Traffic Prediction Problem Research Articles

A Spatial–Temporal Multitask Collaborative Learning Model for Multistep Traffic Flow Prediction

Traffic flow prediction is a fundamental capability for successful deployment of intelligent transportation systems. Traditionally, multiple related prediction tasks are undertaken individually, without considering the relationships among the tasks. This paper presents a spatial–temporal multitask collaborative learning model for multistep traffic flow prediction. The novel approach learns multiple related prediction tasks collaboratively by extracting and utilizing appropriate shared information across tasks. First, each traffic flow prediction problem is formulated as a supervised machine-learning task. Next, the sparse features shared across multiple tasks are learned by solving a regularized optimization problem. To deal with the non-convex and non-smooth challenges, the optimization problem is then transformed into an equivalent convex problem. Finally, the global optimal solution of the convex problem is found by solving a variation of this problem using an alternating minimization algorithm. The proposed model incorporates both the spatial correlation between different observed stations and the intrinsic relationship between different traffic flow parameters, as well as the coarse-grain temporal correlation between different days in a week and the fine-grain temporal correlation between different prediction steps. Application of the proposed model to a real case study for SR180-E freeway in Fresno, California showed its effectiveness, robustness and advantages for multistep traffic flow prediction.

Big data‐driven machine learning‐enabled traffic flow prediction

AbstractReal‐time effective traffic flow big data prediction network has important application significance. Over the past few years, traffic flow data have been exploding and we have entered the big data era. The key challenge of traffic flow prediction network is how to construct an adaptive model relying on historical data. Existing big data‐driven traffic flow prediction networking approaches mainly use shallow learning, and there are unsatisfying for many realistic applications, which inspire us to rethink the traffic flow big data prediction problem with deep learning. In this paper, we propose a novel prediction approach based on machine learning. In addition to the minimum prediction error as the goal, we present the long short‐term memory model, which is a typical machine learning algorithm with deep learning network. This method is applied into the real‐world traffic big data from performance measurement system. Experimental results show that the proposed machine learning algorithm has more applicability and higher performance, compared with shallow machine learning prediction network.

Deep Multi-View Spatial-Temporal Network for Taxi Demand Prediction

Taxi demand prediction is an important building block to enabling intelligent transportation systems in a smart city. An accurate prediction model can help the city pre-allocate resources to meet travel demand and to reduce empty taxis on streets which waste energy and worsen the traffic congestion. With the increasing popularity of taxi requesting services such as Uber and Didi Chuxing (in China), we are able to collect large-scale taxi demand data continuously. How to utilize such big data to improve the demand prediction is an interesting and critical real-world problem. Traditional demand prediction methods mostly rely on time series forecasting techniques, which fail to model the complex non-linear spatial and temporal relations. Recent advances in deep learning have shown superior performance on traditionally challenging tasks such as image classification by learning the complex features and correlations from large-scale data. This breakthrough has inspired researchers to explore deep learning techniques on traffic prediction problems. However, existing methods on traffic prediction have only considered spatial relation (e.g., using CNN) or temporal relation (e.g., using LSTM) independently. We propose a Deep Multi-View Spatial-Temporal Network (DMVST-Net) framework to model both spatial and temporal relations. Specifically, our proposed model consists of three views: temporal view (modeling correlations between future demand values with near time points via LSTM), spatial view (modeling local spatial correlation via local CNN), and semantic view (modeling correlations among regions sharing similar temporal patterns). Experiments on large-scale real taxi demand data demonstrate effectiveness of our approach over state-of-the-art methods.

Cooperative Differential Evolution with Dynamical Population for Short-Term Traffic Flow Prediction Problem

Cooperative Differential Evolution with Dynamical Population for Short-Term Traffic Flow Prediction Problem

Local Gaussian Processes for Efficient Fine-Grained Traffic Speed Prediction

Traffic speed is a key indicator for the efficiency of an urban transportation system. Accurate modeling of the spatiotemporally varying traffic speed thus plays a crucial role in urban planning and development. This paper addresses the problem of efficient fine-grained traffic speed prediction using big traffic data obtained from static sensors. Gaussian processes (GPs) have been previously used to model various traffic phenomena, including flow and speed. However, GPs do not scale with big traffic data due to their cubic time complexity. In this work, we address their efficiency issues by proposing local GPs to learn from and make predictions for correlated subsets of data. The main idea is to quickly group speed variables in both spatial and temporal dimensions into a finite number of clusters, so that future and unobserved traffic speed queries can be heuristically mapped to one of such clusters. A local GP corresponding to that cluster can then be trained on the fly to make predictions in real-time. We call this method localization. We use non-negative matrix factorization for localization and propose simple heuristics for cluster mapping. We additionally leverage on the expressiveness of GP kernel functions to model road network topology and incorporate side information. Extensive experiments using real-world traffic data collected in the two U.S. cities of Pittsburgh and Washington, D.C., show that our proposed local GPs significantly improve both runtime performances and prediction accuracies compared to the baseline global and local GPs.

Pruned Fast Learning Fuzzy Approach for Data-Driven Traffic Flow Prediction

Traffic flow prediction plays an important role in intelligent transportation systems. With the rapid growth of traffic flow data, fast and accurate traffic flow prediction methods are now required. In this paper, we propose a novel fast learning data-driven fuzzy approach for the traffic flow prediction problem. In the proposed approach, to achieve fast learning, an extreme learning machine is utilized to optimize the consequent parameters of the fuzzy rules. Further, a fuzzy rule pruning strategy that involves measuring the firing levels of the fuzzy rules is presented to obtain reduced fuzzy inference systems. To evaluate the performance of the proposed approach, it was experimentally applied to traffic flow prediction and its results compared with those of widely used methods. The experimental results verify that the proposed approach can achieve satisfactory performance. The comparisons show that the proposed approach can obtain better (sometimes similar) performances, but with a simpler structure, fewer parameters, and much faster learning speed than the other methods.

AN ARCHITECTURE OF DEEP LEARNING METHOD TO PREDICT TRAFFIC FLOW IN BIG DATA

The invent of IEEE 802.11p as a communication standard, specific network protocol called vehicular adhoc network (VANET) based on mobile adhoc network ( MANET) along with sensor technology has put a strong foundation to visualize as well as make a reality of various intelligent transport applications & systems (ITAS) for safety and comfort. The success of such conceptualized applications depends upon how precisely as well reliability (in term of timeliness) the “traffic flow prediction” is done. The constitute of traffic data is characterized as Big Data type, therefore existing traffic prediction models are not in a capacity to provide the accurate result for various ITAS as the existing models consider low traffic data which lacks the insight of Big Data . In order to overcome these limitations, the synopsis aims to solve the “traffic flow prediction problem” by a novel mechanism of rigorous-learning based prediction model (RLBPM) using big traffic data. The RLBPM will exploit the spatial-temporal correlation statistics and for the purpose of learning algorithm “stacked auto encoder (SAE) model” will be used. The outcome of the RLBPM is expected to perform more superior as compared to the existing traffic flow prediction model. The implementation strategy involved mathematical modeling and simulation using Matlab.

Trend Modeling for Traffic Time Series Analysis: An Integrated Study

This paper discusses the trend modeling for traffic time series. First, we recount two types of definitions for a long-term trend that appeared in previous studies and illustrate their intrinsic differences. We show that, by assuming an implicit temporal connection among the time series observed at different days/locations, the PCA trend brings several advantages to traffic time series analysis. We also describe and define the so-called short-term trend that cannot be characterized by existing definitions. Second, we sequentially review the role that trend modeling plays in four major problems in traffic time series analysis: abnormal data detection, data compression, missing data imputation, and traffic prediction. The relations between these problems are revealed, and the benefit of detrending is explained. For the first three problems, we summarize our findings in the last ten years and try to provide an integrated framework for future study. For traffic prediction problem, we present a new explanation on why prediction accuracy can be improved at data points representing the short-term trends if the traffic information from multiple sensors can be appropriately used. This finding indicates that the trend modeling is not only a technique to specify the temporal pattern but is also related to the spatial relation of traffic time series.

A novel hybrid prediction algorithm to network traffic

Network traffic describes the characteristics and users’ behaviors of communication networks. It is a crucial input parameter of network management and network traffic engineering. This paper proposes a new prediction algorithm to network traffic in the large-scale communication network. First, we use signal analysis theory to transform network traffic from time domain to time-frequency domain. In the time-frequency domain, the network traffic signal is decomposed into the low-frequency and high-frequency components. Second, the gray model is exploited to model the low-frequency component of network traffic. The white Gaussian noise model is utilized to describe its high-frequency component. This is reasonable because the low-frequency and high-frequency components, respectively, represent the trend and fluctuation properties of network traffic, while the gray model and white Gaussian noise model can well capture the characteristics. Third, the prediction models of low-frequency and high-frequency components are built. The hybrid prediction algorithm is proposed to overcome the problem of network traffic prediction in the communication network. Finally, network traffic data from the real network is used to validate our approach. Simulation results indicate that our algorithm holds much lower prediction error than previous methods.

Design of a Novel Variable Bit Rate Video Traffic Prediction Model Based on Neural Networks

The video business has gradually become the main business of the network traffic, video traffic is Variable Bit Rate (VBR), which has nonlinear and sudden features, a single prediction model is not fit for describing those features, and prediction accuracy is not high. Because of wavelet analysis has advantages of multi-resolution and dealing with unexpectedness, and the neural network has better nonlinear fitting characteristics, in order to improve the prediction performance, the paper researched on the problem of VBR video traffic prediction based on neural networks, a novel prediction model AMWM is proposed, the model firstly introducing BP neural network to multi-fractal wavelet model, and designing prediction method, which introduced multi-fractal wavelet model to model VBR video traffic, and then applying BP neural network to forecast scale coefficient decomposed, and forecast multiplier using AR model, and predict traffic generated by wavelet reconstruction. Finally, the model is built and simulated. The experimental result shows that prediction performance based AMWM is better compared with multi-fractal wavelet model.

An algorithm for traffic flow parameters estimation and prediction using composition of machine learning methods and time series models

A problem of traffic flow analysis and prediction in city transport network is considered in this paper. The proposed algorithm uses GPS / GLONASS data of public transport location as input. Projecting this information on a transport network graph, as well as using additional filtering, we estimate traffic flow parameters. These parameters are used for short-term (up to 1 hour) prediction of road conditions in the city transport network. There is proposed a new method which consists of several steps to construct prediction. First, the transport graph is divided into a number of subgraphs by a territorial basis. Second, we use a dimension reduction method based on principal components analysis to describe the spatio-temporal distribution of traffic flow condition in the subgraphs. Third, an elementary prediction for each of the subgraphs is formed using the potential functions method with the measure of the subgraphs descriptions closeness introduced by analogy with bilateral filtering and support vector machine. Fourth, the additional elementary prediction is calculated using the known scalar and vector Box–Jenkins time series prediction models. Fifth, we construct the result prediction for each of the subgraphs using an adaptive linear composition of elementary predictions. At last, the traffic flow parameters are calculated as a linear combination of predictions for subgraphs of the city transport network. We have also made experimental investigations of transport network in Samara to evaluate the prediction accuracy of the proposed algorithm. The advantages of the proposed solution in comparison with existing ones are provided.

Traffic Flow Prediction With Big Data: A Deep Learning Approach

Accurate and timely traffic flow information is important for the successful deployment of intelligent transportation systems. Over the last few years, traffic data have been exploding, and we have truly entered the era of big data for transportation. Existing traffic flow prediction methods mainly use shallow traffic prediction models and are still unsatisfying for many real-world applications. This situation inspires us to rethink the traffic flow prediction problem based on deep architecture models with big traffic data. In this paper, a novel deep-learning-based traffic flow prediction method is proposed, which considers the spatial and temporal correlations inherently. A stacked autoencoder model is used to learn generic traffic flow features, and it is trained in a greedy layerwise fashion. To the best of our knowledge, this is the first time that a deep architecture model is applied using autoencoders as building blocks to represent traffic flow features for prediction. Moreover, experiments demonstrate that the proposed method for traffic flow prediction has superior performance.

Dynamic Network Traffic Flow Prediction Model based on Modified Quantum-Behaved Particle Swarm Optimization

This paper aims at effectively predicting the dynamic network traffic flow based on quantum-behaved particle swarm optimization algorithm. Firstly, the dynamic network traffic flow prediction problem is analyzed through formal description. Secondly, the structure of the network traffic flow prediction model is given. In this structure, Users can used a computer to start the traffic flow prediction process, and data collecting module can collect and return the data through the destination device. Thirdly, the dynamic network traffic flow prediction model is implemented based on BP Neural Network. Particularly, in this paper, the BP Neural Network is trained by a modified quantum-behaved particle swarm optimization(QPSO). We modified the QPSO by utilizing chaos signals to implement typical logistic mapping and pursuing the fitness function of a particle by a set of optimal parameters. Afterwards, based on the above process, dynamic network traffic flow prediction model is illustrated. Finally, a series of experiments are conduct to make performance evaluation, and related analyses for experimental results are also given

Using Hybrid Fuzzy Neural Network to Improve the Accuracy of Air Traffic Flow Forecasts

Air traffic is increasing worldwide at a steady annual rate, and airport congestion is already a major issue for air traffic controllers. The traditional method of traffic flow prediction is difficult to adapt to complex air traffic conditions. Due to its self-learning, self-organizing, self-adaptive and anti-jamming capability, the hybrid fuzzy neural network can predict more effectively the air traffic flow than the traditional methods can. A good method for training is an important problem in the prediction of air traffic flow with neural network. This paper will try to find a new model to solve the traffic flow prediction problem by hybrid fuzzy neural network.

Research of Air Traffic Flow Forecasts Based on BP Neural Network

Air traffic is increasing worldwide at a steady annual rate, and airport congestion is already a major issue for air traffic controllers. The traditional method of traffic flow prediction is difficult to adapt to complex air traffic conditions. Due to its self-learning, self-organizing, self-adaptive and anti-jamming capability, the neural network can predict more effectively the air traffic flow than the traditional methods can. A good method for training is an important problem in the prediction of air traffic flow with neural network. This paper will try to find a new model to solve the traffic flow prediction problem by back propagation neural network.

Network Traffic Prediction Based on the Wavelet Analysis and Hopfield Neural Network

Build a mathematical model is the key problem of network traffic prediction. Traditional single network flow model of is not simulate the complex characteristics of network traffic. Therefore, a network traffic prediction hybrid model based on αTrous wavelet analysis and Hopfield neural network is proposed in this paper, which can be used to predict the network traffic flow. First, network traffic is normalized and adopt αTrous wavelet transform; And then reconstruct the wavelet single, and predict through sending low frequency components into AR model and sending the high frequency component into Hopfield neural network model; Last, The predictive value are obtained by composing the components. Simulation results show that the model improves the prediction accuracy, and has the good adaptability to the network . Index Terms—Network traffic, neural network, wavelet, Hopfield.

Road Traffic Noise Prediction with Neural Networks - A Review

This paper aims to summarize the findings of research concerning the application of neural networks in traffic noise prediction. Noise is an environmental agent, regarded as a stressful stimulus. Noise exposure causes changes at different levels in living beings, such as the cardiovascular, endocrine and nervous system. Study of traffic noise prediction models began in 1950s to predict a single vehicle sound pressure level at the road side. After that, several traffic noise prediction models such as FHWA, CORTN, STOP and GO, MITHRA, ASJ etc. were developed depending upon various parameters and conditions. Complexity of error identification by means of classical approaches has led to researchers and designers to explore the possibility of neural solution to the problem of traffic noise prediction. Present study is focused on review of various neural network models developed for road traffic noise prediction.

Variational Inference for Infinite Mixtures of Gaussian Processes With Applications to Traffic Flow Prediction

This paper proposes a new variational approximation for infinite mixtures of Gaussian processes. As an extension of the single Gaussian process regression model, mixtures of Gaussian processes can characterize varying covariances or multimodal data and reduce the deficiency of the computationally cubic complexity of the single Gaussian process model. The infinite mixture of Gaussian processes further integrates a Dirichlet process prior to allowing the number of mixture components to automatically be determined from data. We use variational inference and a truncated stick-breaking representation of the Dirichlet process to approximate the posterior of hidden variables involved in the model. To fix the hyperparameters of the model, the variational EM algorithm and a greedy algorithm are employed. In addition to presenting the variational infinite-mixture model, we apply it to the problem of traffic flow prediction. Experiments with comparisons to other approaches show the effectiveness of the proposed model.

MBFP generalized association rule mining and classification in traffic volume prediction

AbstractGenetic network programming (GNP)‐based time‐related association rules mining method provides a useful mean to investigate future traffic volume of road networks and hence helps us to develop traffic navigation system. Further improvements have been proposed in this paper about the time‐related association rule mining using generalized GNP with multibranches and full‐paths (MBFP) algorithm. For fully utilizing the potential ability of GNP structure, the mechanism of generalized GNP with MBFP is studied. The aim of this algorithm is to better handle association rule extraction from the databases with high efficiency in a variety of time‐related applications, especially in the traffic volume prediction problems. The generalized algorithm which can find the important time‐related association rules is described, and experimental results are presented considering a traffic prediction problem. © 2011 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

Time Related Class Association Rule Mining and Its Application to Traffic Prediction

In this paper, an algorithm capable of finding important time related association rules is proposed where Genetic Network Programming (GNP) with Attribute Accumulation Mechanism (AAM) and Extraction Mechanism at Stages (EMS) is used. Then, the classification system based on extracted time related association rules is proposed to estimate to which class the current traffic data belong. Using this kind of classification mechanism, the traffic prediction is available since the rules extracted are based on time sequences. And, we also present the experimental results on the traffic prediction problem.