Wavelet‐Based Denoising for Traffic Volume Time Series Forecasting with Self‐Organizing Neural Networks

Abstract: In their goal to effectively manage the use of existing infrastructures, intelligent transportation systems require precise forecasting of near‐term traffic volumes to feed real‐time analytical models and traffic surveillance tools that alert of network links reaching their capacity. This article proposes a new methodological approach for short‐term predictions of time series of volume data at isolated cross sections. The originality in the computational modeling stems from the fit of threshold values used in the stationary wavelet‐based denoising process applied on the time series, and from the determination of patterns that characterize the evolution of its samples over a fixed prediction horizon. A self‐organizing fuzzy neural network is optimized in its configuration parameters for learning and recognition of these patterns. Four real‐world data sets from three interstate roads are considered for evaluating the performance of the proposed model. A quantitative comparison made with the results obtained by four other relevant prediction models shows a favorable outcome.     

Near-Term Travel Speed Prediction Utilizing Hilbert–Huang Transform

Abstract:  Accurate short-term prediction of travel speed as a proxy for time is central to many Intelligent Transportation Systems, especially for Advanced Traveler Information Systems and Advanced Traffic Management Systems. In this study, we propose an innovative methodology for such prediction. Because of the inherently direct derivation of travel time from speed data, the study was limited to the use of speed only as a single predictor. The proposed method is a hybrid one that combines the use of the empirical mode decomposition (EMD) and a multilayer feedforward neural network with backpropagation. The EMD is the key part of the Hilbert–Huang transform, which is a newly developed method at NASA for the analysis of nonstationary, nonlinear time series. The rationale for using the EMD is that because of the highly nonlinear and nonstationary nature of link speed series, by decomposing the time series into its basic components, more accurate forecasts would be obtained. We demonstrated the effectiveness of the proposed method by applying it to real-life loop detector data obtained from I-66 in Fairfax, Virginia. The prediction performance of the proposed method was found to be superior to previous forecasting techniques. Rigorous testing of the distribution of prediction errors revealed that the model produced unbiased predictions of speeds. The superiority of the proposed model was also verified during peak periods, midday, and night. In general, the method was accurate, computationally efficient, easy to implement in a field environment, and applicable to forecasting other traffic parameters.     

An Approach to Dynamical Classification of Daily Traffic Patterns

This article proposes a prototype of an urban traffic control system based on a prediction‐after‐classification approach. In an off‐line phase, a repository of traffic control strategies for a set of (dynamic) traffic patterns is constructed. The core of this stage is the k‐means algorithm for daily traffic pattern identification. The clustering method uses the input attributes flow, speed, and occupancy and it transforms the dynamic traffic data at network level in a pseudo‐covariance matrix, which collects the dynamic correlations between the road links. A desirable number of traffic patterns is provided by Bayesian Information Criterion and the ratio of change in dispersion measurements. In an on‐line phase, the current daily traffic pattern is predicted within the repository and its associated control strategy is implemented in the traffic network. The dynamic prediction scheme is constructed on the basis of an existing static prediction method by accumulating the trials on set of patterns in the repository. This proposal has been assessed in synthetic and real networks testing its effectiveness as a data mining tool for the analysis of traffic patterns. The approach promises to effectively detect the current daily traffic pattern and is open to being used in intelligent traffic management systems.     

Dynamic origin‐destination flow estimation using automatic vehicle identification data: A 3D convolutional neural network approach

Dynamic origin‐destination (OD) flow estimation is one of the most fundamental problems in traffic engineering. Despite numerous existing studies, the OD flow estimation problem remains challenging, as there is large dimensional difference between the unknown values to be estimated and the known traffic observations. To meet the needs of active traffic management and control, accurate time‐dependent OD flows are required to understand time‐of‐day traffic flow patterns. In this work, we propose a three‐dimensional (3D) convolution‐based deep neural network, “Res3D,” to learn the high‐dimensional correlations between local traffic patterns presented by automatic vehicle identification observations and OD flows. In this paper, a practical framework combining simulation‐based model training and few‐shot transfer learning is introduced to enhance the applicability of the proposed model, as continuously observing OD flows could be expensive. The proposed model is extensively tested based on a realistic road network, and the results show that for significant OD flows, the relative errors are stable around 5%, outperforming several other models, including prevalent neural networks as well as existing estimation models. Meanwhile, corrupted and out‐of‐distribution samples are generated as real‐world samples to validate Res3D's transferability, and the results indicated a 60% improvement with few‐shot transfer learning. Therefore, this proposed framework could help to bridge the gaps between traffic simulations and empirical cases.     

Short‐Term Traffic Speed Prediction for an Urban Corridor

Short‐term traffic speed prediction is one of the most critical components of an intelligent transportation system (ITS). The accurate and real‐time prediction of traffic speeds can support travellers’ route choices and traffic guidance/control. In this article, a support vector machine model (single‐step prediction model) composed of spatial and temporal parameters is proposed. Furthermore, a short‐term traffic speed prediction model is developed based on the single‐step prediction model. To test the accuracy of the proposed short‐term traffic speed prediction model, its application is illustrated using GPS data from taxis in Foshan city, China. The results indicate that the error of the short‐term traffic speed prediction varies from 3.31% to 15.35%. The support vector machine model with spatial‐temporal parameters exhibits good performance compared with an artificial neural network, a k‐nearest neighbor model, a historical data‐based model, and a moving average data‐based model.     

A genetic-algorithms based evolutionary computational neural network for modelling spatial interaction dataNeural network for modelling spatial interaction data

Building a feedforward computational neural network model (CNN) involves two distinct tasks: determination of the network topology and weight estimation. The specification of a problem adequate network topology is a key issue and the primary focus of this contribution. Up to now, this issue has been either completely neglected in spatial application domains, or tackled by search heuristics (see Fischer and Gopal 1994). With the view of modelling interactions over geographic space, this paper considers this problem as a global optimization problem and proposes a novel approach that embeds backpropagation learning into the evolutionary paradigm of genetic algorithms. This is accomplished by interweaving a genetic search for finding an optimal CNN topology with gradient-based backpropagation learning for determining the network parameters. Thus, the model builder will be relieved of the burden of identifying appropriate CNN-topologies that will allow a problem to be solved with simple, but powerful learning mechanisms, such as backpropagation of gradient descent errors. The approach has been applied to the family of three inputs, single hidden layer, single output feedforward CNN models using interregional telecommunication traffic data for Austria, to illustrate its performance and to evaluate its robustness.     

A methodology for obtaining spatiotemporal information of the vehicles on bridges based on computer vision

Spatiotemporal information of the vehicles on a bridge is important evidence for reflecting the stress state and traffic density of the bridge. A methodology for obtaining the information is proposed based on computer vision technology, which contains the detection by Faster region‐based convolutional neural network (Faster R‐CNN), multiple object tracking, and image calibration. For minimizing the detection time, the ZF (Zeiler & Fergus) model with five convolutional layers is selected as the shared part between Region Proposal Network and Fast R‐CNN in Faster R‐CNN. An image data set including 1,694 images is established about eight types of vehicles for training Faster R‐CNN. Combined with the detection of each frame of the video, the methods of multiple object tracking and image calibration are developed for acquiring the vehicle parameters, including the length, number of axles, speed, and the lane that the vehicle is in. The method of tracking is mainly based on the judgment of the distances between the vehicle bounding boxes in virtual detection region. As for image calibration, it is based on the moving standard vehicles whose lengths are known, which can be regarded as the 3D template to calculate the vehicle parameters. After acquiring the vehicles' parameters, the spatiotemporal information of the vehicles can be obtained. The proposed system has a frame rate of 16 fps and only needs two cameras as the input device. The system is successfully applied on a double tower cable‐stayed bridge, and the identification accuracies of the types and number of axles are about 90 and 73% in the virtual detection region, and the speed errors of most vehicles are less than 6%.     

Freeway Travel Time Prediction Using Takagi–Sugeno–Kang Fuzzy Neural Network

This article presents a Takagi–Sugeno–Kang Fuzzy Neural Network (TSKFNN) approach to predict freeway corridor travel time with an online computing algorithm. TSKFNN, a combination of a Takagi–Sugeno–Kang (TSK) type fuzzy logic system and a neural network, produces strong prediction performance because of its high accuracy and quick convergence. Real world data collected from US‐290 in Houston, Texas are used to train and validate the network. The prediction performance of the TSKFNN is investigated with different combinations of traffic count, occupancy, and speed as input options. The comparison between online TSKFNN, offline TSKFNN, the back propagation neural network (BPNN) and the time series model (ARIMA) is made to evaluate the performance of TSKFNN. The results show that using count, speed, and occupancy together as input produces the best TSKFNN predictions. The online TSKFNN outperforms other commonly used models and is a promising tool for reliable travel time prediction on a freeway corridor.     

Slope stability analysis based on big data and convolutional neural network

The Limit Equilibrium Method (LEM) is commonly used in traditional slope stability analyses, but it is time-consuming and complicated. Due to its complexity and nonlinearity involved in the evaluation process, it cannot provide a quick stability estimation when facing a large number of slopes. In this case, the convolutional neural network (CNN) provides a better alternative. A CNN model can process data quickly and complete a large amount of data analysis in a specific situation, while it needs a large number of training samples. It is difficult to get enough slope data samples in practical engineering. This study proposes a slope database generation method based on the LEM. Samples were amplified from 40 typical slopes, and a sample database consisting of 20000 slope samples was established. The sample database for slopes covered a wide range of slope geometries and soil layers’ physical and mechanical properties. The CNN trained with this sample database was then applied to the stability prediction of 15 real slopes to test the accuracy of the CNN model. The results show that the slope stability prediction method based on the CNN does not need complex calculation but only needs to provide the slope coordinate information and physical and mechanical parameters of the soil layers, and it can quickly obtain the safety factor and stability state of the slopes. Moreover, the prediction accuracy of the CNN trained by the sample database for slope stability analysis reaches more than 99%, and the comparisons with the BP neural network show that the CNN has significant superiority in slope stability evaluation. Therefore, the CNN can predict the safety factor of real slopes. In particular, the combination of typical actual slopes and generated slope data provides enough training and testing samples for the CNN, which improves the prediction speed and practicability of the CNN-based evaluation method in engineering practice.     

Wavelet Packet-Autocorrelation Function Method for Traffic Flow Pattern Analysis

Abstract:   Accurate and timely forecasting of traffic flow is of paramount importance for effective management of traffic congestion in intelligent transportation systems. A detailed understanding of the properties of traffic flow is essential for building a reliable forecasting model. The discrete wavelet packet transform (DWPT) provides more coefficients than the conventional discrete wavelet transform (DWT), representing additional subtle details of a signal. In wavelet multiresolution analysis, an important decision is the selection of the decomposition level. In this research, the statistical autocorrelation function (ACF) is proposed for the selection of the decomposition level in wavelet multiresolution analysis of traffic flow time series. A hybrid wavelet packet-ACF method is proposed for analysis of traffic flow time series and determining its self-similar, singular, and fractal properties. A DWPT-based approach combined with a wavelet coefficients penalization scheme and soft thresholding is presented for denoising the traffic flow. The proposed methodology provides a powerful tool in removing the noise and identifying singularities in the traffic flow. The methods created in this research are of value in developing accurate traffic-forecasting models .     

Deep convolutional neural networks for uncertainty propagation in random fields

The development of a reliable and robust surrogate model is often constrained by the dimensionality of the problem. For a system with high‐dimensional inputs/outputs (I/O), conventional approaches usually use a low‐dimensional manifold to describe the high‐dimensional system, where the I/O data are first reduced to more manageable dimensions and then the condensed representation is used for surrogate modeling. In this study, a new solution scheme for this type of problem based on a deep learning approach is presented. The proposed surrogate is based on a particular network architecture, that is, convolutional neural networks. The surrogate architecture is designed in a hierarchical style containing three different levels of model structures, advancing the efficiency and effectiveness of the model in the aspect of training. To assess the model performance, uncertainty quantification is carried out in a continuum mechanics benchmark problem. Numerical results suggest the proposed model is capable of directly inferring a wide variety of I/O mapping relationships. Uncertainty analysis results obtained via the proposed surrogate have successfully characterized the statistical properties of the output fields compared to the Monte Carlo estimates.     

Structure‐aware 3D reconstruction for cable‐stayed bridges: A learning‐based method

A powerful deep learning‐based three‐dimensional (3D) reconstruction method for reconstructing structure‐aware semantic 3D models of cable‐stayed bridges is proposed herein. Typically, conventional bridge semantic 3D model reconstruction methods are not robust when low‐quality point clouds are used. Furthermore, they are suited particularly for their respective fields and less generalized for cable‐stayed bridges. Hence, a structure‐aware learning‐based cable‐stayed bridge 3D reconstruction framework is proposed. The encoder part of the network uses both multiview images and a photogrammetric point cloud as input, whereas the decoder part uses a recursive binary tree network to model a high‐level structural relation graph and low‐level 3D geometric shapes. Two actual cable‐stayed bridges are employed as examples to evaluate the proposed method. Test results demonstrate that the proposed method successfully reconstructs the bridge model with structural components and their relations. Quantitative results indicate that the predicted models achieved an average F₁ score of 99.01%, a Chamfer distance of 0.0259, and a mesh‐to‐cloud distance of 1.78 m. The achieved result is similar to that obtained using the manual reconstruction approach in terms of component‐wise accuracy, and it is considerably better than that obtained using the manual approach in terms of spatial accuracy. In addition, the proposed recursive binary tree network is robust to noise and partial scans. The potential applications of the obtained 3D bridge models are discussed.     

A deep marked graph process model for citywide traffic congestion forecasting

Forecasting citywide traffic congestion on large road networks has long been a nontrivial research problem due to the challenge of modeling complex evolution patterns of congestion in highly stochastic traffic environments. Arguing that purely data-driven methods may not perform well for congestion forecasting, we propose a deep marked graph process model for predicting the congestion indices and the occurrence time of traffic congestion events for complex signalized road networks. Traffic congestion is considered as a nonrigorous spatiotemporal extreme event. We extend the traditional point process model by integrating a specially designed spatiotemporal graph convolutional network. This hybrid strategy takes advantage of the simple form of the point process model as well as the ability of graph neural networks to emulate the evolution of congestion. Experiments on real-world congestion data sets show that the proposed method outperforms state-of-the-art baseline methods, yielding satisfactory prediction results on a large signalized road network with superior computational efficiency.     

The Fast Prefix Coding Algorithm (FPCA) for 3D Pavement Surface Data Compression

The enormous data inflow during three‐dimensional (3D) pavement surface data collection requires an efficient compression system for 3D data. However, with respect to the phase of lossless encoding, the commonly used Huffman Coding is inefficient in terms of speed and memory usage for encoding 3D pavement surfaces. The Fast Prefix Coding Algorithm (FPCA) is proposed in the article as an effective substitute of Huffman Coding at the stage of lossless encoding. It is demonstrated in the article that the FPCA is much faster and more memory efficient than Huffman Coding, while outperforming Shannon–Fano Coding in terms of both redundancy and time efficiency. The FPCA‐based coding approach is a modification of the baseline JPEG algorithm to support 3D pavement data whose dynamic range is more than 12 bits. The presented modifications include algorithms for Quantization, Run‐Length Encoding and Entropy Coding without limiting data depth in terms of dynamic range. Compared with the baseline JPEG approach, the proposed coding system is able to restrict the data loss more successfully and can achieve a significantly higher level of time efficiency and compression ratio (over than 30:1 for most of the evaluated 3D images). With parallel computing techniques, encoding full‐lane width pavement in 3D and at 1 mm resolution with an up‐to‐date desktop computer can be conducted at 150 MPH or even higher speed.     

A three‐dimensional distance transform for optimizing constrained mountain railway alignments

Railway alignment optimization is considered one of the most complicated and time‐consuming problems in railway planning and design. It requires searching among the infinite potential alternatives in huge three‐dimensional (3D) search spaces for a near‐optimal alignment, while considering complex constraints and a nonlinear objective function. In mountainous regions, the complex terrain and constructions require additional and more complex constraints than in topographically simpler regions. In this paper, the authors solve this problem with an algorithm based on a 3D distance transform (3D‐DT). Compared with previous two‐dimensional distance transform (2D‐DT) methods developed in this field, the feasible search spaces of 3D‐DT are greatly increased. Consequently, this new method can find more alternatives with higher qualities. In this approach, an erythrocyte‐shaped 3D neighboring mask is developed to narrow local search spaces and speed up the search process. Besides, a stepwise‐backstepping strategy is designed to dynamically determine feasible 3D search spaces and efficiently search the study area. During the 3D‐DT search process, multiple constraints, including geometric, construction, and location constraints, are effectively handled. After the 3D‐DT search, a genetic algorithm is employed to optimize the 3D‐DT paths into final alignments. Finally, this novel approach is applied to an actual case in a complex mountainous region. The comprehensive cost of the best solution generated by 3D‐DT is 16% below a manual solution produced by very experienced human designers. Furthermore, the total number of feasible alternatives found by 3D‐DT is 4.3 times greater than by 2D‐DT. The comprehensive cost of the best 3D‐DT solution is 10% below the best one generated by 2D‐DT.     

A graph deep learning method for short‐term traffic forecasting on large road networks

Short‐term traffic flow prediction on a large‐scale road network is challenging due to the complex spatial–temporal dependencies, the directed network topology, and the high computational cost. To address the challenges, this article develops a graph deep learning framework to predict large‐scale network traffic flow with high accuracy and efficiency. Specifically, we model the dynamics of the traffic flow on a road network as an irreducible and aperiodic Markov chain on a directed graph. Based on the representation, a novel spatial–temporal graph inception residual network (STGI‐ResNet) is developed for network‐based traffic prediction. This model integrates multiple spatial–temporal graph convolution (STGC) operators, residual learning, and the inception structure. The proposed STGC operators can adaptively extract spatial–temporal features from multiple traffic periodicities while preserving the topology information of the road network. The proposed STGI‐ResNet inherits the advantages of residual learning and inception structure to improve prediction accuracy, accelerate the model training process, and reduce difficult parameter tuning efforts. The computational complexity is linearly related to the number of road links, which enables citywide short‐term traffic prediction. Experiments using a car‐hailing traffic data set at 10‐, 30‐, and 60‐min intervals for a large road network in a Chinese city shows that the proposed model outperformed various state‐of‐the‐art baselines for short‐term network traffic flow prediction.     

Fuzzy Modeling Approach for Combined Forecasting of Urban Traffic Flow

Abstract:   This article addresses the problem of the accuracy of short-term traffic flow forecasting in the complex case of urban signalized arterial networks. A new, artificial intelligence (AI)-based approach is suggested for improving the accuracy of traffic predictions through suitably combining the forecasts derived from a set of individual predictors. This approach employs a fuzzy rule-based system (FRBS), which is augmented with an appropriate metaheuristic (direct search) technique to automate the tuning of the system parameters within an online adaptive rolling horizon framework. The proposed hybrid FRBS is used to nonlinearly combine traffic flow forecasts resulting from an online adaptive Kalman filter (KF) and an artificial neural network (ANN) model. The empirical results obtained from the model implementation into a real-world urban signalized arterial demonstrate the ability of the proposed approach to considerably overperform the given individual traffic predictors .     

混沌局域法与神经网络组合供水量预测

城市供水量是非线性、非平稳时间序列,组合预测模型能获得更高精度预测结果。通过深入分析混沌局域法与神经网络预测模型特点,提出了一种新的组合预测模型。首先,应用混沌局域法对城市日供水量进行初预测,然后,应用神经网络对预测结果进行修正。由于所提出的组合模型利用了混沌局域法及神经网络进行优势互补,能同时提高预测精度与计算效率。为验证所提出组合预测模型的可行性,采用某市7a实测供水量数据,对混沌局域法、BPNN、RBF及GRNN神经网络4种单一预测模型及相应的3种组合模型预测精度进行定量分析,结果表明,组合预测模型精度都高于对应单一预测模型,混沌局域法与GRNN神经网络组合模型预测精度最高,且运算时间远低于单一神经网络模型运算时间。     

Digital image correlation-based structural state detection through deep learning

This paper presents a new approach for automatical classification of structural state through deep learning. In this work, a Convolutional Neural Network (CNN) was designed to fuse both the feature extraction and classification blocks into an intelligent and compact learning system and detect the structural state of a steel frame; the input was a series of vibration signals, and the output was a structural state. The digital image correlation (DIC) technology was utilized to collect vibration information of an actual steel frame, and subsequently, the raw signals, without further pre-processing, were directly utilized as the CNN samples. The results show that CNN can achieve 99% classification accuracy for the research model. Besides, compared with the backpropagation neural network (BPNN), the CNN had an accuracy similar to that of the BPNN, but it only consumes 19% of the training time. The outputs of the convolution and pooling layers were visually displayed and discussed as well. It is demonstrated that: 1) the CNN can extract the structural state information from the vibration signals and classify them; 2) the detection and computational performance of the CNN for the incomplete data are better than that of the BPNN; 3) the CNN has better anti-noise ability.     

Deep learning-based evaluation of factor of safety with confidence interval for tunnel deformation in spatially variable soil

The random finite difference method (RFDM) is a popular approach to quantitatively evaluate the influence of inherent spatial variability of soil on the deformation of embedded tunnels. However, the high computational cost is an ongoing challenge for its application in complex scenarios. To address this limitation, a deep learning-based method for efficient prediction of tunnel deformation in spatially variable soil is proposed. The proposed method uses one-dimensional convolutional neural network (CNN) to identify the pattern between random field input and factor of safety of tunnel deformation output. The mean squared error and correlation coefficient of the CNN model applied to the newly untrained dataset was less than 0.02 and larger than 0.96, respectively. It means that the trained CNN model can replace RFDM analysis for Monte Carlo simulations with a small but sufficient number of random field samples (about 40 samples for each case in this study). It is well known that the machine learning or deep learning model has a common limitation that the confidence of predicted result is unknown and only a deterministic outcome is given. This calls for an approach to gauge the model's confidence interval. It is achieved by applying dropout to all layers of the original model to retrain the model and using the dropout technique when performing inference. The excellent agreement between the CNN model prediction and the RFDM calculated results demonstrated that the proposed deep learning-based method has potential for tunnel performance analysis in spatially variable soils.