Spatiotemporal memory decoupled transformer for traffic flow forecasting

Traffic flow forecasting is an important problem for the successful deployment of intelligent transportation systems, which has been studied for more than two decades. In recent years, deep learning methods are emerging to serve as the benchmark tool for traffic flow forecasting due to its superior prediction performance. However, most studies are based on simple deep learning methods that can not capture inter- and intra-day traffic patterns as well as the correlation between contextual factors like the weather and the traffic flow. In this paper, we propose a novel deep-learning-based method for daily traffic flow forecasting where incorporating contextual factors and traffic flow patterns can be critical. First, a particular convolutional neural network (CNN) is deployed to extract inter- and intra- day traffic flow patterns. Then extracted features are fed into long short-term memory (LSTM) units to learn the intra-day temporal evolution of traffic flow. Finally, contextual information of historical days is integrated to enhance the prediction performance. Through a real-data case study, we show that the proposed approach achieves over 90% prediction accuracy which greatly outperforms existing benchmark methods and its forecasting performance is robust under various scenarios.

Spatial–temporal dependence and similarity aware traffic flow forecasting

Classification of traffic flow patterns is important in many areas in traffic management and planning, traffic flow prediction, real-time control and datasbase design. This paper aims at clustering and modeling the traffic flow patterns from large amount of off-line traffic data. Compared with the research based on temporal context factors, urban function zones and ecologies and other geograhpical context factors complicates the analysis of this reserch. Based on the pre-classification of traffic flow data, we constructed a traffic flow similarity measurement method and applied an unsupervised learning method to achieve clustering. And then, we combine the clustering reseults with geographic context information and perform correlation analysis between them. We believe that the proposed traffic flow patterns extraction and modeling methodology, combined with the empirical analysis, help us to better understand the traffic flow patterns of large scale urban roads in modern metropolis.

In metropolitan areas, common traffic issues include traffic congestion, traffic accidents, air pollution, and energy consumption occur. To resolve this issues, Intelligent Transportation Systems (ITS) have been evolved by many researchers. One of the important sub-systems in the development of ITS is a Traffic Management System (TMS) which attempts to reduce a traffic congestion. In fact, TMS itself relies on the estimation of traffic flow, therefore providing such an accurate traffic flow prediction is needed. For this reason, we aim to provide an accurate traffic flow prediction to facilitate this system. In this works, a Supervised Deep Learning Based Traffic Flow Prediction (SDLTFP) was proposed which is a type of fully-connected deep neural network (FC-DNN). Timely prediction is also a major issue in guaranteeing reliable traffic flow prediction. However, training a deep network could be time-consuming, and overfitting is might be happening, especially when feeding small data into the deep architecture. The network is learned perfectly during the training, but in testing with the new data, it could fail to generalize the model. We adopt the Batch Normalization (BN) and Dropout techniques to help the network training. SGD and momentum are carried out to update the weight. We then take advantage of open data as historical traffic data which are then used to predict future traffic flow with the proposed method and model above. Experiments show that the Mean Absolute Percentage Error (MAPE) for our traffic flow prediction is within 5 % using sample data and between 15% to 20% using out of the sample data. Training a deep network faster with BN and Dropout reduces the overfitting.

Motivated by the need for accurate traffic flow prediction in transportation management, we propose a functional data method to analyze traffic flow patterns and predict future traffic flow. In this study we approach the problem by sampling traffic flow trajectories from a mixture of stochastic processes. The proposed functional mixture prediction approach combines functional prediction with probabilistic functional classification to take distinct traffic flow patterns into account. The probabilistic classification procedure, which incorporates functional clustering and discrimination, hinges on subspace projection. The proposed methods not only assist in predicting traffic flow trajectories, but also identify distinct patterns in daily traffic flow of typical temporal trends and variabilities. The proposed methodology is widely applicable in analysis and prediction of longitudinally recorded functional data.

Accurate traffic flow forecasting is crucial to the development of intelligent transportation systems (ITS). Based on statistical learning theory, support vector machine (SVM) has better generalization performance and can guarantee global minima for given training data. However, the good generalization performance of SVM highly depends on the construction of kernel function. An effective multi-scale Marr wavelet kernel which we combine the wavelet theory with SVM is presented in this paper. The forecasting performance is evaluated by real-time traffic flow data of highway in Los Angeles, USA and a variety of experiments are carried out. Compared to wavelet kernel function and RBF kernel function, the multi-scale wavelet kernel function has much more precise forecasting rate and higher efficiency, especially for boundary approximation.

Double-deck elevator system (DDES) has been invented firstly as a solution to improve the transportation capacity of elevator group systems in the up-peak traffic pattern. The transportation capacity could be even doubled when DDES runs in a pure up-peak traffic pattern where two connected cages stop at every two floors in an elevator round trip. However, the specific features of DDES make the elevator system intractable when it runs in some other traffic patterns. Moreover, since almost all of the traffic flows vary continuously during a day, an optimized controller of DDES is required to adapt the varying traffic flow. In this paper, we have proposed a controller adaptive to traffic flows for DDES using genetic network programming (GNP) based on our past studies in this field, where the effectiveness of DDES controller using GNP has been verified in three typical traffic patterns. A traffic flow judgment part was introduced into the GNP framework of DDES controller, and the different parts of GNP were expected to be functionally localized by the evolutionary process to make the appropriate cage assignment in different traffic flow patterns. Simulation results show that the proposed method outperforms a conventional approach and two heuristic approaches in a varying traffic flow during the work time of a typical office building.

The double‐deck elevator system (DDES) has been invented firstly as a solution to improve the transportation capacity of elevator group systems in the up‐peak traffic pattern. The transportation capacity could be even doubled when DDES runs in a pure up‐peak traffic pattern where two connected cages stop at every two floors in an elevator round trip. However, the specific features of DDES make the elevator system intractable when it runs in some other traffic patterns. Moreover, since almost all the traffic flows vary continuously during a day, an optimized controller of DDES is required to adapt to the varying traffic flow. In this paper, we have proposed a controller adaptive to traffic flows for DDES using Genetic Network Programming (GNP) based on our past studies in this field, where the effectiveness of DDES controller using GNP has been verified in three typical traffic patterns. A new traffic flow judgment part was introduced into the GNP framework of DDES controller in this paper, and the different parts of GNP were expected to be functionally localized by the evolutionary process to make the appropriate cage assignment in different traffic flow patterns. Simulation results show that the proposed method outperforms a conventional approach and two heuristic approaches in a varying traffic flow during the working time of a typical office building. Copyright © 2008 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

Timely and accurate traffic flow information is essential for the realization of intelligent transportation systems (ITS). For the existing traffic flow prediction models, only the temporal features were extracted but the spatial features were neglects. A short-term traffic flow prediction model based on Pearson Correlation Coefficient (PCC) and Bi-directional Long Short-Term Memory Network (BiLSTM) model was proposed. The PCC was used to mine the spatial dependence of the road network, and the temporal features of traffic flow were mined using BiLSTM model. The extracted temporal and spatial features were combined to achieve short-term flow prediction. The experimental results show that the PCC-BiLSTM model has higher prediction accuracy and lower error than the prediction models considering only temporal features, which verifies the effectiveness of traffic flow prediction in consideration of both temporal and spatial characteristics.

Accurate traffic flow prediction is crucial for improving transportation efficiency. To improve the accuracy of traffic flow prediction, we developed a traffic flow prediction framework—namely, traffic flow multicomponent network—that appropriately processes the noise, volatility, and nonlinearity in traffic flow data. This framework comprises three components: a factor selection component, traffic flow decomposition component, and traffic flow prediction component. The factor selection component considers the dynamic effects of weather‐related, environmental, and spatiotemporal factors on traffic flow; it then extracts and analyzes factors exhibiting strong correlations with traffic flow. The traffic flow decomposition component optimizes the parameters of variational mode decomposition on the basis of the envelope entropy by using the sparrow search algorithm; it then transforms traffic flow into multiple intrinsic mode functions to enable accurate traffic flow prediction. Finally, the traffic flow prediction component constructs dynamic feature matrices by using a bidirectional gated recurrent unit model to identify relationships within the data. Moreover, it uses an attention mechanism to assign different weights to different features on the basis of the importance of these features to traffic flow prediction, thereby enabling the efficient processing of a large volume of data. The performance of the proposed framework was examined in experiments conducted on large volumes of traffic flow data with different time granularities. The results indicated that the proposed framework achieved high prediction accuracy and stability for various time granularities, data samples, dataset sizes, and noise conditions. Moreover, it generally outperformed existing traffic flow prediction models under all experimental conditions.

Accurate short-term traffic flow prediction is the foundation of the efficient and proactive management of freeway networks, especially on the abnormal traffic states. The relationship between traffic flow on the current section and the upstream stations can be used for predicting short-term traffic flow. In this paper, we reveal this relationship by the traffic flow structure pattern. The structure pattern can be drawn from real freeway toll data and a few video detective cameras on the freeway segments. Based on the stability pattern, a new traffic flow prediction algorithm has been proposed. Experimental based on real data showed that the prediction method based on structure pattern is an effective approach for traffic flow prediction, especially on the abnormal traffic state.

This paper presents a methodology for the mathematical characterization of three-dimensional (3-D) airspace traffic flows from flight position data. The methodology may be applied to flight data in the en route and terminal-area environments for such airspace applications as the creation of a traffic flow network model, route design based upon overlays of current traffic patterns, and benefits assessment. The methodology is applied to arrivals to the Philadelphia International (PHL) airport and to departures from the Dulles International (IAD) airport. The methodology selects a flight that best characterizes the 3-D distribution of flight position data for flights with a common origin and destination. This characteristic flight, called the typical flight, is most centrally located in an airspace volume about a traffic flow. The typical track respects traffic flow patterns present in the data. The typical track algorithm also provides a means of classifying different traffic flow patterns within an overall traffic flow which can be related to assessing airspace efficiency. This paper presents a method for converting the selected typical track to a route within the context of an area navigation (RNAV) procedure design. An example of using the typical track to categorize flight patterns for the PHL arrival data set shows how this may guide RNAV benefits assessment in a Terminal Radar Approach Control (TRACON).

One-step vs horizon-step training strategies for multi-step traffic flow forecasting with direct particle swarm optimization grid search support vector regression and long short-term memory

Transport plays a fundamental role in the economic development for big cities. Quito, the capital of Ecuador, presents serious mobility problems, with a significant amount of traffic both in the city center and in its main access roads. The detailed proposal in this article aims to link Quito with several new generation technologies such as Internet of Things (Iot), Sensor Web Enablement (SWE) standard for heterogeneous sensor communication, and a system of real-time notifications using the communications protocol Message Queue Telemetry Transport (MQTT). These technologies will be used for the development of an Intelligent Transport System (ITS) that takes advantage of the massive deployment of smartphones in the society as main sources of information, including Arduino and Raspberry modules to check efficient communication. This ITS bases its operation on the paradigm of the Crowdsensing, which allows to reflect more accurately the reality of the object of study when more collaborators use the system; reaching up to 30% improvement in the situational consciousness of the environment variable, the traffic.

As one of the most widely used applications of the Internet of Things (IoT), intelligent transportation system (ITS) is of great significance for urban traffic planning, traffic control, and traffic guidance. However, widespread traffic congestion occurs with the increased number of vehicles. The traffic flow prediction is a good idea for traffic congestion. Therefore, many schemes have been proposed for accurate and real-time traffic flow prediction, but there still exist many issues, including low accuracy, weak adaptability and inferior real-time. Meanwhile, the complex spatial and temporal dependencies in traffic flow are still challenging. To address the above issues, we propose a novel spatial-temporal Chebyshev graph neural network model (ST-ChebNet) for traffic flow prediction to capture the spatial-temporal features, which can ensure accurate traffic flow prediction. Concretely, we first add a fully connected layer to fuse the features of traffic data into a new feature to generate a matrix, and then the long short-term memory (LSTM) model is adopted to learn traffic state changes for capturing the temporal dependencies. Then, we use the Chebyshev graph neural network (ChebNet) to learn the complex topological structures in the traffic network for capturing the spatial dependencies. Eventually, the spatial features and the temporal features are fused to guarantee the traffic flow prediction. The experiments show that ST-ChebNet can make accurate and real-time traffic flow prediction compared with other eight baseline methods on real-world traffic data sets PeMS.