Microscopic Car-following Model Research Articles

Cooperative control of dynamic CAV dedicated lanes and vehicle active lane changing in expressway bottleneck areas

Bottleneck areas on expressways plague the operational efficiency of entire road systems. In mixed traffic flow environments consisting of connected and autonomous vehicles (CAVs) and connected human-driven vehicles (CHVs), it is believed that road capacity can be improved to relieve traffic congestion in bottleneck areas by setting CAV dedicated lanes (CDLs) on expressways. Existing static CDL setup methods provide fixed exclusive right-of-way for CAVs but cannot accommodate dynamic changes in traffic demand. To address this issue, utilizing lane control signals and CAV active lane-changing technology, a cooperative method involving dynamic CAV dedicated lane control (CDLC) and active lane-changing control (LCC) in the bottleneck areas of expressways is proposed in this paper. First, a fundamental diagram of traffic flow with CDLs is established based on the microscopic car-following model of mixed traffic flow, which can be used to determine the conditional thresholds for setting the CDLs of multilane expressways. Second, an expressway traffic model with CDLs is established by improving the lane-level cell transmission model. Finally, cooperative control of dynamic CAV dedicated lanes and active lane changing (CDLLC) is proposed based on the model predictive control (MPC) framework for solving the congestion problem at expressway bottlenecks in a mixed driving environment. The simulation results show that the proposed CDLLC strategy can effectively reduce the total vehicle travel time and decrease the risk of congestion propagation at expressway bottlenecks when compared to the LCC-only strategy.

Integrated self-consistent macro-micro traffic flow modeling and calibration framework based on trajectory data

Calibrating microscopic car-following (CF) models is crucial in traffic flow theory as it allows for accurate reproduction and investigation of traffic behavior and phenomena. Typically, the calibration procedure is a complicated, non-convex optimization issue. When the traffic state is in equilibrium, the macroscopic flow model can be derived analytically from the corresponding CF model. In contrast to the microscopic CF model, calibrated based on trajectory data, the macroscopic representation of the fundamental diagram (FD) primarily adopts loop detector data for calibration. The different calibration approaches at the macro- and microscopic levels may lead to misaligned parameters with identical practical meanings in both macro- and micro-traffic models. This inconsistency arises from the difference between the parameter calibration processes used in macro- and microscopic traffic flow models. Hence, this study proposes an integrated multiresolution traffic flow modeling framework using the same trajectory data for parameter calibration based on the self-consistency concept. This framework incorporates multiple objective functions in the macro- and micro-dimensions. To expeditiously execute the proposed framework, an improved metaheuristic multi-objective optimization algorithm is presented that employs multiple enhancement strategies. Additionally, a deep learning technique based on attention mechanisms was used to extract stationary-state traffic data for the macroscopic calibration process, instead of directly using the entire aggregated data. We conducted experiments using real-world and synthetic trajectory data to validate our self-consistent calibration framework.

Geometry of commutes in the universality of percolating traffic flows.

Traffic congestion is a major problem in megacities which increases vehicle emissions and degrades ambient air quality. Various models have been developed to address the universal features of traffic jams. These models range from microscopic car-following models to macroscopic collective dynamic models. Here, we study the macrostructure of congested traffic influenced by the complex geometry of the commute. Our main focus is on the dynamics of traffic patterns in Paris and Los Angeles, each with distinct urban structures. We analyze the complexity of the giant traffic clusters based on a percolation framework during rush hours in the mornings, evenings, and holidays. We uncover that the universality described by several critical exponents of traffic patterns is highly correlated with the geometry of commute and the underlying urban structure. Our findings might have broad implications for developing a greener, healthier, and more sustainable future city.

Safe Data-Driven Lane Change Decision Using Machine Learning in Vehicular Networks

This research proposes a unique platform for lane change assistance for generating data-driven lane change (LC) decisions in vehicular networks. The goal is to reduce the frequency of emergency braking, the rate of vehicle collisions, and the amount of time spent in risky lanes. In order to analyze and mine the massive amounts of data, our platform uses effective Machine Learning (ML) techniques to forecast collisions and advise the driver to safely change lanes. From the unprocessed large data generated by the car sensors, kinematic information is retrieved, cleaned, and evaluated. Machine learning algorithms analyze this kinematic data and provide an action: either stay in lane or change lanes to the left or right. The model is trained using the ML techniques K-Nearest Neighbor, Artificial Neural Network, and Deep Reinforcement Learning based on a set of training data and focus on predicting driver actions. The proposed solution is validated via extensive simulations using a microscopic car-following mobility model, coupled with an accurate mathematical modelling. Performance analysis show that KNN yields up to best performance parameters. Finally, we draw conclusions for road safety stakeholders to adopt the safer technique to lane change maneuver.

Fault Tolerance Analysis of Car-Following Models for Autonomous Vehicles

Microscopic car-following models can be applied in Autonomous Vehicles (AVs) to control the real-time longitudinal interactions among individual vehicles. Moreover, they can have a vital role in Advanced Vehicle Control and Safety Systems (AVCSSs) such as collision warning, adaptive cruise control, lane guidance driver assistance, brake assist as well as in modeling simulation of safety studies and capacity analysis in transportation science. In reality, sensor measurements are generally inaccurate. Surprisingly no comparative assessment of the car-following models in presence of sensor (measurement) errors for AVs exist till now to the best of our knowledge. Therefore, in this paper we evaluate the prominent car-following models for Avs in presence of sensor errors in terms of safety, trip times, flow and fuel efficiency through simulations. We show that sensor errors significantly impact safety and flow in all models, while they do increase trip times and fuel consumption of some of the models. None of the models is completely fault-tolerant and suitable for AVs as some models produce collisions and/or negative velocity while all models violate traffic light. Nonetheless, the k-leader Fuel-efficient Traffic Model (kFTM) is the most fault-tolerant negative velocity and collision free model having reasonable trip times and energy consumption among the investigated models.

System-size dependence of a jam-absorption driving strategy to remove traffic jam caused by a sag under the presence of traffic instability

Sag is a road section where a downhill changes into an uphill, and is a highway bottleneck. We consider a system in which all vehicles are connected, and run on a single-lane road with a sag. We propose a simple strategy for removing each traffic jam caused by the sag. Our strategy assigns a vehicle upstream of the jam front to perform the jam-absorption driving (JAD): running toward the predicted goal, and finally removing the jam. We use a microscopic car-following model possessing the traffic instability, an acceleration model against the road gradient of a sag, and an instantaneous fuel consumption model. Our main goal is to elucidate the influence of the system size (the number of vehicles in the system) on our strategy. By increasing the system size from 500 to 10000 vehicles, we have found the following results for the average total travel time per vehicle, and the average total fuel consumption per vehicle. Our strategy can reduce the former with a slightly increasing rate of reduction. Our strategy can reduce the latter with a rate of reduction which decreases and becomes roughly constant. Optimal spatiotemporal scales of JAD for the former and the latter become roughly constant, respectively. Minimizing the former and the latter simultaneously is not possible. Not only vehicular traffic flow, but also the collective dynamics of other self-driven particles (such as ships, swarm robots, and pedestrians) lined up in a single column, and passing through a bottleneck can be modeled using our strategy.

An s-shaped three-parameter (S3) traffic stream model with consistent car following relationship

In this study, a new s-shaped three-parameter (S3) traffic flow model is proposed to represent the relationships between three fundamental variables (i.e., flow, speed, and density) in highway traffic. An s-shaped speed-density function is proposed to capture the speed-density relationship under a wide range of possible densities. A consistent car-following model was derived in terms of the proposed s-shaped speed-density function. Both the S3 macroscopic model and the derived microscopic car-following model were calibrated using real-world traffic data.

A Vehicle Guidance Model with a Close-to-Reality Driver Model and Different Levels of Vehicle Automation

This paper presents a microscopic vehicle guidance model which adapts to different levels of vehicle automation. Independent of the vehicle, the driver model built is different from the common microscopic simulation models that regard the driver and the vehicle as a unit. The term “Vehicle Guidance Model” covers, here, both the human driver as well as a combination of human driver and driver assistance system up to fully autonomously operated vehicles without a (human) driver. Therefore, the vehicle guidance model can be combined with different kinds of vehicle models. As a result, the combination of different types of driver (human/machine) and different types of vehicle (internal combustion engine/electric) can be simulated. Mainly two parts constitute the vehicle guidance model in this paper: the first part is a traditional microscopic car-following model adjusted according to different degrees of automation level. The adjusted model represents the automation level for the present and the near and the more distant future. The second part is a fuzzy control model that describes how humans adjust the pedal position when they want to reach a target speed with their vehicle. An experiment with 34 subjects was carried out with a driving simulator based on the experimental data and the fuzzy control strategy was determined. Finally, when comparing the simulated model data and actual driving data, it is found that the fuzzy model for the human driver can reproduce the behavior of human participants almost accurately.

A Macroscopic Model of Heterogeneous Traffic Flow Based on the Safety Potential Field Theory

The macroscopic traffic flow model is the cornerstone of traffic flow theory and can be applied to investigate manytraffic planning and management issues. However, the macroscopic model of the heterogeneous traffic flow has not been profoundly studied. This paper developed a novel macroscopic model of the heterogeneous traffic flow under a partly connected and automated environment based on the safety potential field (SPF) theory. The SPF model was applied to describe the microscopic driving behaviour of traditional human-driven vehicles (HVs) and connected and automated vehicles (CAVs). A fundamental diagram (FD) model was derived according to the correlation with the microscopic car-following model under a steady-state condition. To verify the correlation between the FD and the microscopic simulation results, a Monte-Carlo-simulation was also performed. Finally, the macroscopic model of the heterogeneous traffic flow was validated through a simulation on a road section with a bottleneck under different CAVs market penetration rates ranging from 0.0 to 1.0. The results indicate that introducing the CAVs into the pure traffic flow of HVs may slightly have a negligible impact on road capacity when the CAVs penetration rate at a lower level, but the impact increases significantly as the CAVs penetration rate increased.

Hybrid Solution Combining Kalman Filtering with Takagi-Sugeno Fuzzy Inference System for Online Car-Following Model Calibration.

Nowadays, the intelligent transportation concept has become one of the most important research fields. All of us depend on mobility, even when we talk about people, provide services, or move goods. Researchers have tried to create and test different transportation models that can optimize traffic flow through road networks and, implicitly, reduce travel times. To validate these new models, the necessity of having a calibration process defined has emerged. Calibration is mandatory in the modeling process because it ensures the achievement of a model closer to the real system. The purpose of this paper is to propose a new multidisciplinary approach combining microscopic traffic modeling theory with intelligent control systems concepts like fuzzy inference in the traffic model calibration. The chosen Takagi–Sugeno fuzzy inference system proves its adaptive capacity for real-time systems. This concept will be applied to the specific microscopic car-following model parameters in combination with a Kalman filter. The results will demonstrate how the microscopic traffic model parameters can adapt based on real data to prove the model validity.

A macroscopic traffic flow model considering the velocity difference between adjacent vehicles on uphill and downhill slopes

In this paper, we deduced a macroscopic traffic model on the uphill and downhill slopes by employing the transformation relation from microscopic variables to macroscopic ones based on a microscopic car-following model considering the velocity difference between adjacent vehicles. The angle [Formula: see text] of the uphill and downhill and the gravitational force have a great impact upon the stability of traffic flow. The linear stability analysis for macroscopic traffic model yielded the stability condition. The Korteweg–de Vries (KdV) equation is derived by nonlinear analysis and the corresponding solution to the density wave near the neutral stability line is obtained. By using the upwind finite difference scheme for simulation, the spatiotemporal evolution patterns of traffic flow on the uphill and downhill are attained. The unstable region is shrunken with slope of the gradient increasing and backward-traveling density waves gradually decrease and even disappear on uphill. Conversely, the unstable region on downhill is extended and density waves propagate quickly backward to the whole road with slope of the gradient increasing.

A two-dimensional car-following model for two-dimensional traffic flow problems

This paper proposes a two-dimensional car-following model to tackle traffic flow problems where considering continuum lateral distances enables a simpler or more natural mathematical formulation compared to traditional car-following models. These problems include (i) the effects of lateral friction often observed in HOV lanes and diverge bottlenecks, (ii) the relaxation phenomenon at merge bottlenecks, (iii) the occurrence of accidents due to lane changing, and (iv) traffic models for autonomous vehicles (AVs). We conjecture that traditional car-following models, where the lateral dimension is discretized into lanes, struggle with these problems and one has to resort to ad-hoc rules conceived to directly achieve the desired effect, and that are difficult to validate.We argue that the distance maintained by drivers in order to avoid collisions in all directions plays a fundamental role in all these problems. To test this hypothesis, we propose a simple two-dimensional microscopic car-following model based on the social force paradigm, and build simulation experiments that reproduce these phenomena. These phenomena are reproduced as an indirect consequence of the model’s formulation, as opposed to ad-hoc rules, thus shedding light on their causes.A better understanding of the behavior of human drivers in the lateral dimension can be translated to improving autonomous driving algorithms so that they are human-friendly. In addition, since AV technology is proprietary, we argue that the proposed model should provide a good starting point for building AV traffic flow models when real data becomes available, as these data come from sensors that cover two-dimensional regions.

A Second Order Traffic Flow Model with Lane Changing

This paper concerns modeling and computation of traffic flow for a single in-lane flow as well as multilane flow with lane changing. We consider macroscopic partial differential equation models of two types: (i) First Order Models: equilibrium models, scalar models expressing car mass conservation; and (ii) Second Order Models: dynamic models, $$2 \times 2$$ hyperbolic systems expressing mass conservation as well as vehicle acceleration rules. A new second order model is proposed in which the acceleration terms take lead from microscopic car-following models, and yield a nonlinear hyperbolic system with viscous and relaxation terms. Lane changing conditions are formulated and mass/momentum inter-lane exchange terms are derived. Numerical results are shown, illustrating the merit of the models in describing a rich array of realistic traffic scenarios including varying road conditions, lane closure, and stop-and-go flow patterns.

Efficient calibration of microscopic car-following models for large-scale stochastic network simulators

This paper proposes a simulation-based optimization methodology for the efficient calibration of microscopic traffic flow models (i.e., car-following models) of large-scale stochastic network simulators. The approach is a metamodel simulation-based optimization (SO) method. To improve computational efficiency of the SO algorithm, problem-specific and simulator-specific structural information is embedded into a metamodel. As a closed-form expression is sought, we propose adopting the steady-state solution of the car-following model as an approximation of its simulation-based input-output mapping. This general approach is applied for the calibration of the Gipps car-following model embedded in a microscopic traffic network simulator, on a large network. To this end, a novel formulation for the traffic stream models corresponding to the Gipps car-following law is provided.The proposed approach identifies points with good performance within few simulation runs. Comparing its performances to that of a traditional approach, which does not take advantage of the structural information, the objective function is improved by two orders of magnitude in most experiments. Moreover, this is achieved within tight computational budgets, i.e., few simulation runs. The solutions identified improve the fit to the field measurements by one order of magnitude, on average. The structural information provided to the metamodel is shown to enable the SO algorithm to become robust to both the quality of the initial points and the simulator stochasticity.

Modeling the Imperfect Driver: Incorporating Human Factors in a Microscopic Traffic Model

In this paper, we present the enhanced human driver model (EHDM), a time-continuous microscopic car-following model, which considers numerous characteristics attributable to the human driver. It is based on the popular intelligent driver model and the human driver model (HDM), a meta-model which enables the integration of human driving behavior into classical follow-the-leader models. We extend the HDM by adding variable, situation-dependent reaction times, different types of driver distraction, and driving errors to the model. Furthermore, we show that the EHDM provides a more thorough representation of human driving behavior compared with the HDM by performing dynamic traffic simulations and by validating both models with the aid of real-word vehicle data captured in an extensive field test.

Car-following model with explicit reaction-time delay: linear stability analysis of a uniform solution on a ring

A microscopic car-following model with an explicit reaction-time delay is considered. The acceleration of every vehicle depends on its actual speed, the predecessor speed, and the distance between them. The acceleration function explicitly includes the driver’s reaction-time delay. On the one hand, this is an undoubted advantage of the model, and on the other hand, it significantly complicates the mathematical analysis. The subject of the study is the stability analysis of the uniform solutions on a ring. Using the linear stability analysis, we obtained the conditions on the model parameters including the driver’s reaction time, for which perturbations around the uniform solution decay. The calculations show that there are values for the model parameters, which satisfy the obtained conditions and at the same time provide a realistic dynamic of the vehicles. An important result is the fact that this model is able to reproduce such phenomenon as propagation of stop-and-go waves, which are widely known in traffic flow theory and are observed in practice. The latter is another advantage of the proposed car-following model.

Impact of driving aggressiveness on the traffic stability based on an extended optimal velocity model

The drivers in actual traffic differ in the aggressiveness in driving behaviors, which will finally be reflected in the interaction between vehicles and fluctuations in flows. In this paper, we try to study the effect of driving aggressiveness on the traffic stability by proposing an extended microscopic car-following model, in which the optimal velocity is reconstructed to divide the drivers in traffic system into two groups according to driving aggressiveness of each individual. The stability condition of the proposed model is derived to explore its ability against a small perturbation by use of the linear stability theory. We obtain the neutral stability lines for different percentages of drivers with a higher driving aggressiveness, finding that the traffic flow trends to stable with the increase in the percentage for higher driving aggressiveness drivers when the average headway is less than a critical value or greater than another critical value, but when the average headway falls into the intermediate range between the two critical values, the traffic flow becomes more and more unstable with increase in the percentage of drivers with a higher driving aggressiveness. Finally, numerical simulations are conducted to verify these theoretical results and examine how the percentage of vehicles driven by higher driving aggressiveness drivers affects the traffic flux of the vehicle system.

Jam-absorption driving with a car-following model

Jam-absorption driving (JAD) refers to the action performed by a single car to dynamically change its headway to remove a traffic jam. Because of its irregular motion, a car performing JAD perturbs other cars following it, and these perturbations may grow to become the so-called secondary traffic jams. A basic theory for JAD (Nishi et al. 2013) does not consider accelerations of cars or the stability of traffic flow. In this paper, by introducing car-following behaviors, we implement these elements in JAD. Numerous previous studies on the instability of traffic flow showed that even in a region whose density is below a critical density, perturbation may grow if its initial magnitude is large. According to these previous studies, we expect that the perturbations caused by JAD, if they are sufficiently small, do not grow to become secondary traffic jams. Using a microscopic car-following model, we numerically confirmed that the stability of a flow obeying the model depends on the magnitude of JAD perturbations. On the basis of this knowledge, numerical results indicate that parameter regions exist where JAD allows traffic jams to be removed without causing secondary traffic jams. Moreover, JAD is robust against a parameter of acceleration in the model, as well as the choice of car-following models.

Estimating risk effects of driving distraction: A dynamic errorable car-following model

This paper aims to estimate the risk effects of distracted driving, by incorporating a dynamic, data-driven car-following model in an algorithmic framework. The model was developed to predict the situational risk associated with distracted driving. To obtain longitudinal driving patterns, this paper analyzed and synthesized the NGSIM naturalistic driver and traffic database, through a dynamic time warping algorithm, to identify essential driver behavior and characteristics. Cognitive psychology concepts, distracted driving simulator, and experimental data were adapted to examine the probabilistic nature of distracted driving due to internal vehicle distractions. An extended microscopic car-following model was developed and validated, which can be fully integrated with the naturalistic data and incorporate the probabilities of driver distraction.

Calibrating Car-Following Model Considering Measurement Errors

Car-following model has important applications in traffic and safety engineering. To enhance the accuracy of model in predicting behavior of individual driver, considerable studies strive to improve the model calibration technologies. However, microscopic car-following models are generally calibrated by using macroscopic traffic data ignoring measurement errors-in-variables that leads to unreliable and erroneous conclusions. This paper aims to develop a technology to calibrate the well-known Van Aerde model. Particularly, the effect of measurement errors-in-variables on the accuracy of estimate is considered. In order to complete calibration of the model using microscopic data, a new parameter estimate method named two-step approach is proposed. The result shows that the modified Van Aerde model to a certain extent is more reliable than the generic model.