Driving Behavior Characteristics Research Articles

Field experiments on longitudinal characteristics of human driver behavior following an autonomous vehicle

Although mixed traffic, including both autonomous vehicles (AV) and human-driven vehicles (HV), is expected to prevail in the foreseeable future, our current understanding of the longitudinal characteristics of mixed traffic is limited and, in particular, lacks evidence from field experiments. To bridge this gap, we designed and conducted a set of field experiments to reveal differences in car-following behaviors between a human driver following-AV and following-HV on both constant speed traffic characteristics with discrete speeds ({10,20,…,60}km/h) and dynamic car-following behaviors with continuous speeds (within 0–60 km/h) in both the indifferentiable and differentiable appearance settings of the AV. We recruited 10 drivers for the experiment (14 runs for each driver and collected position and speed data of the tested vehicles along their complete trajectories based on vehicle gaps, headways, and standard deviations of vehicle speed. A K-means clustering algorithm was applied to classify drivers based on their responses in following-AV vs. following-HV with both constant speed and dynamic speed characteristics. The analyses of the differentiable appearance setting show that different drivers exhibit different behaviors in following-AV vs. following-HV: some are AV-believers, some are AV-skeptics, and the others are insensitive. Yet in the indifferentiable appearance setting, there is no significant difference between following a lead AV and following a lead HV. This reveals that drivers’ response to the lead vehicle depends on their subjective trusts on AV technologies rather than the actual driving behavior. The results suggest that, depending on the characteristics and composition of the drivers, classic car-following behavior in pure HV traffic may need to be updated for modeling mixed traffic in the near future.

An investigation on coordination of lane departure warning based on driver behaviour characteristics

As an important part of Advanced Driver Assistance Systems (ADAS), Lane Departure Warning System (LDWS) plays a significant role in lane departure prevention and reducing traffic accidents caused by lane departure. In order to improve the warning effect of the system as well as driver acceptance, this paper describes an LDWS algorithm for personalised driving assistance. The proposed combination algorithm consists of a multi-mode Time to Lane Crossing (TLC) and a Future Offset Distance (FOD) based on driver behaviour characteristics. To detect driver's lane change intention, the steering behaviour has been developed incorporating vehicle states and road curvature. Driving simulator tests are conducted to validate the lane departure warning algorithm with multi-mode based on TLC and FOD under various driving situations. The obtained test results are consistent with the expected performance.

An investigation on coordination of lane departure warning based on driver behaviour characteristics

As an important part of Advanced Driver Assistance Systems (ADAS), Lane Departure Warning System (LDWS) plays a significant role in lane departure prevention and reducing traffic accidents caused by lane departure. In order to improve the warning effect of the system as well as driver acceptance, this paper describes an LDWS algorithm for personalised driving assistance. The proposed combination algorithm consists of a multi-mode Time to Lane Crossing (TLC) and a Future Offset Distance (FOD) based on driver behaviour characteristics. To detect driver's lane change intention, the steering behaviour has been developed incorporating vehicle states and road curvature. Driving simulator tests are conducted to validate the lane departure warning algorithm with multi-mode based on TLC and FOD under various driving situations. The obtained test results are consistent with the expected performance.

Adaptations in driver behaviour characteristics during control transitions from full-range Adaptive Cruise Control to manual driving: an on-road study

Adaptive Cruise Control (ACC) can reduce traffic congestion and accidents. In dense traffic flow conditions and when changing lanes, drivers prefer to deactivate the ACC. These control transitions between automation and manual driving could impact driver behaviour characteristics. However, few studies have analysed the magnitude and duration of these adaptations. This research aims at quantifying the adaptations in speed, acceleration, distance headway and relative speed when drivers resume manual control. We collected driver behaviour data in an on-road experiment with full-range ACC during peak hours in Munich. We analysed these data using linear mixed-effects models to identify statistically significant changes in driver behaviour characteristics after drivers resumed manual control (transition period). The results reveal that the speed decreased significantly after the system was deactivated and it increased significantly after the system was overruled by pressing the gas pedal. These adaptations might have a substantial impact on traffic efficiency and safety.

A Study on an Anthropomorphic Car-Following Strategy Framework of the Autonomous Coach in Mixed Traffic Flow

In this study, to explore the demand characteristics of autonomous coaches in mixed traffic flow, two sections of an expressway were selected for vehicle experiments. With personification as the goal, sensors of the ego-vehicle were used to collect naturalistic car-following behavior data of surrounding coaches. After analyzing the naturalistic car-following data, car-driving behavior characteristics of the coach drivers were acquired. The analysis results indicate that the overall car-following processes of coaches tend to be relatively stable, and most of the car-following processes are in the conditions of higher velocity, smaller acceleration, and smaller relative velocity than the front vehicle. The expected following spacing and time headway of the coach drivers increase with the increase of velocity. By analysis and comparison with existing safe following models, it is found that while the safety and the warning rate based on the car-following coach safety distance model are higher, the warning rate of the car-following time headway model at the relatively higher velocity is lower. The coach safety distance model established in this study considers the car-following risk and the acceptance rate of the drivers to the warning system. This model conforms to the driver’s car-following characteristics and fulfills the requirements of collision avoidance. Applying the safety distance model to the autonomous coach would effectively improve the anthropopathic behavior characteristics of the autonomous coach. Additionally, the abnormality caused by the interference of other vehicles in mixed traffic flow of autonomous and human vehicles is successfully avoided under the premise of ensuring the safe driving of the autonomous coach.

Applying Derived Distribution Method to Microlevel Driving Behavior Characteristics to Quantify Uncertainties in Traffic Stream Flow and Density

AbstractThe flows and densities of traffic streams play an important role in defining the performance of roadways and corresponding improvement strategies. Traffic flows and densities are the outco...

A Study of Bad Driving Behavior Based on Improved K-Means Clustering and Neural Network

With the popularization of the Internet of Vehicles technology, excavating useful information from a large amount of driving behavior data to evaluate the driver's safe driving behaviors in real time, accurately and efficiently, is of great significance to improve the safety management level and transportation efficiency of road transportation process. Based on the driver's driving behavior characteristics and driving industry management standards, this paper proposes a bad driving behavior evaluation method based on improved k-means clustering and neural network. It uses the improved k-means clustering method to select typical sample points from the characteristic parameters extracted from the vehicle GPS positioning platform. The Backpropagation neural network algorithm is designed to learn the clustering results, and the online classification evaluation of bad driving behavior is realized. It provides a new direction for the transportation vehicle management department to carry out safety management of the road transportation process.

Traffic Safety Evaluation Approach for Reconstructed Freeway Medium Strip Opening Area

The remaining of the medium strip of the freeway in reconstruction and widening project reduces the cost of renovation, however brings potential driving risks. In this study, the characteristics of traffic flow changes caused by the opening of the medium strip were analyzed. The safety of the opening length of the medium strip was evaluated based on the interweaving and conflicting driving behavior characteristics, and the opening length of the interweaving area of the medium strip was calculated. In addition, the location of the exit traffic sign was advised for this new traffic flow characteristic. Appropriate opening length of the medium strip and the optimized traffic signs can effectively guide drivers to enter and exit the closed area of the central separation zone safely and comfortably.

Incorporating behavioral variables into crash count prediction by severity: A multivariate multiple risk source approach

The frequency and severity of traffic crashes have commonly been used as indicators of crash risk on transport networks. Comprehensive modeling of crash risk should account for both frequency and injury severity—capturing both the extent and intensity of transport risk for designing effective safety improvement programs. Previous research has revealed that crashes are correlated across severity categories because of the combined influence of risk factors, observed or unobserved. Moreover, crashes are the outcomes of a multitude of factors related to roadway design, traffic operations, pavement conditions, driver behavior, human factors, and environmental characteristics, or in more general terms: factors reflect both engineering and non-engineering risk sources. Perhaps not surprisingly, engineering risk sources have dominated the list of variables in the mainstream modeling of crashes whereas non-engineering sources, in particular, behavioral factors, are crucially omitted. It is plausible to assume that crash contributing factors from the same risk source affect crashes in a similar manner, but their influences vary across different risk sources. Conventional crash frequency modeling hypothesizes that the total crash count at any roadway site is well-approximated by a single risk source to which several explanatory variables contribute collaboratively. The conventional formulation is not capable of accounting for variations between risk sources; therefore, is unable to discriminate distinct impacts between engineering variables and non-engineering variables. To address this shortcoming, this study contributes to the development of multivariate multiple risk source regression, a robust modeling technique to model crash frequency and severity simultaneously.The multivariate multiple risk source regression method applied in this study can effectively capture the correlation between severity levels of crash counts while identifyinging the varying effects of crash contributing factors originated from distinct sources. Using crashes on Wisconsin rural two-lane highways, two risk sources – engineering and behavioral – were employed to develop proposed models. The modeling results were compared with a single equation negative binomial (NB) model, and a univariate multiple risk source model. The results show that the multivariate multiple risk source model significantly outperforms the other models in terms of statistical fit across several measures. The study demonstrates a unique approach to explicitly incorporating behavioral factors into crash prediction models while taking crash severity into consideration. More importantly, the parameter estimates provide more insight into the distinct sources of crash risk, which can be used to further inform safety practitioners and guide roadway improvement programs.

Effect of Characteristics of Driver Behavior on Traffic Safety

Effect of Characteristics of Driver Behavior on Traffic Safety

A graphical modeling method for individual driving behavior and its application in driving safety analysis using GPS data

Due to differences in driving skills and personal characteristics among drivers, the behaviors of drivers when faced with various driving environments differ, causing different levels of driving safety concerns. In past research, the measurement of safety-related driving behavior mostly focused on classification, while few studies were concerned with individual driving behavior characteristics. However, it is important for drivers to recognize and correct their dangerous behaviors and optimize their driving. This paper presents a graphical method for modeling individual driving behaviors, and the results can be used in driving safety analysis. Based on the assumption that drivers have specific driving habits, typical driving patterns during driving are first detected and extracted. These typical driving patterns are then sorted according to their frequencies, forming a driving behavior graph that can directly illustrate each driver’s behavior features. Furthermore, a quantitative analysis method for evaluating driving safety based on the behavior graph is provided. To verify the proposed method, a case study focusing on vehicles’ longitudinal motion was conducted using GPS data collected from Beijing taxis. The results demonstrated that the graphical method can describe the individual features of a driver’s longitudinal acceleration behavior and distinguish differences among drivers. The development of this method can help understand the individual features of driving behaviors and further support measures to optimize driving safety.

Analysis of Accident Severity for Curved Roadways Based on Bayesian Networks

Crashes that occur on curved roadways are often more severe than straight road accidents. Previously, most studies focused on the associations between curved sections and roadway geometric characteristics. In this study, significant factors such as driver behavior, roadway features, vehicle factors, and environmental characteristics are identified and involved in analyzing traffic accident severity. Bayesian network analysis was conducted to deal with data, to explore the associations between variables, and to make predictions using these relationships. The results indicated that factors including point of impact, site of location, accident side of road, alcohol/drugs condition, etc., are relatively critical in crashes on horizontal curves. Accident severity increases when crashes occur on bridges. The sensitivity of accident severity to vehicle use, traffic control, point of impact, and alcohol/drugs condition is relatively high. Moreover, a combination of negative factors will aggravate accident severities. The results also proposed some suggestions regarding the design of vehicles, as well as the construction and improvement of curved roadways.

Identifying motorcycle high-risk traffic scenarios through interactive analysis of driver behavior and traffic characteristics

Although the importance of human factors to crash occurrence has been demonstrated previously, the roles played by human factors in motorcycle killed and severely injured (KSI) crashes have remained unclear. One aim of our study is therefore to empirically determine the relative contribution of illegal behavior to motorcycle KSI crashes, conditional on real-world collisions between motorcycles and motor vehicles. Given that a crash is typically the synthetical result of human, vehicle, roadway, and environmental factors, another aim is to identify high-risk scenarios where inappropriate behavior is more likely to result in severe injuries for motorcyclists through interactions with other related factors. Based on a comprehensive dataset of 4587 police-reported crashes involving motorcycles during 2015–2017 in Hunan province, China, a data mining technique namely classification and regression tree was elaborately employed. Our results demonstrated the illegal behavior of the striking motor-vehicle drivers as one of the most dominant factors contributory to motorcycle KSI crashes, with a normalized importance value of 36.9%. We also confirmed collision object (i.e., collision with heavy or light vehicles) and helmet use of motorcyclists as determinants influencing motorcycle rider injury severities. Two types of extreme high-risk traffic scenarios were identified accordingly. A motorcycle rider was hit at weekends by a heavy motor-vehicle driver who was driving without license, driving a substantial vehicle, speeding, changing lanes illegally or driving in the wrong direction, and a motorcyclist was hit on weekdays by a heavy motor-vehicle driver aged 18–34 or 45–54, who was driving without license, driving a substantial vehicle, speeding, changing lanes illegally or driving in the wrong direction. Our findings are expected to shed more light on a deeper understanding of the illegal driving behavior as causation of motorcycle KSI crashes.

Impact of heterogeneity of car-following behavior on rear-end crash risk

An increasing number of vehicles travel on freeways result not only in traffic congestions but also accidents. Rear-end crashes in freeways can be collectively attributed to drivers, vehicles, and road infrastructure, but driving behavior plays a key role in influencing car-following safety. This study aims to investigate the impact of heterogeneity of driving behavior on rear-end crash risk. Driving behavior depends on perceived risk levels, acceleration and deceleration habits, and driver reaction characteristics. Thus, the influencing factors of rear-end crash risk were initially analyzed by using the desired safety margin (DSM) model. Subsequently, five driving behavior parameters, including upper and lower limits of DSM, sensitivity coefficients of acceleration and deceleration, and response time, were calibrated by using the vehicle trajectories from the Next Generation Simulation I-80 datasets. Simulation experiments were designed to evaluate the impact of heterogeneity of car-following behavior on rear-end crash risk. Results showed that decreasing the lower (or upper) limit of the DSM, increasing the response time, increasing the sensitivity coefficient for acceleration, or decreasing the sensitivity coefficient for deceleration can increase rear-end crash risk. In addition, if stable and unstable driving styles coexist, then their proportions have important influences on rear-end crash risk. These results imply that two critical factors affect shock waves, namely, driving behavior characteristics and proportion of different driving styles. Thus, a potential strategy for the adjustment of the proportions of unstable driving styles can attenuate shock waves and reduce rear-end crash risk to a certain extent. Moreover, a wide extent of driving behavior heterogeneity can attenuate shock waves and subsequently reduce rear-end crash risk. Overall, driving behavior heterogeneity has an important impact on rear-end crash risk. Exploring the effect of each driving behavior parameter on rear-end crash probability is useful for urban road traffic control, and it can provide improved understanding of abnormal driving behavior characteristics to minimize rear-end crash risks.

Simulated Analysis of Modeling of Driving Behavior Characteristics Based on Satellite Positioning Data

To promote the road transportation security, it’s necessary to study the modeling method of driving behavior characteristics. The traffic flow model realized by current modeling methods of driving behavior characteristics has a low accuracy in warning results. Therefore, based on satellite positioning data, a modeling method of driving behavior characteristics is proposed in this paper. Firstly, the dynamic model and kinetic model of traffic flow are built through the flow, speed and density parameters; then the response time, minimum safe distance and stability parameters of driving behavior are taken as the identification index of driving behavior to identify the driving behavior of drivers; according to the identification results, the psychological field theory and satellite positioning data are combined to build the model of driving behavior characteristics, and finally, warning the drivers according to their psychology and the actual situation of road. Experimental results demonstrate that the proposed method can accurately measure the traffic flow and speed, and the score of drivers’ behavior obtained has high accuracy, which verified again the high accuracy of traffic flow model and warning results of the proposed method.

Research on lane-keeping control strategy for bus

Owing to the frequent occurrence of commercial vehicle lane departure, a novel lane departure warning (LDW) system based on driver behaviour characteristics and a lane-keeping assist system (LKAS) based on electro-hydraulic steering system for buses are presented in this paper. Considering the differences between driver behaviour characteristics, a newly designed dynamic warning algorithm is proposed to predict the lane departure risk. Then, for the LKAS control, a simulation model of the electro-hydraulic steering system for buses is introduced for the active-steering-assist control and a multi-mode active control algorithm is designed for the lane keeping. To verify the effectiveness of the proposed control strategy, the co-simulation softwares of TruckSim, AMESim and MATLAB/Simulink are employed for the experiment validation. After extended simulation, it is found that the proposed LDW algorithm demonstrates an effective lane departure detection, and when the imminent lane departure is detected, the multi-mode active control algorithm can rapidly make adjustments to avoid the bus departing from the original lane.

Research on lane-keeping control strategy for bus

Owing to the frequent occurrence of commercial vehicle lane departure, a novel lane departure warning (LDW) system based on driver behaviour characteristics and a lane-keeping assist system (LKAS) based on electro-hydraulic steering system for buses are presented in this paper. Considering the differences between driver behaviour characteristics, a newly designed dynamic warning algorithm is proposed to predict the lane departure risk. Then, for the LKAS control, a simulation model of the electro-hydraulic steering system for buses is introduced for the active-steering-assist control and a multi-mode active control algorithm is designed for the lane keeping. To verify the effectiveness of the proposed control strategy, the co-simulation softwares of TruckSim, AMESim and MATLAB/Simulink are employed for the experiment validation. After extended simulation, it is found that the proposed LDW algorithm demonstrates an effective lane departure detection, and when the imminent lane departure is detected, the multi-mode active control algorithm can rapidly make adjustments to avoid the bus departing from the original lane.

Determination of priority stream volumes for capacity calculation of minor traffic streams for intersections with bending right-of-way

Abstract In Slovakia, unsignalized intersections with bending right-of-way (major street not leading straight, but bending) are a common part of the road network. Their specific mode of right-of-way results in different priority ranks of traffic streams and more complicated traffic situation. The current regulations do not provide procedures to determine the capacity of this kind of intersections. There are needed modifications to determine a priority traffic volume due to a new hierarchy of traffic stream priorities. The priority stream volume, together with driver behavior characteristics, is an important input for calculation of minor stream potential capacity. It is determined for each minor traffic stream individually and presents the sum of the conflicting major stream volumes. However, in some cases also potential conflicting traffic streams, when they share a lane with conflicting traffic streams, are considered because they affect minor-stream drivers. Therefore, for the new calculation proposal of priority stream volumes, the influences of these potentially conflicting streams have to be determined. For this purpose, the traffic surveys at two four-leg intersections with bending right-of-way were done. According to survey results, the potential conflicting major streams and their impact to the minor-stream drivers are determined and the proposal of priority stream volumes is established and presented in this paper.

Critical Gaps at Unsignalized Intersections with Bending Right-of-Way

The capacity calculation procedure for unsignalized intersections is based on the gap-acceptance theory in most of existing capacity regulations and it relies on one of the important parameters - critical gap. However, the capacity calculation procedure and values of critical gaps according to these regulations are valid only for intersections with standard right-of-way (major street leading straight). Nevertheless, in Slovakia, intersections with bending right-of-way (major street not leading straight, but bending) can be encountered. The specific mode of right-of-way results in different priority ranks of traffic movements (set by traffic rules of driving), more complicated traffic situation and therefore, different driver behaviour characteristics. To examine the gap acceptance behaviour of drivers under these specific conditions, an unsignalized four-leg intersection with bending right-of-way located in an urban area of Zilina, Slovakia, was selected. Three different methods (Raff, Wu, and MLM Troutbeck) were used for critical gap estimation from the field data. In the article, results of critical gaps for three through movements of different priority rank (major-street through movement of Rank 2 and minor-street through movements of Rank 3 and 4) are presented. The results show, that the values of critical gaps differ depending on the method by about 3-5 % only, which is not significant. Troutbeck ´s MLM method gives the highest values. The priority rank of movement has the greatest impact on the result. The values of critical gap for major-street through movement of Rank 2 are the smallest; they are approximately 1.3-2.1 s smaller than the values for minor-street through movements of Rank 3 or 4. The highest values of critical gap have been estimated for minor-street through movement of Rank 4 and they are higher compared to the current Slovak regulations TP 102 values for the same priority rank.

Assessment and Enhancement Frameworks for System Reliability Performance Using Different PEV Charging Models

This paper presents a comprehensive reliability framework for incorporating different plug-in electric vehicle (PEV) charging load models into the evaluation of generation adequacy. The proposed framework comprises special treatment and innovative models to achieve an accurate determination of the impact of PEV load models on reliability. First, a goodness-of-fit statistical model determines the probability distribution functions (PDFs) that best reflect the main characteristics of driver behavior. Second, robust and detailed stochastic methods are developed for modeling different charging scenarios (uncontrolled charging and charging based on time-of-use (TOU) pricing). These models are based on the use of a Monte Carlo simulation in conjunction with the fitted PDFs to generate and assess a large number of possible scenarios, while handling the uncertainties associated with driver behavior, penetration levels, charging levels, battery capacities, and customer response to TOU pricing. A novel reliability-based framework for the application of dynamic critical event call programs for use with PEV charging loads is also proposed. The effectiveness of the proposed framework with respect to improving system reliability is demonstrated using several case studies applied on the IEEE Reliability Test System.