Vehicle Driver Behavior Research Articles

Fuzzy modelling and inference for physics-aware road vehicle driver behaviour model calibration

Traffic simulation is a crucial tool for testing and evaluating hypotheses in both the design of autonomous driving systems and the optimization of urban transportation, where modelling and calibrating driver behaviour is key. In this article, we introduce a novel approach to calibrating driver models by combining fuzzy logic and inference to produce a plausible parametrization that, simultaneously, reproduces drivers’ behaviour peculiarities and road traffic characteristics. The calibration system uses solely vehicle trajectories that describe cars’ movements through the road network, with a resolution of time and space enough to allow for determining the lane of the road used, speed, and acceleration at every second of the survey period. This data is processed by a lightweight modelling and inference system capable of interpreting the input data through the lens of domain knowledge and physics laws in order to learn plausible driver model parameters using fuzzy computation. The outcome is a set of parameters for mathematical models which emulate the social and physical features of the behaviour of a driver-vehicle unit. Albeit this study limits itself to determining parameters for car-following and lane-change models, the systematic approach for calibration can be seen as a framework where also other driver model components can be calibrated (e.g. gap-acceptance). We evaluate our system on both synthetic and real data while comparing it with other approaches for calibration. We summarize our findings using a thorough analysis where we point out the most relevant aspects to consider when calibrating driver models.

Vehicle Trajectory Data Mining for Artificial Intelligence and Real-Time Traffic Information Extraction

It aims to improve the efficiency of information collection and extraction in the current intelligent transportation system, and accurately mine the vehicle trajectory data By using Artificial Intelligence (AI) and Deep Learning methods, the trajectory data generated during vehicle driving are deeply mined and analyzed, and the characteristics of driving behavior of vehicle drivers are modeled and analyzed in detail. Then, a method of mining driving behavior characteristics based on Convolutional Neural Network (CNN) and vehicle trajectory is proposed. Based on the mathematical principle of wavelet packet and Least Square Support Vector Machine (LSSVM), a combined model of trajectory mining is constructed and applied to the short-term prediction of traffic flow. The traffic flow of Binjiang Road and Renmin Road in Guangzhou, Guangdong Province from August 19 to August 21, 2021 is predicted to verify the accuracy of the trajectory mining combined model. The results show that the combination model of data mining has good fitting effect, and the average accuracy is above 0.8. Besides, the effectiveness of the Deep Learning model in driver behavior classification is verified. The accuracy of the classification model is 75.2% for trajectory, and that is 76.8% for driver behavior characteristics. It is of great significance to effectively utilize the knowledge data in Intelligent Transportation System (ITS) and extract valuable information from it, which has certain reference value for the subsequent refined prediction of vehicle behavior.

Trajectory Simulation of Emergency Vehicles and Interactions with Surrounding Traffic

Emergency services play an important role in the life of a city and are subject to constant public scrutiny. The efficient dispatch of emergency vehicles (EMVs) requires realistic shortest-path algorithms involving the movement of EMVs within an urban network under emergency conditions. Trip-time estimates used in shortest-path algorithms would be much more precise if it were possible to model more realistically the interactions between EMVs and surrounding traffic, as well as the reactions of other vehicles in the presence of an EMV. Therefore, EMV trajectories should be studied at the microscopic level to accurately model the impact of EMV travel along a path shared with other vehicles. In this research, we develop three models to incorporate specific non-EMV reactions associated with changing lanes, mounting the sidewalk, and approaching an intersection, plus two algorithms to actuate traffic lights at signalized intersections. These models and algorithms were coded in commercial microscopic traffic simulation software through the implementation of an application programming interface (API) designed to overcome the limitations of the software to realistically simulate disturbed traffic conditions and anomalous nonemergency vehicle driver behaviour observed in the presence of an EMV. Basic information about these real-world effects was gleaned from video footage recorded in Santiago, Chile, by traffic cameras, fire truck-mounted cameras, and truck-originated GPS pulses. To validate the design, a real EMV trip captured by the footage was simulated by the API. The simulation considerably reduced the degree of error in delineating the path followed by the EMV compared to the default simulations generated by most commercially available software, thereby demonstrating that the API can provide highly accurate estimates of EMV trip times in an emergency context.

Assessment of selected methods of estimating the maximum back-of-queue size on a signal-controlled intersection approach

The article presents the characteristics and evaluation of the accuracy of estimating the maximum length of the queue of vehicles at signalized intersections by commonly used methods in the world. The analyzes were based on the latest editions of the guidelines in the United States, Canada, Australia, Germany and Poland. In order to carry out accuracy analyzes, traffic tests were carried out at 5 intersection inlets in three different cities in Poland (Bydgoszcz, Torun, Warsaw), covering all phases of vehicle queue formation during individual periods of the signaling cycle (effective red and green signal). In total, the analysis had the results of tests from 81 hours of observation ~23,000 behaviors of vehicle drivers. Based on the analyzes it was found in particular: (1) slight differences in the construction of mathematical models of the considered calculation methods, with the exception of the US HCM model from 2016; (2) small errors in estimating the maximum queue length in unsaturated vehicle flow states (~3-5 vehicles/cycle and ~2-6 vehicles/15 minutes interval); (3) quite large errors in estimating the maximum queue length in saturated and oversaturated vehicle flow states (~11-16 vehicles/cycle and ~15-18 vehicles/15 minutes interval); (4) the main impact on estimation errors in oversaturation traffic states is not taking into account or incorrect determination of the 'so-called' initial queue length of the period preceding the analyzed period.

A Novel Driver Abnormal Behavior Recognition and Analysis Strategy and Its Application in a Practical Vehicle

In this work, a novel driver abnormal behavior analysis system based on practical facial landmark detection (PFLD) and you only look once version 5 (YOLOv5) were developed to solve the recognition and analysis of driver abnormal behaviors. First, a library for analyzing the abnormal behavior of vehicle drivers was designed, in which the factors that cause an abnormal behavior of drivers were divided into three categories according to the behavioral characteristics including natural behavioral factors, unnatural behavioral factors, and passive behavioral factors. Then, different neural network models were established through the representation of the actual scene of the three behaviors. Specifically, the abnormal driver behavior caused by natural behavioral factors was identified by a PFLD neural network model based on facial key point detection, and the abnormal driver behavior caused by unnatural behavioral factors and passive behavioral factors were identified by a YOLOv5 neural network model based on target detection. In addition, in a test of the driver abnormal behavior analysis system in an actual vehicle, the precision rate was greater than 95%, which meets the requirements of practical application.

Analysis of right-turn vehicular driving paths at uncontrolled T-intersections

Turning vehicle maneuvers highly influence vehicular safety at uncontrolled T-intersections due to their complexity and the inattentive behaviour of vehicle drivers. It was observed that the drivers while maneuvering from a minor road to a major road (viz., left-side driving in an Indian context) were distracted from conventional driving paths, which may considerably increase the risk of a crash. The present study examined the impact of various factors on driving path distractions. To fulfil the objective, a Binary Logit Model (BLM) and machine learning techniques, viz. Artificial Neural Networks (ANN), Support Vector Machines (SVM), and Random Forest (RF) models, were developed by considering driving path distractions as a dependent variable and the remaining set variables as independent variables. The model results revealed that driving path distractions are highly sensitive to the turning vehicle type, maneuvering speed of turning vehicles, running speed of through vehicles on a major road, through traffic volume, vehicle gap, waiting time, and right-turning traffic volume. Further, the study concluded that maneuvering speed, waiting time, and adequate available vehicular gaps significantly impact the behaviour of vehicular drivers when changing lanes. Field engineers could use the results of this study to start taking control measures at uncontrolled T-intersections.

Fuzzy Modeling and Inference for Physics-Aware Road Vehicle Driver Behavior Model Calibration

Fuzzy Modeling and Inference for Physics-Aware Road Vehicle Driver Behavior Model Calibration

Evidence of inconsistent results using current eye tracking glance and visit analysis standards

Eye tracking glance analyses have been used to study the behavior of motor vehicle drivers for at least two decades and eye tracking is now emerging as an experimental protocol in construction research. Previous studies have identified relationships between driver glance behavior and performance, so existing motor vehicle driver glance analysis methods were applied to search for analogous relationships with construction craftworkers during a model assembly task. Eye tracking analysis software vendors have yet to adopt standardized eye tracking event detection algorithms, and researchers have historically reported insufficient details to render most eye tracking analyses reproducible. As a result, the purpose of the present work is to address the gap in knowledge in understanding the challenges encountered as a result of a lack of standardization. The methodology applied existing glance analysis methods to a study incorporating mobile eye tracking glasses, which made the experimental environment less controlled than a typical glance analysis study using a screen-based eye tracking system, such as a motor vehicle driver behavior study. A new noise filtering parameter, maximum off-stimulus fixations, was also introduced. The results show that the absence of standardized eye tracking parameters, such as minimum fixation duration, can lead to different statistical results. The study's primary contribution to the body of knowledge is that is identifies the need for standardization of eye tracking analyses to better ensure that future eye tracking studies be more consistent and reproducible.

Virtual Pseudonym-Changing and Dynamic Grouping Policy for Privacy Preservation in VANETs.

Location privacy is a critical problem in the vehicular communication networks. Vehicles broadcast their road status information to other entities in the network through beacon messages. The beacon message content consists of the vehicle ID, speed, direction, position, and other information. An adversary could use vehicle identity and positioning information to determine vehicle driver behavior and identity at different visited location spots. A pseudonym can be used instead of the vehicle ID to help in the vehicle location privacy. These pseudonyms should be changed in appropriate way to produce uncertainty for any adversary attempting to identify a vehicle at different locations. In the existing research literature, pseudonyms are changed during silent mode between neighbors. However, the use of a short silent period and the visibility of pseudonyms of direct neighbors provides a mechanism for an adversary to determine the identity of a target vehicle at specific locations. Moreover, privacy is provided to the driver, only within the RSU range; outside it, there is no privacy protection. In this research, we address the problem of location privacy in a highway scenario, where vehicles are traveling at high speeds with diverse traffic density. We propose a Dynamic Grouping and Virtual Pseudonym-Changing (DGVP) scheme for vehicle location privacy. Dynamic groups are formed based on similar status vehicles and cooperatively change pseudonyms. In the case of low traffic density, we use a virtual pseudonym update process. We formally present the model and specify the scheme through High-Level Petri Nets (HLPN). The simulation results indicate that the proposed method improves the anonymity set size and entropy, provides lower traceability, reduces impact on vehicular network applications, and has lower computation cost compared to existing research work.

A Back-of-Queue Model of a Signal-Controlled Intersection Approach Developed Based on Analysis of Vehicle Driver Behavior

In smart cities, it is expected that transport, communication as well as the movement of people and goods will take place in the shortest possible time while maintaining a high level of safety. In recent years, due to the significant increase in the number of passengers and vehicles on the road and the capacity limitations of transport networks, it has become necessary to use new technologies for intelligent control and traffic management. Intelligent transport systems use advanced technologies in the field of data gathering, information processing, and traffic control to meet current transport needs. To be able to effectively control and manage road traffic, it is necessary to have reliable mathematical models that allow for a faithful representation of the real traffic conditions. Models of this type are usually the basis of complex algorithms used in practice in road traffic control. The application of appropriate models reflecting the behavior of road users contributes to the reduction of congestion, the vehicles travel time on the transport network, fuel consumption and the emissions, which in turn support broadly understood energy savings. The article proposes a model that allows for the estimation of the maximum queue size at the signal-controlled intersection approach (so-called: maximum back-of-queue). This model takes into account the most important traffic characteristics of the vehicles forming this queue. The verification allowed for the conclusion that the proposed model is characterized by high compliance with the actual traffic and road conditions at the intersections with signal controllers located in built-up areas in Poland. The obtained compliance confirms the possibility of using the model for practical applications in calculating the maximum back-of-queue at signal-controlled intersections located in built-up areas in Poland.

Impact on average vehicle speed with the introduction of educational actions and optical character recognition equipment in the Federal District, Brazil

Every year, traffic accidents leave millions of people dead and injured worldwide. Speeding is one of the factors that most contribute to accident severity. The implementation of electronic road speed control equipment promotes a reduction in traffic accidents and their effects. Despite the effectiveness of those devices, some drivers have the habit of braking before and accelerating after checkpoints. A possible solution for the problem of that speeding-up effect and its consequences may be through controlling the average speed. The objective of this work is to verify the behavior of vehicle drivers regarding the average speed between control points when subjected to educational actions. For this, two educational actions were applied to sections of three highways of the Federal District, Brazil. The first action warned vehicle drivers that there was equipment on the highway for monitoring average speed between checkpoints and the second action was the sending of correspondence to vehicle owners who still traveled above the tolerated average speed, after the first educational action. The results of the second educational action indicated very little variation in the average speed when compared with the data of the first educational action.

A Novel Generation-Adversarial-Network-Based Vehicle Trajectory Prediction Method for Intelligent Vehicular Networks

Prediction of the future location of vehicles and other mobile targets is instrumental in intelligent transportation system applications. In fact, networking schemes and protocols based on machine learning can benefit from the results of such accurate trajectory predictions. This is because routing decisions always need to be made for the future scenario due to the inevitable latency caused by the processing and propagation of the routing request and response. Thus, to predict the high-precision trajectory beyond the state of the art, we propose a generative adversarial network (GAN)-based vehicle trajectory prediction method, GAN-VEEP, for urban roads. The proposed method consists of three components: 1) vehicle coordinate transformation for data set preparation; 2) neural network prediction model trained by GAN; and 3) vehicle turning model to adjust the prediction process. The vehicle coordinate transformation model is introduced to deal with the complex spatial dependence in the urban road topology. Then, the neural network prediction model learns from the behavior of vehicle drivers. Finally, the vehicle turning model can refine the driving path based on the driver’s psychology. Compared with its counterparts, the experimental results show that GAN-VEEP exhibits higher effectiveness in terms of the average accuracy, mean absolute error, and root-mean-squared error.

Big data platform of traffic violation detection system: identifying the risky behaviors of vehicle drivers

Since the traffic data has a high volume, high diversity and high speed of production, the traditional systems cannot process them accurately. In this paper, a big data based system was designed and implemented for identifying the offenders’ behaviors. The proposed Traffic Violation Detection System (TVD system) included four main phases which were described using the MAPE methodology. In the monitor phase, unstructured data, such as videos captured by traffic control cameras as well as the images and the descriptions provided by traffic officers, were collected. In the analysis phase, the knowledge base of unsafe driving behaviors was created and classified. In this phase, a standard Work Breakdown Structure for unsafe behaviors was created by the experts in traffic control. In the Plan phase, in order to detect unsafe driving behaviors related to police descriptions for collected images, and to detect unsafe behaviors from video cameras, Behavior-Based Safety process with Map/ Reduce technique and Vector Space Model (VSM) were employed. In the last phase, all types of data, including the structured data and the multimedia/unstructured data, along with the types and details of violations, were stored on the Hadoop Distributed File System. The prototype of the proposed TVD system was successfully implemented for a few common violations. The results showed that by applying big data technologies, the driving violations could be detected more accurately using the combination of the structured and unstructured data. The results indicate that compared to the sequential program, Hadoop only with a single slave-node decreases the processing time of big data by more than 70%. Also, by increasing the number of slave nodes from 1 to 7 in the police descriptions and images of surveillance cameras, the processing time reduces by 60.87% and 70%, respectively. Thus, the TVD performance increases by more than 75% as the number of data nodes boosts. Based on the results, it can be concluded that by identifying unsafe driving behaviors, it is possible to diminish traffic accidents and the damage caused by them at a satisfactory level. Also, the authors decided to compare these studies in terms of qualitative criteria, such as fieldwork and behavioral identification, and quantitative criteria.

Road safety in Poland, the Czech Republic and Slovakia. Presentation of international research results

The safety of road users is one of the most important elements of the functioning of every society. The data published by the European Commission shows that on roads in the EU in 2017 was 25.3 thousand deaths occuried in road accidents. To reduce the number of road accident victims, threats in this area should be identified and appropriate solutions implemented. The article presents factors in traffic accidents, elements of human mentality which have a significant impact on driving a vehicle, factors influencing a driver’s action in traffic and correlation of the system of safety management including the procedure of risk management. The article presents partial results of international scientific research carried out in Poland, Czech Republic and Slovak Republic. In the survey participated respondents from ten cities. Opinions were obtained from 642 people on the subject of individual personality factors determine the behaviour of vehicle drivers. Furthermore were presented respondents´ answers regarding legal changes relating to the proposal penalties for drivers participating in road traffic under the influence of alcohol and drugs. The article has been prepared within the frames of realizing a research project called „Simulator supporting police officers’ training in the implementation of activities during a road accident” no. DOB-BIO9/06/01/2018, which has received funding from the Polish National Centre for Research and Development.

A Novel Driver Performance Model Based on Machine Learning

Models of road vehicle driver behaviour are widely used in several disciplines, like driver distraction and autonomous driving. In this paper, a novel driver performance model, which is unique for every driver, is introduced. The driver is modelled with machine learning algorithms, namely artificial neural network and adaptive neuro-fuzzy inference system. Every model is trained and validated with the data collected during the real-time driver-in-the-loop experiment on a vehicle simulator for each driver separately. In total, 18 participants contributed to the experiment. Although the prediction accuracy of the models depends on the algorithm specifications, the artificial neural network was slightly more accurate in driver performance prediction comparing to the adaptive neuro-fuzzy inference system. The driver models may be used in detection of driver distraction induced by in-vehicle information system.

Повышение качества транспортных услуг при автоперевозках населения на дальние расстояния

The article justifies the role of self-control by transportation organizations and monitoring by tour operators and insurance companies of the behavior of drivers, their health and technical condition of vehicles. The article reveals the methodology of improving the quality of social services for the public by monitoring the behavior, health of drivers of vehicles and the technical state of vehicles in longdistance, international transport of the population (including tourists). The methodology is aimed at identifying possible dangers, discomfort in the process of providing services to the population and proactive purposeful actions to eliminate or minimize the negative consequences. The methodology contains a set of measures and recommendations that improve the quality of tourism road transportation, and allow car carriers, tour operators and travel agents significantly to reduce financial costs by avoiding traffic accidents through the fault of drivers of car carriers. The author develops and analyzes the concepts of behavior, the health status of drivers of the vehicle, the signs of their manifestation. The methodology includes recommendations for transportation organizations, tour operators (travel agents) for tests of the behavior of vehicle drivers through a «mysterious passenger». The author suggests ways to organize quality control of pre-trip medical examination of drivers. The author recommends for auto carriers introduce measures aimed at maintaining the normal technical condition of vehicles. The practical significance of the study lies in the possibility of implementing the methodology in the work of auto carriers, tour operators, travel agents not only in the Russian Federation, but also the host party abroad.

A new insight on the risky behavior of motorists at railway level crossings: An observational field study

Accidents at railway level crossings (LXs) give rise to serious material and human damage. Particularly, collisions between trains and motorized vehicles are the most critical accidents occurring at LXs. It is worth noticing that violations committed by vehicle drivers are the primary cause of such accidents. The present study is a tentative to acquire a better understanding of risky behavior of vehicle drivers while crossing LXs during the closure cycle. Namely, risk analysis based on field measurement conducted at four automated LXs with two half barriers is performed. We focus on vehicle driver behavior during the LX closure cycle while distinguishing between different phases. In fact, the closure cycle is divided into three phases which are “Ph2 Red Flash and Siren”, “Ph3 Barriers Coming Down” and “Ph4 Barriers Down”; and vehicle driver behavior in each phase as time increases is scrutinized respectively. Particularly, zigzag scenarios are detected, using an original experimental setting that we have implemented, and analyzed in detail. The main findings based on the analysis demonstrate that the peak of violation rate in the morning is later than the actual rush hour in the morning; a distinct peak of the violation rate shows on Friday, while the violation rate on weekend is fairly low; the relative violation rate of vehicles with high speed decreases continuously as time advances from Ph2 to Ph3 in the daytime; the violation rate during Ph4 decreases as Ph4 duration is prolonged, which contradicts a general speculation that a higher rate of zigzag violations would appear as the duration of Ph4 is extended. These findings open the way towards determining the impacting factors which have an important contribution to the vehicle driver decision-making in this context (e.g., traffic density, time schedule and phase duration). In addition, the outputs of the present study are conducive to identifying potential interventions to improve safety at LXs.

Simulation of electric vehicle driver behaviour in road transport and electric power networks

The integration of electric vehicles (EVs) will affect both electricity and transport systems and research is needed on finding possible ways to make a smooth transition to the electrification of the road transport. To fully understand the EV integration consequences, the behaviour of the EV drivers and its impact on these two systems should be studied. This paper describes an integrated simulation-based approach, modelling the EV and its interactions in both road transport and electric power systems. The main components of both systems have been considered, and the EV driver behaviour was modelled using a multi-agent simulation platform. Considering a fleet of 1000EV agents, two behavioural profiles were studied (Unaware/Aware) to model EV driver behaviour. The two behavioural profiles represent the EV driver in different stages of EV adoption starting with Unaware EV drivers when the public acceptance of EVs is limited, and developing to Aware EV drivers as the electrification of road transport is promoted in an overall context. The EV agents were modelled to follow a realistic activity-based trip pattern, and the impact of EV driver behaviour was simulated on a road transport and electricity grid. It was found that the EV agents’ behaviour has direct and indirect impact on both the road transport network and the electricity grid, affecting the traffic of the roads, the stress of the distribution network and the utilization of the charging infrastructure.

The many reasons your mileage may vary: Toward a unifying typology of eco-driving behaviors

The role of vehicle driver behavior has been ignored in prior energy and environmental policy making. Laboratory procedures that produce the fuel economy estimates posted on every new car sold in the US are designed to preclude the effects of differences between drivers. Yet, every vehicle states the caveat, “Actual results will vary for many reasons, including driving conditions and how you drive and maintain your vehicle.” Eco-driving as means of strategically taking advantage of this variability has been inconsistently defined in conceptual analyses and variously operationalized in empirical analyses. The present research clarifies, synthesizes, and expands on prior definitions of eco-driving to develop a comprehensive and precise definition and typology of eco-driving behaviors. The resultant typology includes six mutually exclusive classes of behavior: driving, cabin comfort, trip planning, load management, fueling, and maintenance. This typology establishes a basis for systematic research to determine energy and climate impacts and develop effective policies and interventions for different types of eco-driving.

Effect of a community-based pedestrian injury prevention program on driver yielding behavior at marked crosswalks

BackgroundFew studies have comprehensively evaluated the effectiveness of multi-faceted interventions intended to improve pedestrian safety. “Watch for Me NC” is a multi-faceted, community-based pedestrian safety program that includes widespread media and public engagement in combination with enhanced law enforcement activities (i.e., police outreach and targeted pedestrian safety operations conducted at marked crosswalks) and low-cost engineering improvements at selected crossings. The purpose of this study was to estimate the effect of the law enforcement and engineering improvement components of the program on motor vehicle driver behavior, specifically in terms of increased driver yielding to pedestrians in marked crosswalks. MethodsThe study used a pre-post design with a control group, comparing crossing locations receiving enforcement and low-cost engineering treatments (enhanced locations) with locations that did not (standard locations) to examine changes in driver yielding over a 6-month period from 2013 to 2014. A total of 24,941 drivers were observed in 11,817 attempted crossing events at 16 crosswalks in five municipalities that were participating in the program. Observations of real pedestrians attempting to use the crosswalks (“naturalistic” crossing) were supplemented by observations of trained research staff attempting the same crossings following an established protocol (“staged” crossings). Generalized estimating equations (GEE) were used to model driver yielding rates, accounting for repeated observations at the crossing locations and controlling other factors that affect driver behavior in yielding to pedestrians in marked crosswalks. ResultsAt crossings that did not receive enhancements (targeted police operations or low-cost engineering improvements), driver yielding rates did not change from before to after the Watch for Me NC program. However, yielding rates improved significantly (between 4 and 7 percentage points on average) at the enhanced locations. This was true for both naturalistic and staged crossings. ConclusionsThis study provides evidence that enhanced enforcement and low-cost engineering improvements, as a part of a broader program involving community-based outreach, can increase driver yielding to pedestrians in marked crosswalks. These data are important for the staff and decision-makers involved in pedestrian safety programs to gain a better understanding of the different engineering and behavioral mechanisms that could be used to improve driver yielding rates.