A traffic state prediction method for urban road network based on lstm-cnn
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A technology of traffic status and prediction method, applied in the field of intelligent transportation, can solve problems such as only considering the time dimension, achieve good prediction effect and reduce the effect of redundant information input
Active Publication Date: 2021-10-01
SUN YAT SEN UNIV
View PDF0 Cites 0 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
[0007] In order to solve the shortcomings of the prior art that cannot use the nonlinear relationship of traffic state evolution and only consider the characteristics of the time dimension, the present invention provides a traffic state prediction method for urban road networks based on LSTM-CNN
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0069] like figure 1 , figure 2 , image 3 As shown, a LSTM-CNN-based urban road network traffic state prediction method includes the following steps:
[0070] A LSTM-CNN-based traffic state prediction method for urban road network, the specific steps are as follows:
[0071] Step S1: Estimate and calculate the historical vehicle speed of the road network;
[0072] Step S2: Construct an adjacency matrix based on the obtained historical vehicle speed information;
[0073] Step S3: constructing the LSTM-CNN neural network;
[0074] Step S4: using the adjacency matrix as an input to train the LSTM-CNN neural network, and using the gradient descent method to update and optimize the parameters of the LSTM-CNN neural network;
[0075] Step S5: Step S5 is performed iteratively until the LSTM-CNN neural network converges;
[0076] Step S6: Input the adjacency matrix of real-time vehicle speed information to the LSTM-CNN neural network to predict the road network traffic status....
Embodiment 2
[0122] Such as figure 1 , figure 2 , image 3 , Figure 4 As shown, in this embodiment, the step size of the prior data and the prediction result are both set to 5 minutes. In order to test the reliability of the prediction effect of the model under different prior conditions, this setting has prior information of four lengths: 15min, 30min, 45min and 60min, and the corresponding step size is H={3,6,9,12}, The prediction lengths are 5min, 15min and 30min, and the corresponding step lengths are P={1,3,6}.
[0123] This embodiment is built based on the Tensorflow deep learning framework. After screening, the model parameters are determined as: in the convolutional network, use 6 convolution kernels with a size of (3,3), and use an average pooling layer with a size of (2,2 ), the number of neurons in the fully connected layer is 400, in the long short-term memory neural network, a layer of LSTM network is used, the number of nodes is 256, the final output is through a layer ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
PUM
Login to view more
Abstract
The present invention is oriented to the prediction of the traffic state of the urban road network, and will be described with the vehicle speed of the road section as the prediction object. First, the road network adjacency matrix is filled with the speed of the road section to represent the traffic state of the road network in a certain period of time, and the spatial characteristics of the traffic state are learned through the convolutional neural network, and then the long-short-term memory neural network is used to receive the spatial characteristics of different periods of time, and the characteristics of the time dimension are introduced. Learning, and finally combine the two kinds of feature information to predict the average travel speed of each section of the road network. The present invention proposes LSTM-CNN based on the road network adjacency matrix, which can better grasp the time-space nonlinear relationship of the road network traffic state compared with the traditional statistical method. Compared with the current deep learning method, the road network adjacency matrix based The traffic state input of the road network, while retaining the spatial characteristics of the road network traffic state, reduces the input of redundant information, reduces, passes, and effectively learns the characteristics of the road network traffic state, which has a better prediction effect.
Description
technical field [0001] The present invention relates to the field of intelligent transportation, and more specifically, relates to an LSTM-CNN-based urban road network traffic state prediction method. Background technique [0002] In recent years, urban traffic congestion has occurred frequently and has become common in economically developed cities. In order to improve road operating conditions and improve residents' living standards, the development of urban intelligent transportation systems (Intelligent Transportation System, referred to as ITS) has become an urgent need for urban development. With the continuous improvement of information, sensing, communication, and computer technologies, as a product of the comprehensive application of these technologies, ITS is in a stage of rapid development, playing an increasingly important role in allocating urban road resources and improving road network traffic efficiency character of. Accurate prediction of the traffic status...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
Application Information
Patent Timeline
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to view more 
Login to view more