Search

Traffic forecasting plays a critical role in intelligent transportation systems (ITS) in smart cities. Travelers as well as urban managers rely on reliable traffic information to make their decisions for route choice and traffic management. However, noisy or missing traffic data poses a problem for accurate and robust traffic forecasting. While data-driven models such as deep neural networks can achieve high prediction accuracy with complete datasets, sensor malfunctions, and environmental effects degrade the performance of such models, as these models rely heavily on precise traffic measurements for model training and estimation. Consequently, incomplete traffic data poses a challenge for robust model design that can make accurate traffic forecasts with noisy/missing data. This research proposes the Robust Spatiotemporal Graph Convolutional Network (RT-GCN), a traffic prediction model that handles noise perturbations and missing data using a Gaussian distributed node representation and a variance based attention mechanism. Through experiments conducted on four real-world traffic datasets using diverse noisy and missing scenarios, the proposed RT-GCN model has demonstrated its ability to handle noise perturbations and missing values and provide high accuracy prediction. • Gaussian distribution representation enhance inherent robustness of neural network. • Variance-based attention mechanism reduce the propagation of perturbation. • Batch Random Noise help improve robustness during training phase. • Models are tested on both noisy and missing datasets. • Diverse perturbation scenarios are considered in experiments. [ABSTRACT FROM AUTHOR]

Traffic prediction is significantly important for performance analysis and network planning in Software Defined Networking (SDN). However, to effectively predict network traffic in current networks is very difficult and nearly prohibitive. As a new cutting-edge network technology, SDN decouples the control and data planes of network switch devices to enable the flexibility of network measurements and managements. The SDN architecture of the flow-based forwarding idea brings forth a promising of network traffic capture and prediction. We propose a lightweight traffic prediction algorithm for SDN applications. Firstly, different from traditional network traffic measurements, our method uses the flow-based forwarding idea in SDN to extract traffic statistic from data plane. The statistical variable describes network flow information forwarded in SDN and enables more accurate measurements of flow traffic via a direct and low-overhead way compared with traditional traffic measurements. Secondly, based on the temporal nature of traffic, the time-correlation theory is utilized to model flow traffic, where the time-series analysis theory and regressive modeling approach are used to characterize network traffic in SDN. A fully new method is proposed to perform traffic prediction. Thirdly, we propose the flow-based forwarding traffic prediction algorithm to forecast to SDN traffic. The detailed algorithm process is discussed and analyze. Finally, sufficient experiments are presented and designed to validate the proposed method. Simulation results show that our method is feasible and effective. [ABSTRACT FROM AUTHOR]

Software Defined Networking (SDN) architecture separates control plane and data plane, making network flexible and programmable. Since the large number of devices connected to the Internet of things (IoT) networks, the SDN-based network architecture makes the deployment and configuration of IoT much easier. In the IoT network, the fine-grained network traffic is critical to network management, then we propose a novel scheme to measure the fine-grained network traffic in the SDN-based IoT networks. In SDN-based IoT networks, the controller is very easy to be attacked, we introduce the blockchain technology into the measurement framework to ensure the security and consistency of the statistics. To measure flow traffic with low overhead and high accuracy, we collect the statistics of coarse-grained traffic of flows and fine-grained traffic of links, and model the network traffic as an ARIMA model and forecast the network traffic with the coarse-grained measurement of flows. Then, we propose an objective function to decrease the estimation errors. Due to the objective function is an NP-hard problem, we present a heuristic algorithm to obtain the optimal solution of the fine-grained measurement. Finally, we conduct some simulations to verify the validity of the proposed measurement scheme. Simulation results show that our approach is feasible and effective. [ABSTRACT FROM AUTHOR]

Travel time is an important index for managers to evaluate the performance of transportation systems and an intuitive measure for travelers to choose routes and departure times. An important part of the literature focuses on predicting instantaneous travel time under recurrent traffic conditions to disseminate traffic information. However, accurate travel time prediction is important for assessing the effects of abnormal traffic conditions and helping travelers make reliable travel decisions under such conditions. This study proposes an online travel time prediction model with emphasis on capturing the effects of anomalies. The model divides a path into short links. A Functional Principal Component Analysis (FPCA) framework is adopted to forecast link travel times based on historical data and real-time measurements. Furthermore, a probabilistic nested delay operator is used to calculate path travel time distributions. To ensure that the algorithm is fast enough for online applications, parallel computation architecture is introduced to overcome the computational burden of the FPCA. Finally, a rolling horizon structure is applied to online travel time prediction. Empirical results for Guangzhou Airport Expressway indicate that the proposed method can capture an abrupt change in traffic state and provide a promising and reliable travel time prediction at both the link and path levels. In the case where the original FPCA is modified for parallelization, accuracy and computational effort are evaluated and compared with those of the sequential algorithm. The proposed algorithm is found to require only a piece rather than a large set of traffic incident records. [ABSTRACT FROM AUTHOR]

The study aims to develop a novel evolution model of speed perturbations for quantifying the probability of the traffic flow breakdown on congested freeway flow. The model is applicable to freeway operation and management. Time headway, driver behavior, and state transitions with stochastic nature are the most important factors for the breakdown and recovery of a disturbance. Two types of drivers, 'adventurers' and 'conservatives,' and two styles of reacting behavior, 'copy' and 'shift' are defined to characterize driving behavior. The time headway distributions for arriving and discharging traffic are distinguished for the existence of the 'hysteresis' phenomenon. The Monte Carlo simulation is carried out to obtain the estimation of the probability of breakdown (PB). The calculated PB is a monotonically increasing function of freeway flow with an 'S' shape. The simulation results indicate that the proposed model can describe the changes in spatial extent and magnitude of a disturbance, which have potential use for future works and field applications. [ABSTRACT FROM AUTHOR]

Real-time information about the dynamic variation of turning rates at intersections is essential for achieving a more efficient traffic control and management system. Limitations in obtaining measurements of turning movements have drawn attention to the need for alternative methods for estimation of turning rates. In a large-scale urban network, characteristics such as disparate signal plans and various intersection layouts make defining a generic model for the entire network challenging. For this reason, most of the proposed methods are impractical for real-field implementation. The application of existing optimization-based methods is also limited by computational complexity, the need for pre-defined assumptions and insufficient stationary traffic measurements (e.g. loop detectors and plate recognition cameras) in urban networks. In this study, we take advantage of probe vehicles to propose a hybrid method that noticeably reduces the need for stationary measurements in the network. This hybrid method consists of two core stages: signal processing, and physical data-driven modeling to consecutively estimate turning rates. In the signal processing phase, we use wavelet transformation to extract fundamental components of inferred floating car turning rates. These components are then used as inputs to a turning rate estimation model. Using these decomposed components of floating car turning rates instead of raw input helps to balance the effect of low and high-frequency signal components and leads to a more accurate estimation. Investigation of the proposed method in a microscopic simulation environment reveals the ability of the hybrid model to detect turning rate variations even when a severe incident occurs. • A hybrid data-driven method is used for network-wide turning rate estimation. • The method makes use of floating car data for turning rate estimation. • A small proportion of stationary measurements is needed for the training purpose. • Deployment of wavelet transformation allows access to more accurate information. • The method makes a balance between the effect of low and high-frequency changes. [ABSTRACT FROM AUTHOR]

• Estimate the traffic demand integrating information provided by traditional and new technologies. Explore the inherent properties of extended coverage information. • Deal with splitting ratios of demand flows among paths identified by coupled Bluetooth sections. • Derivate the analytical solving procedure. • Increase of the intercepted path flows or spatial coverage leads to more accurate estimates. • Assess the effectiveness of the proposed method on a study network in Rome, Italy. Recent advances in technology have made available numerous new monitoring systems that collect updated traffic measurements both in fixed locations and over specific corridors or paths. Such recent technological developments point to challenging and promising opportunities for Origin-Destination (OD) traffic demand estimation and forecast. Therefore, the aim of this paper is to study how to exploit available information detected by new monitoring devices in the estimation of traffic demand. Starting from the formulation proposed by Spiess (1987, 1990), in this paper a new method to estimate the traffic demand by means of Bluetooth data is proposed. It explores inherent properties of this information in an off-line (and static) context, where mathematical formulation of the estimation problem can be derived. The effectiveness of the proposed method has been investigated in an extensive plan of experiments carried out both on test networks and on a study network consisting in a part of the city of Rome, Italy, obtaining promising results in both applications. [ABSTRACT FROM AUTHOR]

Traffic flow measurement is very important for traffic management systems. However, the existing traditional measurement approaches are highly time-consuming and expensive to continuously gather the required data and to maintain the corresponding equipment, such as loop detectors and video cameras. On the other hand, many services on the web propose to estimate automobile travel time taking into account traffic conditions thanks to crowd sourced data (Floating Car Data). This work proposes to reconstruct, from estimated travel time, traffic flows using machine learning method. In particular, we evaluate the capacity of Gaussian Process Regressor (GPR) to address this issue. After obtaining estimated travel time on a given route, a clustering process shows that travel duration profiles in each day can be associated to different "types of day". Then, different regressors are trained in order to estimate traffic flows from travel duration. In the "multi-model" variant, we trained a Regressor for each type of day. Conversely, in the "single model" variant, only one Regressor is trained (the type of day is not taken into account). This is an innovative work to estimate and reconstruct the traffic flow in transportation networks with machine learning method from aggregated Floating Car Data (FCD). A series of experiments are conducted to compare the estimated traffic flows, obtained by the proposed single model and multi-model, and the real ones from actual sensors. The obtained results show that both single model and multi-models can capture the tendency of real traffic flows. Furthermore, the performance can be improved by regulating parameters in GPR machine learning model, such as half width of sample window and sample size (a whole week or only weekdays), and multi-models can highly increase the performance compared with the single model. Therefore, the proposed GPR machine learning and FCD based new method can replace those traditional loop detectors for the measurement of traffic flow. • Application of Machine Learning based model to rebuild the traffic flow relationship. • Construction and training of multi-models to reduce estimated traffic flow error. • Demonstration of Gaussian Process Regressor to achieve the lowest error. [ABSTRACT FROM AUTHOR]

Abstract: Recent advancement in traffic surveillance systems has allowed for obtaining more detailed vehicular movement such as individual vehicle trajectory data. Understanding the characteristics of interactions between leading vehicle and following in the traffic flow stream is a backbone for designing and evaluating more sophisticated traffic and vehicle control strategies. This study proposes a methodology for estimating rear-end crash potential, as a probabilistic measure, in real time based on the analysis of vehicular movements. The methodology presented in this study consists of two components. The first estimates the probability that a vehicle''s trajectory belonging to either ‘changing lane’ or ‘going straight’. A binary logistic regression (BLR) is used to model the lane-changing decision of the subject vehicle. The other component derives crash probability by an exponential decay function using time-to-collision (TTC) between the subject vehicle and the front vehicle. Also, an aggregated measure, crash risk index (CRI) is used in the analysis to accumulate rear-end crash potential for each subject vehicle. The result of this study can be used in developing traffic control and information systems, in particular, for crash prevention. [Copyright &y& Elsevier]

Abstract: In this paper we explicitly construct the entropy solutions for the Lighthill–Whitham–Richards (LWR) traffic flow model with a flow–density relationship which is piecewise quadratic, continuous, concave, but not differentiable at the junction points where two quadratic polynomials meet, and with piecewise linear initial condition and piecewise constant boundary conditions. The proposed model is a generalization of the well-known piecewise linear flow–density relationship in the LWR model. As observed traffic flow data can be well fitted with such continuous piecewise quadratic functions, the explicitly constructed solutions provide a fast and accurate solution tool which may be used for predicting traffic or as a diagnosing tool to test the performance of numerical schemes. We implement these explicit entropy solutions for two representative traffic flow cases and also compare them with numerical solutions obtained by a high order weighted essentially non-oscillatory (WENO) scheme. [Copyright &y& Elsevier]

This paper presents a case study of real-time traffic state estimation. The adopted general approach to the design of universal traffic state estimators for freeway stretches is based on stochastic macroscopic traffic flow modeling and extended Kalman filtering, which are outlined in the paper. The reported investigations were conducted by use of eight-hour traffic measurement data collected from a freeway stretch of 4.1 km close to Munich, Germany. Some key issues are carefully investigated, including the tracking capability of the designed traffic state estimator, significance of the online model parameter estimation, sensitivity of the estimator to the initial values of the estimated model parameters as well as to the related noise standard deviation values, and the capability of the estimator to handle biased flow measurements. The achieved results are quite satisfactory. [ABSTRACT FROM AUTHOR]

Given traffic flow measurements for one highway, how to forecast its flow in future periods? Recent works in traffic forecast propose burdensome procedures by depending on additional data that is not always available, like traffic measurements from other roads linked to the one of interest, social media, trajectory and car accident data, geographical and socio-demographic attributes, driver behavior information and weather forecast. The most accurate algorithms force anyone to monitor an entire network of highways, even when there is a single highway of interest. This procedure is commonly unaffordable. How to obtain highly accurate results without using any additional data? We answer the question with AdaptFlow : a novel, adaptive algorithm able to accurately forecast traffic flow by individually monitoring highways that are connected to each other in a complex network using local flow measurements only. We performed experiments on large datasets from highways in UK and USA. Our AdaptFlow notably outperformed well-known related works on many settings. For example, it achieved 95.5% accuracy on average when forecasting the next 15 minutes flow of the UK highways, leading to an error rate that is 36% smaller than the one of the most accurate related work that does not use additional data. [ABSTRACT FROM AUTHOR]

Historical measurements are usually used to build assimilation models in sequential data assimilation (S-DA) systems. However, they are always disturbed by local noises. Simultaneously, the accuracy of assimilation model construction and assimilation forecasting results will be affected. The fast Fourier transform (FFT) method can be used to acquire de-noised historical traffic flow measurements to reduce the influence of local noises on constructed assimilation models and improve the accuracy of assimilation results. In the practical signal de-noising applications, the FFT method is commonly used to de-noise the noisy signal with known noise frequency. However, knowing the noise frequency is difficult. Thus, a proper cutoff frequency should be chosen to separate high-frequency information caused by noises from the low-frequency part of useful signals under the unknown noise frequency. If the cutoff frequency is too high, too much noisy information will be treated as useful information. Conversely, if the cutoff frequency is too low, part of the useful information will be lost. To solve this problem, this paper proposes an adaptive cutoff frequency selection (A-CFS) method based on cross-validation. The proposed method can determine a proper cutoff frequency and ensure the quality of de-noised outputs for a given dataset using the FFT method without noise frequency information. Experimental results of real-world traffic flow data measurements in a sub-area of a highway near Birmingham, England, demonstrate the superior performance of the proposed A-CFS method in noisy information separation using the FFT method. The differences between true and predicted traffic flow values are evaluated using the mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage (MAPE) values. Compared to the results of the two commonly used de-noising methods, i.e., discrete wavelet transform (DWT) and ensemble empirical mode decomposition (EEMD) methods, the short-term traffic flow forecasting results of the proposed A-CFS method are much more reliable. In terms of the MAE value, the average relative improvements of the assimilation model built using the proposed method are 19.26%, 3.47%, and 4.25%, compared to the model built using raw data, DWT method, and EEMD method, respectively; the corresponding average relative improvements in RMSE are 19.05%, 5.36%, and 3.02%, respectively; lastly, the corresponding average relative improvements in MAPE are 18.88%, 2.83%, and 2.28%, respectively. The test results show that the proposed method is effective in separating noises from historical measurements and can improve the accuracy of assimilation model construction and assimilation forecasting results. [ABSTRACT FROM AUTHOR]

Freeway work zone with lane closure can lead to disruption to local traffic and cause significant impacts on mobility, safety and environmental sustainability. To mitigate traffic congestion near work zone area, many variable speed limits (VSL) control approaches have been developed. However, VSL control system, as a critical transportation management system, is prone to the occurrence of traffic sensor faults. Faulty sensors can cause great deviations of traffic measurements and system degradation. Therefore, this study aims to develop a fault-tolerant VSL control strategy for freeway work zone with the consideration of the mainline sensor fault and ramp sensor fault. To analyze the traffic dynamics near work zone area, a traffic flow model has been built first. Then a sliding mode controller in the previous study has been utilized for VSL control. In addition to the traffic states estimated by a Kalman filter, two observers have been developed to provide analytical redundancy of traffic states estimation. By comparing the logarithm of the likelihood estimations from the Kalman filter and two observers, a fault diagnosis scheme has been designed to detect and identify the faults of mainline sensors and ramp sensors. Then the VSL controller can be reconfigured accordingly in case of sensor faults. The proposed system is implemented and evaluated under a realistic freeway work zone environment using traffic simulator SUMO. The results demonstrate that the developed system can accurately detect and identify the sensor faults in real time. Consistent improvements of mobility, safety and sustainability are also achieved under fault-free and sensor faults scenarios. [ABSTRACT FROM AUTHOR]

Mathematically describing the physical process of a sequential data assimilation system perfectly is difficult and inevitably results in errors in the assimilation model. Filter divergence is a common phenomenon because of model inaccuracies and affects the quality of the assimilation results in sequential data assimilation systems. In this study, an approach based on an L1-norm constraint for filter-divergence suppression in sequential data assimilation systems was proposed. The method adjusts the weights of the state-simulated values and measurements based on new measurements using an L1-norm constraint when filter divergence is about to occur. Results for simulation data and real-world traffic flow measurements collected from a sub-area of the highway between Leeds and Sheffield, England, showed that the proposed method produced a higher assimilation accuracy than the other filter-divergence suppression methods. This indicates the effectiveness of the proposed approach based on the L1-norm constraint for filter-divergence suppression. [ABSTRACT FROM AUTHOR]

Collection of Annual Average Daily Traffic (AADT) is of major importance for a number of applications in road transport urban and environmental studies. However, traffic measurements are undertaken only for a part of the road network with minor roads usually excluded. This paper suggests a methodology to estimate AADT in England and Wales applicable across the full road network, so that traffic for both major and minor roads can be approximated. This is achieved by consolidating clustering and regression modelling and using a comprehensive set of variables related to roadway, socioeconomic and land use characteristics. The methodological output reveals traffic patterns across urban and rural areas as well as produces accurate results for all road classes. Support Vector Regression (SVR) and Random Forest (RF) are found to outperform the traditional Linear Regression, although the findings suggest that data clustering is key for significant reduction in prediction errors. [ABSTRACT FROM AUTHOR]

Transportation planning requires the use of accurate traffic data to produce estimates of traffic volume predictions over time and space. The annual average daily traffic (AADT) data is an important component of transportation design, operation, policy analysis, and planning. The use of traffic volume forecasting models for the characterization, analysis, and estimation of transportation data has proven to be a useful method for reducing high costs, overcoming spatial constraints, and limiting the errors associated with data collection and analysis in transportation planning. The geostatistical kriging technique is a viable method for modeling and forecasting AADT. The degree to which the technique of kriging can be useful in forecasting AADT depends highly on an understanding of the decision-making variables, the relationship between the variables, and the practical limitations of the various kriging techniques and variogram models. This paper applied three different linear kriging techniques [simple kriging (SK), ordinary kriging (OK), and universal kriging (UK)] and five variogram models (nugget effect, spherical, exponential, Gaussian, and power) to characterize and interpolate the annual average daily traffic of Washington State. The statistical errors (i.e., mean error, root-mean-square error, average standard error, mean standardized error, and root-mean-square standardized error) of the resulting output were then compared to determine the most suitable combination of linear spatial interpolation and variogram algorithms for each of the data sets. Ordinary and/or universal kriging combined with the exponential variogram were the most appropriate methods for the 2008 AADT data set, whereas ordinary kriging combined with the spherical variogram and simple kriging combined with the spherical variogram were the most suitable methods for the 2009 and 2010 data sets, respectively. Results from this study suggest that using the same combination of kriging and variogram algorithms to characterize and interpolate different AADT data sets (2008, 2009, and 2010) could lead to suboptimal results. One reason for the lack of optimal results is for instance, the directional variation in the 2009 and 2010 data sets which undermine the assumption of mean stationarity in the use of ordinary kriging for modeling. [ABSTRACT FROM AUTHOR]