Complex Traffic Phenomena Research Articles

Traffic behavior analysis of the urban expressway ramp based on continuous cellular automata

Urban expressway is an essential part of the urban transportation network, and its role is to speed up traffic operation and relieve traffic congestion. To investigate the complex traffic flow phenomena at urban expressway ramps, we use the Continuous Cellular Automata (CCA) model to model and analyze the urban expressway with entrance and exit ramps in a unified manner. Considering the vehicle's merging, diverging and following behaviors on urban expressway with an entrance and exit ramp, we design the vehicle's main road merging rules, ramp inflow and outflow rules, Cooperative Car-Following for Lane Change process analysis (LC-CCF) model, and improved Nagatani lane change rules to control the overall road vehicle operation. Based on this, we propose a CCA model for urban expressway ramp traffic flow. To clarify the model, The Chengdu Second Ring Elevated Road with entrance and exit ramps as a scenario is used for experiments. The results show that: (1) The spatio-temporal diagrams show a significant metastable congestion phenomenon and distinct congestion boundaries primarily located near the merge and divergence area. Congestion tends to start at the merge and divergence area and then spread outward. (2) The ramp disturbance has an increasingly pronounced effect on average speed and density as Pin−main and Pin−ramp increases. (3) The density-lane change frequency relationship tends to increase and then decrease, with an increase in the frequency of lane changes increasing the interference between vehicles in different lanes and affecting the average traffic speed. The lane change frequency is the main reason for the differences in traffic flow between lanes. (4) The CCA model can effectively characterize congested traffic flows on urban expressway ramps.

Novel Traffic Conflict-Based Framework for Real-Time Traffic Safety Evaluation Under Heterogeneous and Weak Lane-Discipline Traffic

The present study proposed a real-time traffic safety evaluation framework using macroscopic flow variables. To this end, open-access extended vehicle trajectories were employed. Rear-end traffic conflicts and macroscopic traffic flow variables were derived from the trajectory data and were integrated for real-time safety evaluation. The Proportion of Stopping distance ( PSD) accounts for all types of interactions (both safe and unsafe) in the traffic stream; therefore, the same was adopted to analyze the rear-end traffic conflicts. A macroscopic indicator termed “time spent in conflict ( TSC)” was derived to evaluate the rear-end traffic conflicts. Machine learning models, namely, Random Forest (RF), Support Vector Machines (SVM), and eXtreme Gradient Boosting (XGB), were employed to predict TSCs using macroscopic traffic flow variables. The results revealed that the TSC computed based on PSD exhibits a reliable and explainable relationship with the macroscopic traffic flow variables. TSC computed based on PSD revealed that intermediately congested traffic flow conditions are critical in traffic safety and can be attributed to complex traffic phenomena such as traffic hysteresis, traffic oscillations, and increased speed variance. Moreover, a stable relation between traffic safety and traffic flow was suggested for varying threshold values. Among different machine learning models, the RF model was observed as the best-fitted model to predict TSC based on macroscopic traffic variables. TSC quantifies the safety status of a given traffic flow condition, where a higher value of TSC for a particular traffic flow condition indicates that vehicles prevail in the conflicting scenario for a longer time and, therefore, reflect higher operational risk. The developed machine learning model can be employed to predict TSC (operational risk) in real time using the macroscopic traffic flow variables and, therefore, facilitate traffic safety monitoring.

Empirical investigation of lane usage, lane changing and lane choice phenomena in a multimodal urban arterial

The combination of multimodal interactions, dynamic congestion and lane changing phenomena in an urban environment can have a strong impact on traffic flow dynamics, active road capacity and road safety. However, while there is a vast understanding of microscopic traffic models for motorways with respect to lane changes and lane choices, there are not many empirical observations of the phenomena at the urban scale. Recently, new data collection techniques with unmanned aerial systems have emerged offering unique opportunities to study complex traffic phenomena. This paper investigates the interrelation between lane changing and lane choice using the pNEUMA dataset in one of the busiest arterials in Athens, Greece. The main concept relies on the definition of two layers regarding how the multiple lanes of congested arterials are actually being used. Specifically, we show that the marked lanes on the arterial (marked layer) are influenced by the frictions created by static and moving bottlenecks (bus and taxi stops, illegal parking, etc.). Using an existing lane detection algorithm, we show that there is an active layer which is different to the marked lanes on the road affecting capacity in the macroscopic level and driving behavior in the microscopic level. The analysis on lane changing highlights the conflicting areas where more lane changes can occur. Furthermore, the particularity of Powered-Two-Wheelers’ lane changing behavior is examined and evidence on the PTW filtering phenomenon is provided. Following, the analysis on driver lane choice shows how turning vehicles approach the target lane and the differences between left- and right-turning vehicles.

Front-tracking transition system model for traffic state reconstruction, model learning, and control with application to stop-and-go wave dissipation

Connected and Autonomous Vehicles is a technology that will be disruptive for all layers of traffic control. The Lagrangian, in-the-flow nature of their operation offers untapped new potentials for sensing and actuation, but also presents new fundamental challenges. In order to use these vehicles for traffic state reconstruction and control, we need suitable traffic models, which should be computationally efficient and able to represent complex traffic phenomena. To this end, we propose the Front-tracking Transition System Model, a cell-free modelling approach that can incorporate Lagrangian measurements, and has a structure that yields itself to on-line model learning and control. The model is formulated as a transition system, and based on the front-tracking method for finding entropy solutions to the Lighthill–Whitham–Richards model. We characterize the solution of this model and show that it corresponds to the solution of the underlying PDE traffic model. Algorithms for traffic state reconstruction and model learning are proposed, exploiting the model structure. The model is then used to design a prediction-based control law for stop-and-go wave dissipation using randomly arriving Connected and Autonomous Vehicles. The proposed control framework is able to estimate the traffic state and model, adapt to changes in the traffic dynamics, and achieve a reduction in vehicles’ Total Time Spent.

Bifurcation analysis based on new macro two-velocity difference model

Based on the new macroscopic two-velocity difference model, this paper analyzes the linear stability of the new model and studies the nonlinear bifurcation theory. First, the linear stability analysis method is used to study the stability conditions of the shock wave in the model. Then, considering the long wave model in the coarse-grained scale, the reduced perturbation method is used to analyze the characteristics of the traffic flow in the metastable region, and the solitary wave solution of the Korteweg-de Vries (KdV) equation in the metastable region is derived. In addition, by using the bifurcation analysis method, the type, and stability of the equilibrium solution are discussed and the existing conditions of the saddle-node bifurcation are proven. Then, taking the saddle-node bifurcation as the starting point, we draw the density space-time diagram and phase plane diagram of the system. It is proven that the newly proposed model can describe complex traffic phenomena such as stop-and-go and sudden changes in stability, which is of great help to solve traffic congestion.

Microscopic traffic behavior modeling and safety evaluation in mixed traffic area of bus station

With the development of Chinese cities, the mixed behavior at bus stops during the early commuting period is severe, which seriously affects the safety of passengers. Thus, in this paper, we propose a cellular automaton (CA) model to simulate bus arrival, non-motor vehicles movement and passenger boarding processes in mixed traffic area of bus station. Data is collected to explore the passengers’ movements and the complex traffic phenomena (e.g., bus's two driving routes and different number of arriving buses simultaneously) at the station. Based on collected data, the parameters are calibrated and the simulation scenario is built. Numerous simulations are conducted to further analyze the impact of microscopic behaviors of buses and passengers on safety at bus station. Moreover, the most dangerous areas of bus station are delineated under the extended CA model. Comparison between two cases with and without non-motor vehicles reveals that the non-motor vehicles have prominent negative impacts on safety. Finally, some suggestions for bus drivers, passengers and managers are proposed from the perspective of safety. The above results provide a theoretical foundation for the effective management of passengers and bus drivers at bus station.

A continuous traffic flow model considering predictive headway variation and preceding vehicle’s taillight effect

The car taillights in the deceleration process of preceding vehicle may greatly affect the acceleration and deceleration processes of the following vehicle. Additionally, the driver can predict the headway at the next time period from surrounding traffic information, and adjust the vehicle acceleration based on the difference between the predicted headway and the current headway information. To analyze these combined effects, we propose a new continuous traffic flow model taking into account the predictive headway variation and preceding vehicle’s taillight. The stability condition and KdV–Burgers equation of the continuum model are derived in linear and nonlinear stability analysis. The density wave solution obtained by solving the KdV–Burgers equation can be used to describe the formation and propagation mechanism of traffic jams near stable conditions. The complex traffic phenomena such as shock waves, rarefaction waves and local cluster effects are reproduced by simulation. Results show that the taillight information of preceding vehicle and the driver’s prediction time step both contribute greatly to the stability of the traffic flow and energy consumption.

A new heterogeneous traffic flow model based on lateral distance headway

Based on a density gradient model proposed recently by Imran and Khan, a new heterogeneous traffic flow model considering time and lateral distance is proposed. The type and stability of the equilibrium solution of the model are discussed by using the differential equation theory, and the global distribution structure of the trajectory in the phase plane is analyzed. In addition, the density wave stability conditions and saddle-node bifurcation conditions of the model are studied, and the solitary wave solutions of the KdV equation in the metastable region are derived by using the reduced perturbation method. The numerical results show that the new model cannot only reproduce the spatiotemporal oscillation phenomenon when walking and stopping, but also describe the sudden change behavior of traffic near the critical point of saddle-node bifurcation. It is shown that the model can reproduce some complex traffic phenomena qualitatively.

Noise-induced instability of a class of stochastic higher order continuum traffic models

Traffic flow dynamics are complex and nonlinear, and depend on the interactions of many vehicles moving together. Especially, the stochastic driving behavior is a significant reason for many complex traffic phenomena such as the capacity drop or traffic instabilities without lane changing. To date, there has not been a satisfactory general theory that can be consistently applied to represent such traffic flow conditions. While there have been efforts to introduce stochastic behavior in deterministic traffic flow models, these attempts have mainly focused on the small stochastic fluctuations around the theoretical equilibrium fundamental diagram or model parameters. Little attention has been given to the physical relevance of the stochastic component, and the existing models do not consider large deviations from the equilibrium fundamental diagram that occur in the transient phases of traffic flow (i.e acceleration and/or deceleration). The goal of this paper is thus to improve current continuum traffic models, with the primary focus on the stochastic behavior arising in the acceleration and/or deceleration process. Our analytical results show noise-induced (long-wave length) stop-and-go waves in the low speed regime, which conform to the recent empirical findings. In addition, it is found that the stochasticity can lead to situations in which a bottleneck sometimes is activated and at other times is not activated under identical traffic demand.

Performance of continuum models for realworld traffic flows: Comprehensive benchmarking

Numerous continuum models have been proposed for freeway traffic, but their performance for real-world traffic flows has not been rigorously evaluated and compared in the literature. This paper conducts a comprehensive benchmarking study on single-pipe continuum models by using traffic data from the German A5 autobahn. Model families are examined, and the best representative models are selected within the families. A set of benchmarking criteria is designed, ranging from the operational measures (e.g., delay and travel time) as well as the complex traffic phenomena (i.e., oscillations, capacity drop, and hysteresis phenomena). Suitable real-world traffic scenarios are carefully selected concerning the benchmarking criteria, and the selected models are comprehensively assessed for real-world scenarios. The paper comprehensively discusses the performances of each model regarding each aspect of the benchmarking criteria.

A boarding model for heterogeneous passengers on the platform of high-speed railway station

ABSTRACT Boarding behavior on the platform of high-speed railway (HSR) station, as a critical problem that has not been fully explored and understood, is investigated in this paper. An extended cellular automaton (CA) model is proposed to study the impacts of passengers’ heterogeneities during the boarding process. Some numerical tests are conducted to explore the passengers’ movements and the complex traffic phenomena (e.g., passenger's detour behavior, local congestion and overtaking behavior) on the platform. The numerical results show that the proportion of the passengers carrying luggage has significant impacts on the boarding process (e.g., the degree of congestion, inflow rate at the entrance and route choice). Moreover, the heterogeneous speeds increase the total boarding time and decrease the boarding efficiency comparing with the homogenous speeds. The above results are helpful for the effective management of pedestrian flow on the HSR platform and the optimal design of HSR platform layout.

Towards linking driving complexity to crash risk

The purpose of this article is to present insights into the relationship between complex traffic flow phenomena on urban motorways and crash risk. Unstable or congested flow can trigger low speed/high density clusters (e.g. nucleations or shockwaves) creating ‘surprise elements’, therefore sharply increasing the cognitive workload for motorists. When combined with reduced road space and freedom to perform needed manoeuvres (e.g. lane changes), conditions can exceed the physical or mental capability and hence increase the likelihood of human error. There is overwhelming evidence that high traffic density drastically increases the crash risk. Some density concentrations can be avoided through appropriate planning and real-time traffic control, resulting in a reduction in crashes. Modern measurement devices allow for the analysis of individual vehicle behaviours such as ‘Brake’, ‘Speed alert’ or ‘Lane change’ events and show promise in providing robust data to further exploring what makes dense traffic complex. This allows establishing relationships between “events as elementary units of exposure” and crash occurrence resulting in a new way of understanding crash rates. These relationships are important to predict crashes, identify high-risk locations, and establish suitable measures for crash reduction.

Dynamic model of high speed following traffic flow

For the physical phenomenon of high-speed car following in the road traffic flow, all the vehicles with small spacing do not run at low speed. The speeds of the vehicles are significantly higher than those they normally should be when they are in this vehicles’s density. There are more than 7% of high-speed following vehicles in the measured data. At present, the traditional traffic flow model cannot simulate the phenomenon of high-speed car following, so a new nonlinear dynamic mathematical model is needed to describe and analyze the physical phenomenon. In order to study the physical phenomenon of high-speed car following in road traffic, a traffic flow dynamics model for simulating the phenomena is proposed, which combines with the prospect theory and takes into account some factors such as driver’s decision-making behavior and randomization. It is called HCCA (high-speed car following celluar automat) model. In the model, the prospect theory is used to analyze the driver’s lane changing behavior under the uncertain conditions. Combined with the characteristics of the radical driver, the dynamic prediction speed is considered for the front car followed by the radical driver, and the HCCA dynamics rules of high-speed following traffic flow mechanics are defined. By the computer numerical simulation, the evolution mechanism and the characteristics of high-speed car-following flow are studied. The results show that compared with the symmetric two-lane cellular automata (STCA) dynamic model, the HCCA dynamic model established in this paper can simulate abundant traffic physical phenomena, and reproduce complex traffic phenomena such as free flow, synchronous flow and wide-range motion congestion. Finally, the phenomenon of high-speed car following is simulated and the results of high-speed car following rate over 7% with small spacing are in good agreement with the measured results. It overcomes the shortage that traditional STCA model cannot simulate the synchronous flow. It is found that the larger the proportion of radical drivers, the larger the high-speed car following rate and traffic flow with small spacing are under the same road density. The high-speed car following traffic flow mechanics model proposed in this paper has a certain reference significance and practical value for analyzing the physical phenomenon of high-speed car following and enriching the traffic flow theory.

Development of an Asymmetric Car-Following Model and Simulation Validation

Numerous car-following models have been developed since the 1950s. However, there still exist many traffic phenomena that cannot be demonstrated using the existing models. Therefore, this research proposed a new car-following model, the Asymmetric car-following (ACF) model based on the understanding of driver’s asymmetric behavior, which can explain complex traffic phenomena. We established the asymmetric car-following (ACF) rule under the vehicle’s safety constraints using eight parameters that indicate the driver’s characteristics and vehicle’s performances. To evaluate the ACF model, we performed the simulation for car-following pairs and conducted a comparison analysis with the existing models: Newell, Gipps, GM, and IDM. As a result, the proposed ACF model showed good fitness with the empirical trajectory and the apparent asymmetric behavior compared to others. For further investigation in the congested traffic stream, we simulated a group of vehicles by adding an error term to represent the driver’s unexpected behavior. The simulation showed growth, propagation, and dissipation of the stop-and-go traffic. These results proved that the ACF model has the strength to elaborate on various traffic phenomena, such as traffic hysteresis and stop-and-go traffic.

Cross-Comparison and Calibration of Two Microscopic Traffic Simulation Models for Complex Freeway Corridors with Dedicated Lanes

Realistic microscopic traffic simulation is essential for prospective evaluation of the potential impacts of new traffic control strategies. Freeway corridors with interacting bottlenecks and dedicated lanes generate complex traffic flow phenomena and congestion patterns, which are difficult to reproduce with existing microscopic simulation models. This paper discusses two alternative driving behavior models that are capable of modeling freeways with multiple bottlenecks and dedicated lanes over an extended period with varying demand levels. The models have been calibrated using archived data from a complicated 13-mile long section of the northbound SR99 freeway near Sacramento, California, for an 8-hour time period in which the traffic fluctuated from free-flow to congested conditions. The corridor includes multiple bottlenecks, multiple entry and exit ramps, and an HOV lane. Calibration results show extremely good agreement between field data and model predictions. The models have been cross-validated and produced similar macroscopic traffic performance. The main behavior that should be captured for successful modeling of such a complex corridor includes the anticipative and cooperative driver behavior near merges, lane preference in presence of dedicated lanes, and variations in desired headway along the corridor.

Statistical analysis and modeling of pedestrian flow in university canteen during peak period

This paper uses the second canteen of Beihang University to study the statistical features of pedestrian flow during the lunch peak period. We first use camera to collect the pedestrian flow data, then analyze the statistical features of data, thirdly propose a model to explore the pedestrian flow during the lunch peak period and finally testify the simulation results and the experiment data. The results show that the proposed model can perfectly reproduce the complex traffic phenomena that occur at university canteen during the peak period. The results can help administrators design some effective evacuation strategies when some emergencies occur at university canteen.

On the Modeling of Near-Field Scattering of Vehicles in Vehicle-to-X Wireless Channels Based on Scattering Centers

Efficient and reliable vehicle-to-X (V2X) wireless communication requires a deep understanding of the associated propagation channels. However, V2X channels are complicated due to the complex traffic phenomena, and thus extensive propagation studies are required to characterize these channels. This paper considers the effects of vehicle scattering on V2X channels. Most existing techniques that address vehicle scattering are imperfect, due to deficiencies in their computational accuracy and efficiency. In this paper, the tradeoff between computational accuracy and efficiency is addressed by proposing an analytical model based on the concept of scattering centers (SCs). The model can efficiently predict near-field bi-static scattering of vehicles with good accuracy. Also, to make the model widely acceptable, a general methodology of extracting SCs is carefully derived. Then, three standard vehicles (car, van, and truck) are employed for evaluating the performance of the model. The results indicate that the model is useful for reconstructing vehicle scattering in V2X channels.

A cellular automation model accounting for bicycle’s group behavior

Recently, bicycle has become an important traffic tool in China, again. Due to the merits of bicycle, the group behavior widely exists in urban traffic system. However, little effort has been made to explore the impacts of the group behavior on bicycle flow. In this paper, we propose a CA (cellular automaton) model with group behavior to explore the complex traffic phenomena caused by shoulder group behavior and following group behavior on an open road. The numerical results illustrate that the proposed model can qualitatively describe the impacts of the two kinds of group behaviors on bicycle flow and that the effects are related to the mode and size of group behaviors. The results can help us to better understand the impacts of the bicycle’s group behaviors on urban traffic system and effectively control the bicycle’s group behavior.

A cellular automata model for traffic flow based on kinetics theory, vehicles capabilities and driver reactions

In this paper, a reliable cellular automata model oriented to faithfully reproduce deceleration and acceleration according to realistic reactions of drivers, when vehicles with different deceleration capabilities are considered is presented. The model focuses on describing complex traffic phenomena by coding in its rules the basic mechanisms of drivers behavior, vehicles capabilities and kinetics, while preserving simplicity. In particular, vehiclés kinetics is based on uniform accelerated motion, rather than in impulsive accelerated motion as in most existing CA models. Thus, the proposed model calculates in an analytic way three safe preserving distances to determine the best action a follower vehicle can take under a worst case scenario. Besides, the prediction analysis guarantees that under the proper assumptions, collision between vehicles may not happen at any future time. Simulations results indicate that all interactions of heterogeneous vehicles (i.e., car–truck, truck–car, car–car and truck–truck) are properly reproduced by the model. In addition, the model overcomes one of the major limitations of CA models for traffic modeling: the inability to perform smooth approach to slower or stopped vehicles. Moreover, the model is also capable of reproducing most empirical findings including the backward speed of the downstream front of the traffic jam, and different congested traffic patterns induced by a system with open boundary conditions with an on-ramp. Like most CA models, integer values are used to make the model run faster, which makes the proposed model suitable for real time traffic simulation of large networks.

Decision dynamics of departure times: Experiments and modeling

A fundamental problem in traffic science is to understand user-choice behaviors that account for the emergence of complex traffic phenomena. Despite much effort devoted to theoretically exploring departure time choice behaviors, relatively large-scale and systematic experimental tests of theoretical predictions are still lacking. In this paper, we aim to offer a more comprehensive understanding of departure time choice behaviors in terms of a series of laboratory experiments under different traffic conditions and feedback information provided to commuters. In the experiment, the number of recruited players is much larger than the number of choices to better mimic the real scenario, in which a large number of commuters will depart simultaneously in a relatively small time window. Sufficient numbers of rounds are conducted to ensure the convergence of collective behavior. Experimental results demonstrate that collective behavior is close to the user equilibrium, regardless of different scales and traffic conditions. Moreover, the amount of feedback information has a negligible influence on collective behavior but has a relatively stronger effect on individual choice behaviors. Reinforcement learning and Fermi learning models are built to reproduce the experimental results and uncover the underlying mechanism. Simulation results are in good agreement with the experimentally observed collective behaviors.