CN113256985A - Traffic congestion prediction method and device and electronic equipment - Google Patents
Traffic congestion prediction method and device and electronic equipment Download PDFInfo
- Publication number
- CN113256985A CN113256985A CN202110716417.8A CN202110716417A CN113256985A CN 113256985 A CN113256985 A CN 113256985A CN 202110716417 A CN202110716417 A CN 202110716417A CN 113256985 A CN113256985 A CN 113256985A
- Authority
- CN
- China
- Prior art keywords
- data
- road
- lane
- period
- processing
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 55
- 238000012545 processing Methods 0.000 claims abstract description 72
- 239000011159 matrix material Substances 0.000 claims description 66
- 230000015654 memory Effects 0.000 claims description 33
- 230000009467 reduction Effects 0.000 claims description 26
- 239000013598 vector Substances 0.000 claims description 23
- 230000004927 fusion Effects 0.000 claims description 20
- 238000012549 training Methods 0.000 claims description 15
- 238000010801 machine learning Methods 0.000 claims description 10
- 238000013528 artificial neural network Methods 0.000 claims description 9
- 238000007499 fusion processing Methods 0.000 claims description 7
- 230000000306 recurrent effect Effects 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 13
- 230000008569 process Effects 0.000 description 13
- 238000004590 computer program Methods 0.000 description 11
- 238000013527 convolutional neural network Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000012806 monitoring device Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 102000002274 Matrix Metalloproteinases Human genes 0.000 description 1
- 108010000684 Matrix Metalloproteinases Proteins 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008033 biological extinction Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000003631 expected effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000011022 operating instruction Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 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
-
- 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
-
- 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
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Marketing (AREA)
- Biophysics (AREA)
- General Business, Economics & Management (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (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)
Abstract
The application discloses a traffic jam prediction method and device and electronic equipment. The method of the present application comprises: carrying out road condition similarity processing on the acquired road characteristic data of the traffic road conditions between every two lanes to obtain lane similarity data, wherein the road characteristic data comprises TMC data of a plurality of periods before a period to be predicted; processing the lane similarity data by adopting a first preset network model, extracting lane space topological structure characteristics from the lane similarity data, and outputting to obtain a first processing result; processing the first processing result by adopting a second preset network model, and outputting to obtain a second processing result; and determining a congestion prediction result of at least one lane in the road in the period to be predicted according to the second processing result. According to the technical scheme, the accuracy of traffic jam prediction can be improved.
Description
Technical Field
The present application relates to the field of new generation information technologies, and in particular, to a traffic congestion prediction method, an apparatus, and an electronic device.
Background
Tmc (traffic Message channel) is short for real-time traffic condition information, and can reflect the traffic state of the road in the electronic map area in real time. The TMC information is sent to the electronic map of the terminal to be displayed, so that a traveler can be prompted to avoid a crowded road section, and a driving route can be reasonably planned.
While there have been many studies relating to the prediction of traffic congestion, the study of accurately predicting congestion on roads remains a challenging task. In the related art, the problem of low prediction accuracy is common.
Disclosure of Invention
The objective of the present application is to solve at least one of the above technical drawbacks, and particularly to provide a technical solution for predicting traffic congestion based on lane similarity data carrying spatial and temporal characteristics of a lane, so as to improve the prediction accuracy.
The embodiment of the application adopts the following technical scheme:
in one aspect of the present application, a traffic congestion prediction method is provided, including: carrying out road condition similarity processing on the acquired road characteristic data of the traffic road conditions between every two lanes to obtain lane similarity data, wherein the road characteristic data comprises TMC data of a plurality of periods before a period to be predicted; processing the lane similarity data by adopting a first preset network model, extracting lane space topological structure characteristics from the lane similarity data, and outputting to obtain a first processing result, wherein the first preset network model is a graph convolution neural network obtained by using multiple groups of image data through machine learning training, and the multiple groups of image data all comprise image information of the TMC data; processing the first processing result by adopting a second preset network model, and outputting to obtain a second processing result, wherein the second preset network model is a recurrent neural network obtained by using multiple groups of image data with sequence relation through machine learning training, and the multiple groups of image data with sequence relation all comprise image information of TMC data with preset time sequence relation; and determining a congestion prediction result of at least one lane in the road in the period to be predicted according to the second processing result.
In some embodiments, the obtaining the road characteristic data of the traffic road condition is subjected to road condition similarity processing between two lanes to obtain lane similarity data, including: respectively performing feature dimension reduction processing on the road feature data corresponding to each period to obtain dimension-reduced road feature data corresponding to each period; respectively carrying out data fusion processing on the dimensionality-reduced road characteristic data corresponding to each period to obtain fused road characteristic data corresponding to each period; and calculating the road condition similarity between every two lanes according to the fused road characteristic data corresponding to each period to obtain lane similarity data.
In some embodiments, the performing feature dimension reduction processing on the road feature data corresponding to each time period respectively to obtain dimension-reduced road feature data corresponding to each time period includes: respectively calculating covariance matrixes of the feature matrixes corresponding to the periods to obtain eigenvalues and eigenvectors of the covariance matrixes, wherein the road feature data corresponding to the periods respectively form a feature matrix; sorting the eigenvalues from big to small, and selecting K eigenvectors corresponding to the largest K eigenvalues as row vectors respectively to construct an eigenvector matrix; and obtaining the dimension reduction feature matrix corresponding to each period according to the product of the feature vector matrix and the feature matrix.
In some embodiments, the performing data fusion processing on the dimensionality reduced road characteristic data corresponding to each time period to obtain fused road characteristic data corresponding to each time period includes: performing matrix multiplication on the dimensionality reduction characteristic matrix corresponding to each period and the characteristic value matrix corresponding to the dimensionality reduction characteristic matrix to realize data fusion, and taking the calculated column vector as the road characteristic data after fusion of each lane in the corresponding period; the dimensional characteristic matrix is an M multiplied by K dimensional matrix, the characteristic value matrix is a K multiplied by 1 dimensional matrix formed by K characteristic values, and M is the number of lanes.
In some embodiments, calculating road condition similarity between two lanes according to the fused road characteristic data corresponding to each period to obtain lane similarity data, including: taking the lane marks as keywords, and taking data with the same lane marks in the fused road characteristic data corresponding to each period as time sequence data of corresponding lanes; and calculating lane similarity data related to road conditions between every two lanes according to the time sequence data of the lanes.
In some embodiments, processing the lane similarity data using a first predetermined network model includes: and processing the lane similarity data by adopting the trained GCN.
In some embodiments, processing the first processing result by using a second preset network model includes: and processing the first processing result by adopting the trained LSTM network or GRU network.
In some embodiments, determining a congestion prediction result of at least one lane in the road for the time period to be predicted according to the second processing result includes: comparing the second processing result with the congestion index; when the traffic jam index is exceeded, the prediction result of the lane in the period to be predicted is traffic jam; and when the congestion index is not exceeded, the prediction result of the lane in the period to be predicted is smooth or slow running.
In another aspect of the present application, a traffic congestion prediction apparatus is provided for implementing the above traffic congestion prediction method.
In still another aspect of the present application, there is provided an electronic device including: a memory storing computer-executable instructions; a processor, the computer executable instructions, when executed by an electronic device, cause the processor to perform the above described traffic congestion prediction method.
The embodiment of the application adopts at least one technical scheme which can achieve the following beneficial effects:
according to the lane similarity data, the lane similarity data capable of representing the road condition similarity degree between every two lanes is calculated through TMC data, the lane space topological structure features and the time correlation features are extracted on the basis of the lane similarity data, the accuracy and the integrity of the extracted features are improved, so that the lane congestion condition can be predicted according to the lane space topological structure features and the time correlation features, the prediction accuracy is improved, and traffic dredging is facilitated based on the prediction result.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
fig. 1 is a flowchart illustrating a traffic congestion prediction method according to an embodiment of the present disclosure;
fig. 2 is a schematic diagram of a network structure for predicting traffic congestion according to an embodiment of the present disclosure;
fig. 3 is a block diagram illustrating a configuration of a traffic congestion prediction apparatus according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be described in detail and completely with reference to the following specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
An embodiment of the present application discloses a traffic congestion prediction method, which may be executed by any electronic device, for example, the traffic congestion prediction method of the embodiment of the present application may be executed on a terminal or a server, and an execution subject may predict a traffic congestion condition by obtaining lane similarity data.
Fig. 1 is a flowchart of a traffic congestion prediction method according to an embodiment of the present disclosure, and as shown in fig. 1, the method according to the present disclosure includes the following steps:
step S110, the road characteristic data of the acquired traffic road conditions is subjected to road condition similarity processing between every two lanes to obtain lane similarity data, and the road characteristic data comprises TMC data of a plurality of periods before the period to be predicted.
And step S120, processing the lane similarity data by adopting a first preset network model, extracting the spatial topological structure characteristics of the lane from the lane similarity data, and outputting to obtain a first processing result, wherein the first preset network model is a graph convolution neural network obtained by using multiple groups of image data through machine learning training, and the multiple groups of image data all comprise the image information of the TMC data.
Step S130, processing the first processing result by using a second preset network model, and outputting to obtain a second processing result, where the second preset network model is a recurrent neural network obtained by using multiple sets of image data with sequence relation through machine learning training, and the multiple sets of image data with sequence relation all include image information of TMC data with preset time sequence relation.
And step S140, determining a congestion prediction result of at least one lane in the road in the period to be predicted according to the second processing result.
As shown in fig. 1, in the embodiment, lane similarity data capable of representing the degree of road condition similarity between every two lanes is calculated through TMC data, lane spatial topological structure features and time correlation features are extracted based on the lane similarity data, and the accuracy and the integrity of the extracted features are improved, so that lane congestion conditions can be predicted according to the lane spatial topological structure features and the time correlation features, the prediction accuracy is improved, and traffic dredging is facilitated based on the prediction result.
In order to clarify the technical solutions provided by the present application, the following explains the solutions provided by the present application by specific examples:
assuming that the period to be predicted is (T + 1) periodThe previous periods include a current period T and a plurality of historical periods before the current period T, which may be P historical periods before the current period T, assuming that the road characteristic data corresponding to the current period T is characterized by
The road characteristic data corresponding to a plurality of historical periods is characterized in that
. The period T and P periods before the period T are history periods of the period to be predicted (T + 1) compared with the period to be predicted (T + 1). In this case, the acquired road characteristic data of the traffic condition includes:
。
preferably, when traffic congestion prediction is performed, the road characteristic data is extracted from TMC data of a plurality of periods within a half day before the period to be predicted. Of course, the road characteristic data may also be extracted from TMC data of a plurality of periods within a period of 8 hours, 6 hours, or the like before the period to be predicted.
In this example, the road characteristic data is extracted from TMC data representing traffic conditions, and the road characteristic data includes road condition information, including but not limited to: average traffic volume per hour, average speed of road segment, longitude, latitude, altitude, average speed of lane, lane width, number of lanes, lane curvature, road direction (left turn, right turn, straight), etc.
In addition, the road characteristic data also includes user driving behavior information, vehicle information, traffic light information, and the like; the user driving behavior information comprises but is not limited to user active data, refueling data, violation data and the like in travel service; vehicle information includes, but is not limited to, vehicle brand, vehicle age, maximum speed per hour, vehicle type, vehicle length, vehicle width, average vehicle speed, miles, maintenance times, and the like; traffic light information includes, but is not limited to, the presence or absence of a traffic light, red road light status, length of time the traffic light changes, and the like.
When acquiring the road characteristic data, it may be determined which lanes to extract the road characteristic data according to the requirement, for example, if the traffic congestion condition of the whole road network is predicted, the road characteristic data of each lane in the whole road network is acquired, and if the traffic congestion condition of a certain lane is predicted, the road characteristic data of the surrounding lane having a connection relation with the lane is acquired.
After acquiring the road characteristic data of the traffic road condition, calculating lane similarity data by the following method:
respectively performing feature dimension reduction processing on the road feature data corresponding to each period to obtain dimension-reduced road feature data corresponding to each period; respectively carrying out data fusion processing on the dimensionality-reduced road characteristic data corresponding to each period to obtain fused road characteristic data corresponding to each period; and calculating the road condition similarity between every two lanes according to the fused road characteristic data corresponding to each period to obtain lane similarity data.
As shown in fig. 2, acquiring road characteristic data of a plurality of periods includes:
here, the road characteristic data corresponding to the respective periods are respectively associated to form a characteristic matrix, i.e.
The feature matrixes are P +1, and each row vector corresponds to multi-dimensional road feature data of one lane in each feature matrix.
Respectively calculating covariance matrixes of the feature matrixes corresponding to the periods to obtain eigenvalues and eigenvectors of the covariance matrixes; sorting the eigenvalues from big to small, and selecting K eigenvectors corresponding to the largest K eigenvalues as row vectors respectively to construct an eigenvector matrix; supposing that the covariance matrix has J eigenvalues, J is larger than K, the J eigenvalues are sorted from large to small, the largest K eigenvalues are selected, and the value of K can be set according to experienceIn general, K can be set to a value between 6 and 10, preferably K = 8. Constructing a feature vector matrix by respectively using the selected K feature vectors as row vectors, and obtaining a dimension reduction feature matrix corresponding to each period according to the product of the feature vector matrix and the feature matrix
。
The data fusion is realized by carrying out matrix multiplication on the dimensionality reduction characteristic matrix corresponding to each time period and the characteristic value matrix corresponding to the dimensionality reduction characteristic matrix, and the calculated column vector is used as road characteristic data after the fusion of each lane in the corresponding time period.
Since the dimensional feature matrix is an M × K dimensional matrix, M is the number of lanes, and the eigenvalue matrix is a K × 1 dimensional matrix formed by K eigenvalues. The eigenvalue matrix is multiplied by the eigen matrix to obtain an M × 1-dimensional matrix (hereinafter, abbreviated as a fusion matrix)
) Fusion matrix
Is represented as M row vectors
Each row vector comprises an element. Therefore, the fusion process of fusing the multi-dimensional road characteristic data included in the lane into the one-dimensional characteristic data is completed, and the one-dimensional characteristic data capable of carrying the road characteristics of the lane to the maximum extent is obtained.
By the scheme, the fused road characteristic data of each lane corresponding to the P +1 period is obtained. On the basis, lane marks are used as keywords, for example, the LinkID of the lane is used as the mark, data with the same lane mark in the road characteristic data corresponding to each period after fusion is used as time series data of the corresponding lane, and similarity data about road conditions between every two lanes is calculated according to the time series data of the lane.
For example,in the feature matrix
In the method, the row vector of the feature matrix of each period is the road feature data corresponding to each lane, so that when the feature matrix is constructed, the corresponding relationship between the row vector of the feature matrix and the lane identification can be obtained, for example, the first row vector represents the road feature data corresponding to LinkID _1, and the second row vector represents the road feature data … corresponding to LinkID _2, so that after the PCA dimension reduction calculation and data fusion, linkids corresponding to M row vectors of the fusion matrix one by one can be obtained, and thus the fusion matrix of P +1 periods is calculated
And then, respectively forming time sequence data of each lane by using P +1 matrix elements with the same LinkID in the P +1 fusion matrices, and calculating similarity data about road conditions between every two lanes according to the time sequence data of the lanes to obtain the similarity data with time characteristics and space characteristics of the lanes.
After the similar data are obtained, the first preset network model and the second preset network model can be sequentially adopted to process the similar data.
The spatial correlation is a key for predicting traffic congestion, and a CNN (Convolutional Neural Networks) can acquire local spatial features, but the application range is regular images and grids. However, the urban road is a network map and is not a regular two-dimensional grid, which means that CNN is not suitable for a complex urban network map and cannot grasp spatial correlation very accurately.
The inventor considers that a filter is constructed by utilizing a Fourier domain in GCN (Graph Convolutional neural Network), the filter acts on nodes of a Graph and a first-order field of the Graph, the Graph Convolutional neural Network is a multilayer Graph Convolutional neural Network, each Convolutional layer only processes first-order neighborhood information, information transmission of multi-order neighborhoods can be realized by superposing a plurality of Convolutional layers, the spatial characteristics among the nodes are captured, the corresponding spatial correlation is obtained by establishing codes for the topological structure of a road Network and the attributes of urban roads, and after the spatial correlation is established by carrying out PCA dimension reduction, data fusion, similarity and other processing on road characteristic data of traffic road conditions, the GCN is used for accurately learning the spatial characteristics.
While the time correlation is another key factor for traffic congestion prediction, the inventor considers that the LSTM (Long Short-Term Memory Network) can solve the problems of gradient extinction and gradient explosion in the Long sequence training process relative to the RNN (Recurrent Neural Network), and the LSTM can have better performance in longer sequences. The GRU (Gated recurred Unit) is a variant of the LSTM network and can also solve the long dependency problem in RNN networks.
Based on the above consideration, the first preset network model is a GCN network obtained by machine learning training using a plurality of sets of image data, the plurality of sets of image data each include image information of TMC data, the second preset network model is an LSTM or GRU obtained by machine learning training using a plurality of sets of image data having a sequence relationship, and the plurality of sets of image data having a sequence relationship each include image information of TMC data having a preset time sequence relationship.
Optionally, a prediction network is constructed by using a GCN network and an LSTM network, or a prediction network is constructed by using a GCN network and a GRU network, and after the prediction network is constructed, training data is constructed, where road characteristic data corresponding to a plurality of historical periods may be obtained, and the road characteristic data corresponding to the plurality of historical periods may be subjected to, for example, PCA dimension reduction, data fusion, similarity, and the like described above, to obtain lane similarity data of the plurality of historical periods as training data, and a congestion label corresponding to each training data is constructed. Thus, training data is input into the prediction network, the prediction network is trained based on the output of the prediction network and the corresponding congestion label, after the expected effect is achieved, the trained GCN network is adopted to process the lane similarity data, and the trained LSTM network or GRU network is adopted to process the first processing result so as to obtain a second processing result.
And after the second processing result is obtained, comparing the second processing result with the congestion index, wherein when the second processing result exceeds the congestion index, the prediction result of the lane in the period to be predicted is congestion, and when the second processing result does not exceed the congestion index, the prediction result of the lane in the period to be predicted is smooth or slow. Here, the congestion index may be set according to a statistical experiment.
Based on the same idea as the method, another embodiment of the present application further discloses a traffic congestion prediction device. Fig. 3 is a block diagram of a traffic congestion prediction apparatus according to an embodiment of the present invention, and as shown in fig. 3, the traffic congestion prediction apparatus according to the present embodiment includes:
the first calculating unit 310 is configured to perform road condition similarity processing between every two lanes on the acquired road characteristic data of the traffic road condition to obtain lane similarity data, where the road characteristic data includes TMC data of multiple periods before a period to be predicted;
a second calculating unit 320, configured to process the lane similarity data by using a first preset network model, extract a spatial topological structure feature of a lane from the lane similarity data, and output a first processing result, where the first preset network model is a graph convolution neural network obtained through machine learning training by using multiple sets of image data, and the multiple sets of image data all include image information of the TMC data;
a third calculating unit 330, configured to process the first processing result by using a second preset network model, and output a second processing result, where the second preset network model is a recurrent neural network obtained through machine learning training using multiple sets of image data with sequence relationships, and the multiple sets of image data with sequence relationships all include image information of TMC data with a preset time sequence relationship;
and the congestion prediction unit 340 is configured to determine a congestion prediction result of at least one lane in the road in the period to be predicted according to the second processing result.
In some embodiments, the first computing unit 310 includes a dimension reduction model, a fusion module, and a similarity module;
the dimension reduction model is used for carrying out feature dimension reduction processing on the road feature data in multiple periods to obtain the road feature data subjected to dimension reduction in multiple periods;
the fusion module is used for carrying out data fusion processing on the road characteristic data subjected to the dimensionality reduction in multiple periods to obtain the road characteristic data subjected to fusion in multiple periods;
and the similarity module is used for calculating the road condition similarity between every two lanes according to the road characteristic data fused in multiple periods to obtain lane similarity data.
In some embodiments, the dimension reduction model is further configured to calculate covariance matrices of feature matrices of a plurality of time periods respectively, and obtain eigenvalues and eigenvectors of the covariance matrices, where the road feature data of each time period correspondingly form a feature matrix; sorting the eigenvalues from big to small, and selecting K eigenvectors corresponding to the largest K eigenvalues as row vectors respectively to construct an eigenvector matrix; and obtaining the dimension reduction feature matrix of a plurality of periods according to the product of the feature vector matrix and the feature matrix.
In some embodiments, the fusion module is further configured to perform matrix multiplication calculation on the dimensionality reduction feature matrix of each time period and the feature value matrix corresponding to the dimensionality reduction feature matrix to realize data fusion, and use the calculated column vector as the fused road feature data corresponding to each lane of each time period; the dimensional characteristic matrix is an M multiplied by K dimensional matrix, the characteristic value matrix is a K multiplied by 1 dimensional matrix formed by K characteristic values, and M is the number of lanes.
In some embodiments, the similarity module is further configured to use the lane identifier as a keyword, and use data with the same lane identifier in the road characteristic data after the multiple time periods are fused as time series data of a corresponding lane; and calculating lane similarity data related to road conditions between every two lanes according to the time sequence data of the lanes.
In some embodiments, the second computing unit 320 is configured to process the lane similarity data using a trained GCN network.
In some embodiments, the third computing unit 330 is configured to process the first processing result by using a trained LSTM network or a GRU network.
In some embodiments, the congestion prediction unit 340 is configured to compare the second processing result with the congestion indicator; when the traffic jam index is exceeded, the prediction result of the lane in the period to be predicted is traffic jam; and when the congestion index is not exceeded, the prediction result of the lane in the period to be predicted is smooth or slow running.
It can be understood that the monitoring device can implement each step of the monitoring method provided in the foregoing embodiments, and the explanations related to the monitoring method are applicable to the monitoring device, and are not repeated here.
Fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present application. Referring to fig. 4, at a hardware level, the electronic device includes a processor, a memory, and optionally a network interface. The Memory may include a Memory, such as a Random-Access Memory (RAM), and may further include a non-volatile Memory, such as at least 1 disk Memory. Of course, the electronic device also includes hardware required for other services.
The processor, the network interface, and the memory may be connected to each other via an internal bus, which may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one double-headed arrow is shown in FIG. 4, but that does not indicate only one bus or one type of bus.
And the memory is used for storing programs. In particular, the program may include program code comprising computer operating instructions. The memory may include both memory and non-volatile storage and provides instructions and data to the processor.
The processor reads a corresponding computer program from the nonvolatile memory into the memory and then runs the computer program to form the traffic jam prediction device on a logic level. And a processor executing a program stored in the memory to implement the traffic congestion prediction method as described above.
The traffic congestion prediction method disclosed in the embodiment of fig. 1 of the present application may be applied to or implemented by a processor. The processor may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or instructions in the form of software. The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components. The various methods, steps, and logic blocks disclosed in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present application may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software module may be located in ram, flash memory, rom, prom, or eprom, registers, etc. storage media as is well known in the art. The storage medium is located in the memory, and the processor reads the information in the memory and combines the hardware to complete the steps of the traffic jam prediction method.
An embodiment of the present application also provides a computer-readable storage medium storing one or more programs, where the one or more programs include instructions, which when executed by an electronic device including a plurality of application programs, enable the traffic congestion prediction method shown in fig. 1 to be implemented.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A method for predicting traffic congestion, comprising:
carrying out road condition similarity processing on the acquired road characteristic data of the traffic road conditions between every two lanes to obtain lane similarity data, wherein the road characteristic data comprises TMC data of a plurality of periods before a period to be predicted;
processing the lane similarity data by adopting a first preset network model, extracting lane space topological structure characteristics from the lane similarity data, and outputting to obtain a first processing result, wherein the first preset network model is a graph convolution neural network obtained by using multiple groups of image data through machine learning training, and the multiple groups of image data all comprise image information of the TMC data;
processing the first processing result by adopting a second preset network model, and outputting to obtain a second processing result, wherein the second preset network model is a recurrent neural network obtained by using multiple groups of image data with sequence relation through machine learning training, and the multiple groups of image data with sequence relation all comprise image information of TMC data with preset time sequence relation;
and determining a congestion prediction result of at least one lane in the road in the period to be predicted according to the second processing result.
2. The method of claim 1, wherein the step of performing road condition similarity processing between two lanes on the acquired road characteristic data of the traffic road conditions to obtain lane similarity data comprises:
respectively performing feature dimension reduction processing on the road feature data corresponding to each period to obtain dimension-reduced road feature data corresponding to each period;
respectively carrying out data fusion processing on the dimensionality-reduced road characteristic data corresponding to each period to obtain fused road characteristic data corresponding to each period;
and calculating the road condition similarity between every two lanes according to the fused road characteristic data corresponding to each period to obtain lane similarity data.
3. The method of claim 2, wherein performing feature dimension reduction processing on the road feature data corresponding to each time period to obtain dimension-reduced road feature data corresponding to each time period comprises:
respectively calculating covariance matrixes of the feature matrixes corresponding to the periods to obtain eigenvalues and eigenvectors of the covariance matrixes, wherein the road feature data corresponding to the periods respectively form a feature matrix;
sorting the eigenvalues from big to small, and selecting K eigenvectors corresponding to the largest K eigenvalues as row vectors respectively to construct an eigenvector matrix;
and obtaining the dimension reduction feature matrix corresponding to each period according to the product of the feature vector matrix and the feature matrix.
4. The method of claim 3, wherein performing data fusion processing on the reduced-dimension road characteristic data corresponding to each time period to obtain fused road characteristic data corresponding to each time period comprises:
performing matrix multiplication on the dimensionality reduction characteristic matrix corresponding to each period and the characteristic value matrix corresponding to the dimensionality reduction characteristic matrix to realize data fusion, and taking the calculated column vector as the road characteristic data after fusion of each lane in the corresponding period;
the dimensional characteristic matrix is an M multiplied by K dimensional matrix, the characteristic value matrix is a K multiplied by 1 dimensional matrix formed by K characteristic values, and M is the number of lanes.
5. The method of claim 4, wherein calculating the road condition similarity between two lanes according to the fused road characteristic data corresponding to each period to obtain lane similarity data comprises:
taking the lane marks as keywords, and taking data with the same lane marks in the fused road characteristic data corresponding to each period as time sequence data of corresponding lanes;
and calculating lane similarity data related to road conditions between every two lanes according to the time sequence data of the lanes.
6. The method of claim 1, wherein processing the lane similarity data using a first predetermined network model comprises:
and processing the lane similarity data by adopting the trained GCN.
7. The method of claim 1, wherein processing the first processing result using a second predetermined network model comprises:
and processing the first processing result by adopting the trained LSTM network or GRU network.
8. The method of claim 1, wherein determining a congestion prediction result of at least one lane in the road for the time period to be predicted according to the second processing result comprises:
comparing the second processing result with the congestion index;
when the traffic jam index is exceeded, the prediction result of the lane in the period to be predicted is traffic jam; and when the congestion index is not exceeded, the prediction result of the lane in the period to be predicted is smooth or slow running.
9. A traffic congestion prediction apparatus for implementing the method as claimed in any one of claims 1 to 8.
10. An electronic device, comprising:
a memory storing computer-executable instructions;
a processor that, when executed by an electronic device, causes the processor to perform the method of any of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110716417.8A CN113256985B (en) | 2021-06-28 | 2021-06-28 | Traffic congestion prediction method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110716417.8A CN113256985B (en) | 2021-06-28 | 2021-06-28 | Traffic congestion prediction method and device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113256985A true CN113256985A (en) | 2021-08-13 |
| CN113256985B CN113256985B (en) | 2021-09-17 |
Family
ID=77189826
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110716417.8A Active CN113256985B (en) | 2021-06-28 | 2021-06-28 | Traffic congestion prediction method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113256985B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114822019A (en) * | 2022-03-31 | 2022-07-29 | 阿里巴巴（中国）有限公司 | Traffic information processing method and device |
| CN115359444A (en) * | 2022-10-18 | 2022-11-18 | 智道网联科技(北京)有限公司 | Road congestion prediction method and device |
| CN115359662A (en) * | 2022-10-18 | 2022-11-18 | 智道网联科技(北京)有限公司 | Lane congestion prediction method and device |
| CN115457766A (en) * | 2022-08-31 | 2022-12-09 | 华迪计算机集团有限公司 | Method and system for predicting road congestion state |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105279965A (en) * | 2015-09-30 | 2016-01-27 | 百度在线网络技术（北京）有限公司 | Road section combination method and device based on road conditions |
| CN107045788A (en) * | 2017-06-28 | 2017-08-15 | 北京数行健科技有限公司 | Traffic Forecasting Methodology and device |
| CN108510126A (en) * | 2018-04-11 | 2018-09-07 | 南京邮电大学 | A kind of Predictive Methods of Road Accidents based on PCA and BP neural network |
| CN110047275A (en) * | 2018-01-13 | 2019-07-23 | 丰田自动车株式会社 | Association and similarity study between the observation result of the vehicle of multiple connections |
| CN110264709A (en) * | 2019-05-06 | 2019-09-20 | 北京交通大学 | The prediction technique of the magnitude of traffic flow of road based on figure convolutional network |
| CN110288170A (en) * | 2019-07-29 | 2019-09-27 | 重庆电子工程职业学院 | Prediction model construction method, traffic flow forecasting method, device and electronic equipment |
| WO2019195363A1 (en) * | 2018-04-03 | 2019-10-10 | Zoox, Inc. | Detecting errors in sensor data |
| CN111583630A (en) * | 2020-04-10 | 2020-08-25 | 河北德冠隆电子科技有限公司 | Brand-new road high-precision map rapid generation system and method based on space-time trajectory reconstruction |
| CN112216101A (en) * | 2020-09-08 | 2021-01-12 | 吉林大学 | Traffic prediction method and system based on elastic learning framework |
| CN112435462A (en) * | 2020-10-16 | 2021-03-02 | 同盾控股有限公司 | Method, system, electronic device and storage medium for short-time traffic flow prediction |
| CN112562333A (en) * | 2020-12-01 | 2021-03-26 | 智道网联科技(北京)有限公司 | Road congestion processing method and device based on intelligent traffic |
| CN112652167A (en) * | 2020-12-15 | 2021-04-13 | 东北大学 | Traffic speed prediction method based on multi-graph convolution neural network |
| CN112669606A (en) * | 2020-12-24 | 2021-04-16 | 西安电子科技大学 | Traffic flow prediction method for training convolutional neural network by utilizing dynamic space-time diagram |
| CN112863180A (en) * | 2021-01-11 | 2021-05-28 | 腾讯大地通途（北京）科技有限公司 | Traffic speed prediction method, device, electronic equipment and computer readable medium |
| CN112949828A (en) * | 2021-03-04 | 2021-06-11 | 湖南大学 | Graph convolution neural network traffic prediction method and system based on graph learning |
-
2021
- 2021-06-28 CN CN202110716417.8A patent/CN113256985B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105279965A (en) * | 2015-09-30 | 2016-01-27 | 百度在线网络技术（北京）有限公司 | Road section combination method and device based on road conditions |
| CN107045788A (en) * | 2017-06-28 | 2017-08-15 | 北京数行健科技有限公司 | Traffic Forecasting Methodology and device |
| CN110047275A (en) * | 2018-01-13 | 2019-07-23 | 丰田自动车株式会社 | Association and similarity study between the observation result of the vehicle of multiple connections |
| WO2019195363A1 (en) * | 2018-04-03 | 2019-10-10 | Zoox, Inc. | Detecting errors in sensor data |
| CN108510126A (en) * | 2018-04-11 | 2018-09-07 | 南京邮电大学 | A kind of Predictive Methods of Road Accidents based on PCA and BP neural network |
| CN110264709A (en) * | 2019-05-06 | 2019-09-20 | 北京交通大学 | The prediction technique of the magnitude of traffic flow of road based on figure convolutional network |
| CN110288170A (en) * | 2019-07-29 | 2019-09-27 | 重庆电子工程职业学院 | Prediction model construction method, traffic flow forecasting method, device and electronic equipment |
| CN111583630A (en) * | 2020-04-10 | 2020-08-25 | 河北德冠隆电子科技有限公司 | Brand-new road high-precision map rapid generation system and method based on space-time trajectory reconstruction |
| CN112216101A (en) * | 2020-09-08 | 2021-01-12 | 吉林大学 | Traffic prediction method and system based on elastic learning framework |
| CN112435462A (en) * | 2020-10-16 | 2021-03-02 | 同盾控股有限公司 | Method, system, electronic device and storage medium for short-time traffic flow prediction |
| CN112562333A (en) * | 2020-12-01 | 2021-03-26 | 智道网联科技(北京)有限公司 | Road congestion processing method and device based on intelligent traffic |
| CN112652167A (en) * | 2020-12-15 | 2021-04-13 | 东北大学 | Traffic speed prediction method based on multi-graph convolution neural network |
| CN112669606A (en) * | 2020-12-24 | 2021-04-16 | 西安电子科技大学 | Traffic flow prediction method for training convolutional neural network by utilizing dynamic space-time diagram |
| CN112863180A (en) * | 2021-01-11 | 2021-05-28 | 腾讯大地通途（北京）科技有限公司 | Traffic speed prediction method, device, electronic equipment and computer readable medium |
| CN112949828A (en) * | 2021-03-04 | 2021-06-11 | 湖南大学 | Graph convolution neural network traffic prediction method and system based on graph learning |
Non-Patent Citations (2)
| Title |
|---|
| DAOGUANG LIU.ETC: ""A Method For Short-Term Traffic Flow Forecasting Based On GCN-LSTM"", 《2020 INTERNATIONAL CONFERENCE ON COMPUTER VISION, IMAGE AND DEEP LEARNING》 * |
| 白亚男: ""基于大数据的实时交通流预测方法研究"", 《中国优秀博硕士学位论文全文数据库（硕士） 工程科技II辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114822019A (en) * | 2022-03-31 | 2022-07-29 | 阿里巴巴（中国）有限公司 | Traffic information processing method and device |
| CN115457766A (en) * | 2022-08-31 | 2022-12-09 | 华迪计算机集团有限公司 | Method and system for predicting road congestion state |
| CN115457766B (en) * | 2022-08-31 | 2023-08-08 | 华迪计算机集团有限公司 | Method and system for predicting road congestion state |
| CN115359444A (en) * | 2022-10-18 | 2022-11-18 | 智道网联科技(北京)有限公司 | Road congestion prediction method and device |
| CN115359662A (en) * | 2022-10-18 | 2022-11-18 | 智道网联科技(北京)有限公司 | Lane congestion prediction method and device |
| CN115359662B (en) * | 2022-10-18 | 2023-01-10 | 智道网联科技(北京)有限公司 | Lane congestion prediction method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113256985B (en) | 2021-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113256985B (en) | Traffic congestion prediction method and device and electronic equipment | |
| Chu et al. | Deep multi-scale convolutional LSTM network for travel demand and origin-destination predictions | |
| Gong et al. | Online spatio-temporal crowd flow distribution prediction for complex metro system | |
| WO2022105437A1 (en) | Path planning method and apparatus, and electronic device | |
| CN112307978B (en) | Target detection method and device, electronic equipment and readable storage medium | |
| CN109034086B (en) | Vehicle weight identification method, device and system | |
| Sun et al. | Passenger flow prediction of subway transfer stations based on nonparametric regression model | |
| Katariya et al. | Deeptrack: Lightweight deep learning for vehicle trajectory prediction in highways | |
| CN111860763A (en) | Model training method, parameter prediction method, model training device, parameter prediction device, electronic equipment and storage medium | |
| CN111414852A (en) | Image prediction and vehicle behavior planning method, device and system and storage medium | |
| CN113837635A (en) | Risk detection processing method, device and equipment | |
| Takimoto et al. | Predicting traffic accidents with event recorder data | |
| Taylor et al. | Data mining for vehicle telemetry | |
| CN116635872A (en) | Antagonistic semi-supervised one-time learning | |
| Boujemaa et al. | Toward road safety recommender systems: Formal concepts and technical basics | |
| CN115062779A (en) | Event prediction method and device based on dynamic knowledge graph | |
| Rahman et al. | Attention based deep hybrid networks for traffic flow prediction using google maps data | |
| CN115131634A (en) | Image recognition method, device, equipment, storage medium and computer program product | |
| Zou et al. | Traffic incident classification at intersections based on image sequences by HMM/SVM classifiers | |
| Cirstea et al. | Towards Spatio-Temporal Aware Traffic Time Series Forecasting--Full Version | |
| CN116324902A (en) | Detecting objects and determining the behavior of objects | |
| Putatunda et al. | Travel time prediction in real time for GPS taxi data streams and its applications to travel safety | |
| Wang et al. | An End‐to‐End Lane Detection Model with Attention and Residual Block | |
| CN114708110A (en) | Joint training method and device for continuous guarantee behavior prediction model and electronic equipment | |
| CN113192315B (en) | Traffic flow distribution prediction method, prediction device and terminal equipment |
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 |