Deep Spatio-Temporal Modified-Inception with Dilated Convolution Networks for Citywide Crowd Flows Prediction

2019 ◽  
Vol 34 (08) ◽  
pp. 2052003
Author(s):  
Yan Kang ◽  
Bing Yang ◽  
Hao Li ◽  
Tie Chen ◽  
Yachuan Zhang
Keyword(s):  
Traffic Flow ◽  
Traffic Safety ◽  
Road Traffic ◽  
Network Models ◽  
Traffic Data ◽  
Traffic Flow Prediction ◽  
Dilated Convolution ◽  
Flow Prediction ◽  
Traffic Capacity ◽  
Spatio Temporal

Traffic flow prediction has great significance for improving road traffic capacity and traffic safety. However, traffic flow in a certain area is usually affected by some factors such as weather, holidays and neighboring areas. So, traffic situation is complicated and traffic flow prediction is difficult. How to use existing traffic data information to predict future traffic flow is the key to this problem. In this paper, we develop an accurate prediction model based on dilated convolution — ST-MINet (Deep Spatio-Temporal Modified-Inception with Dilated convolution Networks). We fully consider the complexity, nonlinearity and uncertainly of traffic network by summarizing various network models such as ResNet and Inception. So, we use the deep space-time residual network to ensure the convolution accuracy of the information’s position distribution on the basis of existing networks. Then, we add the cavity convolution to the model, which can effectively control the field of view of the convolution kernel. In the experimental part, we compare ten classical algorithms with our ST-MINet, it shows that our model has higher accuracy than others.

scholarly journals FR-MTL: Traffic Flow Prediction Using Fused Ridge Denoising and Multi-Task Learning

2020 ◽  
Vol 24 (7) ◽  
pp. 829-836
Author(s):  
Di Yang ◽  
◽  
Ningjia Qiu ◽  
Peng Wang ◽  
Huamin Yang
Keyword(s):  
Traffic Flow ◽  
Prediction Models ◽  
Road Traffic ◽  
Transportation Systems ◽  
Natural Order ◽  
Traffic Data ◽  
Traffic Flow Prediction ◽  
Task Learning ◽  
Flow Prediction ◽  
Highly Correlated

Traffic flow prediction is one of the fundamental components in Intelligent Transportation Systems (ITS). Many traffic flow prediction models have been developed, but with limitation of noise sensitivity, which will result in poor generalization. Fused Lasso, also known as total variation denoising, penalizes L1-norm on the model coefficients and pairwise differences between neighboring coefficients, has been widely used to analyze highly correlated features with a natural order, as is the case with traffic flow. It denoises data by encouraging both sparsity of coefficients and their differences, and estimates the coefficients of highly correlated variables to be equal to each other. However, for traffic data, the same coefficients will lead to overexpression of features, and losing the trend of time series of traffic flow. In this work, we propose a Fused Ridge multi-task learning (FR-MTL) model for multi-road traffic flow prediction. It introduces Fused Ridge for traffic data denoising, imposes penalty on L2-norm of the coefficients and their differences. The penalty of L2-norm proportionally shrinks coefficients, and generates smooth coefficient vectors with non-sparsity. It has both capability of trend preservation and denoising. In addition, we jointly consider multi-task learning (MTL) for training shared spatiotemporal information among traffic roads. The experiments on real traffic data show the advantages of the proposed model over other four regularized baseline models, and on traffic data with Gaussian noise and missing data, the FR-MTL model demonstrates potential and promising capability with satisfying accuracy and effectiveness.

MS-Net: Multi-Source Spatio-Temporal Network for Traffic Flow Prediction

2021 ◽  
pp. 1-14
Author(s):  
Shen Fang ◽  
Veronique Prinet ◽  
Jianlong Chang ◽  
Michael Werman ◽  
Chunxia Zhang ◽  
...  
Keyword(s):  
Traffic Flow ◽  
Temporal Network ◽  
Traffic Flow Prediction ◽  
Flow Prediction ◽  
Spatio Temporal

scholarly journals Improved Spatio-Temporal Residual Networks for Bus Traffic Flow Prediction

2019 ◽  
Vol 9 (4) ◽  
pp. 615 ◽  
Author(s):  
Panbiao Liu ◽  
Yong Zhang ◽  
Dehui Kong ◽  
Baocai Yin
Keyword(s):  
Traffic Flow ◽  
Traffic Congestion ◽  
Road Traffic ◽  
Transportation Network ◽  
Temporal Correlation ◽  
Traffic Flow Prediction ◽  
Temporal Correlations ◽  
Building Models ◽  
Spatio Temporal ◽  
Specific Scenario

Buses, as the most commonly used public transport, play a significant role in cities. Predicting bus traffic flow cannot only build an efficient and safe transportation network but also improve the current situation of road traffic congestion, which is very important for urban development. However, bus traffic flow has complex spatial and temporal correlations, as well as specific scenario patterns compared with other modes of transportation, which is one of the biggest challenges when building models to predict bus traffic flow. In this study, we explore bus traffic flow and its specific scenario patterns, then we build improved spatio-temporal residual networks to predict bus traffic flow, which uses fully connected neural networks to capture the bus scenario patterns and improved residual networks to capture the bus traffic flow spatio-temporal correlation. Experiments on Beijing transportation smart card data demonstrate that our method achieves better results than the four baseline methods.

scholarly journals Incorporation of Duffing Oscillator and Wigner-Ville Distribution in Traffic Flow Prediction

2017 ◽  
Vol 29 (1) ◽  
pp. 13-22 ◽  
Author(s):  
Anamarija L. Mrgole ◽  
Drago Sever
Keyword(s):  
Traffic Flow ◽  
Road Traffic ◽  
Duffing Oscillator ◽  
Traffic Load ◽  
The Road ◽  
Traffic Flow Prediction ◽  
Pattern Recognition Methods ◽  
Flow Prediction ◽  
Time Period ◽  
Specific Section

The main purpose of this study was to investigate the use of various chaotic pattern recognition methods for traffic flow prediction. Traffic flow is a variable, dynamic and complex system, which is non-linear and unpredictable. The emergence of traffic flow congestion in road traffic is estimated when the traffic load on a specific section of the road in a specific time period is close to exceeding the capacity of the road infrastructure. Under certain conditions, it can be seen in concentrating chaotic traffic flow patterns. The literature review of traffic flow theory and its connection with chaotic features implies that this kind of method has great theoretical and practical value. Researched methods of identifying chaos in traffic flow have shown certain restrictions in their techniques but have suggested guidelines for improving the identification of chaotic parameters in traffic flow. The proposed new method of forecasting congestion in traffic flow uses Wigner-Ville frequency distribution. This method enables the display of a chaotic attractor without the use of reconstruction phase space.

Sign in / Sign up

Export Citation Format

Share Document