Entire Traffic Flow Research Articles

Self-adaptive equation embedded neural networks for traffic flow state estimation with sparse data

The data-driven approach in intelligent traffic systems has achieved successive breakthroughs, thanks to the ever-increasing volume of traffic data. Nonetheless, in practical scenarios, the collected data often contain some issues, e.g., missing values, significantly impacting the accuracy and efficiency of the algorithms. To enhance the precision of traffic estimation utilizing the sparse data, we have developed a physics-informed neural network (PINN) based algorithm in the line with the traffic flow theory and deep learning principles. In contrast to the conventional PINNs, our approach uniquely incorporates a self-adaptive macro model for mixed flow into the network's architecture, serving as an embedded source of physics information. With this algorithm, we can capture the dynamic behavior of an entire traffic flow including its spatiotemporal evolution with sparse traffic data such as initial and boundary value information. To realize the model's adaptability, we have revised the macro model by inverting its parameters and incorporating a data-driven nonlinear element, which simplifies the intricate macro model structure. The network's effectiveness has been validated through the experiments conducted on a mixed traffic flow system experiencing local agglomeration and real-world data, demonstrating its capability for precise traffic simulation, efficient traffic flow prediction, and interpretability. Our study offers a novel insight for data-driven traffic flow state estimation.

Choice-based macroscopic lane-change prediction model for weaving areas

This paper introduces two macroscopic lane change models, separately for small vehicles (passenger cars and vans) as well as heavy vehicles. The model is structured using a Nested Logit discrete choice framework. The model was estimated and validated using unique trajectory data collected from a busy weaving section in Antwerp, Belgium. The models exhibit an average and maximum rate of inaccurate prediction results, amounting to 11% and 16%, respectively. Additionally, the model underestimates the total number of lane change maneuvers, with a margin of no more than 4.1–7.7%. This study aims to fill the gap in macroscopic lane change models, which mainly depend on theoretical underpinnings, by including empirical data from the analysis of more than 31,000 observed maneuvers. Furthermore, it differentiates among three maneuver intents, a unique alternative-specific discrete choice framework. This differentiation allows for the consideration of drivers’ systematic taste variations. In contrast to other models that utilize empirical datasets and are heavily dependent on expensive video-based data, including the entire traffic flow, this study introduces a novel approach for estimating the position and frequency of maneuvers by using just a small fraction (1–2%) of vehicle trajectories, supplemented by loop detector data, therefore eliminating the need for information from surrounding vehicles. Additionally, the work’s uniqueness is its validation methods extending beyond just numerical evaluations. The practical usability and usefulness of the model are shown in real-world traffic circumstances. Finally, the model is integrated in the context of a macroscopic simulation that represents a highly consistent lane balance based on the Scalable Quality Value statistics.

Probabilistic Analysis of Random Check Intrusion Detection System

The ubiquitous adoption of network-based technologies has left organizations vulnerable to malicious attacks. It has become vital to have effective intrusion detection systems (IDS) that protect the network from attacks. In this paper, we study the intrusion detection problem through the lens of probability theory. We consider a situation where a network receives random malicious signals at discrete time instances, and an IDS attempts to capture these signals via a random check process. We aim to develop a probabilistic framework for intrusion detection under the given scenario. Concretely, we calculate the detection rate of a network attack by an IDS and determine the expected number of detections. We perform extensive theoretical and experimental analyses of the problem. The results presented in this paper would be helpful tools for designing and analyzing intrusion detection systems. We propose a probabilistic framework that could be useful for IDS experts; for a network-based IDS that monitors in real-time, analyzing the entire traffic flow can be computationally expensive. By probabilistically sampling only a fraction of the network traffic, the IDS can still perform its task effectively while reducing the computational cost. However, checking only a fraction of the traffic increases the possibility of missing an attack. This research can help IDS designers achieve appropriate detection rates while maintaining a low false alarm rate. The groundwork laid out in this paper could be used for future research on understanding the probabilities related to intrusion detection.

Impact Analysis of Driver Takeover Under a Connected Environment: A Study of the External Effects on Vehicle Conversions

The moment we face cooperative adaptive cruise control (CACC), driver takeover, that is, drivers actively take over vehicles to avoid unknown risks, is a necessary concern in reality, especially with the distrust or low acceptance of automated driving technology in the early stage of CACC development. However, the impact of driver takeover on traffic flow has not been accurately presented, as researchers only focused on the taken-over vehicle and neglected the external effects on surrounding vehicles. Departing from the literature, this study provides a novel external perspective for the impact analysis of driver takeover under a connected environment, that is, it considers the resulting communication heterogeneity and specifies multi-class conversions in the entire traffic flow. To be specific, we subdivide the dynamic process of driver takeover into 10 vehicle conversions considering vehicle degradation and platoon reorganization, then analyze their contributions to traffic stability, emissions, and fuel consumption. What is also innovative is that two forms of driver takeover are first developed in this study, which is determined by whether drivers continue opening the communication device after the takeover. On this basis, this study indicates that driver takeover has the potential to cause greater heterogeneity of the traffic flow in contrast to that in the literature and proposes theoretical models for the changed traffic flow configuration and traffic stability after the takeover. Finally, numerical experiments verify the theoretical results with discussions on multiple factors, for example, vehicle proportion, platoon length, and driving behaviors. The findings extend the external effects of driver takeover on multi-class vehicle conversions in heterogeneity traffic flow and provide a more realistic basis for the evaluation of driver takeover.

Blending of Floating Car Data and Point-Based Sensor Data to Deduce Operating Speeds under Different Traffic Flow Conditions

Nowadays, smart mobility can rely on innovative tools for the knowledge of road system conditions, like operating speed data extracted from the so-called Floating Car Data (FCD). Probe vehicles in the traffic flow send to operation centres a large amount of travel information, collected through GPS detection systems, especially with regard to geolocation, date and time, direction and speed. As the sample deriving from these vehicles represents a tiny portion of the entire vehicular fleet, in this paper an analysis and a comparison with data obtained by point-based traffic sensors is proposed. Therefore, the study analyses data collected by inductive loop detectors and microwave radar sensors, that provide information on the entire traffic flow in the time domain, in particular with the aim to identify free flow speed time bands. Afterwards, by means of the fusion between the results obtained from the data coming from these point-based control units and the ones coming from the probe vehicles, a comparison of the operating speeds in the two conditions of constrained and unconstrained traffic flow is performed.

A PDE-ODE model for traffic control with autonomous vehicles

<abstract><p>We consider a partial differential equation - ordinary differential equation system to describe the dynamics of traffic flow with autonomous vehicles. In the model, the bulk flow of human drivers is represented by a scalar conservation law, while each autonomous vehicle is described by an ordinary differential equation. The coupled PDE-ODE model is introduced, and existence of solutions for this model is shown, along with a proposed algorithm to construct approximate solutions. Next, we propose a control strategy for the speeds of the autonomous vehicles to minimize the average fuel consumption of the entire traffic flow. Existence of solutions for the optimal control problem is proved, and we numerically show that a reduction in average fuel consumption is possible with an AV acting as a moving bottleneck.</p></abstract>

Ecological Driving for Connected and Automated Vehicles at Unsaturated Intersections Considering Queue Effects

In this paper, an ecological driving (eco-driving) algorithm considering queue effects is proposed for connected and automated vehicles (CAVs) at unsaturated intersections in order to reduce fuel consumption and travel time. Firstly, after the traffic flow parameters are obtained using vehicle-to-infrastructure communication technology, the kinematic shockwave model is used to predict the vehicle queue length at the saturated intersections. Secondly, to decrease fuel consumption, a fuel-saving optimization problem is formulated using the estimated queue length, and for real-time control, the formulated optimization problem is decomposed into two subproblems depending on whether the CAV will stop to queue. Then, to reduce fuel consumption and travel time simultaneously, the eco-driving algorithm is designed, where the trajectory re-optimization process is implemented in order not to block the upstream vehicles. Finally, extensive simulations are carried out on VISSIM to demonstrate the control performance of the proposed eco-driving algorithm on CAVs and on the entire traffic flow. Simulation results show that the proposed eco-driving algorithm can significantly decrease the fuel consumption and travel time of both CAVs and the traffic flow, and higher market penetration rate of CAVs can result in better control performance.

Ego-efficient lane changes of connected and automated vehicles with impacts on traffic flow

Connected and automated vehicles (CAVs) enabled by wireless communication and vehicle automation are believed to revolutionize the form and operation of road transport in the next decades. This paper addresses traffic flow effects of CAVs, and focuses on their lane-changing impacts on the mixed traffic flow of CAVs and human-driven vehicles (HVs). At present technical paths towards the development and deployment of CAVs are still uncertain. With CAV technologies getting matured, CAVs are supposed to provide rides of higher efficiency than HVs, beyond improved safety and comfort. In heavy traffic, this would only be achievable via agile and flexible lane changes of CAVs, because longitudinal acceleration would be unhelpful or even impossible in mixed traffic. Such lane changes are expected to be ego-efficient in that they serve solely CAVs’ interests without much considering surrounding vehicles, as long as safety constraints are not violated. As road resources are limited, the growth of the CAV population adopting such ego-efficient lane-changing strategies would probably lead to renowned “Tragedy of the Commons”. In this context, this paper considers three important prospective questions:A: How to determine an ego-efficient lane-changing strategy for CAVs?B: With increasingly more CAVs introduced each adopting the ego-efficient lane-changing strategy, what is the impact on traffic flow?C: How to determine a system-efficient lane-changing strategy for CAVs?These forward-looking issues are addressed from the perspectives of microscopic traffic simulation and reinforcement learning. Without any constraint on the lane-changing incentive, the developed lane-changing strategy was found to be beneficial for CAVs and the entire traffic flow only if the market penetration rate (MPR) of CAVs is less than 50%. With an appropriate constraint placed, however, the lane-changing strategy was found to become consistently beneficial for the entire traffic flow at any MPR. These findings suggest that CAVs may not simply be a magic cure for traffic problems that the society is currently facing, unless some upper-level coordination may be proposed for CAVs to benefit not only themselves but also the entire traffic. This is also consistent with what “Tragedy of the Commons” suggests.

An analytical model for quantifying the efficiency of traffic-data collection using instrumented vehicles

Emerging instrumented vehicles (IVs), when equipped with high-precision positioning devices (e.g., DGPS) and ranging sensors (e.g., radar, LIDAR, cameras), are capable of generating high-quality traffic data. This study evaluates the efficiency of such data-collection procedures; this remains largely unknown because variable penetration rates of IVs rarely arise in the real world. We propose an analytical model that establishes a quantitative relationship between the ratio of collected trajectory points to total traffic trajectory points (RCT) and the IV penetration rate, according to stochastic geometry theory. With this, the data-collection efficiency (DCE) of IVs can be effectively evaluated. A simulation approach is developed to generate IV data and thereby validate the proposed analytical model, using a comprehensive set of traffic scenarios; these data consist of eight micro-trajectory datasets from the next-generation simulation (NGSIM) program and four typical sensor IV deployments. The numerical analysis demonstrates that the model perfectly reflects the simulated IV data for all traffic scenarios. In addition, an analytical comparison of the DCEs for fixed sensors, probe vehicles, and IVs reveals that IVs are the most efficient method for collecting traffic data in road networks. This study proposes a theory to predict the percentage of trajectory points that can be collected by a certain percentage of IVs within the entire traffic flow.

An Analytical Model for Quantifying the Efficiency of Traffic-Data Collection Using Instrumented Vehicles

Emerging instrumented vehicles (IVs), when equipped with high-precision positioning devices (e.g., DGPS) and ranging sensors (e.g., radar, LIDAR, cameras), are capable of generating high-quality traffic data. This study evaluates the efficiency of such data-collection procedures; this remains largely unknown because variable penetration rates of IVs rarely arise in the real world. We propose an analytical model that establishes a quantitative relationship between the ratio of collected trajectory points to total traffic trajectory points (RCT) and the IV penetration rate, according to stochastic geometry theory. With this, the data-collection efficiency (DCE) of IVs can be effectively evaluated. A simulation approach is developed to generate IV data and thereby validate the proposed analytical model, using a comprehensive set of traffic scenarios; these data consist of eight micro-trajectory datasets from the next-generation simulation (NGSIM) program and four typical sensor IV deployments. The numerical analysis demonstrates that the model perfectly reflects the simulated IV data for all traffic scenarios. In addition, an analytical comparison of the DCEs for fixed sensors, probe vehicles, and IVs reveals that IVs are the most efficient method for collecting traffic data in road networks. This study proposes a theory to predict the percentage of trajectory points that can be collected by a certain percentage of IVs within the entire traffic flow.

Detection of anomalous vehicles using physics of traffic

The world is embracing the presence of connected autonomous vehicles which are expected to play a major role in the future of intelligent transport systems. Given such connectivity, vehicles in the networks are vulnerable to making incorrect decisions due to anomalous data. No sophisticated attacks are required; just a vehicle reporting anomalous speeds would be enough to disrupt the entire traffic flow. Detection of such anomalies is vital to ensure the security of a vehicular network. We propose the use of traffic flow theory for anomalous data detection in vehicular networks, by evaluating the consistency of microscopic parameters which are derived by traffic flow theory (i.e. speed and space-headway) with macroscopic views of traffic under different traffic conditions. Though a little attention has been given to using traffic flow properties to determine anomalous basic safety message (BSM) data, the fundamental nature of traffic flow properties makes it a robust assessment tool. Usually, traffic flow data are determined through roadside units (RSUs) such as cameras and loop detectors; they are financially impractical to roll out on an entire network. Therefore, the method proposed in this study establishes traffic flow data that are used as “ground truth” through RSUs if available, or by the vehicles' own sensor systems. The numerical results indicate that the proposed method provides extremely reliable and consistent predictions of anomalous BSM data. The more the road segment is congested, the higher the accuracy of the anomalous space-headway detection. The anomalous speed detection performs robustly well across all the traffic conditions. The study also finds that both global and local ground truths provide consistent results.

A minority of self-organizing autonomous vehicles significantly increase freeway traffic flow

This study investigates the dynamics of traffic containing human-driven vehicles along with a fraction of self-organized artificial intelligence (AI) autonomous vehicles (AVs) on multilane freeways. We propose guidelines for the development of AI agents, such that a small fraction of AVs forms local constellations that significantly accelerate the entire traffic flow while reducing fuel consumption and increasing safety. Specifically, we report a 40% enhancement in traffic flow efficiency and up to a 28% reduction in fuel consumption even when only 5% of vehicles are autonomous. This scenario does not require changes to current infrastructure or communication between vehicles; it only requires proper regulations. The results indicate that more efficient, safer, faster, and greener traffic flow can be realized in the near future.

Vehicle trajectory reconstruction for signalized intersections: A hybrid approach integrating Kalman Filtering and variational theory

Trajectory data provide a rich source of information on spatial and temporal speed fluctuations of individual vehicles. To assemble a complete picture of traffic operations, obtaining vehicle trajectories of the entire traffic flow is essential. The current techniques cannot fulfill this need sufficiently. To this end, we developed a hybrid approach for reconstructing vehicle trajectories at signalized intersections that accounts for queue dynamics. Specifically, an extended variational solution network was established and Kalman Filtering was applied to capture and reproduce the stochastic properties of queue boundary curves. The proposed method was demonstrated using both NGSIM data and simulation data. Results showed that the hybrid approach reproduces accurately the shockwave boundaries and trajectory reconstruction. By fusing different sources of traffic data, the proposed method does not rely on the common assumption of uniform arrival between adjacent probe vehicles, which is the main drive for estimation improvement.

До питання стійкості метробуса у гальмівному режимі

According to the recommended ratio between the number of buses of different classes used in large cities, three-quarters of the bus fleet of a city with a population of more than 1 million inhabitants should be buses of large and especially large class. The greatest effect of using buses of large and especially large class is achieved in the BRT system. The use of multi-link both freight and passenger trains should be confirmed not only by progressive transport legislation, but also by solving certain technical problems aimed at ensuring the high safety of trains and the entire traffic flow. At the same time, it is very important to ensure the stability of the vehicle in boundary modes so that it remains within the overall corridor. Conducted studies to date have proved that the most dangerous in terms of ensuring the stability of the road train in emergency braking is the onboard non-uniformity of braking forces, and with increasing the initial speed of braking, non-uniformity of action of the brake mechanisms on wheels of the same axles of cars and road trains, significantly angular deviations. It is shown that a generic parameter characterizing the stability of straight-line movement of a three-lane metro bus in braking mode should be taken to be the initial speed of movement at which the bus and trailer links do not extend beyond the width of the lane. According to this criterion, the initial braking speeds are obtained, which ensure the stability of the metrobus with both a guided and an unmanaged second trailer. These velocities were 12.5 and 16.5 m / s, respectively, when the coefficient of adhesion of the wheels to the support surface within j = 0.7 and the moment of non-uniformity of the braking forces on the axles of the trailers within 8000 Nm were respectively. This must be taken into account when determining the stability of the metrobus in braking mode.

Optimal driving strategies for traffic control with autonomous vehicles

This article considers the possibility of using a small number of autonomous vehicles (AV) for traffic control of the predominantly human-piloted traffic. Specifically, we consider the control of the AV to act as a moving bottleneck, which will be used to optimize traffic flow properties such as fuel consumption of the combined human-piloted and autonomous traffic flow. We use a coupled partial differential equation (PDE)-ordinary differential equation (ODE) framework to model the bulk traffic flow using a PDE, and the trajectory of an autonomous vehicle in the flow using an ODE, depending on the downstream traffic density. The autonomous vehicle acts on the traffic flow as a moving bottleneck via a moving flux constraint. Using this modeling framework, we consider an optimal control problem which consists in finding the optimal AV trajectory to minimize fuel consumption of the entire traffic flow. We prove existence of optimal AV trajectories and we present two different optimal driving strategies depending on the initial traffic conditions.

Optimal Formation of Autonomous Vehicles in Mixed Traffic Flow

Platooning of multiple autonomous vehicles has attracted significant attention in both academia and industry. Despite its great potential, platooning is not the only choice for the formation of autonomous vehicles in mixed traffic flow, where autonomous vehicles and human-driven vehicles (HDVs) coexist. In this paper, we investigate the optimal formation of autonomous vehicles that can achieve an optimal system-wide performance in mixed traffic flow. Specifically, we consider the optimal H2 performance of the entire traffic flow, reflecting the potential of autonomous vehicles in mitigating traffic perturbations. Then, we formulate the optimal formation problem as a set function optimization problem. Numerical results reveal two predominant optimal formations: uniform distribution and platoon formation, depending on traffic parameters. In addition, we show that 1) the prevailing platoon formation is not always the optimal choice; 2) platoon formation might be the worst choice when HDVs have a poor string stability behavior. These results suggest more opportunities for the formation of autonomous vehicles, beyond platooning, in mixed traffic flow.

Eco approaching at an isolated signalized intersection under partially connected and automated vehicles environment

This research proposed an eco-driving system for an isolated signalized intersection under partially Connected and Automated Vehicles (CAV) environment. This system prioritizes mobility before improving fuel efficiency and optimizes the entire traffic flow by optimizing speed profiles of the connected and automated vehicles. The optimal control problem was solved using Pontryagin’s Minimum Principle. Simulation-based before and after evaluation of the proposed design was conducted. Fuel consumption benefits range from 2.02% to 58.01%. The CO2 emissions benefits range from 1.97% to 33.26%. Throughput benefits are up to 10.80%. The variations are caused by the market penetration rate of connected and automated vehicles and v/c ratio. No adverse effect is observed. Detailed investigation reveals that benefits are significant as long as there is CAV and they grow with CAV’s market penetration rate (MPR) until they level off at about 40% MPR. This indicates that the proposed eco-driving system can be implemented with a low market penetration rate of connected and automated vehicles and could be implemented in a near future. The investigation also reveals that the proposed eco-driving system is able to smooth out the shock wave caused by signal controls and is robust over the impedance from conventional vehicles and randomness of traffic. The proposed system is fast in computation and has great potential for real-time implementation.

Dynamic Resource Allocation for Immediate and Advance Reservation in Space-Division-Multiplexing-Based Elastic Optical Networks

Numerous studies have investigated elastic optical networks (EONs) with the aim of expanding the transmission capacities of core networks. To achieve this goal, it is necessary to solve the spectrum resource wastage problem caused by spectrum fragmentation. Moreover, due to the potentially high traffic demands in future networks, it is important to handle requests that need to be reserved immediately (immediate reservation, IR) as well as those that can be reserved in advance (advance reservation, AR). In networks that support the coexistence of IR and AR requests, IR service degradation by AR requests is a challenging issue because AR requests tend to reserve future resources, which causes a lack of current resources to meet IR requests. Therefore, we address the problem of spectrum fragmentation and the service-level control of IR and AR requests by routing and spectrum allocation (RSA). First, we summarize related research into EONs and resource- allocation methods for IR and AR requests. Next, we propose a novel dynamic RSA method to reduce spectrum fragmentation and control the service level of IR and AR requests in terms of bandwidth blocking probability (BBP) in EONs considering the multiplexing effect of spatial channels. Finally, we evaluate the proposed method based on computer simulations and our results demonstrate that the proposed method can improve the BBP for the entire traffic flow by reducing spectrum fragmentation, as well as the service control of AR requests and IR requests under various network conditions.

Real‐virtual fusion model for traffic animation

AbstractIn this paper, we present an innovative, animated traffic simulation method that we designed to feature an enhanced sense of reality and diversity of traffic flows. Instead of the typical one‐off initialization, our simulation method includes continuous, real trajectory data input providing an interactive control function that maximizes the characteristics of real‐world traffic flows. Our fusion models represent a comprehensive integration of the interactions among real‐data‐driven and virtual vehicles, thus depicting accurately the irregularity of traffic flows. Test results showed that animations generated via our proposed method depict inverse and irregular vehicle driving behaviors throughout the entire traffic flow.

Comparison of Variability of Individual Vehicle Delay and Average Control Delay at Signalized Intersections

Two types of delay variability have been commonly referred to at signalized intersections: arrival-time-dependent individual vehicle delay variability and average control delay variability. The former focuses on the delay experienced by individual vehicles, whereas the latter deals with the average consequence of signal control for the entire traffic flow during a given evaluation period. With a number of methods available for estimation, little research has been conducted to enable a comprehensive comparison of these two variability measures. This issue is addressed by a comparison of these two types of delay variability at isolated, fixed-time controlled intersections based on two analytical models accounting for the stochastic nature of traffic arrivals and overflow queues. The results show that in undersaturated conditions, individual vehicle delay tends to have a wider distribution and a larger variance compared with average control delay. In oversaturated conditions, the shapes of both types of delay distributions shift to the higher values with larger variances. Different arrival processes are found to have a significant influence on the pairwise delay distributions. In addition, the width of the delay distribution based on percentiles is used to quantify the delay uncertainty under different degrees of saturation. Compared with average control delay, individual vehicle delay is found to have a larger uncertainty in undersaturated conditions, whereas it becomes relatively certain in oversaturated conditions. All the analyses have special practical implications for enriched knowledge of stochastic arrival and departure processes at signals.