Combined impact of road and traffic characteristic on driver behavior using smartphone sensor data

Analysis of mobile phone use engagement during naturalistic driving through explainable imbalanced machine learning

While driver distraction remains an issue in modernized societies, technological advancements in data collection, storage and analysis provide the means for deeper insights of this complex phenomenon. In this research, factors influencing when driver distraction through mobile phone use occurs during naturalistic driving are investigated. Naturalistic data from a 6-stage, 230-driver experiment are exploited, in which drivers installed a non-intrusive driving recording application in their devices and conducted their trips normally across a 21-month timespan, coupled with corresponding questionnaire data. The various experiment stages involved providing progressively more behavioral feedback to drivers while continuing to record them. Subsequently, supervised Machine Learning XGBoost algorithms were employed to model the contributions of naturalistic driving and questionnaire features to the decision to engage mobile phone use. Mobile phone use percentages were heavily skewed towards zero, therefore imbalanced ML with a minority-oversampling approach in a binary format was employed. To increase the explainability offered by the algorithm, SHAP values were calculated for the informative features. Results indicate that the decision of drivers to use a mobile while driving is governed by a number of complex, non-linear relationships. Total trip distance is the most significant predictor variable by a wide margin, with mean SHAP values of 0.79 towards affecting the model decisions for the probability of mobile phone use of each driver. However, other variables influence the final predictions as well, such as the number of tickets in the last three years (m.SHAP = 0.30), declared mobile phone use (m.SHAP = 0.26), the amount and variety of provided feedback (m.SHAP = 0.17) (i.e. experiment phase number) and family member numbers (m.SHAP = 0.09) decrease the probability of using a mobile phone while driving. Conversely, increases in driver experience (m.SHAP = 0.22), driver age (m.SHAP = 0.11), engine capacity (m.SHAP = 0.11) and total kilometers driven annually (m.SHAP = 0.08) increase the probability of using a mobile phone in naturalistic driving conditions. SHAP dependency plots reveal non-linear effects present in almost all variables. Fuel consumption had a particularly strong non-linear effect, as higher values of this variable lead to both higher and lower probability of drivers using a mobile phone, deviating from the safer average. Legislation, campaigns and enforcement measures can be restructured to take advantage of gains margins in terms of understanding and predicting driver distraction behavior, as explored in the present study.

Investigating the impact of environmental and temporal features on mobile phone distracted driving behavior using phone use data

Mobile phone distracted driving (MPDD) is one of the most significant and common factors in distraction-affected crashes. In previous studies, MPDD has been described as a self-selected behavior that affects driving performance, rather than a multidimensionally impacted behavior. In this study, the researchers hypothesized that external environmental features significantly impacted MPDD and tested this hypothesis by structural equation modeling (SEM). Three external latent variables (road, operation, and control factors) were measured at different times during weekdays in urban areas of Texas by integrating a large number of mobile phone sensor data and roadway inventory data. A structural model was developed to test the relationship between the latent variables and the rate of drivers involved in MPDD (MPDDR) on the roadway during different time periods. Finally, the data summary and model results revealed significant temporal effects. Standardized estimates from the SEM results revealed the positive impact of roads factors in the morning peak that broader shoulders, wider medians, and smaller curve radians were correlated with higher MPDDR in the morning peak hours; the negative impact of operation factors that higher average annual daily truck traffic (truck AADT) were associated with lower MPDDR significantly. And the impact of control factors on MPDDR is positive. In other words, the road segments with a large number of traffic signals in urban areas had a higher MPDDR than those without traffic signals. These findings could assist transportation and legislation agencies in the development of appropriate countermeasures or enforcement tactics and implement them effectively to reduce the occurrence of MPDD. In addition, this study provides a novel perspective close to the actual consideration of drivers about using mobile phones while driving, in the context of MPDD research, rather than comparing driver groups and vehicle performance.

Correlation Analysis of In-Vehicle Sensors Data and Driver Signals in Identifying Driving and Driver Behaviors

Today's cars have dozens of sensors to monitor vehicle performance through different systems, most of which communicate via vehicular networks (CAN). Many of these sensors can be used for applications other than the original ones, such as improving the driver experience or creating new safety tools. An example is monitoring variables that describe the driver's behavior. Interactions with the pedals, speed, and steering wheel, among other signals, carry driving characteristics. However, not always all variables related to these interactions are available in all vehicles; for example, the excursion of the brake pedal. Using an acquisition module, data from the in-vehicle sensors were obtained from the CAN bus, the brake pedal (externally instrumented), and the driver's signals (instrumented with an inertial sensor and electromyography of their leg), to observe the driver and car information and evaluate the correlation hypothesis between these data, as well as the importance of the brake pedal signal not usually available in all car models. Different sets of sensors were evaluated to analyze the performance of three classifiers when analyzing the driver's driving mode. It was found that there are superior results in classifying identity or behavior when driver signals are included. When the vehicle and driver attributes were used, hits above 0.93 were obtained in the identification of behavior and 0.96 in the identification of the driver; without driver signals, accuracy was more significant than 0.80 in identifying behavior. The results show a good correlation between vehicle data and data obtained from the driver, suggesting that further studies may be promising to improve the accuracy of rates based exclusively on vehicle characteristics, both for behavior identification and driver identification, thus allowing practical applications in embedded systems for local signaling and/or storing information about the driving mode, which is important for logistics companies.

Bayesian spatial correlation, heterogeneity and spillover effect modeling for speed mean and variance on urban road networks

Analyzing speed mean and variance is vital to safety management in urban roadway networks. However, modeling speed mean and variance on structured roads could be influenced by the spatial effects, which are rarely addressed in the existing studies. The inadequacy may lead to biased conclusions when considering vehicle speed as a surrogate safety measure. The current study focuses on developing a Bayesian modeling approach with three types of spatial effects, i.e., spatial correlation, spatial heterogeneity, and spillover effect. To capture the spatial correlation, the study employs the intrinsic conditional autoregressive (ICAR) models, spatial lag models (SLM), and spatial error models (SEM). Spatial heterogeneity and spillover effect are considered by the random parameters approach and spatially lagged covariates (SLCs). Speed data are collected from the float cars running on 134 urban arterials in Chengdu, China. The results indicate that the random parameters ICAR model with SLCs (RPICAR-SLC) outperforms others in terms of goodness-of-fit, accuracy, and efficiency for modeling speed mean, while the random parameters ICAR model (RPICAR) is the best for modeling speed variance. Moreover, RPICAR-SLC and RPICAR models are beneficial to address spatial correlation of residuals, explaining the unobserved influence among the observations, and are less likely to cause biased or overestimated parameters. The study also discusses how traffic conditions, road characteristics, traffic management strategies, and facilities on roadway networks influence speed mean and variance. The findings highlight the importance of multi-type spatial effects on modeling speed mean and variance along the structured roadways.

Estimating mountainous freeway crash rate: Application of a spatial model with three-dimensional (3D) alignment parameters

The road alignment is a three-dimensional (3D) curve in nature. In this study, we quantitatively examine the effect of 3D road alignment on traffic safety on mountainous freeways. Geometric parameters of 3D curvature and torsion in mathematics are derived to characterize the 3D road curve. Based on the coordination of different horizontal and vertical elements, 3D road alignment is divided into twelve types of combined alignment. For each alignment combination, the 3D curvature and torsion are calculated according to the differential geometry theory. Regarding crash statistical modeling, the Bayesian spatial Tobit (BST) model is developed to accommodate possible spatial correlation of traffic crashes among adjacent freeway segments. The Bayesian Tobit (BT) model is also built for comparison. A 118-km mountainous freeway associated road geometric features, traffic volume with three years of crash data is used as a case study. The result from the model comparison shows the BST model outperforms the BT model in terms of goodness-of-fit. Parameter estimation result for the BST model shows that the differences of average 3D curvature (and torsion) between adjacent segments have statistically significant effects on the crash rate of the segment, indicating it is necessary to consider three-dimensional alignment parameters in estimating mountainous freeway crash rate. Moreover, by comparing the predicted crash rate calculated by the BST model and the observed crash rate, the result shows the proposed BST model can provide a reliable prediction for freeway crash rates of different combined alignments. This study provides new insight on the effect of road geometric design on traffic safety but also deepens our understanding of spatial correlations in freeway crash modeling.

Evaluation of 1D and 2D Deep Convolutional Neural Networks for Driving Event Recognition

Driving event detection and driver behavior recognition have been widely explored for many purposes, including detecting distractions, classifying driver actions, detecting kidnappings, pricing vehicle insurance, evaluating eco-driving, and managing shared and leased vehicles. Some systems can recognize the main driving events (e.g., accelerating, braking, and turning) by using in-vehicle devices, such as inertial measurement unit (IMU) sensors. In general, feature extraction is a commonly used technique to obtain robust and meaningful information from the sensor signals to guarantee the effectiveness of the subsequent classification algorithm. However, a general assessment of deep neural networks merits further investigation, particularly regarding end-to-end models based on Convolutional Neural Networks (CNNs), which combine two components, namely feature extraction and the classification parts. This paper primarily explores supervised deep-learning models based on 1D and 2D CNNs to classify driving events from the signals of linear acceleration and angular velocity obtained with the IMU sensors of a smartphone placed in the instrument panel of the vehicle. Aggressive and non-aggressive behaviors can be recognized by monitoring driving events, such as accelerating, braking, lane changing, and turning. The experimental results obtained are promising since the best classification model achieved accuracy values of up to 82.40%, and macro- and micro-average F1 scores, respectively, equal to 75.36% and 82.40%, thus, demonstrating high performance in the classification of driving events.

A Proactive Recognition System for Detecting Commercial Vehicle Driver’s Distracted Behavior

Road traffic accidents regarding commercial vehicles have been demonstrated as an important culprit restricting the steady development of the social economy, which are closely related to the distracted behavior of drivers. However, the existing driver’s distracted behavior surveillance systems for monitoring and preventing the distracted behavior of drivers still have some shortcomings such as fewer recognition objects and scenarios. This study aims to provide a more comprehensive methodological framework to demonstrate the significance of enlarging the recognition objects, scenarios and types of the existing driver’s distracted behavior recognition systems. The driver’s posture characteristics were primarily analyzed to provide the basis of the subsequent modeling. Five CNN sub-models were established for different posture categories and to improve the efficiency of recognition, accompanied by a holistic multi-cascaded CNN framework. To suggest the best model, image data sets of commercial vehicle driver postures including 117,410 daytime images and 60,480 night images were trained and tested. The findings demonstrate that compared to the non-cascaded models, both daytime and night cascaded models show better performance. Besides, the night models exhibit worse accuracy and better speed relative to their daytime model counterparts for both non-cascaded and cascaded models. This study could be used to develop countermeasures to improve driver safety and provide helpful information for the design of the driver’s real-time monitoring and warning system as well as the automatic driving system. Future research could be implemented to combine the vehicle state parameters with the driver’s microscopic behavior to establish a more comprehensive proactive surveillance system.

Quantifying the effect of roadway, driver, vehicle, and location characteristics on the frequency of longitudinal and lateral accelerations

The purpose of this study is to understand and quantify the simultaneous effects of roadway speed category, driver age, driver gender, vehicle class, and location on the rates of longitudinal and lateral acceleration epochs. The rate of usual as well as harsh acceleration epochs are used to extract insights on driving risk and driver comfort preferences. However, an analysis of acceleration rates at multiple thresholds incorporating various effects while using a large-scale and diverse dataset is missing. This analysis will fill this research gap. Data from the 2nd Strategic Highway Research Program Naturalistic Driving Study (SHRP2 NDS) was used for this analysis. The rate of occurrence of acceleration epochs was modeled using negative binomial distribution based generalized linear mixed effect models. Roadway speed category, driver age, driver gender, vehicle class, and location were used as the fixed effects and the driver identifier was used as the random effect. Incidence rate ratios were then calculated to compare subcategories of each fixed effect. Roadway speed category has the strongest effect on longitudinal and lateral accelerations of all magnitudes. Acceleration epoch rates consistently decrease as the roadway speed category increases. The difference in the rates depends on the threshold and is up to three orders of magnitude. Driver age is another significant factor with clear trends for longitudinal and lateral acceleration epochs. Younger and older drivers experience higher rates of longitudinal accelerations and decelerations. However, the rate of lateral accelerations consistently decreases with age. Vehicle class also has a significant effect on the rate of harsh accelerations with minivans consistently experiencing lower rates.

Profiling drivers to assess safe and eco-driving behavior - A systematic review of naturalistic driving studies

Road accidents and vehicular emissions are two significant issues related to road transportation, affecting both human life and the environment. Prior research suggests that driver behavior is a crucial factor in the majority of road crashes and is a significant factor influencing fuel consumption and vehicle emission. Significant improvement in driving behavior can be achieved by providing feedback to drivers about their driving behavior. An increasing interest among researchers to identify risky and non-economical driving maneuvers has led to the development of driver behavior profiling, i.e., rating/categorizing drivers into different categories based on how they drive. To get an insight into different parameters and methodology adopted by researchers for categorizing drivers into different categories, this paper presents a systematic review of studies on driver behavior profiling. In the present paper, PRISMA approach was adopted to shortlist the most relevant studies for systematic review out of 1231 initial studies, which were extracted using the relevant keywords. The findings from our study suggest that the selection of parameters for profiling the driver will depend on the application of the profiling scheme, type of device used for extracting data, and importance of parameter in rating criteria. Further, the findings suggest that significant improvement in driving behavior can be achieved by providing feedback to the drivers about their driving behavior and by implementing usage-based insurance schemes. It is also suggested that future studies shall focus on using smartphone devices for the collection of driver data as smartphones are nowadays easily accessible to everyone.

Identifying the impact of the COVID-19 pandemic on driving behavior using naturalistic driving data and time series forecasting

INTRODUCTION

COVID-19 has disrupted daily life and societal flow globally since December 2019; it introduced measures such as lockdown and suspension of all non-essential movements. As a result, driving activity was also significantly affected. Still, to-date, a quantitative assessment of the effect of COVID-19 on driving behavior during the lockdown is yet to be provided. This gap forms the motivation for this paper, which aims at comparing observed values concerning three indicators (average speed, speeding, and harsh braking), with forecasts based on their corresponding observations before the lockdown in Greece.

METHOD

Time series of the three indicators were extracted using a specially developed smartphone application and transmitted to a back-end platform between 01/01/2020 and 09/05/2020, a time period containing normal operations, COVID-19 spreading, and the full lockdown period in Greece. Based on the collected data, XGBoost was employed to identify the most influential COVID-19 indicators, and Seasonal AutoRegressive Integrated Moving Average (SARIMA) models were developed for obtaining forecasts on driving behavior.

RESULTS

Results revealed the intensity of the impact of COVID-19 on driving, especially on average speed, speeding, and harsh braking per 100 km. More specifically, speeds were found to increase by 2.27 km/h on average compared to the forecasted evolution, while harsh braking/100 km increased to almost 1.51 on average. On the bright side, road crashes in Greece were reduced by 49% during the months of COVID-19 compared to the non-COVID-19 period.

Mobile applications for road traffic health and safety in the mirror of the Haddon's matrix

BACKGROUND

Road traffic accidents have been one of the leading causes of death. Despite the increasing trend of road traffic apps, there is no comprehensive analysis of their features and no taxonomy for the apps based on traffic safety theories. This study aimed to explore the characteristics of available mobile apps on road traffic health/safety and classify them with emphasis on Haddon's matrix.

METHODS

The researchers examined the mobile applications related to road traffic health/safety using qualitative content analysis. Google Play was searched using a combination of the keywords. Haddon's matrix was applied to analyze and classify those mobile apps residing in the categories of Road Traffic health & Safety, and Road Traffic Training.

RESULTS

Overall, 913 mobile apps met the inclusion criteria and were included in the final analysis. Classification of the apps based on their features resulted in 4 categories and 21 subcategories. A total number of 657 mobile apps were classified based on Haddon's matrix. About 45.67% of these apps were categorized as the road traffic health & safety group.

CONCLUSIONS

Haddon's matrix appears to have the potential to reveal the strengths and weaknesses of existing mobile apps in the road traffic accident domain. Future development of mobile apps in this domain should take into account the existing gap.

Spatial analysis of harsh driving behavior events in urban networks using high-resolution smartphone and geometric data

The aim of the present study is to conduct spatial analysis of harsh events of driving behavior across road segments of an urban road network. The adopted approach involved automating the segment characteristic extraction process for the urban network study area. Subsequently, naturalistic driving big data from an innovative smartphone application were map-matched to the segments that each driver traversed, and thus geometrical, road network and driver behavior spatial data frames were obtained per road segment. Global and local Moran's I coefficients were calculated based on a nearest-neighbour scheme, and indicated the presence of a certain degree of positive spatial autocorrelation both for harsh brakings (HBs) and for harsh accelerations (HAs). Furthermore, the creation of empirical and theoretical spherical variograms indicated that on average, about 190 m from each road segment centroid there is no observable spatial autocorrelation for HBs; the respective distance is 200 m for HAs. Geographically Weighted Poisson Regression (GWPR) models were used to model harsh event frequencies. Segment length and pass count are positively correlated with HB frequencies, while gradient and neighbourhood complexity are negatively correlated with HB frequencies. Curvature, segment length, pass count and the presence of traffic lights are positively correlated with HA frequencies. Road type and lane number were found to have a more circumstantial effect overall.

Investigation of the speeding behavior of motorcyclists through an innovative smartphone application

OBJECTIVE

The objective of the present study is twofold: (i) to explore the riding behavior of motorcyclists while speeding, based on detailed riding analytics collected by smartphone sensors, and (ii) to investigate whether personalized feedback can improve motorcyclist behavior.

METHODS

In order to achieve the objective, a naturalistic riding experiment with a sample of 13 motorcyclists based on a smartphone application developed within the framework of the BeSmart project was conducted. Using risk exposure and riding behavior indicators calculated from smartphone sensor data, Generalized Linear Mixed-Effects Models are calibrated to correlate the percentage of riding time over the speed limit with other riding behavior indicators. An overall model was developed for all trips, as well as separate models for the parts of trips realized on different road types (urban and rural).

RESULTS

Results indicate that the parameters of trip duration, distance driven during risky hours, morning peak hours and the number of harsh accelerations are all determined as statistically significant and positively correlated with the percentage of speeding time. Additionally, the provision of rider feedback and riding during afternoon peak hours are statistically significant and correlated with decreased percentages of speeding time.

CONCLUSIONS

The outcomes of this study entail both scientific and social impacts. The present research contributes a preliminary example of the quantitative documentation of the impact of personalized rider feedback on one of the most important human risk factors; speeding. The ultimate objective when providing feedback to riders is to: (i) trigger their learning and self-assessment process, thus enabling them to gradually improve their performance and (ii) monitor the shift of riding behavior as the application provides feedback. The present results capture and quantify the positive effects of rider feedback, thus providing needed impetus for larger-scale applications as well as relevant policy interventions.

SafeDrive: Hybrid Recommendation System Architecture for Early Safety Predication Using Internet of Vehicles

The Internet of vehicles (IoV) is a rapidly emerging technological evolution of Intelligent Transportation System (ITS). This paper proposes SafeDrive, a dynamic driver profile (DDP) using a hybrid recommendation system. DDP is a set of functional modules, to analyses individual driver’s behaviors, using prior violation and accident records, to identify driving risk patterns. In this paper, we have considered three synthetic data-sets for 1500 drivers based on their profile information, risk parameters information, and risk likelihood. In addition, we have also considered the driver’s historical violation/accident data-set records based on four risk-score levels such as high-risk, medium-risk, low-risk, and no-risk to predict current and future driver risk scores. Several error calculation methods have been applied in this study to analyze our proposed hybrid recommendation systems’ performance to classify the driver’s data with higher accuracy based on various criteria. The evaluated results help to improve the driving behavior and broadcast early warning alarm to the other vehicles in IoV environment for the overall road safety. Moreover, the propoed model helps to provide a safe and predicted environment for vehicles, pedestrians, and road objects, with the help of regular monitoring of vehicle motion, driver behavior, and road conditions. It also enables accurate prediction of accidents beforehand, and also minimizes the complexity of on-road vehicles and latency due to fog/cloud computing servers.

Characterizing phone usage while driving: Safety impact from road and operational perspectives using factor analysis

Phone use while driving (PUWD) is one of the most crucial factors of distraction related traffic crashes. This study utilized an unsupervised learning method, known as factor analysis, on a unique distracted driving dataset to understand PUWD behavior from the roadway geometry and operational perspectives. The results indicate that the presence of a shoulder, median, and access control on the relatively higher functional class roadways could encourage more PUWD events. The roadways with relatively lower speed limits could have high PUWD event occurrences if the variation in operating speed is high. The results also confirm the correlations between the frequency of PUWD events and the frequency of distracted crashes. This relationship is strong on urban roadways. For rural roadways, this correlation is only strong on the roadways with a large amount of PUWD events. The findings could help transportation agencies to identify suitable countermeasures in reducing distraction related crashes. Moreover, this study provides researchers a new perspective to study PUWD behavior rather than only focus on drivers' personalities.

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.

Observations on the Relationship between Crash Frequency and Traffic Flow

This work analyzes the relationship between crash frequency N (crashes per hour) and exposure Q (cars per hour) on the macroscopic level of a whole city. As exposure, the traffic flow is used here. Therefore, it analyzes a large crash database of the city of Berlin, Germany, together with a novel traffic flow database. Both data display a strong weekly pattern, and, if taken together, show that the relationship N(Q) is not a linear one. When Q is small, N grows like a second-order polynomial, while at large Q there is a tendency towards saturation, leading to an S-shaped relationship. Although visible in all data from all crashes, the data for the severe crashes display a less prominent saturation. As a by-product, the analysis performed here also demonstrates that the crash frequencies follow a negative binomial distribution, where both parameters of the distribution depend on the hour of the week, and, presumably, on the traffic state in this hour. The work presented in this paper aims at giving the reader a better understanding on how crash rates depend on exposure.