scholarly journals Multi-Range Attentive Bicomponent Graph Convolutional Network for Traffic Forecasting

2020 ◽  
Vol 34 (04) ◽  
pp. 3529-3536 ◽  
Author(s):  
Weiqi Chen ◽  
Ling Chen ◽  
Yu Xie ◽  
Wei Cao ◽  
Yusong Gao ◽  
...  
Keyword(s):  
Road Network ◽  
Weighted Graph ◽  
Interaction Patterns ◽  
Convolutional Network ◽  
The Road ◽  
Traffic Forecasting ◽  
Convolutional Networks ◽  
Traffic Conditions ◽  
Aggregate Information ◽  
Deep Learning Model

Traffic forecasting is of great importance to transportation management and public safety, and very challenging due to the complicated spatial-temporal dependency and essential uncertainty brought about by the road network and traffic conditions. Latest studies mainly focus on modeling the spatial dependency by utilizing graph convolutional networks (GCNs) throughout a fixed weighted graph. However, edges, i.e., the correlations between pair-wise nodes, are much more complicated and interact with each other. In this paper, we propose the Multi-Range Attentive Bicomponent GCN (MRA-BGCN), a novel deep learning model for traffic forecasting. We first build the node-wise graph according to the road network distance and the edge-wise graph according to various edge interaction patterns. Then, we implement the interactions of both nodes and edges using bicomponent graph convolution. The multi-range attention mechanism is introduced to aggregate information in different neighborhood ranges and automatically learn the importance of different ranges. Extensive experiments on two real-world road network traffic datasets, METR-LA and PEMS-BAY, show that our MRA-BGCN achieves the state-of-the-art results.

scholarly journals A3T-GCN: Attention Temporal Graph Convolutional Network for Traffic Forecasting

2021 ◽  
Vol 10 (7) ◽  
pp. 485
Author(s):  
Jiandong Bai ◽  
Jiawei Zhu ◽  
Yujiao Song ◽  
Ling Zhao ◽  
Zhixiang Hou ◽  
...  
Keyword(s):  
Road Network ◽  
Temporal Dynamics ◽  
Technological Problem ◽  
Traffic Flows ◽  
Convolutional Network ◽  
Temporal Dimension ◽  
Time Points ◽  
The Road ◽  
Traffic Forecasting ◽  
Temporal Graph

Accurate real-time traffic forecasting is a core technological problem against the implementation of the intelligent transportation system. However, it remains challenging considering the complex spatial and temporal dependencies among traffic flows. In the spatial dimension, due to the connectivity of the road network, the traffic flows between linked roads are closely related. In the temporal dimension, although there exists a tendency among adjacent time points, the importance of distant time points is not necessarily less than that of recent ones, since traffic flows are also affected by external factors. In this study, an attention temporal graph convolutional network (A3T-GCN) was proposed to simultaneously capture global temporal dynamics and spatial correlations in traffic flows. The A3T-GCN model learns the short-term trend by using the gated recurrent units and learns the spatial dependence based on the topology of the road network through the graph convolutional network. Moreover, the attention mechanism was introduced to adjust the importance of different time points and assemble global temporal information to improve prediction accuracy. Experimental results in real-world datasets demonstrate the effectiveness and robustness of the proposed A3T-GCN. We observe the improvements in RMSE of 2.51–46.15% and 2.45–49.32% over baselines for the SZ-taxi and Los-loop, respectively. Meanwhile, the Accuracies are 0.95–89.91% and 0.26–10.37% higher than the baselines for the SZ-taxi and Los-loop, respectively.

scholarly journals The Development of a Concept Model of All-Weather Safe Winter Tires

2019 ◽  
pp. 3-11
Author(s):  
I.V. Balabin ◽  
O.I. Balabin ◽  
I.S. Chabunin
Keyword(s):  
Conceptual Model ◽  
Russian Federation ◽  
Road Safety ◽  
Road Network ◽  
Winter Temperature ◽  
The Road ◽  
Concept Model ◽  
Traffic Conditions ◽  
Proposed Model ◽  
The Russian Federation

The article presents issues related to improving safety and efficiency of operation of mobile machines in the constantly changing, winter temperature and traffic conditions. The authors develop a conceptual model of winter all-weather tires able to adapt to various road conditions such as when the road is covered with a layer of ice or compacted snow, or when the road is free from snow and ice. The use of such winter all weather tires will improve the road safety by contributing to increasing the life of tires and preserving the road network. The proposed model has no foreign analogues and is protected by a patent of the Russian Federation.

Low-carbon VRP for cold chain logistics considering real-time traffic conditions in the road network

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Qinyang Bai ◽  
Xaioqin Yin ◽  
Ming K. Lim ◽  
Chenchen Dong
Keyword(s):  
Road Network ◽  
Cold Chain ◽  
Time Varying ◽  
Low Carbon ◽  
Content Type ◽  
Distribution Costs ◽  
The Road ◽  
Traffic Conditions ◽  
Real Time Traffic ◽  
Distribution Cost

PurposeThis paper studies low-carbon vehicle routing problem (VRP) for cold chain logistics with the consideration of the complexity of the road network and the time-varying traffic conditions, and then a low-carbon cold chain logistics routing optimization model was proposed. The purpose of this paper is to minimize the carbon emission and distribution cost, which includes vehicle operation cost, product freshness cost, quality loss cost, penalty cost and transportation cost.Design/methodology/approachThis study proposed a mathematical optimization model, considering the distribution cost and carbon emission. The improved Nondominated Sorting Genetic Algorithm II algorithm was used to solve the model to obtain the Pareto frontal solution set.FindingsThe result of this study showed that this model can more accurately assess distribution costs and carbon emissions than those do not take real-time traffic conditions in the actual road network into account and provided guidance for cold chain logistics companies to choose a distribution strategy and for the government to develop a carbon tax.Research limitations/implicationsThere are some limitations in the proposed model. This study assumes that there are only one distribution and a single type of vehicle.Originality/valueExisting research on low-carbon VRP for cold chain logistics ignores the complexity of the road network and the time-varying traffic conditions, resulting in nonmeaningful planned distribution routes and furthermore low carbon cannot be discussed. This study takes the complexity of the road network and the time-varying traffic conditions into account, describing the distribution costs and carbon emissions accurately and providing the necessary prerequisites for achieving low carbon.

scholarly journals Feature Extraction and Representation of Urban Road Networks Based on Travel Routes

2020 ◽  
Vol 12 (22) ◽  
pp. 9621
Author(s):  
Shichen Huang ◽  
Chunfu Shao ◽  
Juan Li ◽  
Xiong Yang ◽  
Xiaoyu Zhang ◽  
...  
Keyword(s):  
Language Processing ◽  
Traffic Safety ◽  
Link Prediction ◽  
Road Network ◽  
Spatial Information ◽  
Representation Learning ◽  
Similarity Analysis ◽  
Safety Planning ◽  
Convolutional Network ◽  
The Road

Extraction of traffic features constitutes a key research direction in traffic safety planning. In previous traffic tasks, road network features are extracted manually. In contrast, Network Representation Learning aims to automatically learn low-dimensional node representations. Enlightened by feature learning in Natural Language Processing, representation learning of urban nodes is studied as a supervised task in this paper. Following this line of thinking, a deep learning framework, called StreetNode2VEC, is proposed for learning feature representations for nodes in the road network based on travel routes, and then model parameter calibration is performed. We explain the effectiveness of features from visualization, similarity analysis, and link prediction. In visualization, the features of nodes naturally present a clustered pattern, and different clusters correspond to different regions in the road network. Meanwhile, the features of nodes still retain their spatial information in similarity analysis. The proposed method StreetNode2VEC obtains a AUC score of 0.813 in link prediction, which is greater than that obtained from Graph Convolutional Network (GCN) and Node2vec. This suggests that the features of nodes can be used to effectively and credibly predict whether a link should be established between two nodes. Overall, our work provides a new way of representing road nodes in the road network, which have potential in the traffic safety planning field.

scholarly journals DC-STGCN: Dual-Channel Based Graph Convolutional Networks for Network Traffic Forecasting

Electronics ◽  
2021 ◽  
Vol 10 (9) ◽  
pp. 1014
Author(s):  
Chengsheng Pan ◽  
Jiang Zhu ◽  
Zhixiang Kong ◽  
Huaifeng Shi ◽  
Wensheng Yang
Keyword(s):  
Feature Extraction ◽  
Network Traffic ◽  
Traffic Flows ◽  
Convolutional Network ◽  
Temporal Characteristics ◽  
Traffic Forecasting ◽  
Convolutional Networks ◽  
Forecasting Models ◽  
Dual Channel

Network traffic forecasting is essential for efficient network management and planning. Accurate long-term forecasting models are also essential for proactive control of upcoming congestion events. Due to the complex spatial-temporal dependencies between traffic flows, traditional time series forecasting models are often unable to fully extract the spatial-temporal characteristics between the traffic flows. To address this issue, we propose a novel dual-channel based graph convolutional network (DC-STGCN) model. The proposed model consists of two temporal components that characterize the daily and weekly correlation of the network traffic. Each of these two components contains a spatial-temporal characteristics extraction module consisting of a dual-channel graph convolutional network (DCGCN) and a gated recurrent unit (GRU). The DCGCN further consists of an adjacency feature extraction module (AGCN) and a correlation feature extraction module (PGCN) to capture the connectivity between nodes and the proximity correlation, respectively. The GRU further extracts the temporal characteristics of the traffic. The experimental results based on real network data sets show that the prediction accuracy of the DC-STGCN model overperforms the existing baseline and is capable of making long-term predictions.

scholarly journals TRAFFIC FORECASTING ON THE CITY ROAD NETWORK TAKING INTO ACCOUNT THE CAPACITY LIMIT

2021 ◽  
Vol 2 (2) ◽  
pp. 27-33
Author(s):  
Denys Zhezherun
Keyword(s):  
Transport System ◽  
Road Network ◽  
Road Traffic ◽  
Traffic Intensity ◽  
The Road ◽  
Traffic Forecasting ◽  
Planning Decisions ◽  
Road Design ◽  
Traffic Volumes ◽  
On The Road

The purpose of the paper is to present a model of traffic forecasting on the road section based on a model of the transport system. Traffic forecasting is an integral part of the road design process, from investment to the feasibility study of working documentation. The definition of transportation and distribution of cars by sections is based on a set of interrelated factors. Full and reasonable consideration of these factors for complex road networks is possible only with the help of mathematical models and appropriate programs. The accuracy and consistency of the forecast determine the reliability of almost all the main characteristics of the projected object, from the direction of the route and the location of connection points with existing elements of the road network, ending with specific planning decisions for the road objects. Subject of research: a road traffic and a traffic intensity. Knowledge of forecast data on traffic intensity makes it possible to predict the possible mechanisms to solve the above problems. Methodology: analysis and research of methods used to predict traffic volumes. The method of extrapolation and the method of using approximating functions. Goal. The aim of the work is to compare the forecasting methods used to determine traffic on the road. It is also necessary to show the experience of traffic forecasting on the road network from a European country. Conclusion. All methods for predicting the volume and intensity of movement are short-lived, and if some achieve the desired predicted result, it is very vague and needs to be tested with complex and expensive research to determine and process the initial data. To achieve the desired results, it is necessary to apply new methods of forecasting modeling or improvement of already known ones, which would take into account the evolution of the entire transport system and its components. Determining the capacity of highways is necessary perform to identify areas with possible congestion, assessment economy and conditions of movement of vehicles, and also for a choice of methods and means to improve the traffic conditions of all road users.

Benefits from Constructing a Dam Across the Vistula in Siarzewo in Terms of Transport and Environmental Protection

2017 ◽  
Author(s):  
Jan Kempa ◽  
Jacek Chmielewski ◽  
Grzegorz Bebyn
Keyword(s):  
Environmental Impact ◽  
Road Network ◽  
Transport Model ◽  
Investment Project ◽  
Dam Construction ◽  
Transport Activity ◽  
The Road ◽  
Traffic Conditions ◽  
On The Road

This paper presents the results of analyses that concern the benefits from the planned construction of a dam across the Vistula in Siarzewo. The simulated transport model developed in the VISUM environment has been used to determine the forecast traffic intensity, the value of traffic volume indices, transport activity, travel times of drivers and passengers as well as the costs of environmental impact. The above-mentioned characteristics have enabled to determine savings both in terms of traffic costs and environmental impacts resulting from the dam construction. The paper indicates that the implementation of the investment project improves traffic conditions on the road network and reduces the transport environmental impact in Kujawsko-Pomorskie Province. Moreover, it has been found that the revealed effects concern in particular the first years after the launch of the project. The development of the road network diminishes the role of the analysed investment project significantly.

scholarly journals Considerations on the role of modernizing the road infrastructure in the prevention of road accidents

2019 ◽  
Vol 290 ◽  
pp. 06004
Author(s):  
Cristian Deac ◽  
Lucian Tarnu
Keyword(s):  
Road Safety ◽  
Road Network ◽  
Road Traffic ◽  
Road Accidents ◽  
Road Infrastructure ◽  
The Road ◽  
Traffic Conditions ◽  
Pedestrian Crossing

The realizing and improvement of road infrastructure, of modern road networks provides normal, safe and pleasant road traffic conditions and also help prevent road accidents. The road network, with its constructive characteristics, has to offer optimal conditions for the movement of vehicles, pedestrians and other categories of participants in the road traffic. Starting from the case study of a road sector with heavy road traffic, the current paper analyzes the increase in road safety in Romanian localities along European and national roads through the implementation of specific measures such as setting up sidewalks, installing New Jersey median barriers, expanding the road sectors with 2+1 lanes, replacing normal pedestrian crossings with elevated crossings or with pedestrian crossing with mid-road waiting areas etc.

scholarly journals Relation Extraction with Convolutional Network over Learnable Syntax-Transport Graph

2020 ◽  
Vol 34 (05) ◽  
pp. 8928-8935
Author(s):  
Kai Sun ◽  
Richong Zhang ◽  
Yongyi Mao ◽  
Samuel Mensah ◽  
Xudong Liu
Keyword(s):  
Relation Extraction ◽  
Weighted Graph ◽  
Relevant Information ◽  
Irrelevant Information ◽  
Classification Task ◽  
Convolutional Network ◽  
Structure Information ◽  
Convolutional Networks ◽  
Dependency Tree ◽  
Relation Classification

A large majority of approaches have been proposed to leverage the dependency tree in the relation classification task. Recent works have focused on pruning irrelevant information from the dependency tree. The state-of-the-art Attention Guided Graph Convolutional Networks (AGGCNs) transforms the dependency tree into a weighted-graph to distinguish the relevance of nodes and edges for relation classification. However, in their approach, the graph is fully connected, which destroys the structure information of the original dependency tree. How to effectively make use of relevant information while ignoring irrelevant information from the dependency trees remains a challenge in the relation classification task. In this work, we learn to transform the dependency tree into a weighted graph by considering the syntax dependencies of the connected nodes and persisting the structure of the original dependency tree. We refer to this graph as a syntax-transport graph. We further propose a learnable syntax-transport attention graph convolutional network (LST-AGCN) which operates on the syntax-transport graph directly to distill the final representation which is sufficient for classification. Experiments on Semeval-2010 Task 8 and Tacred show our approach outperforms previous methods.

scholarly journals A Calibrated Multiexit Neural Network for Detecting Urothelial Cancer Cells

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
L. Lilli ◽  
E. Giarnieri ◽  
S. Scardapane
Keyword(s):  
Ground Truth ◽  
Cellular Level ◽  
Diagnostic Tools ◽  
Small Scale ◽  
Convolutional Network ◽  
Convolutional Networks ◽  
Confidence Levels ◽  
Temperature Scaling ◽  
Major Shortcoming ◽  
Deep Learning Model

Deep convolutional networks have become a powerful tool for medical imaging diagnostic. In pathology, most efforts have been focused in the subfield of histology, while cytopathology (which studies diagnostic tools at the cellular level) remains underexplored. In this paper, we propose a novel deep learning model for cancer detection from urinary cytopathology screening images. We leverage recent ideas from the field of multioutput neural networks to provide a model that can efficiently train even on small-scale datasets, such as those typically found in real-world scenarios. Additionally, we argue that calibration (i.e., providing confidence levels that are aligned with the ground truth probability of an event) has been a major shortcoming of prior works, and we experiment a number of techniques to provide a well-calibrated model. We evaluate the proposed algorithm on a novel dataset, and we show that the combination of focal loss, multiple outputs, and temperature scaling provides a model that is significantly more accurate and calibrated than a baseline deep convolutional network.

Sign in / Sign up

Export Citation Format

Share Document