CN111862592B - Traffic flow prediction method based on RGCN - Google Patents
Traffic flow prediction method based on RGCN Download PDFInfo
- Publication number
- CN111862592B CN111862592B CN202010461193.6A CN202010461193A CN111862592B CN 111862592 B CN111862592 B CN 111862592B CN 202010461193 A CN202010461193 A CN 202010461193A CN 111862592 B CN111862592 B CN 111862592B
- Authority
- CN
- China
- Prior art keywords
- traffic flow
- road
- matrix
- road traffic
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 31
- 239000011159 matrix material Substances 0.000 claims abstract description 55
- 238000009826 distribution Methods 0.000 claims abstract description 32
- 238000007781 pre-processing Methods 0.000 claims abstract description 10
- 238000005070 sampling Methods 0.000 claims abstract description 10
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 8
- 238000012360 testing method Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 claims description 2
- 230000004913 activation Effects 0.000 claims description 2
- 230000001186 cumulative effect Effects 0.000 claims description 2
- 238000005457 optimization Methods 0.000 claims description 2
- 239000000126 substance Substances 0.000 claims description 2
- 230000006870 function Effects 0.000 description 9
- 238000012549 training Methods 0.000 description 5
- 238000002474 experimental method Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000006698 induction Effects 0.000 description 2
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 description 1
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Data Mining & Analysis (AREA)
- Analytical Chemistry (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Chemical & Material Sciences (AREA)
- Marketing (AREA)
- Biophysics (AREA)
- General Business, Economics & Management (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Tourism & Hospitality (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Primary Health Care (AREA)
- Traffic Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
An RGCN-based traffic flow prediction method comprises the following steps: step 1) selecting similar roads for each road based on the time sequence similarity criterion, and constructing a road traffic network; step 2) acquiring road traffic flow data, preprocessing the data and constructing a road traffic flow state matrix data set; step 3) taking the road traffic network and the road traffic state matrix as the input of a graph convolution layer based on Gaussian distribution, and further extracting the node characteristics of the road traffic network; step 4), taking the characteristics of the graph convolution layer based on Gaussian distribution after sampling as regression prediction layer input, and calculating a prediction result corresponding to the current input; defining a model loss function, and continuously optimizing model parameters by using a back propagation algorithm according to the value of the loss function; and acquiring real-time traffic flow data as model input to realize the prediction of real-time road traffic flow. The invention improves the robustness of the graph convolution network, thereby improving the accuracy of traffic flow prediction.
Description
Technical Field
The invention relates to a traffic flow prediction method based on RGCN, belonging to the field of traffic prediction.
Background
The improvement of urban economic level and the enlargement of urban scale lead people to continuously improve living standard, so that the number of vehicles in the city is also increased sharply, the increase of the vehicles brings many problems, and traffic jam can be regarded as a first problem. The traffic state prediction plays a crucial role in reasonably distributing urban road resources and relieving traffic congestion.
The current mainstream traffic flow prediction method based on data driving mainly comprises the following steps: the method comprises the steps of carrying out differential integration on a moving average autoregressive model, a support vector machine, a convolutional neural network, a long-short term memory neural network and the like, wherein some methods are used for carrying out prediction based on probability statistics, some methods are used for carrying out prediction based on time characteristics, some methods are used for carrying out prediction based on shallow layer neural network extraction characteristics and the like, and although the methods all obtain good prediction results, the methods do not learn the traffic data of a graph structure. The graph convolution network can effectively learn the data of the graph structure, but cannot eliminate the noise in the original data and has instability.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention aims to provide a traffic flow prediction method based on RGCN (robust graph convolution network). According to the method, the Gaussian distribution is used in the map convolution layer to absorb noise in traffic flow data, the robustness of the map convolution network is improved, and therefore the accuracy of traffic flow prediction is improved.
The technical scheme adopted by the invention for solving the technical problems is as follows:
an RGCN-based traffic flow prediction method comprises the following steps:
step 1) constructing a road traffic network based on time sequence similarity: selecting roads similar to the time sequence similarity for each road based on the time sequence similarity criterion, and constructing a road traffic network;
step 2) preprocessing the road traffic flow data and constructing a road traffic flow state matrix data set: acquiring road traffic flow data, preprocessing the data, and constructing a road traffic flow state matrix data set;
and 3) extracting node characteristics by using a graph convolution layer based on Gaussian distribution based on a road traffic network and a road traffic state matrix: taking a road traffic network and a road traffic state matrix as the input of a graph convolution layer based on Gaussian distribution, and further extracting the node characteristics of the road traffic network;
and 4) carrying out traffic flow prediction based on the map convolution layer with Gaussian distribution: taking the characteristics of the graph convolution layer after sampling based on Gaussian distribution as regression prediction layer input, and calculating a prediction result corresponding to the current input; defining a model loss function, and continuously optimizing model parameters by using a back propagation algorithm according to the value of the loss function; and acquiring real-time traffic flow data as model input to realize the prediction of real-time road traffic flow.
The invention provides a traffic flow prediction method based on RGCN, which adopts Gaussian distribution as hidden representation of nodes in each convolution layer, can absorb random noise in traffic flow data, improves the robustness of a graph convolution network, and further improves the accuracy of traffic flow prediction.
The invention has the beneficial effects that: the invention forms a road traffic flow prediction model through the learning training of the RGCN model to the traffic flow data. According to the invention, the road traffic network is constructed according to the similarity of the time sequences, so that the problem that the geographical position of the traffic network is difficult to obtain is overcome, and the effective extraction of the time-space characteristics of the traffic network is realized; the graph convolution network based on Gaussian distribution is used for traffic flow prediction, noise of traffic flow data is absorbed, and robustness and accuracy of road traffic flow prediction are improved.
The short-time traffic flow prediction is used as an important component of a traffic flow induction system, and the performance of the traffic flow induction system can be effectively improved to a certain extent. In addition, the invention can also be used as an effective auxiliary tool for travelers to travel.
Drawings
Fig. 1 is a block diagram of RGCN-based traffic flow prediction.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
Referring to fig. 1, a traffic flow prediction method based on RGCN includes the following steps:
step 1) constructing a road traffic network based on time sequence similarity, wherein the process is as follows:
for a plurality of road traffic sections, acquiring time sequence data of a certain day of the road traffic sections, and constructing a road traffic network based on similarity of the time sequence, for example: given two pieces of time-series data X ═ X (X)1,x2,…,xn)，Y＝(y1,y2,…,yn) Then constructing a distance matrix Mn×nWherein M isi,j＝|xi-yjL. Based on distance matrix Mn×nConstructing a cumulative distance matrix Dn×nWherein D isi,j＝Mi,j+min(Di,j-1,Di-1,j,Di-1,j-1)，Dn,nThe distance is the final distance, and for the road i, the front d roads with the minimum final distance are selected as { j, k, … }, namely the road i is most similar to the road S;
constructing a road traffic network graph G (V, E, A) based on time series similarity, wherein V represents a node, E represents a connecting edge, and A represents an adjacency matrix of a traffic network, wherein Aij1 indicates that road i is similar to road j, otherwise, aij＝0；
Step 2) preprocessing the road traffic flow data and constructing a road traffic flow state matrix, wherein the process is as follows:
2.1: preprocessing road traffic flow data
Preprocessing traffic flow data of multiple roads for multiple days, normalizing the data by using maximum-minimum standardization, and calculating an expression as follows:
wherein x isrealIs the raw flow data of the road, xminIs the minimum value, x, in the road raw flow datamaxThe data is the maximum value in the original road flow data, and x is the road flow data after pretreatment;
2.2: constructing a road traffic flow state matrix
Constructing a road traffic flow state matrix according to the preprocessed road traffic flow data, wherein the road traffic flow state matrix is in the following form:
wherein, the state matrix row vector represents the traffic states of different roads at the same time, the column vector represents the time states of the traffic states of the roads at the same lane and different times, M represents the number of historical traffic state data, N represents the number of roads in the input matrix, and x represents the number of the roads in the input matrixitIndicates the ith laneTraffic state on the road at time t;
step 3) extracting node characteristics by using graph convolution layers based on Gaussian distribution based on road traffic network and road traffic state matrix
The overall framework of RGCN-based road traffic flow prediction is shown in fig. 1.
Defining a node v using a Gaussian distribution of hidden features obtained in the original graph convolution layeriThe hidden features at layer i are:
wherein, mui (l)Denotes the mean vector, diag (σ)i (l)) Representing diagonal variance matrix, σ instead of σ2Represents the variance;
the mean and variance of all nodes are defined as:
M(l)＝[μ1 (l),μ2 (l),…,μN (l)] (4)
Σ(l)＝[σ1 (l),σ2 (l),…,σN (l)] (5)
since the road traffic flow state input matrix X is a feature vector rather than a gaussian distribution, a first-level mean matrix M is defined(1)The variance matrix Σ(1)In the form:
wherein, Wμ，WσThe weights of the mean matrix and the variance matrix, respectively, f and p represent the ELU function and the ReLU function, respectively, as follows:
after passing through the first fully connected layer, the input vector becomes a hidden feature and then gaussian distribution is used, and the graph convolution layer based on gaussian distribution is defined as follows:
wherein the content of the first and second substances,
is a matrix of the auto-correlation,
INis an elementary matrix of the data stream,
is a matrix of degrees and is,
theta is attention weight, theta(l)＝exp(-γΣ(l)) And gamma is a hyperparameter. An indication of a matrix dot product;
step 4) traffic flow prediction is carried out based on the graph convolution layer with Gaussian distribution, and the process is as follows:
4.1: calculating a predicted result
Taking the hidden features as Gaussian distribution to predict the traffic flow, a sampling process is needed to be carried out, and the last layer of map convolution layer based on Gaussian distribution is sampled, wherein the sampling formula is as follows:
wherein L represents the last layer of the convolutional layer. Sampling characteristic Z ═ Z1,z2,…,zN]As input to the all-connected layer, the state at the next time based on the historical traffic flow data input is predicted, and the all-connected layer expression is as follows:
Ypre＝f(Wf·Z+bf) (13)
wherein, WfIs a full connection layer weight matrix, bfIs a bias matrix, where sigmoid function is used as the activation function f, in order to restore the normalized values;
4.2: defining model loss function, optimizing model parameters and realizing real-time traffic flow prediction
To ensure that the learned features obey a gaussian distribution, we apply regularization to only the hidden layer features obtained by the first convolutional layer to constrain:
wherein KL (| ·) is the KL distance between the two distributions;
to prevent model overfitting, L2 regularization was applied to the first layer convolutional layer:
defining the traffic flow state predicted by the model as YpreThe actual traffic flow state is YtrueThe final model loss function L is then of the form:
wherein n represents the number of samples input by the model, and lambda and beta are parameters;
and comparing the traffic flow state predicted by the model with the actual traffic flow state, calculating a loss function L of the model, and then realizing continuous optimization of model parameters by using a back propagation algorithm. And finally, inputting the test set data as a model to realize the prediction of the real-time road traffic flow.
In the invention, gradient calculation and parameter updating in the back propagation algorithm are realized by an Adam optimizer.
Example 2: data in actual experiments
(1) Selecting experimental data
The original traffic flow data set comprises traffic flow data of 323 detectors for 1 month, the traffic flow data in the data set is the traffic flow data of a Seattle expressway part detector, and the sampling interval T is 5 min.
And taking the road traffic flow data of the first three weeks of 323 detectors as a training data set to train model parameters. And (3) taking the road traffic flow data of the last week of the 323 detectors as an experimental data set, and carrying out algorithm verification. The sliding window is set to 10 and the prediction steps are set to 1, respectively.
(2) Parameter determination
The experimental results of the invention are realized based on the tensoflow environment, and the whole experimental parameters are set as follows: the number of layers of the map convolution layer based on the gaussian distribution is 2, the first layer hidden unit is set to 16, the second layer hidden unit is set to 10, the learning rate is 0.001, the training number is 100, the batch size is 64, γ is 1, λ is 0.00005, and β is 0.00005.
(3) Results of the experiment
The invention aims at predicting the short-term traffic flow of multiple road sections, carries out model training through a training set, and carries out the test of the predictive performance of the model through a testing set.
In the experiment, the absolute mean square error (MAE) and the Root Mean Square Error (RMSE) are selected as indexes of the road traffic flow prediction precision, and the calculation formulas are respectively as follows:
wherein, yiIn order to actually observe the flow rate,
the predicted flow output for the model.
Statistical analysis of all experimental road section flow prediction results is shown in table 1:
| model (model) | RMSE | MAE | MAPE(％) |
| GCN | 4.911 | 3.367 | 8.380 |
| RGCN | 4.775 | 3.290 | 8.019 |
Table 1.
Claims (5)
1. An RGCN-based traffic flow prediction method, characterized in that the method comprises the following steps:
step 1) constructing a road traffic network based on time sequence similarity: selecting roads similar to the time sequence similarity for each road based on the time sequence similarity criterion, and constructing a road traffic network;
step 2) preprocessing the road traffic flow data and constructing a road traffic flow state matrix data set: acquiring road traffic flow data, preprocessing the data, and constructing a road traffic flow state matrix data set;
and 3) extracting node characteristics by using a graph convolution layer based on Gaussian distribution based on a road traffic network and a road traffic state matrix: taking a road traffic network and a road traffic state matrix as the input of a graph convolution layer based on Gaussian distribution, and further extracting the node characteristics of the road traffic network;
and 4) carrying out traffic flow prediction based on the map convolution layer with Gaussian distribution: taking the characteristics of the graph convolution layer after sampling based on Gaussian distribution as regression prediction layer input, and calculating a prediction result corresponding to the current input; defining a model loss function, and continuously optimizing model parameters by using a back propagation algorithm according to the value of the loss function; acquiring real-time traffic flow data as model input to realize prediction of real-time road traffic flow;
the process of the step 1) is as follows:
for a plurality of road traffic sections, acquiring time sequence data of a certain day of the road traffic sections, and constructing a road traffic network based on similarity of the time sequence, for example: given two pieces of time-series data X ═ X (X)1,x2,…,xn)，Y＝(y1,y2,…,yn) Then constructing a distance matrix Mn×nWherein M isi,j＝|xi-yjBased on a distance matrix Mn×nConstructing a cumulative distance matrix Dn×nWherein D isi,j＝Mi,j+min(Di,j-1,Di-1,j,Di-1,j-1)，Dn,nThe distance is the final distance, and for the road i, the front d roads with the minimum final distance are selected as { j, k, … }, namely the road i is most similar to the road S;
constructing a road traffic network graph G (V, E, A) based on time series similarity, wherein V represents a node, E represents a connecting edge, and A represents an adjacency matrix of a traffic network, wherein Aij1 indicates that road i is similar to road j, otherwise, aij＝0。
2. An RGCN-based traffic flow prediction method according to claim 1, wherein the process of step 2) is as follows:
2.1: preprocessing road traffic flow data
Preprocessing traffic flow data of multiple roads for multiple days, normalizing the data by using maximum-minimum standardization, and calculating an expression as follows:
wherein x isrealIs the raw flow data of the road, xminIs the minimum value, x, in the road raw flow datamaxThe data is the maximum value in the original road flow data, and x is the road flow data after pretreatment;
2.2: constructing a road traffic flow state matrix
Constructing a road traffic flow state matrix according to the preprocessed road traffic flow data, wherein the road traffic flow state matrix is in the following form:
wherein, the state matrix row vector represents the traffic states of different roads at the same time, the column vector represents the time states of the traffic states of the roads at the same lane and different times, and M represents the historical trafficNumber of state data, N represents the number of roads in the input matrix, then xitIndicating the traffic state on the ith road at time t.
3. An RGCN-based traffic flow prediction method according to claim 1, wherein the process of the step 3) is as follows:
defining a node v using a Gaussian distribution of hidden features obtained in the original graph convolution layeriThe hidden features at layer i are:
wherein, mui (l)Denotes the mean vector, diag (σ)i (l)) Representing diagonal variance matrix, σ instead of σ2Represents the variance;
the mean and variance of all nodes are defined as:
M(l)＝[μ1 (l),μ2 (l),…,μN (l)] (4)
Σ(l)＝[σ1 (l),σ2 (l),…,σN (l)] (5)
since the road traffic flow state input matrix X is a feature vector rather than a gaussian distribution, a first-level mean matrix M is defined(1)The variance matrix Σ(1)In the form:
wherein, Wμ，WσThe weights of the mean matrix and the variance matrix, respectively, f and p represent the ELU function and the ReLU function, respectively, as follows:
after passing through the first fully connected layer, the input vector becomes a hidden feature and then gaussian distribution is used, and the graph convolution layer based on gaussian distribution is defined as follows:
wherein the content of the first and second substances,
is a matrix of the auto-correlation,
INis an elementary matrix of the data stream,
is a matrix of degrees and is,
theta is attention weight, theta(l)＝exp(-γΣ(l)) And γ is a hyperparameter, which indicates the matrix dot product.
4. An RGCN-based traffic flow prediction method according to claim 1, wherein the process of the step 4) is as follows:
4.1: calculating a predicted result
Taking the hidden features as Gaussian distribution to predict the traffic flow, a sampling process is needed to be carried out, and the last layer of map convolution layer based on Gaussian distribution is sampled, wherein the sampling formula is as follows:
wherein L represents the last layer of the convolutional layer, and the sampling characteristic Z is [ Z ]1,z2,…,zN]As input to the all-connected layer, the state at the next time based on the historical traffic flow data input is predicted, and the all-connected layer expression is as follows:
Ypre＝f(Wf·Z+bf) (13)
wherein, WfIs a full connection layer weight matrix, bfIs a bias matrix, where sigmoid function is used as the activation function f, in order to restore the normalized values;
4.2: defining model loss function, optimizing model parameters and realizing real-time traffic flow prediction
To ensure that the learned features obey a gaussian distribution, we apply regularization to only the hidden layer features obtained by the first convolutional layer to constrain:
wherein KL (| ·) is the KL distance between the two distributions;
to prevent model overfitting, L2 regularization was applied to the first layer convolutional layer:
defining the traffic flow state predicted by the model as YpreThe actual traffic flow state is YtrueThe final model loss function L is then of the form:
wherein n represents the number of samples input by the model, and lambda and beta are parameters;
comparing the traffic flow state predicted by the model with the actual traffic flow state, calculating a loss function L of the model, then realizing continuous optimization of model parameters by using a back propagation algorithm, and finally, inputting the test set data as the model to realize the prediction of the real-time road traffic flow.
5. An RGCN-based traffic flow prediction method according to claim 4, characterized in that the gradient calculation and parameter update in the back propagation algorithm are implemented by Adam optimizer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010461193.6A CN111862592B (en) | 2020-05-27 | 2020-05-27 | Traffic flow prediction method based on RGCN |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010461193.6A CN111862592B (en) | 2020-05-27 | 2020-05-27 | Traffic flow prediction method based on RGCN |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111862592A CN111862592A (en) | 2020-10-30 |
| CN111862592B true CN111862592B (en) | 2021-12-17 |
Family
ID=72985881
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010461193.6A Active CN111862592B (en) | 2020-05-27 | 2020-05-27 | Traffic flow prediction method based on RGCN |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111862592B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112489420B (en) * | 2020-11-17 | 2022-06-21 | 中国科学院深圳先进技术研究院 | Road traffic state prediction method, system, terminal and storage medium |
| CN112562339B (en) * | 2020-12-09 | 2022-01-04 | 福州大学 | Robust traffic flow prediction method based on multitask graph convolutional network |
| CN112766551B (en) * | 2021-01-08 | 2022-05-17 | 鹏城实验室 | Traffic prediction method, intelligent terminal and computer readable storage medium |
| CN113379156A (en) * | 2021-06-30 | 2021-09-10 | 南方科技大学 | Speed prediction method, device, equipment and storage medium |
| CN115759484A (en) * | 2023-01-06 | 2023-03-07 | 南京隼眼电子科技有限公司 | Traffic flow prediction method, electronic device and storage medium |
Family Cites Families (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4353862B2 (en) * | 2004-06-30 | 2009-10-28 | パナソニック株式会社 | Traffic flow data prediction apparatus and traffic flow data prediction method |
| JP4175312B2 (en) * | 2004-09-17 | 2008-11-05 | 株式会社日立製作所 | Traffic information prediction device |
| US8266389B2 (en) * | 2009-04-29 | 2012-09-11 | Advanced Micro Devices, Inc. | Hierarchical memory arbitration technique for disparate sources |
| CN101739819A (en) * | 2009-11-19 | 2010-06-16 | 北京世纪高通科技有限公司 | Method and device for predicting traffic flow |
| CN101853573A (en) * | 2010-05-19 | 2010-10-06 | 北京科技大学 | City expressway short-time traffic information predicting system and method |
| US9390622B2 (en) * | 2013-04-16 | 2016-07-12 | International Business Machines Corporation | Performing-time-series based predictions with projection thresholds using secondary time-series-based information stream |
| CN104821082B (en) * | 2015-04-29 | 2017-01-18 | 电子科技大学 | Short-time traffic flow prediction method based on integrated evaluation |
| US20200074301A1 (en) * | 2018-09-04 | 2020-03-05 | Beijing Jingdong Shangke Information Technology Co., Ltd. | End-to-end structure-aware convolutional networks for knowledge base completion |
| CN109887282B (en) * | 2019-03-05 | 2022-01-21 | 中南大学 | Road network traffic flow prediction method based on hierarchical timing diagram convolutional network |
| CN110070715A (en) * | 2019-04-29 | 2019-07-30 | 浙江工业大学 | A kind of road traffic flow prediction method based on Conv1D-NLSTMs neural network structure |
| CN110264709B (en) * | 2019-05-06 | 2021-02-12 | 北京交通大学 | Method for predicting traffic flow of road based on graph convolution network |
| CN110189275A (en) * | 2019-05-29 | 2019-08-30 | 清华大学 | The picture scroll product method and system of robustness enhancing |
| CN110428424B (en) * | 2019-07-02 | 2023-04-07 | 中国航空工业集团公司雷华电子技术研究所 | Radar echo image high-voltage line segmentation method based on deep learning |
| CN110516833A (en) * | 2019-07-03 | 2019-11-29 | 浙江工业大学 | A method of the Bi-LSTM based on feature extraction predicts road traffic state |
| CN110929962A (en) * | 2019-12-13 | 2020-03-27 | 中国科学院深圳先进技术研究院 | Traffic flow prediction method and device based on deep learning |
| CN111126704B (en) * | 2019-12-25 | 2022-07-29 | 河海大学常州校区 | Multi-region precipitation prediction model construction method based on multi-graph convolution and memory network |
-
2020
- 2020-05-27 CN CN202010461193.6A patent/CN111862592B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN111862592A (en) | 2020-10-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111862592B (en) | Traffic flow prediction method based on RGCN | |
| CN110570651B (en) | Road network traffic situation prediction method and system based on deep learning | |
| CN112418547B (en) | Bus stop passenger flow prediction method based on GCN-LSTM combination model | |
| CN111210633B (en) | Short-term traffic flow prediction method based on deep learning | |
| CN109492830B (en) | Mobile pollution source emission concentration prediction method based on time-space deep learning | |
| CN108510741B (en) | Conv1D-LSTM neural network structure-based traffic flow prediction method | |
| CN109285346B (en) | Urban road network traffic state prediction method based on key road sections | |
| CN110070713B (en) | Traffic flow prediction method based on bidirectional nested LSTM neural network | |
| CN109492822B (en) | Air pollutant concentration time-space domain correlation prediction method | |
| CN111563706A (en) | Multivariable logistics freight volume prediction method based on LSTM network | |
| CN114330671A (en) | Traffic flow prediction method based on Transformer space-time diagram convolution network | |
| CN108009674A (en) | Air PM2.5 concentration prediction methods based on CNN and LSTM fused neural networks | |
| CN111785014B (en) | Road network traffic data restoration method based on DTW-RGCN | |
| CN111613054B (en) | Multi-step traffic speed prediction method cooperatively considering space-time correlation and contribution difference | |
| CN111709549A (en) | Navigation reminding method for short-time traffic flow prediction based on SVD-PSO-LSTM | |
| CN110942637B (en) | SCATS system road traffic flow prediction method based on airspace map convolutional neural network | |
| CN106781489A (en) | A kind of road network trend prediction method based on recurrent neural network | |
| CN116721537A (en) | Urban short-time traffic flow prediction method based on GCN-IPSO-LSTM combination model | |
| CN111695722A (en) | Method for predicting short-term passenger flow in holidays of urban rail transit station | |
| CN113205698A (en) | Navigation reminding method based on IGWO-LSTM short-time traffic flow prediction | |
| CN111784017A (en) | Road condition factor regression analysis-based road traffic accident quantity prediction method | |
| CN110889092A (en) | Short-time large-scale activity peripheral track station passenger flow volume prediction method based on track transaction data | |
| CN116311921A (en) | Traffic speed prediction method based on multi-spatial scale space-time converter | |
| CN112766603A (en) | Traffic flow prediction method, system, computer device and storage medium | |
| Zhang et al. | Surface and high-altitude combined rainfall forecasting using convolutional neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |