Examination of the Effect of Task Complexity and Coping Capacity on Driving Risk: A Cross-Country and Transportation Mode Comparative Study.

The i-DREAMS project established a 'Safety Tolerance Zone (STZ)' to maintain operators within safe boundaries through real-time and post-trip interventions, based on the crucial role of the human element in driving behavior. This paper aims to model the inter-relationship among driving task complexity, operator and vehicle coping capacity, and crash risk. Towards that aim, data from 80 drivers, who participated in a naturalistic driving experiment carried out in three countries (i.e., Belgium, Germany, and Portugal), resulting in a dataset of approximately 19,000 trips were collected and analyzed. The exploratory analysis included the development of Generalized Linear Models (GLMs) and the choice of the most appropriate variables associated with the latent variables "task complexity" and "coping capacity" that are to be estimated from the various indicators. In addition, Structural Equation Models (SEMs) were used to explore how the model variables were interrelated, allowing for both direct and indirect relationships to be modeled. Comparisons on the performance of such models, as well as a discussion on behaviors and driving patterns across different countries and transport modes, were also provided. The findings revealed a positive relationship between task complexity and coping capacity, indicating that as the difficulty of the driving task increased, the driver's coping capacity increased accordingly, (i.e., higher ability to manage and adapt to the challenges posed by more complex tasks). The integrated treatment of task complexity, coping capacity, and risk can improve the behavior and safety of all travelers, through the unobtrusive and seamless monitoring of behavior. Thus, authorities should utilize a data system oriented towards collecting key driving insights on population level to plan mobility and safety interventions, develop incentives for road users, optimize enforcement, and enhance community building for safe traveling.

Real-time monitoring of driver distraction: State-of-the-art and future insights.

Driver distraction and inattention have been found to be major contributors to a large number of serious road crashes. It is evident that distraction reduces to a great extent driver perception levels as well as their decision making capability and the ability of drivers to control the vehicle. An effective way to mitigate the effects of distraction on crash probability, would be through monitoring the mental state of drivers or their driving behaviour and alerting them when they are in a distracted state. Towards that end, in recent years, several inexpensive and effective detection systems have been developed in order to cope with driver inattention. This study endeavours to critically review and assess the state-of-the-art systems and platforms measuring driver distraction or inattention. A thorough literature review was carried out in order to compare and contrast technologies that can be used to detect, monitor or measure driver's distraction or inattention. The systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. The results indicated that in most of the identified studies, driver distraction was measured with respect to its impact to driver behaviour. Real-time eye tracking systems, cardiac sensors on steering wheels, smartphone applications and cameras were found to be the most frequent devices to monitor and detect driver distraction. On the other hand, less frequent and effective approaches included electrodes, hand magnetic rings and glasses.

One year of COVID-19: Impacts on safe driving behavior and policy recommendations.

INTRODUCTION: In the unprecedented year of 2020, the rapid spread of COVID-19 disrupted everyday activities worldwide, leading the majority of countries to impose lockdowns and confine citizens in order to minimize the exponential increase in cases and casualties. To date, very few studies have been concerned with the effect of the pandemic on driving behavior and road safety, and usually explore data from a limited time span. METHOD: This study presents a descriptive overview of several driving behavior indicators as well as road crash data in correlation with the strictness of response measures in Greece and the Kingdom of Saudi Arabia (KSA). A k-means clustering approach was also employed to detect meaningful patterns. RESULTS: Results indicated that during the lockdown periods, speeds were increased by up to 6%, while harsh events were increased by about 35% in the two countries, compared to the period after the confinement. However, the imposition of another lockdown did not cause radical changes in Greek driving behavior during the late months of 2020. Finally, the clustering algorithm identified a "baseline," a "restrictions," and a "lockdown" driving behavior cluster, and it was shown that harsh braking frequency was the most distinctive factor. POLICY RECOMMENDATIONS: Based on these findings, policymakers should focus on the reduction and enforcement of speed limits, especially within urban areas, as well as the incorporation of active travelers in the current transport infrastructure.

Data-Driven Estimation of a Driving Safety Tolerance Zone Using Imbalanced Machine Learning.

Predicting driving behavior and crash risk in real-time is a problem that has been heavily researched in the past years. Although in-vehicle interventions and gamification features in post-trip dashboards have emerged, the connection between real-time driving behavior prediction and the triggering of such interventions is yet to be realized. This is the focus of the European Horizon2020 project "i-DREAMS", which aims at defining, developing, testing and validating a 'Safety Tolerance Zone' (STZ) in order to prevent drivers from risky driving behaviors using interventions both in real-time and post-trip. However, the data-driven conceptualization of STZ levels is a challenging task, and data class imbalance might hinder this process. Following the project principles and taking the aforementioned challenges into consideration, this paper proposes a framework to identify the level of risky driving behavior as well as the duration of the time spent in each risk level by private car drivers. This aim is accomplished by four classification algorithms, namely Support Vector Machines (SVMs), Random Forest (RFs), AdaBoost, and Multilayer Perceptron (MLP) Neural Networks and imbalanced learning using the Adaptive Synthetic technique (ADASYN) in order to deal with the unbalanced distribution of the dataset in the STZ levels. Moreover, as an alternative approach of risk prediction, three regression algorithms, namely Ridge, Lasso, and Elastic Net are used to predict time duration. The results showed that RF and MLP outperformed the rest of the classifiers with 84% and 82% overall accuracy, respectively, and that the maximum speed of the vehicle during a 30 s interval, is the most crucial predictor for identifying the driving time at each safety level.

Analysis of the impact of COVID-19 on collisions, fatalities and injuries using time series forecasting: The case of Greece.

The current study aims to investigate the impact of the COVID-19 pandemic on road traffic collisions, fatalities, and injuries using time series analyses. To that aim, a database containing road collisions, fatalities, and slight injuries data from Greece were derived from the Hellenic Statistical Authority (HSA) and covered a ten-year timeframe (from January 2010 to August 2020. The chosen time period contained normal operations, as well as the period of the first COVID-19-induced lockdown period in Greece. Three different Seasonal Autoregressive Integrated Moving Average (SARIMA) time series models were implemented in order to compare the observed measurements to forecasted values that were intended to depict assumed conditions; namely, without the appearance of the COVID-19 pandemic. Modelling results revealed that the total number of road collisions, fatalities, and slightly injured were decreased, mainly due to the sharp traffic volume decrease. However, the percentage reduction of the collision variables and traffic volume were found to be disproportionate, which probably indicates that more collisions occurred with regard to the prevailing traffic volume. An additional finding is that fatalities and slightly injured rates were significantly increased during the lockdown period and the subsequent month. Overall, it can be concluded that a worse performance was identified in terms of road safety. Since subsequent waves of COVID-19 cases and other pandemics may reappear in the future, the outcomes of the current study may be exploited for the improvement of road safety from local authorities and policymakers.

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.

Post-trip safety interventions: State-of-the-art, challenges, and practical implications.

INTRODUCTION: Currently, risky driving behaviour is a major contributor to road crashes and as a result, wide array of tools have been developed in order to record and improve driving behaviour. Within that group of tools, interventions have been indicated to significantly enhance driving behaviour and road safety. This study critically reviews monitoring technologies that provide post-trip interventions, such as retrospective visual feedback, gamification, rewards or penalties, in order to inform an appropriate driver mentoring strategy delivered after each trip. METHOD: The work presented here is part of the European Commission H2020 i-DREAMS project. The reviewed platform characteristics were obtained through commercially available solutions as well as a comprehensive literature search in popular scientific databases, such as Scopus and Google Scholar. Focus was given on state-of-the-art-technologies for post-trip interventions utilized in four different transport modes (i.e. car, truck, bus and rail) associated with risk prevention and mitigation. RESULTS: The synthesized results revealed that smartphone applications and web-based platforms are the most accepted, frequently and easiest to use tools in cars, buses and trucks across all papers considered, while limited evidence of post-trip interventions in -rail was found. The majority of smartphone applications detected mobile phone use and harsh events and provided individual performance scores, while in-vehicle systems provided delayed visual reports through a web-based platform. CONCLUSIONS: Gamification and appropriate rewards appeared to be effective solutions, as it was found that they keep drivers motivated in improving their driving skills, but it was clear that these cannot be performed in isolation and a combination with other strategies (i.e. driver coaching and support) might be beneficial. Nevertheless, as there is no holistic and cross-modal post-trip intervention solution developed in real-world environments, challenges associated with post-trip feedback provision and suggestions on practical implementation are also provided.

Prediction of rear-end conflict frequency using multiple-location traffic parameters.

Traffic conflicts are heavily correlated with traffic collisions and may provide insightful information on the failure mechanism and factors that contribute more towards a collision. Although proactive traffic management systems have been supported heavily in the research community, and autonomous vehicles (AVs) are soon to become a reality, analyses are concentrated on very specific environments using aggregated data. This study aims at investigating -for the first time- rear-end conflict frequency in an urban network level using vehicle-to-vehicle interactions and at correlating frequency with the corresponding network traffic state. The Time-To-Collision (TTC) and Deceleration Rate to Avoid Crash (DRAC) metrics are utilized to estimate conflict frequency on the current network situation, as well as on scenarios including AV characteristics. Three critical conflict points are defined, according to TTC and DRAC thresholds. After extracting conflicts, data are fitted into Zero-inflated and also traditional Negative Binomial models, as well as quasi-Poisson models, while controlling for endogeneity, in order to investigate contributory factors of conflict frequency. Results demonstrate that conflict counts are significantly higher in congested traffic and that high variations in speed increase conflicts. Nevertheless, a comparison with simulated AV traffic and the use of more surrogate safety indicators could provide more insight into the relationship between traffic state and traffic conflicts in the near future.

A descriptive analysis of the effect of the COVID-19 pandemic on driving behavior and road safety.

The spread of the new coronavirus COVID-19, has led to unparalleled global measures such as lockdown and suspension of all retail, recreation and religious activities during the first months of 2020. Nevertheless, no scientific evidence has been reported so far with regards to the impact on road safety and driving behavior. This paper investigates the effect of COVID-19 on driving behavior and safety indicators captured through a specially developed smartphone application and transmitted to a back-end platform. These indicators are reflected with the spread of COVID-19 and the respective governmental countermeasures in two countries, namely Greece and Kingdom of Saudi Arabia (KSA), which had the most completed routes for users of the smartphone applications. It was shown that reduced traffic volumes due to lockdown, led to a slight increase in speeds by 6-11%, but more importantly to more frequent harsh acceleration and harsh braking events (up to 12% increase) as well mobile phone use (up to 42% increase) during March and April 2020, which were the months where COVID-19 spread was at its peak. On the bright side, accidents in Greece were reduced by 41% during the first month of COVID-19-induced measures and driving in the early morning hours (00:00-05:00) which are considered dangerous dropped by up to 81%. Policymakers should concentrate on establishing new speed limits and ensure larger spaces for cycling and pedestrians in order to enlarge distances between users in order to safeguard both an enhanced level of road safety and the prevention of COVID-19 spread.

A new integrated collision risk assessment methodology for autonomous vehicles.

Real-time risk assessment of autonomous driving at tactical and operational levels is extremely challenging since both contextual and circumferential factors should concurrently be considered. Recent methods have started to simultaneously treat the context of the traffic environment along with vehicle dynamics. In particular, interaction-aware motion models that take inter-vehicle dependencies into account by utilizing the Bayesian interference are employed to mutually control multiple factors. However, communications between vehicles are often assumed and the developed models are required many parameters to be tuned. Consequently, they are computationally very demanding. Even in the cases where these desiderata are fulfilled, current approaches cannot cope with a large volume of sequential data from organically changing traffic scenarios, especially in highly complex operational environments such as dense urban areas with heterogeneous road users. To overcome these limitations, this paper develops a new risk assessment methodology that integrates a network-level collision estimate with a vehicle-based risk estimate in real-time under the joint framework of interaction-aware motion models and Dynamic Bayesian Networks (DBN). Following the formulation and explanation of the required functions, machine learning classifiers were utilized for the real-time network-level collision prediction and the results were then incorporated into the integrated DBN model for predicting collision probabilities in real-time. Results indicated an enhancement of the interaction-aware model by up to 10%, when traffic conditions are deemed as collision-prone. Hence, it was concluded that a well-calibrated collision prediction classifier provides a crucial hint for better risk perception by autonomous vehicles.