Utilizing naturalistic driving data for in-depth analysis of driver lane-keeping behavior in rain: Non-parametric MARS and parametric logistic regression modeling approaches

It is known that adverse weather conditions can affect driver performance due to reduction in visibility and slippery surface conditions. Lane keeping is one of the main factors that might be affected by weather conditions. Most of the previous studies on lane keeping have investigated driver lane-keeping performance from driver inattention perspective. In addition, the majority of previous lane-keeping studies have been conducted in controlled environments such as driving simulators. Therefore, there is a lack of studies that investigate driver lane-keeping ability considering adverse weather conditions in naturalistic settings. In this study, the relationship between weather conditions and driver lane-keeping performance was investigated using the SHRP2 naturalistic driving data for 141 drivers between 19 and 89 years of age. Moreover, a threshold was introduced to differentiate lane keeping and lane changing in naturalistic driving data. Two lane-keeping models were developed using the logistic regression and multivariate adaptive regression splines (MARS) to better understand factors affecting driver lane-keeping ability considering adverse weather conditions. The results revealed that heavy rain can significantly increase the standard deviation of lane position (SDLP), which is a very widely used method for analyzing lane-keeping ability. It was also found that traffic conditions, driver age and experience, and posted speed limits have significant effects on driver lane-keeping ability. An interesting finding of this study is that drivers have a better lane-keeping ability in roadways with higher posted speed limits. The results from this study might provide better insights into understanding the complex effect of adverse weather conditions on driver behavior.     

Weather and road geometry impact on longitudinal driving behavior: Exploratory analysis using an empirically supported acceleration modeling framework

The objective of this paper is to quantify and characterize driver behavior under different roadway geometries and weather conditions. In order to explore how a driver perceives the rapidly changing driving surrounding (i.e. different weather conditions and road geometry configurations) and executes acceleration maneuvers accordingly, this paper extends a Prospect Theory based acceleration modeling framework. A driving simulator is utilized to conduct 76 driving experiments. Foggy weather, icy and wet roadway surfaces, horizontal and vertical curves, and different lane and shoulder widths are simulated while having participants driving behind a yellow cab at speeds/headways of their choice. After studying the driving trends observed in the different driving experiments, the extended Prospect Theory based acceleration model is calibrated using the produced trajectory data. The extended Prospect Theory based model parameters are able to reflect a change in risk-perception and acceleration maneuvering when receiving different parameterized exogenous information. The results indicate that drivers invest more attention and effort to deal with the roadway challenges compared to the effort to deal with the weather conditions. Moreover, the calibrated model is used to simulate a highway segment and observe the produced fundamental diagram. The preliminary results suggest that the model is capable of capturing driver behavior under different roadway and weather conditions leading to changes in capacity and traffic disruptions.     

Modeling of individual vehicle safety and fuel consumption under comprehensive external conditions

This research intends to explore external factors affecting driving safety and fuel consumption, and build a risk and fuel consumption prediction model for individual drivers based on natural driving data. Based on 120 taxi drivers’ natural driving data during 4 months, driving behavior data under various conditions of the roadway, traffic, weather, and time of day are extracted. The driver's fuel consumption is directly collected by the on-board diagnostics (OBD) unit, and safety index is calculated based on Data Threshold Violations (DTV) and Phase Plane Analysis with Limits (PPAL) considering speed, longitudinal and lateral acceleration. By using a linear mixed model explaining the fixed effect of the external conditions and the random effect of the driver, the influences of various external factors on fuel consumption and safety are analyzed and discussed. The prediction model lays a foundation for drivers' fuel consumption and risk prediction in different external conditions, which could help improve individual driving behavior for the benefit of both fuel consumption and safety.     

Mobile phone use by young drivers: effects on traffic speed and headways

Abstract

This paper investigates the effects of mobile phone use while driving on traffic speed and headways, with particular focus on young drivers. For this purpose, a field survey was carried out in real road traffic conditions, in which drivers' speeds and headways were measured while using or not using a mobile phone. The survey took place within a University Campus area, allowing to distinguish between settings approximating to either free flow or interrupted flow conditions. Linear and loglinear regression methods were used to investigate the effects of mobile phone use and several other young driver characteristics, such as gender, driving experience and annual distance travelled, on vehicle speeds and headways. Separate models were developed for average free flow, interrupted flow, as well as for total average speed. Results show that mobile phone use leads to a statistically significant reduction in traffic speeds of young drivers in all types of traffic conditions. Furthermore, male and female drivers reduce their speed similarly when using a mobile phone while driving. However, male drivers using their mobile phone drive at lower speeds than female drivers not using their mobile phones. Sensitivity analysis revealed that, among all explanatory variables, the effect of mobile phone use on speed was most important. Accordingly, vehicle headways appear to increase for drivers using their mobile phone. However, this effect could not be statistically validated, due to the strong correlation between speed and headway.     

Modeling traffic's flow-density relation: Accommodation of multiple flow regimes and traveler types

This research investigates freeway-flow impacts of different traveler types by specifying and applying a latent-segmentation model of congested and uncongested driving behaviors. Drivers in uncongested conditions are assumed to drive at self-chosen speeds, while drivers in congested conditions are assumed to take speed as given and choose a spacing (between their vehicle and the previous vehicle). Several classes of driver-vehicle combinations are distinguished in a data set based on double-loop-detector pulses and a household travel survey. These classifications are made on the basis of vehicle type and gender, leading to class estimates of speeds and spacings. The segmentation model is specified as a logit function of density, weather, and vehicle type, leading to estimates of congested-condition probabilities. Unobserved heterogeneity is incorporated in all models via common error assumptions.Results indicate that segmentation models are promising tools for traffic data analysis and that information on travelers, their vehicles, and weather conditions explains significant variation in flow data. By clarifying a greater understanding of traffic conditions and traveler behavior explains much scatter in the fundamental relation between flow, speed, and density, can assist regions in their traffic-management efforts and engineers in their design of roadway facilities. Ultimately, such improvements to travel networks should enhance quality of life.     

Analysis of volatility in driving regimes extracted from basic safety messages transmitted between connected vehicles

Driving volatility captures the extent of speed variations when a vehicle is being driven. Extreme longitudinal variations signify hard acceleration or braking. Warnings and alerts given to drivers can reduce such volatility potentially improving safety, energy use, and emissions. This study develops a fundamental understanding of instantaneous driving decisions, needed for hazard anticipation and notification systems, and distinguishes normal from anomalous driving. In this study, driving task is divided into distinct yet unobserved regimes. The research issue is to characterize and quantify these regimes in typical driving cycles and the associated volatility of each regime, explore when the regimes change and the key correlates associated with each regime. Using Basic Safety Message (BSM) data from the Safety Pilot Model Deployment in Ann Arbor, Michigan, two- and three-regime Dynamic Markov switching models are estimated for several trips undertaken on various roadway types. While thousands of instrumented vehicles with vehicle to vehicle (V2V) and vehicle to infrastructure (V2I) communication systems are being tested, nearly 1.4 million records of BSMs, from 184 trips undertaken by 71 instrumented vehicles are analyzed in this study. Then even more detailed analysis of 43 randomly chosen trips (N = 714,340 BSM records) that were undertaken on various roadway types is conducted. The results indicate that acceleration and deceleration are two distinct regimes, and as compared to acceleration, drivers decelerate at higher rates, and braking is significantly more volatile than acceleration. Different correlations of the two regimes with instantaneous driving contexts are explored. With a more generic three-regime model specification, the results reveal high-rate acceleration, high-rate deceleration, and cruise/constant as the three distinct regimes that characterize a typical driving cycle. Moreover, given in a high-rate regime, drivers’ on-average tend to decelerate at a higher rate than their rate of acceleration. Importantly, compared to cruise/constant regime, drivers’ instantaneous driving decisions are more volatile both in “high-rate” acceleration as well as “high-rate” deceleration regime. The study contributes to analyzing volatility in short-term driving decisions, and how changes in driving regimes can be mapped to a combination of local traffic states surrounding the vehicle.     

Personalised assistance for fuel-efficient driving

Recent advances in technology are changing the way how everyday activities are performed. Technologies in the traffic domain provide diverse instruments of gathering and analysing data for more fuel-efficient, safe, and convenient travelling for both drivers and passengers. In this article, we propose a reference architecture for a context-aware driving assistant system. Moreover, we exemplify this architecture with a real prototype of a driving assistance system called Driving coach. This prototype collects, fuses and analyses diverse information, like digital map, weather, traffic situation, as well as vehicle information to provide drivers in-depth information regarding their previous trip along with personalised hints to improve their fuel-efficient driving in the future. The Driving coach system monitors its own performance, as well as driver feedback to correct itself to serve the driver more appropriately.     

A Bayesian Network model for contextual versus non-contextual driving behavior assessment

Driving behavior is generally considered to be one of the most important factors in crash occurrence. This paper aims to evaluate the benefits of utilizing context-relevant information in the driving behavior assessment process (i.e. contextual driving behavior assessment approach). We use a Bayesian Network (BN) model that investigates the relationships between GPS driving observations, individual driving behavior, individual driving risks, and individual crash frequency. In contrast to prior studies without context information (i.e. non-contextual approach), the data used in the BN approach is a combination of contextual features in the surrounding environment that may contribute to crash risk, such as road conditions surrounding the vehicle of interest and dynamic traffic flow information, as well as the non-contextual data such as instantaneous driving speed and the acceleration/deceleration of a vehicle. An information-aggregation mechanism is developed to aggregates massive amounts of vehicle GPS data points, kinematic events and context information into drivel-level data. With the proposed model, driving behavior risks for drivers is assessed and the relationship between contextual driving behavior and crash occurrence is established. The analysis results in the case study section show that the contextual model has significantly better performance than the non-contextual model, and that drivers who drive at a speed faster than others or much slower than the speed limit at the ramp, and with more rapid acceleration or deceleration on freeways are more likely to be involved in crash events. In addition, younger drivers, and female drivers with higher VMT are found to have higher crash risk.     

Effect of variable message signs on driver speed behavior on a section of expressway under adverse fog conditions—A driving simulator approach

Variable message signs (VMS) are used to provide dynamic information and one current application is to show different speed limits under different conditions. As speed is an important contributor to road accidents and also affects driver speed behavior, the present study focuses on how effective traffic advisory information is when helping drivers to divert from potentially dangerous conditions. Graphical representation of an Expressway section made it easy to isolate the effects of speed etc. by drivers with information provided through VMS under adverse fog conditions. Understanding and reacting to the VMS system by drivers is essential for its success. If drivers do not react by changing speed behavior then the VMS system will fail and further implementation may cease. In this paper an Analysis of Variance model, which is appropriate to the proposed experimental conditions, is used to study how subjects (drivers) will perceive provided information and also to find the effect of VMS on driver speed behavior on the simulated Expressway section.     

A rule-based neural network approach to model driver naturalistic behavior in traffic

This paper proposes a rule-based neural network model to simulate driver behavior in terms of longitudinal and lateral actions in two driving situations, namely car-following situation and safety critical events. A fuzzy rule based neural network is constructed to obtain driver individual driving rules from their vehicle trajectory data. A machine learning method reinforcement learning is used to train the neural network such that the neural network can mimic driving behavior of individual drivers. Vehicle actions by neural network are compared to actions from naturalistic data. Furthermore, this paper applies the proposed method to analyze the heterogeneities of driving behavior from different drivers’ data.Driving data in the two driving situations are extracted from Naturalistic Truck Driving Study and Naturalistic Car Driving Study databases provided by the Virginia Tech Transportation Institute according to pre-defined criteria. Driving actions were recorded in instrumented vehicles that have been equipped with specialized sensing, processing, and recording equipment.     

Impact of traveler advisory systems on driving speed: some new evidence

This paper explores the effects of driving behavior using in-vehicle and out-of-vehicle traffic advisory information relating to adverse weather and incident conditions. A full-size, fixed-based driving simulator is used to collect data on drivers’ speed behavior under four different advisory-information conditions: in-vehicle messages, out-of-vehicle messages, both types of messages, and no messages. The findings of this study suggest an interesting phenomenon in that, while messages are significant in reducing speeds in the area of adverse conditions, drivers tend to compensate for this speed reduction by increasing speeds downstream when such adverse conditions do not exist. As a result, the net safety effects of such message systems are ambiguous.     

The role of alternative fuel vehicles: Using behavioral and sensor data to model hierarchies in travel

Greater adoption and use of alternative fuel vehicles (AFVs) can be environmentally beneficial and reduce dependence on gasoline. The use of AFVs vis-à-vis conventional gasoline vehicles is not well understood, especially when it comes to travel choices and short-term driving decisions. Using data that contains a sufficiently large number of early AFV adopters (who have overcome obstacles to adoption), this study explores differences in use of AFVs and conventional gasoline vehicles (and hybrid vehicles). The study analyzes large-scale behavioral data integrated with sensor data from global positioning system devices, representing advances in large-scale data analytics. Specifically, it makes sense of data containing 54,043,889 s of speed observations, and 65,652 trips made by 2908 drivers in 5 regions of California. The study answers important research questions about AFV use patterns (e.g., trip frequency and daily vehicle miles traveled) and driving practices. Driving volatility, as one measure of driving practice, is used as a key metric in this study to capture acceleration, and vehicular jerk decisions that exceed certain thresholds during a trip. The results show that AFVs cannot be viewed as monolithic; there are important differences within AFV use, i.e., between plug-in hybrids, battery electric, or compressed natural gas vehicles. Multi-level models are particularly appropriate for analysis, given that the data are nested, i.e., multiple trips are made by different drivers who reside in various regions. Using such models, the study also found that driving volatility varies significantly between trips, driver groups, and regions in California. Some alternative fuel vehicles are associated with calmer driving compared with conventional vehicles. The implications of the results for safety, informed consumer choices and large-scale data analytics are discussed.     

Human-like autonomous car-following model with deep reinforcement learning

This study proposes a framework for human-like autonomous car-following planning based on deep reinforcement learning (deep RL). Historical driving data are fed into a simulation environment where an RL agent learns from trial and error interactions based on a reward function that signals how much the agent deviates from the empirical data. Through these interactions, an optimal policy, or car-following model that maps in a human-like way from speed, relative speed between a lead and following vehicle, and inter-vehicle spacing to acceleration of a following vehicle is finally obtained. The model can be continuously updated when more data are fed in. Two thousand car-following periods extracted from the 2015 Shanghai Naturalistic Driving Study were used to train the model and compare its performance with that of traditional and recent data-driven car-following models. As shown by this study’s results, a deep deterministic policy gradient car-following model that uses disparity between simulated and observed speed as the reward function and considers a reaction delay of 1 s, denoted as DDPGvRT, can reproduce human-like car-following behavior with higher accuracy than traditional and recent data-driven car-following models. Specifically, the DDPGvRT model has a spacing validation error of 18% and speed validation error of 5%, which are less than those of other models, including the intelligent driver model, models based on locally weighted regression, and conventional neural network-based models. Moreover, the DDPGvRT demonstrates good capability of generalization to various driving situations and can adapt to different drivers by continuously learning. This study demonstrates that reinforcement learning methodology can offer insight into driver behavior and can contribute to the development of human-like autonomous driving algorithms and traffic-flow models.     

Customizing driving cycles to support vehicle purchase and use decisions: Fuel economy estimation for alternative fuel vehicle users

Wider deployment of alternative fuel vehicles (AFVs) can help with increasing energy security and transitioning to clean vehicles. Ideally, adopters of AFVs are able to maintain the same level of mobility as users of conventional vehicles while reducing energy use and emissions. Greater knowledge of AFV benefits can support consumers’ vehicle purchase and use choices. The Environmental Protection Agency’s fuel economy ratings are a key source of potential benefits of using AFVs. However, the ratings are based on pre-designed and fixed driving cycles applied in laboratory conditions, neglecting the attributes of drivers and vehicle types. While the EPA ratings using pre-designed and fixed driving cycles may be unbiased they are not necessarily precise, owning to large variations in real-life driving. Thus, to better predict fuel economy for individual consumers targeting specific types of vehicles, it is important to find driving cycles that can better represent consumers’ real-world driving practices instead of using pre-designed standard driving cycles. This paper presents a methodology for customizing driving cycles to provide convincing fuel economy predictions that are based on drivers’ characteristics and contemporary real-world driving, along with validation efforts. The methodology takes into account current micro-driving practices in terms of maintaining speed, acceleration, braking, idling, etc., on trips. Specifically, using a large-scale driving data collected by in-vehicle Global Positioning System as part of a travel survey, a micro-trips (building block) library for California drivers is created using 54 million seconds of vehicle trajectories on more than 60,000 trips, made by 3000 drivers. To generate customized driving cycles, a new tool, known as Case Based System for Driving Cycle Design, is developed. These customized cycles can predict fuel economy more precisely for conventional vehicles vis-à-vis AFVs. This is based on a consumer’s similarity in terms of their own and geographical characteristics, with a sample of micro-trips from the case library. The AFV driving cycles, created from real-world driving data, show significant differences from conventional driving cycles currently in use. This further highlights the need to enhance current fuel economy estimations by using customized driving cycles, helping consumers make more informed vehicle purchase and use decisions.     

Understanding the impacts of mobile phone distraction on driving performance: A systematic review

The use of mobile phones while driving—one of the most common driver distractions—has been a significant research interest during the most recent decade. While there has been a considerable amount research and excellent reviews on how mobile phone distractions influence various aspects of driving performance, the mechanisms by which the interactions with mobile phone affect driver performance is relatively unexamined. As such, the aim of this study is to examine the mechanisms involved with mobile phone distractions such as conversing, texting, and reading and the driving task, and subsequent outcomes. A novel human-machine framework is proposed to isolate the components and various interactions associated with mobile phone distracted driving. The proposed framework specifies the impacts of mobile phone distraction as an inter-related system of outcomes such as speed selection, lane deviations and crashes; human-car controls such as steering control and brake pedal use and human-environment interactions such as visual scanning and navigation. Eleven literature-review/meta-analyses papers and 62 recent research articles from 2005 to 2015 are critically reviewed and synthesised following a systematic classification scheme derived from the human-machine system framework. The analysis shows that while many studies have attempted to measure system outcomes or driving performance, research on how drivers interactively manage in-vehicle secondary tasks and adapt their driving behaviour while distracted is scant. A systematic approach may bolster efforts to examine comprehensively the performance of distracted drivers and their impact over the transportation system by considering all system components and interactions of drivers with mobile phones and vehicles. The proposed human-machine framework not only contributes to the literature on mobile phone distraction and safety, but also assists in identifying the research needs and promising strategies for mitigating mobile phone-related safety issues. Technology based countermeasures that can provide real-time feedback or alerts to drivers based on eye/head movements in conjunction with vehicle dynamics should be an important research direction.     

Development of the World-wide harmonized Light duty Test Cycle (WLTC) and a possible pathway for its introduction in the European legislation

This paper presents the World-wide harmonized Light duty Test Cycle (WLTC), developed under the Working Party on Pollution and Energy (GRPE) and sponsored by the European Union (with Switzerland) and Japan. India, Korea and USA have also actively contributed. The objective was to design the harmonized driving cycle from “real world” driving data in different regions around the world, combined with suitable weighting factors. To this aim, driving data and traffic statistics of light duty vehicles use were collected and analyzed as basic elements to develop the harmonized cycle. The regional driving data and weighting factors were then combined in order to develop a unified database representing the worldwide light duty vehicle driving behavior. From the unified database, short trips were selected and combined to develop a driving cycle as representative as possible of the unified database. Approximately 765,000 km of data were collected, covering a wide range of vehicle categories, road types and driving conditions. The resulting WLTC is an ensemble of three driving cycles adapted to three vehicle categories with different power-to-mass ratio (PMR). It has been designed as a harmonized cycle for the certification of light duty vehicles around the world and, together with the new harmonized test procedures (WLTP), will serve to check the compliance of vehicle pollutant emissions with respect to the applicable emissions limits and to establish the reference vehicle fuel consumption and CO₂ performance.     

Analyzing fluctuations in car-following

Many car-following models predict a stable car-following behavior with a very small fluctuation around an equilibrium value g¹ of the net headway g with zero speed-difference Δv between the following and the lead vehicle. However, it is well-known and additionally demonstrated by data in this paper, that the fluctuations are much larger than these models predict. Typically, the fluctuation in speed difference is around ±2 m/s, while the fluctuation in the net time headway T = g/v can be as big as one or even two seconds, which is as large as the mean time headway itself. By analyzing data from loop detectors as well as data from vehicle trajectories, evidence is provided that this randomness is not due to driver heterogeneity, but can be attributed to an internal stochasticity of the driver itself. A final model-based analysis supports the hypothesis, that the preferred headway of the driver is the parameter that is not kept constant but fluctuates strongly, thus causing the even macroscopically observable randomness in traffic flow.     

Young drivers and graduated licensing: The Ontario case

Despite improvements in road safety over the past several decades, accident rates remain high for young drivers. One accident countermeasure that is expected to improve the safety record of this group is graduated licensing. The philosophy behind this licensing system is that novice drivers, of whom the majority are young, should be restricted to relatively safe driving environments during the initial learning period. Graduated licensing was implemented in the Province of Ontario, Canada in 1994. The objective of this study is to estimate the potential benefits and costs for young drivers associated with two components of the Ontario graduated licensing package: the late-night driving curfew and the high-speed roadway restrictions. Based on accident and travel data for the year 1988, accident-involvement rates per kilometre driven were calculated for different driver groups for various combinations of time of day and roadway speed limit. These rates were then applied to the expected mobility profiles of young drivers affected by graduated licensing. The results of the study support the late-night curfew and suggest that this component of the licensing package should reduce total accident involvements for the affected group by up to 10 percent and fatal accident involvements by up to 24 percent, while reducing their total driving by only four percent. By contrast, the empirical evidence suggest that the high-speed roadway restrictions are likely to increase accident involvements, and thus it is strongly recommended that this component of Ontario's graduated licensing package be changed.     

Eco-driving key factors that influence fuel consumption in heavy-truck fleets: A Colombian case

This research identifies key variables that influence fuel consumption that might be improved through eco-driving training programs under three circumstances that have been scarcely studied before: (a) heavy- and medium-duty truck fleets, (b) long-distance freight transport, and (c) the Latin American region. Based on statistical analyses that include multivariate regression of operational variables on fuel consumption, the impacts of an eco-driving training campaign were measured by comparing ex ante and ex post data. Operational variables are grouped into driving errors, trip conditions, driver behavior, driver profile, and vehicle attributes.The methodology is applied in a freight fleet with nationwide transport operations located in Colombia, where the steepness of its roads plays an important role in fuel consumption. The fleet, composed of 18 trucks, is equipped with state-of-the-art real-time data logger systems. During four months, 517 trips traveling a total distance of 292,512 km and carrying a total of 10,034 tons were analyzed.The results show a baseline average fuel consumption (FC) of 1.716 liters per ton-100 km. A different logistics performance indicator, which measures FC in liters per ton transported each 100 km, shows an average of 3.115. After the eco-driving campaign, reductions of 6.8% and 5.5% were obtained. Drivers’ experience, driving errors, average speed, and weight-capacity ratio, among others, were found to be highly relevant to FC. In particular, driving errors such as acceleration, braking and speed excesses are the most sensitive to eco-driving training, showing reductions of up to 96% on the average number of events per trip.     

Driving with intelligent speed adaptation: Final results of the Belgian ISA-trial

In October 2002 the first ISA-trial in Belgium was started in Ghent. Thirty-four cars and three buses were equipped with the “active accelerator pedal”. In this system a resistance in the accelerator is activated when the driver attempts to exceed the speed limit. If necessary, the driver can overrule the system. The main research goals of the trial in Ghent were to evaluate the effects of ISA on speed-change, traffic safety, drivers’ attitude, behaviour and drivers’ acceptance. To study these effects of the ISA-system both surveys and logged speed data were analyzed. In the surveys drivers noticed that the pedal assisted them well in upholding the speed limits and that the system increased driving comfort. Most important drawbacks were technical issues. Data analysis shows a reduction in the amount of speeding due to the ISA-system. There is however still a large remaining percentage of distance speeding, especially in low speed zones. Differences between drivers are large. For some drivers speeding even increases despite activation of the system. For less frequent speeders average driving speed almost always increases and for more frequent speeders average speed tends to decrease. With the system, less frequent speeders tend to accelerate faster towards the speed limit and drive exactly at the speed limit instead of safely below, which causes average speeds to go up.