Driving Behavior Model Research Articles

A Visual Search Model for In-Vehicle Interface Design

As in-vehicle infotainment systems gain new functionality, their potential to distract drivers increases. Searching for an item on interface is a critical concern because a poorly designed interface that draws drivers’ attention to less important items can extend drivers’ search for items of interest and pull attention away from roadway events. This potential can be assessed in simulator-based experiments, but computational models of driver behavior might enable designers to assess this potential and revise their designs more quickly than if they have to wait weeks to compile human subjects data. One such model, reported in this paper, predicts the sequence of eye fixations of drivers based on a Boolean Map-based Saliency model augmented with top-down feature bias. Comparing the model predictions to empirical data shows that the model can predict search time, especially in cluttered scenes and when a target item is highlighted. We also describe the integration of this model into a web application ( http://distraction.engr.wisc.edu/ ) that can help assess the distraction potential of interface designs.

Modeling the Dynamics of Driver’s Dilemma Zone Perception Using Agent Based Modeling Techniques

Several research efforts investigated and targeted the issue of reducing dilemma zone (DZ) related crashes (where drivers upstream of intersections are uncertain about their decision to stop or go at the onset of yellow). One of the important unanswered questions in the literature is whether driver's perception of DZ changes individually as a function of their safe and unsafe past experience at similar situations. This paper investigates the use of agent-based methods in capturing the effect of driver's learning/dynamic perception of DZ. A driving simulator was used to collect driver behavior data. An actor-critic reinforcement learning algorithm was implemented to model the dynamic behavior of driver in dilemma zone. Fuzzy logic is used to partition traffic state variables and a reinforcement learning technique is used in policy calibration and update. The study results show a close matching between the driver's action from the driving simulator and the model output. The research reported here contributes to improved modeling of driver definition and behavior in dilemma zone, which will have significant impacts on the design of optimal control methods and the assessment of intersection safety. Moreover, it lays the groundwork for several subsequent simulator studies and scenario development in driving simulator to investigate the drivers' behavior at signalized intersections. Keywords: Dilemma zone; machine learning; dynamic perception; driving simulator Copyright c 2016 SERSC Language: en

Driving Behaviors: Models and Challenges for Non-Lane Based Mixed Traffic

Most published microscopic driving behavior models, such as car following and lane changing, were developed for homogeneous and lane-based settings. In the emerging and developing world, traffic is characterized by a wide mix of vehicle types (e.g., motorized and non-motorized, two, three and four wheelers) that differ substantially in their dimensions, performance capabilities and driver behavior and by a lack of lane discipline. This paper presents a review of current driving behavior models in the context of mixed traffic, discusses their limitations and the data and modeling challenges that need to be met in order to extend and improve their fidelity. The models discussed include those for longitudinal and lateral movements and gap acceptance. The review points out some of the limitations of current models. A main limitation of current models is that they have not explicitly considered the wider range of situations that drivers in mixed traffic may face compared to drivers in homogeneous lane-based traffic, and the strategies that they may choose in order to tackle these situations. In longitudinal movement, for example, such strategies include not only strict following, but also staggered following, following between two vehicles and squeezing. Furthermore, due to limited availability of trajectory data in mixed traffic, most of the models are not estimated rigorously. The outline of modeling framework for integrated driver behavior was discussed finally.

Ethnic groups differences in regard to social networks, daily activity patterns, and driving behavior

Research on the impact of the social environment on travel behavior and driving behavior has proliferated. Yet, the individual-level mechanisms involved remain under-theorized and understudied. Studies usually focus on how drivers are influenced directly through the presence of passengers or indirectly through the norms of relevant social groups. The current study introduces a novel dimension by examining the impact of interpersonal ties on daily activity patterns and driving behavior. We analyze the results of a large telephone survey carried out in Northern Israel (n=517). Respondents include Arabs (58%) and Jews (42%), roughly corresponding to the demographic composition of Northern Israel. The results show that individuals with interpersonal ties characterized with a higher frequency of contacts were less likely to engage in leisure trips outside their community, thus reducing their exposure to high-risk driving environments. We found an inverse association between social interactions and speeding violations, although it was only significant for the Arab group, and inverse association between speeding violations and internal leisure trips for the whole sample. Married persons reported significantly higher social interactions than unmarried persons. Arab women have significantly more social interactions than Arab men, and older Arabs reported higher social interactions than younger Arabs, but no gender or age differences were found within the Jewish group. Finally, men drove for longer periods while women tended to have more complex trips. These results stress the need to make explicit the links between driving behavior models and travel behavior models using social variables.

DESIGNING OF DRIVER BEHAVIOR MODEL AND ITS DATA ANALYSIS FOR VEHICULAR APPLICATIONS

Design and Implementation of Evidence Collection System from Event Data Recorder”, deals with collection of Evidence for accident analysis. These evidences are in form of graphs. It includes graph of Speed Vs Time, Engine Temperature Vs Time, Distance Vs Time, Brake Pressure Vs Time, Engine Speed (RPM) Vs Time, Impact Vs Time and Tire Pressure Vs Time. Also it provides single graph merged with all these parameters at the time of accident. One more advantage of system is that, it can also provide Driver Behavior Model. By using this, driver’s driving performance could be found out. This system is nothing but the GUI constructed using Microsoft Visual Studio. All graphs are plotted using this Graphical User Interface. For Data Analysis purpose, B-tree algorithm is used here. Its performance for secondary storage devices and huge data handling capacity makes it more efficient. Data recorded by Event Data Recorder is used here as input. Event Data Recorder is also called as Black box of vehicles. It records vehicular data like speed, date, time, location (latitude, longitude), alcohol content of driver, airbag deployment etc. This data plays an important role detection and analysis of accident if any persists. These EDR are controlled by processors for very accurate operations.

An impact analysis of traffic image information system on driver travel choice

A driver is one of the main components in a transportation system that influences the effectiveness of any active demand management (ADM) strategies. As such, the understanding on driver behavior and their travel choice is crucial to ensure the successful implementation of ADM strategies in alleviating traffic congestion, especially in city centres. This study aims to investigate the impact of traffic information dissemination via traffic images on driver travel choice and decision. A relationship of driver travel choice with respect to their perceived congestion level is developed by an integrated framework of genetic algorithm–fuzzy logic, being a new attempt in driver behavior modeling. Results show that drivers consider changing their travel choice when the perceived congestion level is medium, in which changing departure time and diverting to alternative roads are two popular choices. If traffic congestion escalates further, drivers are likely to cancel their trip. Shifting to public transport system is the least likely choice for drivers in an auto-dependent city. These findings are important and useful to engineers as they are required to fully understand driver (user) sensitivity to traffic conditions so that relevant active travel demand management strategies could be implemented successfully. In addition, engineers could use the relationships established in this study to predict drivers’ response under various traffic conditions when carrying out modeling and impact studies.

Towards Behavioral Modeling of Drivers in Mixed Traffic Conditions

Mixed traffic conditions characterized by the presence of different vehicle types and weak lane discipline are common in the developing world. Different maneuvering capabilities of different vehicle types lead to vehicle-type dependent longitudinal and lateral movement driving behaviors. Weak lane discipline allows drivers to simultaneously look for possible lateral movements while progressing longitudinally. This integrated driving behavior gives rise to various driving phenomena like tailgating, multiple-leader following, swerving and filtering. However, very few studies exist in the literature that attempt to model these typical driving behaviors of mixed traffic streams. Moreover, most of these studies are not often readily available. Thus, a comprehensive review of the studies that consider modeling driver behavior in mixed traffic conditions is presented in this paper. Often, models developed for mixed traffic conditions stemmed out of the basic principles employed in the development of homogeneous traffic models. Hence, this paper first reviews the lane-based macro-, micro- and meso-scopic models in tandem and identifies their strengths and weaknesses in describing the mixed traffic system. A thorough understanding and conceptualization of any system is possible by analyzing large sets of empirical data. Therefore, a brief review on traffic data requirements to understand and model the driver behavior and, data extraction tools to obtain the trajectories of vehicles is presented next. Driver behavioral models form the core of any microscopic traffic simulation model. Core behavioral models such as car-following and lane-changing models which describe the longitudinal and lateral movements of drivers respectively are reviewed next. However, these two behavioral models may not be able to describe the integrated driving behavior independently. Therefore, various integrated driver behavioral models are briefed next. A simulation framework is essential to implement the driver behavioral models and evaluate their capabilities in representing larger scale traffic systems. Hence, a brief review of various simulation frameworks is presented subsequently. Finally, the challenges that are involved in modeling driver behavior in mixed traffic conditions are discussed and some useful research directions are proposed.

Modeling Left-Turn Driving Behavior at Signalized Intersections with Mixed Traffic Conditions

In many developing countries, mixed traffic is the most common type of urban transportation; traffic of this type faces many major problems in traffic engineering, such as conflicts, inefficiency, and security issues. This paper focuses on the traffic engineering concerns on the driving behavior of left-turning vehicles caused by different degrees of pedestrian violations. The traffic characteristics of left-turning vehicles and pedestrians in the affected region at a signalized intersection were analyzed and a cellular-automata-based “following-conflict” driving behavior model that mainly addresses four basic behavior modes was proposed to study the conflict and behavior mechanisms of left-turning vehicles by mathematic methodologies. Four basic driving behavior modes were reproduced in computer simulations, and a logit model of the behavior mode choice was also developed to analyze the relative share of each behavior mode. Finally, the microscopic characteristics of driving behaviors and the macroscopic parameters of traffic flow in the affected region were all determined. These data are important reference for geometry and capacity design for signalized intersections. The simulation results show that the proposed models are valid and can be used to represent the behavior of left-turning vehicles in the case of conflicts with illegally crossing pedestrians. These results will have potential applications on improving traffic safety and traffic capacity at signalized intersections with mixed traffic conditions.

Modeling Driver Behavior in a Connected Environment: Integrated Microscopic Simulation of Traffic and Mobile Wireless Telecommunication Systems

Connected and autonomous vehicles will influence the future of the surface transportation system by enhancing safety, improving mobility, reducing energy consumption, and controlling emissions. The real-time information provided by connected vehicles technology allows drivers to become more aware of the surrounding traffic conditions and to maneuver safely and more efficiently. Furthermore, when coupled with onboard sensing technologies, the connected vehicles technology can improve the efficiency and the reliability of a driverless transportation network. This paper presents a comprehensive simulation framework to model driver behavior in a connected driving environment with connected vehicles. The framework, which consists of a microscopic traffic simulator integrated with a discrete-event communications network simulator, Network Simulator 3, forms a basis for exploration of the properties of the resulting traffic systems and assessment of the system-level impacts of these technologies. Furthermore, the connectivity of a vehicle-to-vehicle and vehicle-to-infrastructure communications network was investigated with the FHWA Next Generation Simulation: US-101 Highway data set (to represent vehicular movements in a highway environment). It was found that signal interference can result in information loss and partial connectivity. Finally, through the implementation of a speed harmonization algorithm, the paper discusses the importance of consideration of telecommunications along with vehicular movements to investigate the effects of connected vehicle applications on mobility and emissions.

A Driving Behaviour Model of Electrical Wheelchair Users.

In spite of the presence of powered wheelchairs, some of the users still experience steering challenges and manoeuvring difficulties that limit their capacity of navigating effectively. For such users, steering support and assistive systems may be very necessary. To appreciate the assistance, there is need that the assistive control is adaptable to the user's steering behaviour. This paper contributes to wheelchair steering improvement by modelling the steering behaviour of powered wheelchair users, for integration into the control system. More precisely, the modelling is based on the improved Directed Potential Field (DPF) method for trajectory planning. The method has facilitated the formulation of a simple behaviour model that is also linear in parameters. To obtain the steering data for parameter identification, seven individuals participated in driving the wheelchair in different virtual worlds on the augmented platform. The obtained data facilitated the estimation of user parameters, using the ordinary least square method, with satisfactory regression analysis results.

Steer or Brake?: Modeling Drivers’ Collision-Avoidance Behavior by Using Perceptual Cues

Driver models have been developed to capture collision-avoidance behaviors, yet there is a lack of understanding of what perceptual processes influence drivers’ choices to brake or steer. A statistical model of these decisions was developed with cluster analysis and multinomial logistic regression with data from a simulator study of drivers’ responses to rear-end collisions. Drivers’ choices of responses were clustered on the basis of the maximum values of the magnitude of braking and steering forces, starting from the time at which the driver looked back to the road, just before initiating the avoidance maneuver, to the end of the maneuver. The clusters identified three types of responses: medium to high braking with medium to high steering, medium to high braking with mild steering, and mild braking with medium to high steering. The perceptual variables such as the optical angle, the expansion rate of the optical angle, and their ratio were used to predict the drivers’ choice of response. The results show that, of the perceptual variables, the combination of optical angle and tau performs as well as or better than others in predicting the choice of response. The mode and timing of an alert from a collision-warning system did not influence the drivers’ choices. These results can inform driver behavior models to guide design and assess benefits of advanced driver assistance systems.

Driver Behavior Characteristics Identification Strategies Based on Bionic Intelligent Algorithms

The exact and reliable understanding of driver behavior characteristics has a large potential contribution to the active safety control system, advanced driver assistance system, and intelligent traffic system. This study presents a driver behavior characteristics identification strategy based on bionic intelligent algorithms. First, a driver behavior data acquisition system is established to be used with test subjects of different driving skill levels for driving data acquisition. The preview optimal curvature model, which directly reflects the driver's behavior characteristics based on its parameters, is then chosen as the ideal driver behavior model and is identified through the genetic algorithm, particle swarm optimization, and back propagation neural network. Moreover, this paper discusses the application of driver behavior characteristics identification in an integrated chassis control system (ICC), which integrates active front steering with electronic stability control. Finally, simulations are performed to verify the identification results and the application in the ICC by PanoSim and MATLAB/Simulink software. Based on the results, the proposed identification strategy precisely characterized the driver behavior. As a result, its application in the ICC improved the path-following ability and vehicle stability performance of the drivers.

Cohort-Based Analysis Structure for Modeling Driver Behavior with an In-Vehicle Data Recorder

In recent years, the increasing availability of in-vehicle data recorder technologies has resulted in more safety studies that have unobtrusively observed the behavior of drivers during actual driving. The research in this paper proposes and applies a flexible analysis structure that can be applied to data that reflect events of interest—including crash data and safety critical event data—from naturalistic driving studies and alert warnings from onboard safety warning systems. The fundamental requirements for the data include a need for GPS and a geographic information system to allow positioning the vehicle on a network and the occurrence and recording of events of interest with attributes, which may include those reflecting the roadway, environment, driver, and event. A distinguishing feature of the formulation is the explicit inclusion of exposure to risk. The analysis approach responds to four challenges facing researchers using in-vehicle data recorder data to assess driving behavior with surrogate safety performance measures: inclusion of exposure to risk; allowance for inclusion of driver, event, environment, and roadway attributes in a structured formulation; modeling of driving exposure in different contexts, including both roadway and environment; and allowance for comparison to a baseline hazard where no events occur. A hierarchical mixed-effects Poisson regression model is used with data from University of Michigan Transportation Institute’s curve speed warning system field operation test in the application of the method. The model identifies drivers who have significantly higher alert frequency than similar drivers, both for all driving and for driving on major and minor arterials. The paper concludes with a discussion of potential applications beyond field operational test settings, which include analysis of crash and other naturalistic driving study–observed events.

Categorization of the lane change decision process on freeways

Traffic operations for new road layouts are often simulated using microscopic traffic simulation packages. These traffic simulation packages usually simulate traffic on freeways by a combination of a car-following model and a lane change model. The car-following models have gained attention of researchers and are well calibrated versus data. The proposed lane change models are often representations of assumed reasonable behavior, not necessarily corresponding to reality. The current simulation packages apply solely one specific type of model for car-following or lane changing for all vehicles during the simulation. This paper investigates the decision process of lane changing maneuvers for a variety of drivers based on a two-stage test-drive. Participants are asked to take a drive on a freeway in the Netherlands in a camera-equipped vehicle. Afterwards, the drivers are asked to comment on their choices related to lane and speed choice, while watching the video. This paper reveals that different drivers have completely different strategies to choose lanes, and the choices to change lane are related to their speed choice. Four distinct strategies are empirically found. These strategies differ not only in parameter values, as is currently being modeled in most simulation packages, but also in their reasoning. Most remarkably, all drivers perceive their strategy as an obvious behavior and expect all other drivers to drive in a similar way. In addition to the interviews of the participants in the test-drive, 11 people who did not take part in the experiment were interviewed and questioned on lane change decisions. Moreover, the findings of this study have been presented to various groups of audience with different backgrounds (about 150 people). Their comments and feedback on the derived driving strategies have added some value to this study. The findings in this paper form a starting point for developing a novel lane change model which considers four different driving strategies among the drivers on freeway. This is a significant contribution in the area of driving behavior modeling, since the existing microscopic simulators consider only one type of lane change models for all drivers during the simulation. This could lead to significant changes in the way lane changes on freeways are modeled.

Impacts of SOC on car-following behavior and travel time in the heterogeneous traffic system

Since the SOC (state of charge) of the battery of each electric vehicle directly determines whether the battery should be charged/swapped, the SOC may affect the electric vehicle’s driving behavior. In this paper, we introduce the SOC of battery into the electric vehicle’s driving behavior model and propose a car-following model for electric vehicles, and then use the proposed model to study the effects of the SOC of battery and battery swap on each vehicle’s driving behavior in the heterogeneous traffic system consisting of traditional vehicles and electric vehicles. The numerical results show that the proposed model can reproduce some complex traffic phenomena resulted by the SOC of battery and battery swap and that the influences on each vehicle’s driving behavior are directly related to the initial traffic state, the electric vehicle’s proportion, and the SOC of battery.

Aggregate driver model to enable predictable behaviour

The categorization of driving styles, particularly in terms of aggressiveness and skill is an emerging area of interest under the broader theme of intelligent transportation. There are two possible discriminatory techniques that can be applied for such categorization; a microscale (event based) model and a macro-scale (aggregate) model. It is believed that an aggregate model will reveal many interesting aspects of human-machine interaction; for example, we may be able to understand the propensities of individuals to carry out a given task over longer periods of time. A useful driver model may include the adaptive capability of the human driver, aggregated as the individual propensity to control speed/acceleration. Towards that objective, we carried out experiments by deploying smartphone based application to be used for data collection by a group of drivers. Data is primarily being collected from GPS measurements including position & speed on a second-by-second basis, for a number of trips over a two months period. Analysing the data set, aggregate models for individual drivers were created and their natural aggressiveness were deduced. In this paper, we present the initial results for 12 drivers. It is shown that the higher order moments of the acceleration profile is an important parameter and identifier of journey quality. It is also observed that the Kurtosis of the acceleration profiles stores major information about the driving styles. Such an observation leads to two different ranking systems based on acceleration data. Such driving behaviour models can be integrated with vehicle and road model and used to generate behavioural model for real traffic scenario.

Multi-step prediction of nonlinear Gaussian Process dynamics models with adaptive Gaussian mixtures

This paper presents an adaptive Gaussian Mixture Model (aGMM) formulation for performing multiple-step probabilistic state predictions using a nonparametric Gaussian Process (GP) regression model. The presented prediction algorithm is applicable to any dynamic system that is challenging to model parametrically, but where data is available. Gaussian mixture elements are propagated through the GP by analytically evaluating expectation integrals for the moments of the output distribution. Two metrics are presented and compared for adaptively splitting the initial state distribution into a sum of Gaussians to reduce the effect of nonlinearities on prediction accuracy: (1) an analytical evaluation of the excess kurtosis which measures the non-Gaussianity of the output distribution, and (2) a weighted least-squares regression model which evaluates the local nonlinearity of the GP mapping with respect to the input distribution. In addition, an on-the-fly data selection method is presented to reduce the computational complexity associated with analytically evaluating the higher-order moments of the GP output distribution. The proposed adaptive GP-aGMM formulation is applied to the case of anticipating driver behavior at road intersections using a GP driver behavior model in combination with a parametric vehicle model. Prediction performance for this scenario is evaluated using driving data collected from three human subjects navigating a standard four-way intersection. Results demonstrate that the presented prediction algorithm is capable of accurately capturing multimodal behavior in the GP training data.

Seemingly irrational driving behavior model: The effect of habit strength and anticipated affective reactions

An increasing amount of evidence suggests that aberrant driving behaviors are not entirely rational. On the basis of the dual-process theory, this study postulates that drivers may learn to perform irrational aberrant driving behaviors, and these behaviors could be derived either from a deliberate or an intuitive decision-making approach. Accordingly, a seemingly irrational driving behavior model is proposed; in this model, the theory of planned behavior (TPB) was adopted to represent the deliberate decision-making mechanism, and habit strength was incorporated to reflect the intuitive decision process. A multiple trivariate mediation structure was designed to reflect the process through which driving behaviors are learned. Anticipated affective reactions (AARs) were further included to examine the effect of affect on aberrant driving behaviors. Considering the example of speeding behaviors, this study developed scales and conducted a two-wave survey of students in two departments at a university in Northern Taiwan. The analysis results show that habit strength consists of multiple aspects, and frequency of past behavior cannot be a complete repository for accumulating habit strength. Habit strength appeared to be a crucial mediator between intention antecedents (e.g., attitude) and the intention itself. Including habit strength in the TPB model enhanced the explained variance of speeding intention by 26.7%. In addition, AARs were different from attitudes; particularly, young drivers tended to perform speeding behaviors to reduce negative feelings such as regret. The proposed model provides an effective alternative approach for investigating aberrant driving behaviors; corresponding countermeasures are discussed.

Stochastic modeling of driver behavior by Langevin equations

A procedure based on stochastic Langevin equations is presented and shows how a stochastic model of driver behavior can be estimated directly from given data. The Langevin analysis allows the separation of a given data-set into a stochastic diffusion- and a deterministic drift field. Form the drift field a potential can be derived. In particular the method is here applied on driving data from a simulator. We overcome typical problems like varying sampling rates, low noise levels, low data amounts, inefficient coordinate systems, and non-stationary situations. From the estimation of the drift- and diffusion vector-fields derived from the data, we show different ways how to set up Monte-Carlo simulations for the driver behavior.

Contrasting models of driver behaviour in emergencies using retrospective verbalisations and network analysis

Automated assistance in driving emergencies aims to improve the safety of our roads by avoiding or mitigating the effects of accidents. However, the behavioural implications of such systems remain unknown. This paper introduces the driver decision-making in emergencies (DDMiEs) framework to investigate how the level and type of automation may affect driver decision-making and subsequent responses to critical braking events using network analysis to interrogate retrospective verbalisations. Four DDMiE models were constructed to represent different levels of automation within the driving task and its effects on driver decision-making. Findings suggest that whilst automation does not alter the decision-making pathway (e.g. the processes between hazard detection and response remain similar), it does appear to significantly weaken the links between information-processing nodes. This reflects an unintended yet emergent property within the task network that could mean that we may not be improving safety in the way we expect.Practitioner Summary: This paper contrasts models of driver decision-making in emergencies at varying levels of automation using the Southampton University Driving Simulator. Network analysis of retrospective verbalisations indicates that increasing the level of automation in driving emergencies weakens the link between information-processing nodes essential for effective decision-making.