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

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

Safety in the road environment: a driver behavioural response perspective

The existing literature on road safety suggests that a driver's perception of safety is an important influence on their driving behaviour. A challenging research question is how to measure the perception of safety given the complex interactions among drivers, vehicles and the road setting. In this paper, we investigate a sample of driver evaluations of the perception of safety associated with a set of typical road environments. A roundabout was selected as the context for the empirical study. Data was obtained by a computerised survey using the video-captured road and traffic situations. A controlled experiment elicited driver responses when faced with a mixture of attributes that describe the roundabout environment. An ordered probit model identified the contribution of each attribute to the overall determination of the perception of safety. An indicator of perceived safety was developed for a number of typical road and traffic situations and for different driver segments.     

Classifiers for driver activity monitoring

The goal of this work is the detection and classification of driver activities in an automobile using computer vision. To this end, this paper presents a novel two-step classification algorithm, namely, an unsupervised clustering algorithm for grouping the actions of a driver during a certain period of time, followed by a supervised activity classification algorithm. The main contribution of this work is the combination of the two methods to provide a computationally fast solution for deployment in real-world scenarios that is robust to illumination and segmentation issues under most conditions experienced in the automobile environment. The unsupervised clustering groups the actions of the driver based on the relative motion detected using a skin-color segmentation algorithm, while the activity classifier is a binary Bayesian eigenimage classifier. Activities are grouped as safe or unsafe and the results of the classification are shown on several subjects obtained from two distinct driving video sequences.     

Probe vehicle-based traffic state estimation method with spacing information and conservation law

This paper proposes a method of estimating a traffic state based on probe vehicle data that contain spacing and position of probe vehicles. The probe vehicles were assumed to observe spacing by utilizing an advanced driver assistance system, that has been implemented in practice and is expected to spread in the near future. The proposed method relies on the conservation law of the traffic flow but is independent of a fundamental diagram. The conservation law is utilized for reasonable aggregation of the spacing data to acquire the traffic state, i.e., a flow, density and speed. Its independence from a fundamental diagram means that the proposed method does not require predetermined nor exogenous assumptions with regard to the traffic flow model parameters. The proposed method was validated through a simulation experiment under ideal conditions and a field experiment conducted under actual traffic conditions; and empirical characteristics of the proposed method were investigated.     

From behavioral psychology to acceleration modeling: Calibration,validation, and exploration of drivers’ cognitive and safety parameters in a risk-taking environment

We investigate a utility-based approach for driver car-following behavioral modeling while analyzing different aspects of the model characteristics especially in terms of capturing different fundamental diagram regions and safety proxy indices. The adopted model came from an elementary thought where drivers associate subjective utilities for accelerations (i.e. gain in travel times) and subjective dis-utilities for decelerations (i.e. loss in travel time) with a perceived probability of being involved in rear-end collision crashes. Following the testing of the model general structure, the authors translate the corresponding behavioral psychology theory – prospect theory – into an efficient microscopic traffic modeling with more elaborate stochastic characteristics considered in a risk-taking environment.After model formulation, we explore different model disaggregate and aggregate characteristics making sure that fidelity is kept in terms of equilibrium properties. Significant effort is then dedicated to calibrating and validating the model using microscopic trajectory data. A modified genetic algorithm is adopted for this purpose while focusing on capturing inter-driver heterogeneity for each of the parameters. Using the calibration exercise as a starting point, simulation sensitivity analysis is performed to reproduce different fundamental diagram regions and to explore rear-end collisions related properties. In terms of fundamental diagram regions, the model in hand is able to capture traffic breakdowns and different instabilities in the congested region represented by flow-density data points scattering. In terms of incident related measures, the effect of heterogeneity in both psychological factors and execution/perception errors on the accidents number and their distribution is studied. Through sensitivity analysis, correlations between the crash-penalty, the negative coefficient associated with losses in speed, the positive coefficient associated with gains in speed, the driver’s uncertainty, the anticipation time and the reaction time are retrieved. The formulated model offers a better understanding of driving behavior, particularly under extreme/incident conditions.     

Behavior-based consistency-seeking models as deployment alternatives to dynamic traffic assignment models

This paper proposes a behavior-based consistency-seeking (BBCS) model as an alternative to the dynamic traffic assignment paradigm for the real-time control of traffic systems under information provision. The BBCS framework uses a hybrid probabilistic–possibilistic model to capture the day-to-day evolution and the within-day dynamics of individual driver behavior. It considers heterogeneous driver classes based on the broad behavioral characteristics of drivers elicited from surveys and past studies on driver behavior. Fuzzy logic and if–then rules are used to model the various driver behavior classes. The approach enables the modeling of information characteristics and driver response to be more consistent with the real-world. The day-to-day evolution of driver behavior characteristics is reflected by updating the appropriate model parameters based on the current day’s experience. The within-day behavioral dynamics are reactive and capture drivers’ actions vis-à-vis the ambient driving conditions by updating the weights associated with the relevant if–then rules. The BBCS model is deployed by updating the ambient driver behavior class fractions so as to ensure consistency with the real-time traffic sensor measurements. Simulation experiments are conducted to investigate the real-time applicability of the proposed framework to a real-world network. The results suggest that the approach can reasonably capture the within-day variations in driver behavior model parameters and class fractions in the traffic stream. Also, they indicate that deployment-capable information strategies can be used to influence system performance. From a computational standpoint, the approach is real-time deployable.     

Towards an understanding of adaptive cruise control

Adaptive cruise control (ACC) provides assistance to the driver in the task of longitudinal control of their vehicle during motorway driving. The system controls the accelerator, engine powertrain and vehicle brakes to maintain a desired time-gap to the vehicle ahead. This research describes the results of a detailed microscopic simulation investigation into the potential impacts of ACC on motorway driving. In addition to simulation, real vehicle driving profiles, obtained from instrumented vehicle experiments in three European countries, have been used to compare real following behaviour with that of a simulated ACC equipped vehicle. This new approach has shown that following with an ACC system can provide considerable reductions in the variation of acceleration compared to manual driving. This indicates a potential comfort gain for the driver and environmental benefits. A number of critical situations in which ACC does not perform well have also been identified. The research also highlights the limitations of microscopic simulation in modelling the impacts of ACC because of the lack of understanding of the interaction between the driver and the ACC system relative to the traffic conditions.     

Manual vs. adaptive cruise control – Can driver’s expectation be matched?

The role of the driver in the longitudinal car following control task will change from operator to supervisor with most of manual control replaced by automation as adaptive cruise control (ACC) technologies become commonplace. The extent to which manual control can be replaced by ACC will be determined by many factors. An important issue is the compatibility between ACC performance and the driver’s expectations.This paper describes the results of a simulation study of the performance of ACC relative to driver expectation. Driver’s expectation is quantitatively defined as the expected deceleration rate for several time-to-collision (TTC) levels, and an absolute minimum TTC that drivers tried to avoid in all cases. A two-level ACC algorithm was used to simulate the performance of an ACC equipped vehicle in various scenarios, and the result was compared to the driver’s expectations. The investigation has focused on scenarios which ACC is able to manage technically, but where driver expectations might be breached.By systematically changing variables such as the parameters of the ACC algorithms, traffic scenarios and time-headway settings, a large number of situations have been tested. The results have revealed that whilst appropriate ACC settings can be found which will meet the driver’s expectations, the ACC settings that are most capable in a range of traffic conditions are not necessarily the most user-friendly. A discussion on the implications of the findings is also presented.     

Trip time-stop time studies of extreme driver behaviors

The focus of the present study has been to investigate the extent to which drivers with “extreme” modes of behavior deviate from the “normal” (chase-car) trend in a given city. The experiment consisted of collecting trip time-stop time data in the Central Business District (CBD) of Roanoke, Virginia, using three vehicles circulating simultaneously in the area. While one of the three vehicles was engaged in the chase-car mode, the other two vehicles were driven aggressively or conservatively. Similar data were also collected in the Austin, Texas, CBD. Using the data for each driver type, three two-fluid trends are established for the Roanoke CBD: an aggressive, a normal (chase-car), and a conservative trend. On a trip time-stop time diagram, while the conservative and aggressive trends are essentially parallel, the conservative trend approaches the normal trend at off-peak periods. The aggressive trend, on the other hand, approaches the normal trend at peak periods. The differences among the trends, however, are by and large statistically significant. Similar results have been obtained in Austin—despite uncertainties at high levels of demand, where the normal trend may cross the aggressive trend. The results also underline the importance of common data collection techniques when comparing the quality of traffic service of various cities using the two-fluid model approach. In assessing the statistical significance of differences among the two-fluid trends of various city networks, the potential impact of driver behavioral variations must be considered.     

基于驾驶员信息处理的道路安全性评价模型

在车辆行驶过程中,行车环境包含了大量的交通信息.道路条件以及管理手段等因素最终影响到驾驶员的信息处理强度.文章通过探讨驾驶员信息处理过程,建立了一种基于驾驶员信息处理强度的道路交通安全评价模型.在示例中,把道路交通信息量定义为道路几何曲率的一次函数,并结合相邻路段运行速度差给出评价指标.     

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

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

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

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

Multi-criteria route planning based on a driver’s preferences in multi-criteria route selection

In this study, some different approaches were designed, implemented, and evaluated to perform multi-criteria route planning by considering a driver’s preferences in multi-criteria route selection. At first, by using a designed neuro-fuzzy toolbox, the driver’s preferences in multi-criteria route selection such as the preferred criteria in route selection, the number of route-rating classes, and the routes with the same rate were received. Next, to learn the driver’s preferences in multi-criteria route selection and to classify any route based on these preferences, a methodology was proposed using a locally linear neuro-fuzzy model (LLNFM) trained with an incremental tree based learning algorithm. In this regard, the proposed LLNFM-based methodology reached better results for running-times, as well as root mean square error (RMSE) estimations in learning and testing processes of training/checking data-set in comparison with those of the proposed adaptive neuro-fuzzy inference system (ANFIS) based methodology. Finally, the trained LLNFM-based methodology was utilized to plan and predict a driver’s preferred routes by classifying Pareto-optimal routes obtained by running the modified invasive weed optimization (IWO) algorithm between an origin and a destination of a real urban transportation network based on the driver’s preferences in multi-criteria route selection.     

An agent-based approach to modelling driver route choice behaviour under the influence of real-time information

This paper presents an agent-based approach to modelling individual driver behaviour under the influence of real-time traffic information. The driver behaviour models developed in this study are based on a behavioural survey of drivers which was conducted on a congested commuting corridor in Brisbane, Australia. Commuters’ responses to travel information were analysed and a number of discrete choice models were developed to determine the factors influencing drivers’ behaviour and their propensity to change route and adjust travel patterns. Based on the results obtained from the behavioural survey, the agent behaviour parameters which define driver characteristics, knowledge and preferences were identified and their values determined. A case study implementing a simple agent-based route choice decision model within a microscopic traffic simulation tool is also presented. Driver-vehicle units (DVUs) were modelled as autonomous software components that can each be assigned a set of goals to achieve and a database of knowledge comprising certain beliefs, intentions and preferences concerning the driving task. Each DVU provided route choice decision-making capabilities, based on perception of its environment, that were similar to the described intentions of the driver it represented. The case study clearly demonstrated the feasibility of the approach and the potential to develop more complex driver behavioural dynamics based on the belief–desire–intention agent architecture.     

A model for the structure of lane-changing decisions

A structure is proposed to connect the decisions which a driver has to make before changing lanes. The model is intended to cover the urban driving situation, where traffic signals, obstructions and heavy vehicles all exert an influence. The structure is designed to ensure that the vehicles in traffic simulations behave logically when confronted with situations commonly encountered in real traffic. The specific mathematical expression of the questions embedded in the decision process and employed in the present implementation of the model are not critical and can be replaced by alternatives, but the heirarchy of the decisions is crucial. On the basis of experience to date, the lane changing model produces a realistic simulation of driver behaviour and has proved very robust under a wide range of conditions.     

Experimental study and modeling of car-following behavior under high speed situation

To investigate the car-following behavior under high speed driving conditions, we performed a set of 11-car-platoon experiments on Hefei airport highway. The formation and growth of oscillations have been analyzed and compared with that in low speed situations. It was found that there is considerable heterogeneity for the same driver over different runs of the experiment. This intra-driver heterogeneity was quantitatively depicted by a new index and incorporated in an enhanced two-dimensional intelligent driver model. Using both the new high-speed and the previous low-speed experimental data, the new and three existing models were calibrated. Simulation results show that the enhanced model outperforms the three existing car-following models that do not take into account this intra-driver heterogeneity in reproducing the essential features of the traffic in the experiments.     

Modeling car-following behavior on urban expressways in Shanghai: A naturalistic driving study

Although car-following behavior is the core component of microscopic traffic simulation, intelligent transportation systems, and advanced driver assistance systems, the adequacy of the existing car-following models for Chinese drivers has not been investigated with real-world data yet. To address this gap, five representative car-following models were calibrated and evaluated for Shanghai drivers, using 2100 urban-expressway car-following periods extracted from the 161,055 km of driving data collected in the Shanghai Naturalistic Driving Study (SH-NDS). The models were calibrated for each of the 42 subject drivers, and their capabilities of predicting the drivers’ car-following behavior were evaluated.The results show that the intelligent driver model (IDM) has good transferability to model traffic situations not presented in calibration, and it performs best among the evaluated models. Compared to the Wiedemann 99 model used by VISSIM®, the IDM is easier to calibrate and demonstrates a better and more stable performance. These advantages justify its suitability for microscopic traffic simulation tools in Shanghai and likely in other regions of China. Additionally, considerable behavioral differences among different drivers were found, which demonstrates a need for archetypes of a variety of drivers to build a traffic mix in simulation. By comparing calibrated and observed values of the IDM parameters, this study found that (1) interpretable calibrated model parameters are linked with corresponding observable parameters in real world, but they are not necessarily numerically equivalent; and (2) parameters that can be measured in reality also need to be calibrated if better trajectory reproducing capability are to be achieved.     

Adaptive cruise control design for active congestion avoidance

We present an adaptive cruise control (ACC) strategy where the acceleration characteristics, that is, the driving style automatically adapts to different traffic situations. The three components of the concept are the ACC itself, implemented in the form of a car-following model, an algorithm for the automatic real-time detection of the traffic situation based on local information, and a strategy matrix to adapt the driving characteristics (that is, the parameters of the ACC controller) to the traffic conditions. Optionally, inter-vehicle and infrastructure-to-car communication can be used to improve the accuracy of determining the traffic states. Within a microscopic simulation framework, we have simulated the complete concept on a road section with an on-ramp bottleneck, using empirical loop-detector data for an afternoon rush-hour as input for the upstream boundary. We found that the ACC vehicles improve the traffic stability and the dynamic road capacity. While traffic congestion in the reference scenario was completely eliminated when simulating a proportion of 25% ACC vehicles, travel times were already significantly reduced for much lower penetration rates. The efficiency of the proposed driving strategy even for low market penetrations is a promising result for a successful application in future driver assistance systems.     

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 1000 EV 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.     

Personalised feedback and eco-driving: An explorative study

Conventional road transport has negative impact on the environment. Stimulating eco-driving through feedback to the driver about his/her energy conservation performance has the potential to reduce CO₂ emissions and promote fuel cost savings. Not all drivers respond well to the same type of feedback. Research has shown that different drivers are attracted to different types of information and feedback. The goal of this paper is to explore which different driver segments with specific psychographic characteristics can be distinguished, how these characteristics can be used in the development of an ecodriving support system and whether tailoring eco-driving feedback technology to these different driver segments will lead to increased acceptance and thus effectiveness of the eco feedback technology. The driver segments are based on the value orientation theory and learning orientation theory. Different possibilities for feedback were tested in an exploratory study in a driving simulator. An explorative study was selected since the choice of the display (how and when the information is presented) may have a strong impact on the results. This makes testing of the selected driver segments very difficult. The results of the study nevertheless suggest that adapting the display to a driver segment showed an increase in acceptance in certain cases. The results showed small differences for ratings on acceptation, ease of use, favouritism and a lower general rating between matched (e.g., learning display with learning oriented drivers) and mismatched displays (e.g., learning display with performance oriented drivers). Using a display that gives historical feedback and incorporates learning elements suggested a non-verifiable increase in acceptance for learning oriented drivers. However historical feedback and learning elements may be less effective for performance oriented drivers, who may need comparative feedback and game elements to improve energy conserving driving behaviour.