Traffic forecasting is a particularly challenging application of spatiotemporal forecasting, due to the time-varying traffic patterns and the complicated spatial dependencies on road networks. To ...address this challenge, we learn the traffic network as a graph and propose a novel deep learning framework, Traffic Graph Convolutional Long Short-Term Memory Neural Network (TGC-LSTM), to learn the interactions between roadways in the traffic network and forecast the network-wide traffic state. We define the traffic graph convolution based on the physical network topology. The relationship between the proposed traffic graph convolution and the spectral graph convolution is also discussed. An L1-norm on graph convolution weights and an L2-norm on graph convolution features are added to the model's loss function to enhance the interpretability of the proposed model. Experimental results show that the proposed model outperforms baseline methods on two real-world traffic state datasets. The visualization of the graph convolution weights indicates that the proposed framework can recognize the most influential road segments in real-world traffic networks.
Understanding how congestion at one location can cause ripples throughout large-scale transportation network is vital for transportation researchers and practitioners to pinpoint traffic bottlenecks ...for congestion mitigation. Traditional studies rely on either mathematical equations or simulation techniques to model traffic congestion dynamics. However, most of the approaches have limitations, largely due to unrealistic assumptions and cumbersome parameter calibration process. With the development of Intelligent Transportation Systems (ITS) and Internet of Things (IoT), transportation data become more and more ubiquitous. This triggers a series of data-driven research to investigate transportation phenomena. Among them, deep learning theory is considered one of the most promising techniques to tackle tremendous high-dimensional data. This study attempts to extend deep learning theory into large-scale transportation network analysis. A deep Restricted Boltzmann Machine and Recurrent Neural Network architecture is utilized to model and predict traffic congestion evolution based on Global Positioning System (GPS) data from taxi. A numerical study in Ningbo, China is conducted to validate the effectiveness and efficiency of the proposed method. Results show that the prediction accuracy can achieve as high as 88% within less than 6 minutes when the model is implemented in a Graphic Processing Unit (GPU)-based parallel computing environment. The predicted congestion evolution patterns can be visualized temporally and spatially through a map-based platform to identify the vulnerable links for proactive congestion mitigation.
Taxi GPS trajectories data contain massive spatial and temporal information of urban human activity and mobility. Taking taxi as mobile sensors, the information derived from taxi trips benefits the ...city and transportation planning. The original data used in study are collected from more than 1100 taxi drivers in Harbin city. We firstly divide the city area into 400 different transportation districts and analyze the origin and destination distribution in urban area on weekday and weekend. The Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm is used to cluster pick-up and drop-off locations. Furthermore, four spatial interaction models are calibrated and compared based on trajectories in shopping center of Harbin city to study the pick-up location searching behavior. By extracting taxi trips from GPS data, travel distance, time and average speed in occupied and non-occupied status are then used to investigate human mobility. Finally, we use observed OD matrix of center area in Harbin city to model the traffic distribution patterns based on entropy-maximizing method, and the estimation performance verify its effectiveness in case study.
•We use taxi GPS data to analyze travel demand distributions.•DBSCAN algorithm is used to cluster pick-up and drop-off locations.•Spatial interaction models are calibrated and compared to study searching behavior.•Travel distance, time and average speed are utilized to explore human mobility.•We construct an entropy-maximizing model to estimate the traffic distribution.
This paper proposes a new method in construction fuzzy neural network to forecast travel speed for multi-step ahead based on 2-min travel speed data collected from three remote traffic microwave ...sensors located on a southbound segment of a fourth ring road in Beijing City. The first-order Takagi-Sugeno system is used to complete the fuzzy inference. To train the evolving fuzzy neural network (EFNN), two learning processes are proposed. First, a K-means method is employed to partition input samples into different clusters and a Gaussian fuzzy membership function is designed for each cluster to measure the membership degree of samples to the cluster centers. As the number of input samples increases, the cluster centers are modified and membership functions are also updated. Second, a weighted recursive least squares estimator is used to optimize the parameters of the linear functions in the Takagi-Sugeno type fuzzy rules. Furthermore, a trigonometric regression function is introduced to capture the periodic component in the raw speed data. Specifically, the predicted performance between the proposed model and six traditional models are compared, which are artificial neural network, support vector machine, autoregressive integrated moving average model, and vector autoregressive model. The results suggest that the prediction performances of EFNN are better than those of traditional models due to their strong learning ability. As the prediction time step increases, the EFNN model can consider the periodic pattern and demonstrate advantages over other models with smaller predicted errors and slow raising rate of errors.
•A Long Short-Term Memory Neural Network (LSTM) is developed for travel speed prediction.•The LSTM NN can capture the long-term temporal dependency for time series.•The LSTM NN can automatically ...determine the optimal time window.•A comparative study suggests that the LSTM NN receives the best performance.
Neural networks have been extensively applied to short-term traffic prediction in the past years. This study proposes a novel architecture of neural networks, Long Short-Term Neural Network (LSTM NN), to capture nonlinear traffic dynamic in an effective manner. The LSTM NN can overcome the issue of back-propagated error decay through memory blocks, and thus exhibits the superior capability for time series prediction with long temporal dependency. In addition, the LSTM NN can automatically determine the optimal time lags. To validate the effectiveness of LSTM NN, travel speed data from traffic microwave detectors in Beijing are used for model training and testing. A comparison with different topologies of dynamic neural networks as well as other prevailing parametric and nonparametric algorithms suggests that LSTM NN can achieve the best prediction performance in terms of both accuracy and stability.
Unmanned aerial vehicles (UAVs) are gaining popularity in traffic monitoring due to their low cost, high flexibility, and wide view range. Traffic flow parameters such as speed, density, and volume ...extracted from UAV-based traffic videos are critical for traffic state estimation and traffic control and have recently received much attention from researchers. However, different from stationary surveillance videos, the camera platforms move with UAVs, and the background motion in aerial videos makes it very challenging to process for data extraction. To address this problem, a novel framework for real-time traffic flow parameter estimation from aerial videos is proposed. The proposed system identifies the directions of traffic streams and extracts traffic flow parameters of each traffic stream separately. Our method incorporates four steps that make use of the Kanade-Lucas-Tomasi (KLT) tracker, k-means clustering, connected graphs, and traffic flow theory. The KLT tracker and k-means clustering are used for interest-point-based motion analysis; then, four constraints are proposed to further determine the connectivity of interest points belonging to one traffic stream cluster. Finally, the average speed of a traffic stream as well as density and volume can be estimated using outputs from previous steps and reference markings. Our method was tested on five videos taken in very different scenarios. The experimental results show that in our case studies, the proposed method achieves about 96% and 87% accuracy in estimating average traffic stream speed and vehicle count, respectively. The method also achieves a fast processing speed that enables real-time traffic information estimation.
Traffic signs recognition (TSR) is an important part of some advanced driver-assistance systems (ADASs) and auto driving systems (ADSs). As the first key step of TSR, traffic sign detection (TSD) is ...a challenging problem because of different types, small sizes, complex driving scenes, and occlusions. In recent years, there have been a large number of TSD algorithms based on machine vision and pattern recognition. In this paper, a comprehensive review of the literature on TSD is presented. We divide the reviewed detection methods into five main categories: color-based methods, shape-based methods, color- and shape-based methods, machine-learning-based methods, and LIDAR-based methods. The methods in each category are also classified into different subcategories for understanding and summarizing the mechanisms of different methods. For some reviewed methods that lack comparisons on public datasets, we reimplemented part of these methods for comparison. The experimental comparisons and analyses are presented on the reported performance and the performance of our reimplemented methods. Furthermore, future directions and recommendations of the TSD research are given to promote the development of the TSD.
•We proposed an efficient and effective data-mining procedure that models the travel patterns of transit riders using the transit smart card data.•Transit riders’ trip chains are identified based on ...the temporal and spatial characteristics of smart card transaction data.•A Density-based Spatial Clustering of Applications with Noise (DBSCAN) algorithm is used to detect each transit rider’s historical travel patterns.•The K-Means++ clustering algorithm and the rough-set theory are jointly applied to clustering and classifying the travel pattern regularities.
To mitigate the congestion caused by the ever increasing number of privately owned automobiles, public transit is highly promoted by transportation agencies worldwide. A better understanding of travel patterns and regularity at the “magnitude” level will enable transit authorities to evaluate the services they offer, adjust marketing strategies, retain loyal customers and improve overall transit performance. However, it is fairly challenging to identify travel patterns for individual transit riders in a large dataset. This paper proposes an efficient and effective data-mining procedure that models the travel patterns of transit riders in Beijing, China. Transit riders’ trip chains are identified based on the temporal and spatial characteristics of their smart card transaction data. The Density-based Spatial Clustering of Applications with Noise (DBSCAN) algorithm then analyzes the identified trip chains to detect transit riders’ historical travel patterns and the K-Means++ clustering algorithm and the rough-set theory are jointly applied to cluster and classify travel pattern regularities. The performance of the rough-set-based algorithm is compared with those of other prevailing classification algorithms. The results indicate that the proposed rough-set-based algorithm outperforms other commonly used data-mining algorithms in terms of accuracy and efficiency.
► Accessibility impacts of China’s high-speed rail are evaluated at macro level. ► Most cities have accessibility gain by HSR compared to their conventional rail access. ► Cities with dense ...population and high GDP level obtain prominent benefits. ► Cities covering longer travel distances have more accessibility benefits by airline compared to HSR.
The large-scale implementation of High-Speed Rail (HSR) network in China not only offers a new option for travelers’ mode choice, but also may influence, or even generate, the redistribution of demographic and economic activities. As has been observed over the past several years in other countries, the impact of HSR spans a wide range. However, few quantitative studies have been conducted to measure this impact. As a new attempt, this study uses accessibility analysis for quantifying the impact of China’s HSR network. Weighted average travel times and travel costs, contour measures, and potential accessibility are employed as indicators of accessibility at the macro or national level. Forty-nine major cities in the HSR network are used in the accessibility analysis. Accessibility quantification and spatial distribution analysis for the study cities are performed on a Geographical Information System (GIS) platform. Accessibilities associated with varying availabilities of HSR, conventional rail, and airline are estimated and compared. The selected indicators and computational methods are found effective in evaluating the accessibility impacts of HSR from different conceptualization strategies and perspectives. They also offer complementary information on accessibility capacity of the study cities created by the HSR network.
The increasing availability of big Automatic Identification Systems (AIS) sensor data offers great opportunities to track ship activities and mine spatial-temporal patterns of ship traffic worldwide. ...This research proposes a data integration approach to construct Global Shipping Networks (GSN) from massive historical ship AIS trajectories in a completely bottom-up way. First, a DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm is applied to temporally identify relevant stop locations, such as marine terminals and their associated events. Second, the semantic meanings of these locations are obtained by mapping them to real ports as identified by the World Port Index (WPI). Stop events are leveraged to develop travel sequences of any ship between stop locations at multiple scales. Last, a GSN is constructed by considering stop locations as nodes and journeys between nodes as links. This approach generates different levels of shipping networks from the terminal, port, and country levels. It is illustrated by a case study that extracts country, port, and terminal level Global Container Shipping Networks (GCSN) from AIS trajectories of more than 4000 container ships in 2015. The main features of these GCSNs and the limitations of this work are finally discussed.