Dynamic Traffic Model Research Articles

MAGT-toll: A multi-agent reinforcement learning approach to dynamic traffic congestion pricing.

Modern urban centers have one of the most critical challenges of congestion. Traditional electronic toll collection systems attempt to mitigate this issue through pre-defined static congestion pricing methods; however, they are inadequate in addressing the dynamic fluctuations in traffic demand. Dynamic congestion pricing has been identified as a promising approach, yet its implementation is hindered by the computational complexity involved in optimizing long-term objectives and the necessity for coordination across the traffic network. To address these challenges, we propose a novel dynamic traffic congestion pricing model utilizing multi-agent reinforcement learning with a transformer architecture. This architecture capitalizes on its encoder-decoder structure to transform the multi-agent reinforcement learning problem into a sequence modeling task. Drawing on insights from research on graph transformers, our model incorporates agent structures and positional encoding to enhance adaptability to traffic flow dynamics and network coordination. We have developed a microsimulation-based environment to implement a discrete toll-rate congestion pricing scheme on actual urban roads. Our extensive experimental results across diverse traffic demand scenarios demonstrate substantial improvements in congestion metrics and reductions in travel time, thereby effectively alleviating traffic congestion.

Dynamic traffic network representation model for improving the prediction performance of passenger flow for mass rapid transit

Accurate machine learning predictions of passenger flow data for mass rapid transit (MRT) systems can considerably improve operational efficiency by enabling better allocation of train and human resources. However, such predictions are challenging because MRT networks have complex structures with route dependence and transfer stations. Although the static state of an MRT network has been computed in previous studies, a comprehensive understanding of an MRT network requires characterizing its dynamics. Therefore, this paper proposes a dynamic traffic network representation (DTNR) model that captures station features from historical traffic flows and geographical information of MRT stations. Furthermore, a multilevel attention network (MLAN) model is proposed to predict MRT passenger flow as a downstream task following the pretraining of the DTNR model. The experimental results of this study indicate that the developed DTNR and MLAN models can accurately predict MRT passenger flow. These models are widely applicable to different MRT systems and passenger flow situations, making them a valuable tool for transportation planners and operators.

Macroscopic Traffic Modeling Using Probe Vehicle Data: A Machine Learning Approach

The macroscopic fundamental diagram (MFD) captures an orderly relationship among traffic flow, density, and speed at the network level. It is a simple yet powerful tool for modeling traffic dynamics in large urban networks with broad application in traffic control and management. However, empirically derived MFDs in urban regions require high-resolution traffic data from the network. Having the network flow and vehicular density estimated at the (granular) census tract level using vehicle probe data, we apply machine learning methods to predict the MFDs across U.S. urban areas and capture the impacts of location-specific input features on the network flow–density relationships at a large scale. The results show that, among the four tested machine learning approaches (Random Forest, XGBoost, Support Vector Machine, and Neural Network), XGBoost delivers the best performance in predicting network traffic flow based on vehicular density and location attributes. Using interaction Shapley Additive explanation (SHAP) values and partial correlation analysis, we examine the factors influencing MFD shapes across different locations. Our empirical findings reveal that across U.S. urban areas, network topology, transportation infrastructure, and land use are primary factors shaping MFD curves, while demand and trip-related factors play a lesser role. Specifically, higher ranking roads, centrality, and development levels correlate positively with network capacity and critical density, whereas negative associations are observed for network connectivity, mixed-use development, and road roughness levels.

Dynamic traffic noise simulation at signal-controlled intersections based on cellular automata model

Predicting and assessing road traffic noise is a crucial task in urban environmental management. To accurately predict traffic noise at signal-controlled intersections, this study introduces a dynamic traffic noise simulation model based on cellular automata. Firstly, a cellular automata traffic flow model for a multi-lane signal-controlled intersection is established. The vehicle noise emission and noise propagation models are incorporated into the dynamic simulation of traffic flow to accurately calculate traffic noise. Subsequently, the simulated data is compared with the measured data. The results demonstrate that the proposed method can effectively predict the statistical distribution of traffic noise at the intersection, with prediction errors of less than 2 dB(A) for Leq, L50, L90, and less than 3 dB(A) for L10. Finally, the proposed method is employed to simulate the operational status and noise emissions of vehicles at intersections under various traffic conditions. Based on the simulation results, the intrinsic connection between traffic operation and noise emission is explored, and the characteristics of traffic noise emission and distribution at intersections are analyzed. The method and its findings are anticipated to serve as a valuable reference for the management of road traffic noise.

Development of Dynamic Micro- and Macroscopic Hybrid Model for Efficient Highway Traffic Simulation

This paper presents a dynamic hybrid traffic simulation model and its improvement. Here, a hybrid traffic simulation model is defined as a combination of multiple traffic models with different resolutions, which are dynamically switched to each other. The proposed model couples a microscopic and a macroscopic model. In a dynamic hybrid model, the application domains of micro- and macroscopic models can be switched during the simulation to achieve overall speed-up while keeping a high resolution of the area of interest. The authors extended an existing dynamic hybrid model to be applicable to highway merging sections. The proposed model was also validated by data obtained on an actual highway.

Investigating the effect of dynamic traffic distribution on network-wide traffic emissions: An empirical study in Ningbo, China.

Urban road traffic is one of the primary sources of carbon emissions. Previous studies have demonstrated the close relationship between traffic flow characteristics and carbon emissions (CO2). However, the impact of dynamic traffic distribution on carbon emissions is rarely empirically studied on the network level. To fill this gap, this study proposes a dynamic network carbon emissions estimation method. The network-level traffic emissions are estimated by combining macroscopic emission models and recent advances in dynamic network traffic flow modeling, namely, Macroscopic Fundamental Diagram. The impact of traffic distribution and the penetration of battery electric vehicles on total network emissions are further investigated using the Monte Carlo method. The results indicate the substantial effect of network traffic distribution on carbon emissions. Using the urban expressway network in Ningbo as an example, in the scenario of 100% internal combustion engine vehicles, increasing the standard deviation of link-level traffic density from 0 to 15 veh/km-ln can result in an 8.9% network capacity drop and a 15.5% reduction in network carbon emissions. This effect can be moderated as the penetration rate of battery electric vehicles increases. Based on the empirical and simulating evidence, different expressway pollution management strategies can be implemented, such as petrol vehicle restrictions, ramp metering, congestion pricing, and perimeter control strategies.

Construction of Dynamic Traffic Pattern Recognition and Prediction Model Based on Deep Learning in the Background of Intelligent Cities

Construction of Dynamic Traffic Pattern Recognition and Prediction Model Based on Deep Learning in the Background of Intelligent Cities

A two‐stage robust optimal traffic signal control with reversible lane for isolated intersections

AbstractThe integrated design of traffic signal control (TSC) and reversible lane (RL) is an effective way to solve the problem of tidal congestion with uncertainty at isolated intersections, because of its advantage in making full use of temporal‐spatial transportation facilities. Considering the contradiction between the dynamic TSC scheme and the fixed RL scheme in one period, a two‐stage optimization method based on improved mean‐standard deviation (MSD) model for isolated intersections with historical and real‐time uncertain traffic flow is proposed. In the first stage, applying the same‐period historical data of multiple days, a robust optimal traffic signal control model with reversible lane based on MSD model (MSD‐RTR model) is put forward to obtain the fixed RL scheme and the compatible initial TSC scheme. A double‐layer nested genetic algorithm (DN‐GA) is designed to solve this model. In the second stage, applying real‐time period data and multi‐day same‐period historical data, a robust optimal dynamic traffic signal control model based on MSD model (MSD‐RDT model) is put forward to obtain the dynamic TSC scheme. Three modes which reflect the different weights of historical period and real‐time period in this MSD‐RDT model are presented to improve the model stability, and a multi‐mode genetic algorithm (MM‐GA) is designed. Finally, a case study is presented to demonstrate the efficiency and applicability of the proposed models and algorithms.

Comparative Evaluation of Road Pricing Schemes: A Simulation Approach (Australian Perspective)

Road network pricing and congestion charging continue to be debated as efficient instruments to address traffic congestion and emissions. For cities where the schemes have not been implemented yet, the impacts of these schemes are typically evaluated using transport simulation models to understand the impacts and design effective solutions before the schemes are deployed. This paper considers a simulation approach for the city of Melbourne in Australia to investigate the potential impacts of road network pricing on reducing private vehicle travel, road congestion, and vehicle emissions. The study uses a dynamic traffic simulation model developed for Melbourne using the AIMSUN modeling tool, which was extended for modeling road user pricing and congestion charging, including considerations and formulations of distance-based, delay-based, joint-distance-and-delay-based, and cordon-based schemes under low-cost, medium-cost, and high-cost regimes. The study’s contributions also include an extension of the modeling framework to include public transport options to allow for providing travelers with the option of choosing an alternative mode of transport if they do not wish to pay. A mesoscopic stochastic route choice modeling approach was adopted to examine the impact of road pricing inside a nominated charging zone within the network. The results showed it would be possible to achieve a reduction of 11% in vehicle count, a 20% reduction in travel time, a 13% reduction in emissions, and a 3% increase in travel speed within the proposed pricing zone under a high-cost pricing scenario. The results also showed a significant reduction in emissions resulting from shifting drivers who are not willing to pay the congestion charge to public transport. When 20% of car drivers shifted to public transport, carbon emissions were reduced by up to 30% and network performance improved by 45%, compared to the baseline scenario without pricing. The findings of this research provide important directions for policymakers in deciding on the type and scope of charging schemes to use and how these could reshape transportation taxation systems by moving away from taxes on vehicles through registration fees and towards user-pay taxations where travelers pay for the amount of travel they do or the pollution and emissions they are responsible for.

Variable speed limit control strategy at the entrance and exit of freeway tunnel

In order to effectively reduce the accident rate of freeway tunnel section and improve the safety factor of tunnel section, variable speed limit control scheme at the entrance and exit of tunnel is presented from the perspective of safety. First, by considering the influence of tunnel section characteristics on vehicle operation, variable speed limit cells and variable length cells are added to improve the cell transmission model and establish a dynamic traffic flow model; Second, aimed at the minimum speed dispersion and the minimum change rate of tunnel illumination, and taken the fixed speed limit and the maximum speed difference as constraints, the variable speed limit control optimization model at the entrance and exit of the tunnel is established; Third, the genetic algorithm is employed to solve the control function with meeting the safety requirement of traffic operation. The results show that compared with the fixed speed limit mode, the implementation of variable speed control in tunnel group sections can significantly reduce the dispersion of running speed, which is conducive to traffic safety and operation efficiency.

When do traffic reports make traffic better?

We present a simple dynamic traffic model to study whether public information about road conditions increases or decreases travel times. We find that the effect depends on comparing (1) the increase in travel time when a road with a pre-existing delay becomes congested, and (2) the corresponding increase on a road with no pre-existing delay. Traffic reports are helpful when the first value is large relative to the second. Surprisingly, they are less useful when total road capacity is sufficient to potentially accommodate all drivers without congestion. In that case, traffic reports are always harmful in the nonatomic limit as individual drivers become negligible.

Road traffic flow prediction based on dynamic spatiotemporal graph attention network

To improve the prediction accuracy of traffic flow under the influence of nearby time traffic flow disturbance, a dynamic spatiotemporal graph attention network traffic flow prediction model based on the attention mechanism was proposed. Considering the macroscopic periodic characteristics of traffic flow, the spatiotemporal features are extracted by constructing spatiotemporal blocks with an adjacent period, daily period, and weekly period respectively. The spatiotemporal block is mainly composed of a two-layer graph attention network and a gated recurrent unit to capture the hidden features of space and time. In space, based on considering adjacent road segments, the Pearson correlation coefficient is used to capture the hidden correlation characteristics between non-adjacent road segments according to a certain time step. In terms of time, due to the random disturbance of traffic flow at the micro level, the attention mechanism is introduced to use the adjacent time as the query matrix to weight the output characteristics of daily cycle and weekly cycle, and the three are connected in series to output the prediction results through the linear layer. Finally, the experimental results on the public data sets show that the proposed model is superior to the six baseline models.

Advanced Noise Indicator Mapping Relying on a City Microphone Network.

In this work, a methodology is presented for city-wide road traffic noise indicator mapping. The need for direct access to traffic data is bypassed by relying on street categorization and a city microphone network. The starting point for the deterministic modeling is a previously developed but simplified dynamic traffic model, the latter necessary to predict statistical and dynamic noise indicators and to estimate the number of noise events. The sound propagation module combines aspects of the CNOSSOS and QSIDE models. In the next step, a machine learning technique-an artificial neural network in this work-is used to weigh the outcomes of the deterministic predictions of various traffic parameter scenarios (linked to street categories) to approach the measured indicators from the microphone network. Application to the city of Barcelona showed that the differences between predictions and measurements typically lie within 2-3 dB, which should be positioned relative to the 3 dB variation in street-side measurements when microphone positioning relative to the façade is not fixed. The number of events is predicted with 30% accuracy. Indicators can be predicted as averages over day, evening and night periods, but also at an hourly scale; shorter time periods do not seem to negatively affect modeling accuracy. The current methodology opens the way to include a broad set of noise indicators in city-wide environmental noise impact assessment.

The Traffic Flow Prediction of Newly-Constructed Expresseway Based on ETC Database

The traffic flow prediction of newly-constructed expressway can provide reference for highway capacity design and highway toll standard formulation, but the traffic flow prediction of newly-constructed expressway is always considered a complicated project. With the wide application of electronic toll collection system, it is possible to effectively predict the traffic flow of newly-constructed expressways. In this paper, We propose a new highway traffic flow prediction method based on ETC database. ETC database can be used to extract OD matrix of expressway network, and the dynamic traffic allocation model is applied to predict the traffic flow of newly-constructed expressway. The experimental results show that the method is feasible and effective.

Intra- and inter-individual risk attitude heterogeneity and the impact on road travel: evidence from a dynamic procedural model

ABSTRACT This paper develops a dynamic procedural model to understand the day-to-day evolution of road traffic and risky mode choice behaviour. We embed a modified Bureau of Public Roads function and a queueing model into a dynamic traffic flow model to estimate information on travel time and in-vehicle crowding. Then, the estimated information is fed into a risky mode choice model, with the joint consideration of intra- and inter-individual risk attitude heterogeneity. In order to reflect what may be happening in the real market, we establish a close loop with individuals’ mode choice behaviours shaped by their travel experiences and historical information, the induced impact on traffic flow, and more importantly, the feedback and interplay between them. Our model delivers realistic and robust results, in which risk attitudes play the most prominent role.

Signal-control refined dynamic traffic graph model for movement-based arterial network traffic volume prediction

Forecasting the forthcoming intersection movement-based traffic volume enables adaptive traffic control systems to dynamically respond to the fluctuation of traffic demands. In this paper, a deep-learning based Signal-control Refined Dynamic Traffic Graph (ScR-DTG) Model is proposed for advancing the network-level movement-based traffic volume prediction task. The proposed model attempts to further improve the-state-of-art and practice algorithms in traffic prediction for arterial network adaptive signal control utilizing tradition traffic flow theory boosted deep-learning methodology. For precisely inferencing the movement-based demand at cycle-to-cycle level, the proposed model incorporates spatial graph convolution inferencing layer and temporal inferencing layer to explore both the intricate spatial temporal dependencies, respectively. A signal control refining module is contrived to deduce the controlled movement saturation flow and introduce the essential control inferences, which is of great significance but frequently neglected in the previous researches. Additionally, according to the real-time movement specified travel time, this paper creatively constructs an adjacent graph with dynamic order for more accurately capturing the ever-changing spatial relevancies. Field experiments with multiple signal schemes were conducted in the downtown area of Zhangzhou (China). The promising results demonstrated the state-of-the-art accuracy than other high-performance volume prediction algorithms. Implementing the proposed model enables to obtain accurate movement-based volume predictions, which would assist the traffic management agencies in adjusting signal timing adaptively and further improve the efficiency of signal intersection.

Internet of Vehicles Data-Oriented Arterial Travel Time Estimation Framework With Dynamic Multigraph Model

The rapid development of Internet of Vehicles (IoV) data powers various online intelligent transportation applications, such as network travel time reporting. However, the accuracy might be severely compromised due to limited probe vehicle sampling frequency. On that account, this article ­proposes a dynamic multigraph model-enabled framework to estimate reliable network travel time, even in low-IoV-frequency arterial corridors. The proposed framework first develops an improved sparse IoV travel time decomposition method. The segment travel time is further divided into the free-flow running time and static and dynamic delays. Second, a dynamic multigraph traffic network model (DMGTN) is developed to aid the proposed decomposition method. The model analyzes complicated spatiotemporal relevance between segments from multiple perspectives: the real-time travel time, congestion level, signal control (which is frequently neglected in previous research), and segment properties. Additionally, two distinct enhanced modules are designed for handing dense and sparse network graphs, respectively. This allows for a more efficient inspection over large-scale intricate arterial networks while maintaining precision. Field implementation is conducted in the downtown area of Zhangzhou, China. Compared to other high-performance baseline models, the designed DMGTN model as well as the proposed decomposition method demonstrate state-of-the-art accuracy and successfully capture travel time variability. The proposed framework better utilizes available IoV data to provide valuable traffic information for commuters and traffic management agencies.

An Approach to Planning Scenic Routes by Integrating Dynamic Traffic Models with A* Algorithm

<div>During the entry and exit of attraction viewing, the rapid generation of travel demand and converging traffic flows in a short period can easily pose safety hazards to people due to its complex terrain. This study aims to propose a path planning method that integrates a dynamic traffic model with the A* algorithm for the planning of scenic routes. The study first combines the cellular transport model (CTM) model with the Greenshield model as its dynamic traffic model and then improves the A* algorithm with the Morphin search tree algorithm (Morphin) as its scenic route planning. The results of the study show that the improved A* algorithm reaches the expected error of 10<sup>−4</sup> after 21 ms using Matlab tests, and simulation tests are conducted in regular and complex sections of the scenic area. The results show that the improved A* algorithm has a significant improvement over the A* algorithm in node selection, and its performance indexes such as the number of inflection points and smoothness also have a significant improvement effect, the improved A* algorithm has a significant decrease in the search area, and the search efficiency of the A* algorithm has also improved greatly.</div>

Dynamic Traffic Allocation Model Considering the Effects of Vehicle Emission Diffusion

Vehicle exhausts diffuse into roadside crowd breathing zones, thereby jeopardising human health. This study applies dynamic traffic distribution theories to comprehensively consider the impact of vehicle emission diffusion. The results provide a theoretical basis for improving the diffusion of urban traffic pollution to benefit the surrounding environment for roadside crowds. Firstly, a multi-vehicle cellular transport model that is suitable for analysing dynamic traffic distribution is constructed considering the distinct emission factors of various types of vehicles. Secondly, a multi-vehicle emission model is established to consider a range of driving conditions. Then, the concept of roadside crowd exposure risk is introduced, and we describe a method for calculating the total amounts of pollutants emitted by vehicles and inhaled by roadside crowds. The impact of vehicle emission diffusion is comprehensively discussed in terms of vehicle emissions and roadside crowd exposure risk. A generalised impedance function considering the influence of vehicle exhaust emission diffusion is also established based on the weighted average of actual vehicle travel time, multi-model emissions and roadside crowd exposure risk. Finally, this generalised impedance function is integrated into the dynamic optimal user allocation model, and a dynamic traffic allocation model considering the influence of vehicle emission control is developed.

Data-driven control of agent-based models: An Equation/Variable-free machine learning approach

We present an Equation/Variable free machine learning (EVFML) framework for the control of the collective dynamics of complex/multiscale systems modeled via microscopic/agent-based simulators. The approach obviates the need for construction of surrogate, reduced-order models. The proposed implementation consists of three steps: (A) from high-dimensional agent-based simulations, machine learning, in particular, non-linear manifold learning (Diffusion Maps (DMs)) identifies a set of coarse-grained variables that parametrize the low-dimensional manifold on which the emergent/collective dynamics evolve. The out-of-sample extension and pre-image problems, i.e. the construction of non-linear mappings from the high-dimensional input space to the low-dimensional manifold and back, are solved by coupling DMs with the Nyström extension and Geometric Harmonics, respectively; (B) having identified the manifold and its coordinates, we exploit the Equation-free approach to perform numerical bifurcation analysis of the emergent dynamics; then (C) based on the previous steps, we design data-driven embedded wash-out controllers that drive the agent-based simulators to their intrinsic, imprecisely known, emergent open-loop unstable steady-states, thus demonstrating that the scheme is robust against numerical approximation errors and modeling uncertainty. The efficiency of the framework is illustrated by controlling emergent unstable (i) traveling waves of a deterministic agent-based model of traffic dynamics, and (ii) equilibria of a stochastic financial market model with mimesis.