Local Traffic Information Research Articles

HLGST: Hybrid local–global spatio-temporal model for travel time estimation using Siamese graph convolutional with triplet networks

Travel time estimation (TTE) is a crucial and challenging task due to the complex spatial and dynamic temporal correlations between local and global traffic regions. Though many existing methods have used multi-graph structures to model traffic variations, the majority of them are unable to capture such sophisticated local and global perspectives dynamically at different time-intervals. Furthermore, these methods have used shallow graphs (i.e., based on speed or distance between traffic nodes), thereby limiting their ability to learn the latent dependencies. To overcome these limitations, we propose a novel Hybrid Local-Global Spatio-Temporal (HLGST) framework for TTE. Specifically, we first introduce a dynamic composition unit that builds local traffic information and multi-dynamic semantic graphs based on the similarities between nodes. Then, the HLGST model learns the local and global dependencies via hybrid correlation method. It mainly comprises two modules: (1) local correlation module that integrates casual TCN layers with a self-attention mechanism to capture local dependencies of traffic patterns. (2) triplet-siamese GCN (TS-GCN) module that is employed to smoothly capture global dependencies based on triplet relationships between multi-dynamical semantic graphs. Moreover, we build a dynamic adaptive learning algorithm to transfer the gained knowledge and leverage it to model multi-objective prediction tasks collaboratively. Extensive experiments conducted on two real-world traffic datasets (Xi’an and Chengdu) demonstrate that our HLGST model outperforms compared baselines and achieves significant improvement.

A general approach to smoothing nonlinear mixed traffic via control of autonomous vehicles

Stop-and-go waves are easily caused by unstable traffic due to the collective behavior of human drivers, resulting in higher vehicle fuel consumption and emissions. In this article, we develop a general approach to synthesizing feedback controllers of autonomous vehicles (AVs) for smoothing unstable nonlinear mixed traffic flow, consisting of AVs and human-driven vehicles (HVs). In the context of traffic smoothing and stabilization, prior studies have mainly focused on analyzing string stability of specific car-following models linearized at flow equilibrium points. By contrast, we are interested to analytically synthesize appropriate feedback controllers of AVs for smoothing nonlinear mixed traffic in its general functional forms, covering a broad class of deterministic car-following models commonly seen in the literature. Specifically, by leveraging feedback control theory AVs are controlled to operate in such a way that they closely track a virtual speed profile, i.e., a subtler version of the disturbance resulting from the immediate preceding vehicle. Thus, traffic waves are reduced when propagating backward across controlled AVs. Based on the general functional form of car-following dynamics, we derive a class of effective additive AV controllers that are proven to be able to ensure convergence in tracking desired speed profiles, leading to smoother traffic. In addition, a set of sufficient conditions is devised for guaranteeing car-following safety. More importantly, head-to-tail string stability of the mixed traffic is likely to be achieved with a sufficient market penetration rate (MPR) of AVs. Notably, unlike many existing studies the feedback controllers synthesized for AVs require only local traffic information without having to rely on high degrees of vehicle connectivity, and the rate of smoothing mixed traffic is readily tunable, which is useful for practical implementations. The proposed approach is further illustrated with a theoretical intelligent driver model (IDM) and commercially available adaptive cruise control (ACC) vehicles represented by a well calibrated IDM. Extensive numerical results are presented to show the effectiveness and robustness of the feedback controllers synthesized for AVs on smoothing nonlinear mixed traffic.

WSNet: A local–global consistent traffic density estimation method based on weakly supervised learning

The available traffic density estimation methods mainly rely on label maps generated by manual annotation to train a network model and enhance the ability of the network model to extract traffic flow feature information. This fully supervised approach requires a lot of manual annotation, which is a time-consuming and laborious task. To solve this problem, this paper proposes a local–global consistent traffic density estimation method based on weakly supervised learning (WSNet). This method extracts the global traffic flow feature information through a trans-traffic module, solves the problem that the traffic flow feature cannot be fully extracted due to the limited receptive fields of convolutional neural networks (CNNs), extracts the local traffic flow feature information through a feedback module, enhances the local representation ability of the traffic flow feature information, and solves the problem of CNNs lacking inductive bias capability due to the use of a transformer only. This method extracts the global traffic flow feature information through the trans-traffic module and the local traffic flow feature information through the feedback module. In addition, a local–global consistency loss function (Lc) is added into the training process and combined with the L1 loss function to strengthen the constraints on traffic density estimation, which effectively improves the accuracy of traffic density estimation. The experimental results show that this method significantly reduces the gap between fully supervised traffic density estimation and weakly supervised traffic density estimation, and the MAE and MSE values of this method are reduced to 4.33 and 5.82, respectively, on the TRANCOS dataset and to 3.90 and 5.8 on the VisDrone2019 Vehicle dataset.

Dramatic Lockdown Fossil Fuel CO2 Decrease Detected by Citizen Science-Supported Atmospheric Radiocarbon Observations.

COVID-19 lockdowns resulted in dramatic changes to fossil fuel CO2 emissions around the world, most prominently in the transportation sector. Yet travel restrictions also hampered observational data collection, making it difficult to evaluate emission changes as they occurred. To overcome this, we used a novel citizen science campaign to detect emission changes during lockdown and engage youth in climate science. Citizen scientists collected grass samples from their garden or local park, from which we analyzed the radiocarbon content to infer the recently added atmospheric fossil fuel CO2 mole fraction at each sampling location. The local fossil fuel CO2 mole fractions during lockdown were compared with a “normal” nonlockdown period. Our results from 17 sites in five cities around New Zealand demonstrate dramatic reductions in traffic emissions of 75 ± 3% during the most severe lockdown restriction period. This is consistent with sparse local traffic count information and a much larger decrease in traffic emissions than reported in global aggregate estimates of emission changes. Our results demonstrate that despite nationally consistent rules on travel during lockdown, emission changes varied by location, with inner-city sites typically dominated by bus traffic showing smaller decreases in emissions than elsewhere.

Design of a Two Layered Blockchain-Based Reputation System in Vehicular Networks

Vehicles on the road, where an Intelligent Transportation System (ITS) is built, can share a lot of traffic information and drive more safely and efficiently through data sharing. Since incorrect information misleads vehicles and causes confusion in traffic, a vehicular trust model is needed to check the message’s trustworthiness while considering the vehicle’s properties and protecting its privacy. In this paper, we proposed a two layered blockchain-based reputation system, which consists of a local one-day message blockchain and a global vehicle reputation blockchain. It can administrate the reputation score securely and preserve the vehicle’s partial privacy. The proposed model efficiently manages local traffic information through the local one-day blockchain, reducing the memory overhead of vehicles. As the vehicle’s actual identity and activities in other areas are hidden by using one-time public keys, partial privacy of the vehicle is preserved. According to the activity of the vehicle, the vehicle’s reputation score is updated and stored permanently in the global reputation blockchain. We also suggested the location-based practical byzantine fault tolerance (LPBFT), a new consensus algorithm for fast block generating. The LPBFT lowers message propagation time through location-based primary node selection and is about 1.4 times faster than existing PBFT. The simulation results show the efficiency and the feasibility of LPBFT and our proposed protocol.

Incentive-based decentralized routing for connected and autonomous vehicles using information propagation

Routing strategies under the aegis of dynamic traffic assignment have been proposed in the literature to optimize system performance. However, challenges have persisted in their deployment ability and effectiveness due to inherent strong assumptions on traveler behavior and availability of network-level real-time traffic information, and the high computational burden associated with computing network-wide flows in real-time. To address these gaps, this study proposes an incentive-based decentralized routing strategy to nudge the network performance closer to the system optimum for the context where all vehicles are connected and autonomous vehicles (CAVs). The strategy consists of three stages. The first stage incorporates a local route switching dynamical system to approximate the system optimal route flow in a local area based on vehicles’ knowledge of local traffic information. This system is decentralized in the sense that it only updates the local route choices of vehicles in this area rather than route choices of all vehicles in the network, which circumvents the high computational burden associated with computing the flows on the entire network. The second stage optimizes the route for each CAV by considering individual heterogeneity in traveler preferences (e.g., the value of time) to maximize the utilities of all travelers in the local area. Constraints are also incorporated to ensure that these routes can achieve the approximated local system optimal flow of the first stage. The third stage leverages an expected envy-free incentive mechanism to ensure that travelers in the local area can accept the optimal routes determined in the second stage. The study analytically discusses the convergence of the local route switching dynamical system. We also show that the proposed incentive mechanism is expected individual rational and budget-balanced, which ensure that travelers are willing to participate and guarantee the balance between payments and compensations, respectively. Further, the conditions for the expected incentive compatibility of the incentive mechanism are analyzed and proved, ensuring behavioral honesty in disclosing information. Thereby, the proposed incentive-based decentralized routing strategy can enhance network performance and user satisfaction under fully connected and autonomous environments.

Congestion recognition for hybrid urban road systems via digraph convolutional network

Congestion recognition is the prerequisite for traffic control and management, vehicle routing, and many other applications in intelligent transportation systems. Different types of roads with traffic facilities provide multi-source heterogeneous field traffic data, which contain the fundamental information and distinct features for congestion recognition. To exploit these traffic big data, in this paper, we propose a machine learning-based framework to tackle the congestion recognition problem. It can be divided in two parts, a digraph-based representation for hybrid urban traffic network and a Dirgraph Convolutional Neural Network (DGCN)-based learning model. At first, the representation incorporates the fundamental traffic variables with the correlation of different traffic flows, and partially decouples the global network topology from local traffic information. And then, to proceed with digraph-based samples, a new type of graph feature extraction method is introduced and the graph Fourier transform is defined accordingly. This distinguishes the proposed model from the conventional graph convolutional networks. Comprehensive experiments are conducted based on real traffic data. The results demonstrate the advantages of the proposed framework over the existing congestion recognition methods.

A gradient boost approach for predicting near-road ultrafine particle concentrations using detailed traffic characterization

This study investigates the influence of meteorology, land use, built environment, and traffic characteristics on near-road ultrafine particle (UFP) concentrations. To achieve this objective, minute-level UFP concentrations were measured at various locations along a major arterial road in the Greater Toronto Area (GTA) between February and May 2019. Each location was visited five times, at least once in the morning, mid-day, and afternoon. Each visit lasted for 30 min, resulting in 2.5 h of minute-level data collected at each location. Local traffic information, including vehicle class and turning movements, were processed using computer vision techniques. The number of fast-food restaurants, cafes, trees, traffic signals, and building footprint, were found to have positive impacts on the mean UFP, while distance to the closest major road was negatively associated with UFP. We employed the Extreme Gradient Boosting (XGBoost) method to develop prediction models for UFP concentrations. The Shapley additive explanation (SHAP) measures were used to capture the influence of each feature on model output. The model results demonstrated that minute-level counts of local traffic from different directions had significant impacts on near-road UFP concentrations, model performance was robust under random cross-validation as coefficients of determination (R2) ranged from 0.63 to 0.69, but it revealed weaknesses when data at specific locations were eliminated from the training dataset. This result indicates that proper cross-validation techniques should be developed to better evaluate machine learning models for air quality predictions.

A Novel VANET-Assisted Traffic Control for Supporting Vehicular Cloud Computing

Vehicular Ad hoc Networks (VANETs) allow for vehicle-to-vehicle and vehicle-to-infrastructure communications using wireless local area network technologies. The distinctive features of their candidate applications (e.g., collision warning and local traffic information for drivers), resources (e.g., computational sources), and their ability to collect various data from their environment (e.g., vehicular traffic flow patterns) make VANETs a rich resource for information and resources. In this paper, we propose a new methodology to use VANETs to optimize signal control at traffic intersections as well as to create Vehicular Cloud (VC) computing environments. We theoretically analyze the traffic flow patterns in a given road intersection by using the diffusion approximation model. We calculate the probability of clearing the intersection and demonstrate the effect of the traffic patterns on the optimal choice of the traffic signal control parameters. Then, we employ our theoretical analysis to propose a potential solution to construct VANET-assisted VCs. Experimental results verify the correctness of our analysis.

Distributed Cooperative Backpressure-Based Traffic Light Control Method

On the foundation of the original backpressure-based traffic light control algorithm, a distributed cooperative backpressure-based traffic light control method is proposed in this paper. The urban traffic network is modeled as a smart agent-controlled queuing network, in which the intersection agents exchange the queue length information and the selected activating light phase information of neighboring intersections through communications and determine the activating light phase at each time slot according to local traffic information. The improved phase pressure computation method considers the phase state of downstream intersections instead of only the queue length of the local intersections. Light phase switching coordination among adjacent intersections is achieved using the consensus-based bundle algorithm, in which the cooperative light phase switching problem is viewed as a task assignment issue among adjacent intersections. Simulation results illustrated that the proposed cooperative backpressure-based traffic light control method obtained better performance than the original backpressure-based and fixed-time traffic control methods.

Distributed Traffic Speed Control for Improving Vehicle Throughput

This article presents a real-time speed control strategy for vehicle throughput improvement on highway. Using Barron-Jensen/Frankowska (B-J/F) solution to Hamilton-Jacobi (HJ) Moskowitz function, we formulate a Convex Quadratic Optimization Problem (CQOP) solved by Quadratic Programming (QP). Unlike with previous investigation on optimal control of scalar conservation law, a distributed framework is constructed based on dual decomposition and subgradient method via networked Road Infrastructures (RIs). Instead of relying on complex RIs network for information collection, the distributed method only depends on local traffic information and iteratively approaches to feasible solutions constrained by flow consensus. For both uniform and non-uniform road conditions, the proposed distributed algorithm is verified to be able to improve the vehicle throughput, reduce traffic delay and alleviate congestion in simulation.

Internet of Vehicles: Sensing-Aided Transportation Information Collection and Diffusion

In view of the emergence and rapid development of the Internet of Vehicles (IoV) and cloud computing, intelligent transport systems are beneficial in terms of enhancing the quality and interactivity of urban transportation services, reducing costs and resource wastage, and improving the traffic management capability. Efficient traffic management relies on the accurate and prompt acquisition as well as diffusion of traffic information. To achieve this, research is mostly focused on optimizing the mobility models and communication performance. However, considering the escalating scale of IoV networks, the interconnection of heterogeneous smart vehicles plays a critical role in enhancing the efficiency of traffic information collection and diffusion. In this paper, we commence by establishing a weighted and undirected graph model for IoV sensing networks and verify its time-invariant complex characteristics relying on a real-world taxi GPS dataset. Moreover, we propose an IoV-aided local traffic information collection architecture, a sink node selection scheme for the information influx, and an optimal traffic information transmission model. Our simulation results and theoretical analysis show the efficiency and feasibility of our proposed models.

Calibration and sensitivity analysis of rut prediction model for semi-rigid pavement using AASHTOWare ME design

Rutting is one of the most serious distresses of semi-rigid asphalt pavements in Jiangsu. In this research, AASHTOWare pavement ME Design software was used for rutting prediction of semi-rigid pavements. A semi-rigid pavement is composed of a flexible layer and a rigid layer. Local calibration was done first by taking into account local materials, traffic information, and environmental conditions. A total of eight sites representing typical mixture types of Jiangsu province were selected to complete the calibration and another four highways were chosen to make the validation. Then the calibrated software was used to identify sensitive factors for rutting prediction. Critical inputs needed for the calibration and sensitive analysis were tested or collected from local project-level Pavement Management System (PMS) database. The modulus of semi-rigid base was back-calculated from deflection basin data. Calibration coefficients for each typical structure and the optimal coefficients for this region were identified, respectively. Results also demonstrate that traffic inputs and thickness of Hot Mixed Asphalt (HMA) layers are more sensitive to the predicted rutting depth than void contents of HMA mixtures.

Congestion management techniques for disruption‐tolerant satellite networks

SummaryDelay and disruption‐tolerant networks are becoming an appealing solution for extending Internet boundaries toward challenged environments where end‐to‐end connectivity cannot be guaranteed. In particular, satellite networks can take advantage of a priori trajectory estimations of nodes to make efficient routing decisions. Despite this knowledge is already used in routing schemes such as contact graph routing, it might derive in congestion problems because of capacity overbooking of forthcoming connections (contacts). In this work, we initially extend contact graph routing to provide enhanced congestion mitigation capabilities by taking advantage of the local traffic information available at each node. However, since satellite networks data generation is generally managed by a mission operation center, a global view of the traffic can also be exploited to further improve the latter scheme. As a result, we present a novel strategy to avoid congestion in predictable delay‐ and disruption‐tolerant network systems by means of individual contact plans. Finally, we evaluate and compare the performance improvement of these mechanisms in a typical low Earth orbit satellite constellation.

CANS: context‐aware traffic estimation and navigation system

Acquiring real-time traffic information is a basic requirement for dynamic vehicular navigation systems. The majority of the current navigation systems are based on static traffic information. Building on mobile crowdsensing technology, the authors propose context-aware traffic estimation and navigation system (CANS), a context-aware system that can estimate traffic state without any requirement for expensive infrastructure. Using only available equipment, it can provide dynamic navigation service to drivers. The proposed system consists of three main components: local traffic estimation, global traffic aggregation, and navigation. In this system, vehicles estimate local traffic state using vehicular contextual information including speed and acceleration by relying on fuzzy logic, and transmit the information to the urban server. The server integrates the received local traffic information from different vehicles and estimates the global traffic state, providing the traffic-aware navigation system to drivers. CANS performance is evaluated for an urban scenario in a traffic flow in Birjand, Iran. The experiment is conducted for evaluating CANS in both traffic congestion estimation and navigation. The results show an accurate estimation of traffic states along urban roads. Compared with previous approaches, CANS overrides them for its reduced travel time.

PALS: Saving Network Power With Low Overhead to ISPs and Applications

Power saving in the network infrastructure has received great attention in recent years. Power-aware traffic management is proposed in many works, in which a subset of routers/links are preferentially used to carry traffic while other links are activated only when traffic load is high. However, it remains challenging how to minimize the overhead to both ISPs and applications, which is important to the successful deployment of power-aware traffic management in a real network. This paper presents PALS, a new Power-Aware Link State routing based traffic management protocol. Compared with previous solutions, PALS remarkably reduces the overheads to ISPs and applications by the following innovations. First, PALS minimizes the forwarding table expansion due to dynamic power-aware routing, by using destination based routing instead of pairwise routing (e.g., MPLS). Second, PALS limits packet reordering for applications, by never splitting traffic between an IE (ingress-egress) router pair to multiple paths. Third, PALS significantly reduces the computation complexity of the power-aware routing algorithm, by running a simple path selection algorithm at each ingress router with the knowledge of local traffic information as well as global link utilization, which are much easier to obtain than global traffic matrix required by the state-of-the-art solutions (e.g., Zhang , IEEE ICNP 2010). Extensive simulations and testbed experiments show that, although bearing the simplicities to minimize the overhead, PALS saves satisfactory network power, with quick response to traffic variance and negligible impact on the packet delivery performance for applications.

Finding the biased-shortest path with minimal congestion in networks via linear-prediction of queue length

In this paper, we propose a biased-shortest path method with minimal congestion. In the method, we use linear-prediction to estimate the queue length of nodes, and propose a dynamic accepting probability function for nodes to decide whether accept or reject the incoming packets. The dynamic accepting probability function is based on the idea of homogeneous network flow and is developed to enable nodes to coordinate their queue length to avoid congestion. A path strategy incorporated with the linear-prediction of the queue length and the dynamic accepting probability function of nodes is designed to allow packets to be automatically delivered on un-congested paths with short traveling time. Our method has the advantage of low computation cost because the optimal paths are dynamically self-organized by nodes in the delivering process of packets with local traffic information. We compare our method with the existing methods such as the efficient path method (EPS) and the optimal path method (OPS) on the BA scale-free networks and a real example. The numerical computations show that our method performs best for low network load and has minimum run time due to its low computational cost and local routing scheme.

IEEE Access Special Section Editorial: Advances in Vehicular Clouds

Connected vehicles are seen as the next frontier for the mobile revolution. By bringing Internet connectivity to vehicles, not only infotainment of the vehicles can be enriched, but also safety and driving experience of the drivers can be enhanced. Connected vehicles can access real-time information about traffic conditions to make the journey more time efficient. They can also share local traffic information in real-time to assist other vehicles making better route planning. Consolidating various transportation data from individual vehicles and other transportation infrastructure can further enhance Intelligent Transportation System (ITS).

Effective local dynamic routing strategy for air route networks

With the rapid development of air transportation, network service ability has attracted a lot of attention in academe. Aiming to improve the throughput of the air route network (ARN), we propose an effective local dynamic routing strategy in this paper. Several factors, such as the routing distance, the geographical distance and the real-time local traffic, are taken into consideration. When the ARN is in the normal free-flow state, the proposed strategy can recover the shortest path routing (SPR) strategy. When the ARN undergoes congestion, the proposed strategy changes the paths of flights based on the real-time local traffic information. The throughput of the Chinese air route network (CARN) is evaluated. Results confirm that the proposed strategy can significantly improve the throughput of CARN. Meanwhile, the increase in the average flying distance and time is tiny. Results also indicate the importance of the distance related factors in a routing strategy designed for the ARN.

Spatial Inference of Traffic Transition Using Micro–Macro Traffic Variables

This paper proposes an online traffic inference algorithm for road segments in which local traffic information cannot be directly observed. Using macro-micro traffic variables as inputs, the algorithm consists of three main operations. First, it uses interarrival time (time headway) statistics from upstream and downstream locations to spatially infer traffic transitions at an unsupervised piece of segment. Second, it estimates lane-level flow and occupancy at the same unsupervised target site. Third, it estimates individual lane-level shockwave propagation times on the segment. Using real-world closed-circuit television data, it is shown that the proposed algorithm outperforms previously proposed methods in the literature.