models with input information of different scales are trained based on transfer learning. Then, a new model called consolidation training (CT) using the Mish is trained based on the weight data from the first step. Finally, a novel method for the visualization of the attention area is proposed. The experimental results demonstrate that the proposed model achieved the highest accuracy (94.72% on the Kaggle-driving test dataset), generalization and real-time performance, the attention area is more anthropomorphic as compared with existing state-of-the-art models.

Experimental Setup of Driver Behavior on Road Darkness Level


 
 
 Most of the previous research on the road darkness effect in Malaysia focused on traffic flow characteristics such as traffic speed and its density. Only limited efforts have been made to assess the road darkness effect on driver behaviours. In this paper, an experimental setup is explained to study driver behaviours when approaching parked or slow- moving vehicles during nighttime. A single-carriageway was selected to conduct the study. The road lighting will be controlled to be switched on or switched off during the experiment. Three obstacles in the shape of a motorcycle, car, and lorry will be used to simulate the parked or slow-moving vehicle. The experimental setup is designed to collect driver behaviour data including the reaction time taken for breaking or slowing down when approaching the obstacle, the distance between the car and the obstacle, the speed of the car and its trajectory. This experimental setup is expected to be used in a study that investigates the effect of dark roads on driver behavior.
 
 

Using single and multiple unmanned aerial vehicles for microscopic driver behaviour data collection at freeway interchange ramps

This paper presents a detailed methodological framework for collecting microscopic driver and vehicle behaviour data over a long road segment with an application to the entire stretch of a freeway ramp segment using single and multiple unmanned aerial vehicles (UAVs). The methodology allows users to collect reliable and complete trajectories of traffic movements in areas with challenging physical characteristics (long road segment, horizontal curvature, changing elevation, and presence of shadow), challenging traffic characteristics (high traffic volume, high speeds, and high-speed changes), and restrictive regulations (UAVs prohibited from hovering over the freeway or the right-of-way). Different UAV setups are recommended and can be used depending on the site conditions. Specific commercial software and procedures used to complete the data collection are explained. The methodology was applied at two ramps and verified with speed data acquired from differential GPS receivers using three different error metrics. The results showed good performance of the proposed methodology, including when aerial videos were taken from oblique angles.

Exploring the Different Patterns for Generation Process of Driving Fatigue Based on Individual Driving Behavior Parameters

Driving fatigue is one of the main causes of traffic accidents in monotonous environments such as grassland highways. However, the process of generation of driving fatigue on grassland highways is still not clear. A driving simulation experiment with 23 participants was performed to collect data on driving behavior, reaction time and electrocardiogram (ECG) results when driving on a grassland highway. The effective feature indicators of driving fatigue based on driving behavior data were calculated by Pearson correlation coefficient and principal component analysis method. The matter-element model based on entropy weight method was used to quantify the generation process of driving fatigue (GPDF). GPDF was classified as different patterns by the eigenvalue of GPDF curves. Reaction time and ECG data were utilized to verify the rationality of GPDF. Results show that there were 13 feature indicators of driving behavior suitable for driving fatigue description. GPDF was not completely consistent among different participants and was classified into three patterns (i.e., mild, moderate and severe fatigue). The mean similarity for GPDF in each pattern was 0.87, 0.61 and 0.50. Validation test demonstrated that driving fatigue detection accuracy by GPDF was 72%. The mean similarity of the GPDF between driving behavior and ECG was 0.72. Driving fatigue tended to occur with driving time of 19 min or 33 min. This study is helpful to understand GPDF on grassland highways from the perspective of individual driving behavior, which would provide suggestions for the reasonable setting of anti-fatigue devices.

Adaptations in driver deceleration behaviour with automatic incident detection: A naturalistic driving study

Traffic congestion and crash rates can be reduced by introducing variable speed limits (VSLs) and automatic incident detection (AID) systems. Previous findings based on loop detector measurements have revealed that drivers reduce their speeds while approaching traffic congestion when the AID system is active. Notwithstanding these behavioural effects, most microscopic traffic flow models assessing the impact of VSLs do not describe driver response accurately.This study analyses the main factors that influence driver deceleration behaviour while approaching traffic congestion with and without VSLs. The Dutch VSL database was linked to the driver behaviour data collected in the UDRIVE naturalistic driving study. Driver engagement in secondary tasks and glance behaviour were extracted from the video data. Linear mixed-effects models predicting the characteristics of deceleration events were estimated.The results show that the maximum deceleration is high when approaching a slower leader, when driving at high speeds and short distance headways, and close to the beginning of traffic congestion. The minimum time headway is short when driving at high speeds and changing lanes. Certain drivers showed higher decelerations and shorter time headways than others. Controlled for these main factors, smaller maximum decelerations were found when the VSLs were present and visible, and when the gantries were within close proximity. These factors could be incorporated into microscopic traffic simulations to evaluate the impact of AID systems on traffic congestion more realistically. Further research is needed to clarify the link between engagement in secondary tasks, glance behaviour and deceleration behaviour.

Research on The Model of UBI Car Insurance Rates Rating Based on CNN-Softmax Algorithm

With the support of vehicular networking technology, the rating of UBI auto insurance rate has a certain guiding significance for accurate pricing of the rate and personalized demand. Based on CNN-Softmax algorithm, a rating model of UBI auto insurance rate is proposed in this paper. The model first performs a series of operations such as convolution, pooling, and non-linear activation function mapping on CNN algorithm to enable it to extract features based on UBI customers’ driving behavior data, and then uses Softmax algorithm to classify customers according to their driving behaviors, thereby obtain UBI customer’s auto insurance rate rating. The empirical results show that compared with, CNN algorith, BP neural network algorithm and SVM algorithm, CNN-Softmax algorithm has a higher discrimination accuracy in the risk assessment of UBI customers’ driving behavior. And it is easy to implement in establishing rating model of UBI auto insurance rate. What’s more, it also has a good robustness, which can achieve better grade evaluations.

Research on Safety Evaluation of Commercial Vehicle Driving Behavior Based on Data Mining Technology

The arrival of big data era of internet of vehicles promotes the rapid development of logistics industry, which also indirectly leads to the high traffic accident rate, resulting in huge casualties and property losses. Driving behavior is considered the most central factor leading to traffic accidents. Therefore, a scientific and effective method for evaluating the safety of commercial vehicle driving behavior is urgently needed. In this study, a comprehensive evaluation model of driving behavior security based on multimembership function is proposed, and entropy weight method (EWM), analytic hierarchy process (AHP), and fuzzy comprehensive evaluation algorithm are integrated. Firstly, the evaluation system of commercial vehicle safety of driving behavior is established. Secondly, the weight vector of each evaluation index is determined by combining EW‐AHP to eliminate the subjectivity of the traditional AHP algorithm. Then, the fuzzy comprehensive evaluation matrix is calculated based on the multimembership function and fuzzy mathematics theory, and the quantitative evaluation of driving behavior safety is realized based on the matrix. Finally, the real road vehicle driving data and driving behavior data are verified by experiments. The experimental results show that the model can accurately and reasonably evaluate the safety of driving behavior, which is of great significance to improve road traffic safety.

A Vehicle Trajectory Extraction Method for Traffic Simulating Modeling

Vehicle driving behavior in urban traffic environment is interactive and complex. In order to simulate the driving behavior of real vehicles, it needs to rely on a large number of real vehicle trajectory data. However, making the simulated model more consistent with the real situation, a large vehicle trajectory dataset is needed. Therefore, this paper proposes a method to automatically extract accurate vehicle trajectory from video captured by UAV camera. The proposed computer vision method obtains the trajectory by detecting the vehicle and then tracking the trajectory, and calculates the relevant parameters of the vehicle, such as position, speed and direction. Therefore, the extracted data can be used as real driving behavior data for simulation modeling. The experimental results show that the proposed method is effective and reliable.

Implementation and Analytics of the Distributed Eco-Driving Simulation iCO2

We describe iCO2, a simulation platform for collecting driving behavior data. It is designed as the first massively multiplayer online game for mobile devices to practice eco-friendly driving. It facilitates the collection of large-scale data on driving behavior to better understand compliance and incentive mechanisms for eco-driving and users' preferences. We present the results of a campaign with iCO2 that used a game promoter to attract 2455 users. The results are described from three angles: (1) types of drivers are identified by clustering driving behavior; (2) types of players are identified by relating players' interaction with game elements and their driving behavior; (3) by looking at longer sessions, we demonstrate that players who show eco-unfriendly behavior at the beginning of the session improve their eco-driving behavior throughout their playtime.

An Acceptance Approach for Novel Technologies in Car Insurance

Background: Unlike other financial services, technology-driven changes in the insurance industry have not been a vastly explored topic in scholarly literature. Incumbent insurance companies have hitherto been holding their positions using the complexity of the product, heavy regulation, and gigantic balance sheets as paramount factors for a relatively slow digitalization and technological transformation. However, new technologies such as car telematic devices have been creating a new insurance ecosystem. The aim of this study is to assess the telematics technology acceptance for insurance purposes. Methods: The study is based on the Unified Theory of Acceptance and Use of Technology (UTAUT). By interviewing 502 new car buyers, we tested the factors that affect the potential usage of telematic devices for insurance purposes. Results: The results indicate that facilitating conditions are the main predictor of telematics use. Moreover, privacy concerns related to the potential abuse of driving behavior data play an important role in technology acceptance. Conclusions: Although novel insurance technologies are mainly presented as user-driven, users (drivers and insurance buyers) are often neglected as an active party in the development of such technologies.

Analyzing driver behavior under naturalistic driving conditions: A review

For a decade, researchers working in the area of road safety have started exploring the use of driving behavior data for a better understanding of the causes related to road accidents. A review of the literature reveals the excellent potential of naturalistic driving studies carried out by collecting vehicle performance data and driver behavior data during normal, impaired, and safety-critical situations. An in-depth understanding of driver behavior helps analyze and implement pre-crash safety measures - the development of enforcement policies, infrastructure design, and intelligent vehicle safety systems. The present paper attempts to review the naturalistic driving studies that have been undertaken so far. The paper begins with an overview of different methods for collecting unobtrusive driver behavior data during their day to day trip, followed by a discussion of various factors affecting driving behavior and their influence on vehicle performance parameters. The paper also discusses the strategies mentioned in the literature for improving driving behavior using naturalistic driving studies to enhance road safety. Some of the major findings of this review suggest that i) driver behavior is a major cause in the majority of the road accidents ii) drivers generally reduce their speed and increases headway as a compensatory measure to reduce the workload imposed during distracting activity and adverse weather conditions iii) mobile phone has emerged as a potential device for collecting naturalistic driving data and, iv) improvement in driving behavior can be achieved by providing feedback to the drivers about their driving behavior. This can be done by implementing usage-based insurance schemes such as pay as you drive (PAYD), pay how you drive (PHYD), and manage how you drive (MHYD). While a considerable amount of research has been done to analyze driving behavior under naturalistic conditions, some areas which are yet to be explored are highlighted in the present paper.

Multitask Learning Assisted Driver Identity Authentication and Driving Behavior Evaluation

The industrial Internet of Things has become the new driving force for the automobile industry, making people's travel increasingly convenient. However, there are still a multitude of challenges that need to be tackled, including but not limited to illegal driver detection, legal driver identification, and driving behavior evaluation. At present, many researchers have attempted to solve issues of illegal driver detection and legal driver identification by using deep learning network, but there are still quite a few limitations in the collection and analysis of driving behavior data. Moreover, the problem of driving behavior evaluation has been paid little attention. Therefore, in this article we conduct a comprehensive study on driving behavior habits and establish a multitask learning (MTL) network to solve the abovementioned problems. First, we collect original data from a real vehicle and extract the driving behavior characteristics. Then, a novel MTL network composed of long short-term memory network, support vector domain description model and feedforward neural network is established, which achieves illegal driver detection, legal driver identification, and driving behavior evaluation. Extensive experiments illustrate that the proposed MTL network not only supports parallel learning to reduce time and space costs, but also has excellent performances and robustness for the three tasks.

Assessing the impact of multi-dimensional driving behaviors on link-level emissions based on a Portable Emission Measurement System (PEMS)

Eco-driving is designed to reduce fuel consumption and emissions by the improvement of driving behaviors. The objective of this work is to propose some eco-driving behavior suggestions by analyzing the impact of driving behaviors on vehicle emissions. A Portable Emission Measurement System (PEMS) was used to collect emissions and microscopic driving behavior data. Then the multi-dimensional characteristics of driving behaviors and corresponding link-level emission characteristics were quantified. The results show that the vehicle activities change significantly when the road traffic tends to be saturated or forced. The change of driving behavior and associated vehicle activities leads to the significant increase of emission factor in saturated flow and forced flow. Traditional eco-driving advice advocates using the highest gear as much as possible to reduce fuel consumption. But the results indicate that using the highest gear can achieve a savings in CO2 while at a cost in NOX. If the time percentage of the highest gear increase by 54%, the CO2 emission factors will reduce by 22% while NOX will increase by 14%. Then a total of six factors were extracted to characterize driving behaviors. The statistical analysis shows that not only the driving operational intensity have an impact on the emissions factors but also the durations and frequencies of individual maneuver states will affect emissions. Decision tree algorithm was used to identify eco-driving behaviors. Finally, a model with 97% accuracy was put forward. The conclusions are helpful to improve traditional eco-driving strategies.

Development of Driver-Behavior Model Based onWOA-RBM Deep Learning Network

Human drivers’ behavior, which is very difficult to model, is a very complicated stochastic system. To characterize a high-accuracy driver behavior model under different roadway geometries, the paper proposes a new algorithm of driver behavior model based on the whale optimization algorithm-restricted Boltzmann machine (WOA-RBM) method. This method establishes an objective optimization function first, which contains the training of RBM deep learning network based on the real driver behavior data. Second, the optimal training parameters of the restricted Boltzmann machine (RBM) can be obtained through the whale optimization algorithm. Finally, the well-trained model can be used to represent the human drivers’ operation effectively. The MATLAB simulation results showed that the driver model can achieve an accuracy of 90%.

Using a multi-user driving simulator system to explore the patterns of vehicle fleet rear-end collisions occurrence under different foggy conditions and speed limits

Vehicle fleet rear-end collisions (FRECs) are an extremely fatal type of traffic collisions on freeway and they usually occur in foggy weather. This study aimed to explore the patterns of vehicle fleet rear-end collisions occurrence under different foggy conditions and speed limits on freeway. A multi-user driving simulator system was used to conduct the experiment and the driving behavior data were collected from eight participants. The experimental results showed that as the fog density increased, the length of vehicle fleet decreased significantly, and drivers tended to keep a more stable car-following distance. The fog weather and short vehicle gap prompted drivers to react faster and brake harder in respond to the leading vehicle’s brake. In spite of the compensational behaviors, more FRECs were observed under heavy fog condition. Lowering speed limit can significantly reduce the FRECs under foggy conditions. As the speed limits reduced, drivers’ brake response time and speed variance significantly reduced. The study also found that drivers’ brake response time was negatively correlated with their positions in the fleet. Drivers in the front positions of the fleet had a longer response time than drivers in the back positions and thus were more likely to encounter collisions. The study generated a better understanding of drivers’ behavioral pattern in a vehicle fleet and the patterns of vehicle fleet rear-end collisions occurrence. The findings also shed lights on the design of driver assistance system for complex driving situations such as freeway driving under adverse weather.

Drug use and driving behaviors among drivers with and without alcohol-related infractions.

Introduction Brazil is one of the countries with the highest rates of alcohol-related traffic infractions, but little is known about the profile of the drivers who commit them. Identifying the characteristics of impaired drivers is essential for planning preventive actions.Objective To compare drug use and driving behavior profiles of drivers with and without alcohol-related infractions.Methods 178 drivers stopped at routine roadblocks were assessed by traffic agents who conducted standard roadblock procedures (document verification; request of a breathalyzer test [BT]). Drug use and driving behavior data were collected through semi-structured interviews. Subjects were divided into three groups: drivers who refused the BT (RDs, n = 72), drivers who tested positive on the BT (PDs, n = 34), and drivers who had committed other infractions (ODs, n = 72).Results The proportion of alcohol use in the last year was higher among RDs (100%) than in the PD and OD groups (97.1% and 72.2% respectively, p < 0.001). Lifetime prevalence of cannabis and cocaine use for the overall sample was 44.3% and 18.2%, respectively. Fewer individuals in the OD group (31.5%) reported having been stopped at roadblocks in the previous year compared to the PDs (55.9%) and RDs (48.6%, p = 0.03). However, a higher proportion of RDs reported drunk driving in the same period (87.5%; PD 69.7%; OD 26.9%; p < 0.001).Conclusion Essential differences among groups were observed. RDs had a higher proportion of alcohol use and drunk driving in the previous year; drivers who fit into this particular group may be unresponsive or less responsive to social deterrence and enforcement actions.