CN115909734B - Vehicle driving data screening and updating system and method based on track big data - Google Patents
Vehicle driving data screening and updating system and method based on track big data Download PDFInfo
- Publication number
- CN115909734B CN115909734B CN202211420559.0A CN202211420559A CN115909734B CN 115909734 B CN115909734 B CN 115909734B CN 202211420559 A CN202211420559 A CN 202211420559A CN 115909734 B CN115909734 B CN 115909734B
- Authority
- CN
- China
- Prior art keywords
- data
- monitoring
- vehicle
- node
- screening
- 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
- 238000012216 screening Methods 0.000 title claims abstract description 81
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000012544 monitoring process Methods 0.000 claims abstract description 128
- 238000012545 processing Methods 0.000 claims abstract description 54
- 238000007781 pre-processing Methods 0.000 claims abstract description 21
- 238000005457 optimization Methods 0.000 claims abstract description 15
- 230000005540 biological transmission Effects 0.000 claims description 81
- 239000013598 vector Substances 0.000 claims description 24
- 238000013145 classification model Methods 0.000 claims description 18
- 238000010586 diagram Methods 0.000 claims description 12
- 230000002159 abnormal effect Effects 0.000 claims description 10
- 238000010606 normalization Methods 0.000 claims description 9
- 230000002776 aggregation Effects 0.000 claims description 6
- 238000004220 aggregation Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 230000008676 import Effects 0.000 claims description 2
- 238000005259 measurement Methods 0.000 claims 1
- 230000009286 beneficial effect Effects 0.000 abstract description 5
- 238000004458 analytical method Methods 0.000 abstract description 3
- 238000010276 construction Methods 0.000 abstract 1
- 230000000694 effects Effects 0.000 description 16
- 230000006870 function Effects 0.000 description 16
- 230000008569 process Effects 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 3
- 238000007405 data analysis Methods 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention relates to the technical field of traffic data processing, and discloses a vehicle driving data screening and updating system and a vehicle driving data screening and updating method based on track big data, wherein the vehicle driving data screening and updating system comprises an intelligent scheduling management platform for screening and updating driving data, a vehicle-mounted end for monitoring vehicle internal information and a road side end for monitoring vehicle external environment; the intelligent scheduling management platform comprises a path optimization module, an association processing module, a data acquisition module, a data preprocessing module, a feature screening module and a model building module; the intelligent traffic data screening and updating monitoring system aims at intelligent traffic data in a large area, realizes low-cost, high-stability and optimized pre-classification monitoring, further improves classification data accuracy through differential analysis based on normalized processing classification characteristic data, and is beneficial to construction of a real-time intelligent traffic data screening and updating monitoring system.
Description
Technical Field
The invention relates to the technical field of traffic data processing, in particular to a vehicle driving data screening and updating system and method based on track big data.
Background
In the traffic management of the operation vehicles, the operation vehicles on the road side and the changed environment are required to be monitored, and after the monitored data are analyzed and calculated in the later period, the operation vehicles in different areas in the city can be dynamically planned, so that good vehicle operation effect is achieved, the crowdedness of the road is reduced, and the convenience degree of people in traveling is improved.
However, at present, there is no systematic data analysis and screening model, which can classify, screen and process the driving data in the area, if the existing data processing system is used for processing the vehicle traffic data, the problems of incomplete data acquisition, poor data transmission effect and low adaptability of data analysis and screening exist, and the desired data is difficult to obtain; meanwhile, in the process of detecting data, due to the fact that the vehicle operation area is large, the problems of more signal nodes, tedious and laborious operation, large error, poor timeliness, limited monitoring range and the like exist in the process of data transmission, and the data information of the traveling in the area is difficult to grasp in time, so that it is important to construct a data monitoring system for intelligent traffic and to screen and update the data in time.
Disclosure of Invention
The present invention has been made in view of the above-described problems existing in traffic data processing.
Therefore, the invention aims to provide a vehicle driving data screening and updating system and a vehicle driving data screening and updating method based on track big data, which construct a data monitoring system of intelligent traffic, optimize and correlate transmission paths of source nodes, perform characteristic screening processing and updating, adapt to information transmission requirements in traffic operation of a large area, have a preprocessing association classification effect, simplify complicated processes of data transmission, facilitate data processing effects of different channels, further improve subsequent screening classification updating effects, reduce cost of monitoring traffic data of the large area and improve operation effects of intelligent traffic.
In order to solve the technical problems, the invention provides the following technical scheme:
on one hand, the invention provides a vehicle driving data screening and updating system based on track big data, which comprises an intelligent scheduling management platform for screening and updating driving data, a vehicle-mounted end for monitoring vehicle internal information and a road side end for monitoring vehicle external environment; the intelligent scheduling management platform comprises a path optimization module, an association processing module, a data acquisition module, a data preprocessing module, a feature screening module and a model building module; the vehicle-mounted terminal is provided with vehicle-mounted OBU equipment, and the vehicle-mounted OBU equipment is used for acquiring data of a vehicle loaded with the vehicle-mounted OBU equipment in a vehicle driving area; the road side end is provided with RSU road side equipment for road vehicle monitoring and platform monitoring equipment for road site monitoring, the RSU road side equipment and the platform monitoring equipment are respectively used for acquiring corresponding road monitoring data in a vehicle driving area, wherein the RSU road side equipment is connected with the platform monitoring equipment, and the RSU road side equipment is connected with the vehicle-mounted OBU equipment;
the intelligent scheduling management platform optimizes a transmission path of an internal node in the acquired vehicle driving area through a path optimization module, and associates the acquired road monitoring data in the vehicle driving area through an association processing module to form vehicle monitoring data in the vehicle driving area; performing data preprocessing on vehicle monitoring data in the vehicle driving area; and carrying out feature screening processing on the vehicle monitoring data in the vehicle driving area after data preprocessing, screening out preset vehicle driving update data meeting preset conditions, carrying out continuous data discretization processing on the vehicle monitoring data in the history vehicle driving area after preprocessing as classification feature data, enabling traffic data of different areas to be discrete data, carrying out normalization processing on the discrete data, enabling the traffic data of different areas to have the same dimension, wherein the normalization processing is shown in a formula (1):
（1） ；
wherein,indicate->Dimension data->；/>Indicate->First->Sample values;and->Respectively represent +.>Maximum and minimum values of samples in the dimensional data; />Indicating>Dimension data->Sample value->Represents normalized->Dimension data->Sample values;
screening traffic data of different areas after normalization processing, and screening out preset vehicle running update data meeting preset conditions as classification characteristic data;
inputting the classification characteristic data into a preset classification model for classification model training and setting a penalty function, when a loss function between a classification result output by the preset classification model and a real label corresponding to the classification characteristic data meets a preset convergence condition, obtaining a target data screening model, wherein the loss function adopts a mean square error as a cost function MSE of an evaluation value, and measuring the difference between an output predicted value and a true value of the target data screening model, as shown in a formula (2):
（2）；
wherein,a single predictor representing the ith sample, < +.>The true value of the i-th sample is represented, and n is the sequence number.
As a preferred embodiment of the present invention, wherein: the method comprises the steps that a transmission path of an internal node in a vehicle running area is obtained and optimized, the intelligent scheduling management platform optimizes the obtained transmission path of the internal node in the vehicle running area through a path optimization module, the method comprises the steps of calling the path optimization module in the intelligent scheduling management platform, wherein a relay forwarding node is selected between monitoring nodes of RSU road side equipment and platform monitoring equipment at a road side end and corresponding sink nodes, and the priority selection is specifically as follows:
acquiring source nodes of road side ends in a vehicle driving area, wherein the source nodes comprise nodes of RSU road side equipment and platform monitoring equipment in road monitoring, namely, all data transmission paths for transmitting road monitoring data to corresponding sink nodes;
calculating a preferred value for each data transmission path, the preferred value being calculated as shown in formula (3):
（3）；
in the method, in the process of the invention,representing a data transmission path for a source node to transmit monitoring data of RSU road side equipment and station monitoring equipment to a corresponding sink nodeIs a preferred value of (2);is a weight coefficient;representing data transmission pathsThe number of nodes is monitored by the RSU road side equipment and the station monitoring equipment;representing data transmission pathsMiddle (f)iThe number of neighbor nodes of the data transmission node;representing data transmission pathsMiddle (f)iPacket loss rate of data transmission node and previous monitoring node;the method comprises the steps of setting a preset packet loss rate threshold value;
and based on the preference value of the data transmission path, performing sequencing preference and then serving as a relay forwarding node of the new source node.
As a preferred embodiment of the present invention, wherein: the method comprises the steps of obtaining a transmission path of an internal node in a vehicle driving area to optimize, and calling an association processing module in the intelligent scheduling management platform, wherein the relay forwarding node is used for carrying out association classification transmission between monitoring nodes of RSU road side equipment and platform monitoring equipment at the road side end and corresponding aggregation nodes, and the method comprises the following specific steps:
acquiring a plurality of source nodes at a road side end in a vehicle driving area, wherein the source nodes comprise nodes of RSU road side equipment and nodes of platform monitoring equipment in road monitoring, namely, all data transmission paths for transmitting road monitoring data to corresponding sink nodes;
obtaining a node data set according to a preset area transmission path, and finding out a double frequent set from the node data set;
calculating the support degree and the confidence degree of each double item set in the double item frequent set, and constructing an association structure diagram among the aggregation nodes of the vehicle driving area according to the support degree and the confidence degree of the double item set;
obtaining important feature vectors and secondary feature vectors of all sink nodes in the association structure diagram, and carrying out feature combination on the important feature vectors and the secondary feature vectors to obtain node feature vectors;
and inputting the association structure diagram and the node feature vector into a GCN model for convolution classification, outputting classification result data, and transmitting the classification result data to an intelligent scheduling management platform.
As a preferred embodiment of the present invention, wherein: and carrying out data preprocessing on the vehicle monitoring data in the vehicle driving area, namely carrying out outlier rejection and missing value processing on the classification result data after the classification result data is transmitted to an intelligent scheduling management platform, forming the vehicle monitoring data in the vehicle driving area in different areas, carrying out feature screening processing, screening out preset vehicle driving updating data meeting preset conditions according to the target data screening model, and updating traffic index data in different areas.
As a preferred embodiment of the present invention, wherein: the outlier rejection includes:
the preset classification result data are stored in a database of the intelligent scheduling management platform, and repeated transmission preset classification result data of the same sink node are deleted; and/or the number of the groups of groups,
setting a similarity threshold of the preset classification result data, and deleting the preset classification result data in the same sink node data similarity threshold; and/or the number of the groups of groups,
deleting the data with one or more missing preset classification result data at the tail of the same sink node data.
As a preferred embodiment of the present invention, wherein: the missing value processing comprises the steps of calculating the mode of data corresponding to the data abnormal item or the data missing item in different sink node data preset item data aiming at the preset classification result data with the data abnormal item or the data missing item in the head or the middle part of the same sink node data, and replacing the corresponding abnormal or missing data with the mode.
On the other hand, the invention provides a method for a vehicle driving data screening and updating system applied to the track big data, which comprises the following steps:
based on source node data acquired by a road side end, the intelligent scheduling management platform optimizes and correlates a transmission path of the source node acquired by the road side end to form vehicle monitoring data in a vehicle driving area;
performing data preprocessing on vehicle monitoring data in the vehicle driving area;
performing feature screening processing based on the preprocessed vehicle monitoring data in the vehicle driving area, and screening out preset vehicle driving update data meeting preset conditions as classification feature data;
and inputting the classification characteristic data into a preset classification model to perform classification model training and setting a punishment function, and obtaining a target data screening model when a loss function between a classification result output by the preset classification model and a real label corresponding to the classification characteristic data meets a preset convergence condition.
The invention has the beneficial effects that: the invention constructs the data monitoring system of intelligent traffic, optimizes and associates the transmission path of the source node, performs characteristic screening treatment and updating, can adapt to the information transmission requirement in the traffic operation of a large area, has the preprocessing association classification effect, can simplify the complexity of data transmission, is beneficial to the data processing effect of different channels, further improves the subsequent screening classification updating effect, reduces the scale of classified data, reduces the complexity of classification calculation, reduces the cost of monitoring the traffic data of the large area and improves the operation effect of intelligent traffic. In summary, the intelligent traffic data in a large area is monitored in a low cost, high stability and optimized pre-classification mode, and based on the normalized classification characteristic data, the accuracy of the classification data is further improved through differential analysis, so that the intelligent traffic data screening and updating monitoring system is constructed in real time.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. Wherein:
FIG. 1 is a schematic view of an application scenario of a vehicle driving data screening and updating system based on track big data;
FIG. 2 is a schematic diagram of a modular structure of an intelligent scheduling management platform in a vehicle driving data screening and updating system based on track big data;
FIG. 3 is a flowchart showing steps of a method for screening and updating vehicle driving data for intelligent transportation according to the present invention.
Reference numerals in the drawings: 10. an intelligent scheduling management platform; 20. a vehicle-mounted end; 30. and a road side end.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. It will be apparent that the described embodiments are some, but not all, embodiments of the invention. All other embodiments, which are obtained by a person skilled in the art based on the described embodiments of the invention, fall within the scope of protection of the invention.
The problems of incomplete data acquisition, poor data transmission effect, low adaptability of data analysis and screening, difficulty in obtaining the desired data and the like in traffic data processing are solved. Based on the method, the invention provides a vehicle driving data screening and updating system and a method based on the track big data.
Referring to fig. 1 and 2, in an embodiment of the present invention, a vehicle driving data screening and updating system based on big track data is provided, which includes an intelligent scheduling management platform 10 for screening and updating driving data, a vehicle-mounted terminal 20 for monitoring vehicle internal information, and a road side terminal 30 for monitoring vehicle external environment; the intelligent scheduling management platform 10 comprises a path optimization module, an association processing module, a data acquisition module, a data preprocessing module, a feature screening module and a model building module; the vehicle-mounted terminal 20 is provided with vehicle-mounted OBU equipment, and the vehicle-mounted OBU equipment is used for acquiring data of a vehicle loaded with the vehicle-mounted OBU equipment in a vehicle driving area; the road side end 30 is provided with RSU road side equipment for road vehicle monitoring and platform monitoring equipment for road site monitoring, the platform monitoring equipment is arranged at a site corresponding to an operation vehicle and used for monitoring waiting personnel data, the RSU road side equipment and the platform monitoring equipment respectively acquire data corresponding to road side sections in a vehicle driving area, wherein the RSU road side equipment is connected with the platform monitoring equipment, and the RSU road side equipment is connected with vehicle-mounted OBU equipment;
the intelligent scheduling management platform 10 optimizes and associates transmission paths of the acquired internal nodes of the vehicle running area based on source node data in the vehicle running area of the road side end 30, optimizes the transmission paths of the acquired internal nodes of the vehicle running area through a path optimization module, and associates the acquired road monitoring data in the vehicle running area through an association processing module to form vehicle monitoring data in the vehicle running area; data preprocessing is carried out on vehicle monitoring data in a vehicle driving area; performing feature screening processing based on the preprocessed vehicle monitoring data in the vehicle driving area, and screening out preset vehicle driving update data meeting preset conditions as classification feature data;
in this embodiment, preferably, feature screening processing is performed based on the preprocessed vehicle monitoring data in the vehicle driving area, and preset vehicle driving update data satisfying a preset condition is screened out as classification feature data, including:
carrying out continuous data discretization processing on the vehicle monitoring data in the preprocessed historical vehicle driving area to ensure that traffic data of different areas are all discrete data, and carrying out normalization processing on the discrete data to ensure that the traffic data of different areas have the same dimension;
screening traffic data of different areas after normalization processing, wherein the screened out preset vehicle running update data meeting preset conditions is used as classification characteristic data, and the normalization processing is shown in a formula (1):
（1）；
wherein,indicate->Dimension data->；/>Indicate->First->Sample values; />And->Respectively represent +.>Maximum and minimum values of samples in the dimensional data; />Indicating>Dimension data->Sample value->Represents normalized->Dimension data->And sample values.
In the embodiment, the classification feature data is input into a preset classification model to perform classification model training and a penalty function is set, and when a loss function between a classification result output by the preset classification model and a real label corresponding to the classification feature data meets a preset convergence condition, a target data screening model is obtained.
In order to further improve the accuracy of classified data, it is emphasized that the loss function of this embodiment adopts the mean square error as the cost function MSE of the comprehensive evaluation value model, and measures the difference between the output predicted value and the actual value of the target data screening model, as shown in the formula (2):
（2）；
wherein,a single predictor representing the ith sample, < +.>The true value of the i-th sample is represented, and n is the sequence number.
The method for obtaining the transmission path of the source node in the vehicle driving area to optimize specifically includes invoking a path optimization module in the intelligent scheduling management platform 10, where the priority selection between the monitoring nodes of the relay forwarding node on the RSU road side equipment and the platform monitoring equipment of the road side end 30 and the corresponding sink node is specifically as follows:
firstly, acquiring source nodes of a road side end 30 in a vehicle driving area, wherein the source nodes comprise nodes of RSU road side equipment and platform monitoring equipment in road monitoring, namely all data transmission paths for transmitting road monitoring data to corresponding sink nodes;
second, a preferred value for each data transmission path is calculated, the preferred value being calculated as shown in formula (3):
（3）；
in the method, in the process of the invention,representing a data transmission path for a source node to transmit monitoring data of RSU road side equipment and station monitoring equipment to a corresponding sink nodeIs a preferred value of (2);is a weight coefficient;representing data transmission pathsThe number of nodes is monitored by the RSU road side equipment and the station monitoring equipment;representing data transmission pathsMiddle (f)iThe number of neighbor nodes of the data transmission node;representing data transmission pathsMiddle (f)iPacket loss rate of data transmission node and previous monitoring node;the method comprises the steps of setting a preset packet loss rate threshold value;
thirdly, based on the preferred value of the data transmission path, the relay forwarding node is used as a new source node after sequencing and preference.
The embodiment combines the above scheme for optimizing the transmission path of the source node in the vehicle driving area, reduces the scale of the classified data, reduces the complexity of classified calculation, and is beneficial to the data processing effect of different channels, and meanwhile, the embodiment further provides a scheme for the associated classified transmission processing between the sink nodes, as follows:
firstly, acquiring a transmission path of an internal node in a vehicle driving area for optimization, and calling an association processing module in the intelligent scheduling management platform 10, wherein the association classification transmission between monitoring nodes of RSU road side equipment and platform monitoring equipment of a relay forwarding node at a road side end 30 and corresponding aggregation nodes is specifically as follows:
secondly, acquiring a plurality of source nodes of the road side end 30 in the vehicle driving area, wherein the source nodes comprise nodes of RSU road side equipment and nodes of platform monitoring equipment in road monitoring, namely all data transmission paths for transmitting road monitoring data to corresponding sink nodes;
thirdly, a node data set is obtained according to a preset area transmission path, and a double-item frequent set is found out from the node data set;
fourthly, calculating the support degree and the confidence degree of each double item set in the double item frequent set, and constructing an association structure diagram among the aggregation nodes of the vehicle driving area according to the support degree and the confidence degree of the double item set;
fifthly, acquiring important feature vectors and secondary feature vectors of all the sink nodes in the association structure chart, and carrying out feature combination on the important feature vectors and the secondary feature vectors to obtain node feature vectors;
sixthly, inputting the association structure diagram and the node feature vector into the GCN model for convolution classification, outputting classification result data, and transmitting the classification result data to the intelligent scheduling management platform 10.
When the intelligent scheduling management platform 10 of the present embodiment performs data preprocessing on the vehicle monitoring data in the vehicle driving area, including abnormal value rejection and missing value processing on the classification result data after the classification result data is transmitted to the intelligent scheduling management platform 10, feature screening processing is performed after the vehicle monitoring data in the vehicle driving area in different areas are formed, and preset vehicle driving update data meeting preset conditions is screened out according to the target data screening model to update traffic index data in different areas.
In addition, specifically, the outlier rejection includes: the preset classification result data are stored in a database of the intelligent scheduling management platform 10, and repeated transmission preset classification result data of the same sink node are deleted; setting a similarity threshold of the preset classification result data, and deleting the preset classification result data in the same sink node data similarity threshold; and deleting the data with one or more missing preset classification result data at the tail of the same sink node data. The missing value processing comprises the steps of calculating the mode of data corresponding to the data abnormal items or the data missing items in different sink node data preset item data according to the preset classification result data with the data abnormal items or the data missing items in the head or the middle of the same sink node data, and replacing the corresponding abnormal or missing data with the mode.
It should be further noted that the classification characteristic data may include operation information of the monitored vehicle in the on-board terminal 20 in the area, for example: scheduling information, GPS information for debugging, attendance information, station entering and exiting information, overtime information, shift starting information, operation vehicle statistics information, IC card data import information, real-time departure information, operation vehicle inquiry information, peak interval inquiry information, first station reporting rate information, time period line average operation speed, time period personnel single completion rate information, time period personnel single completion personnel quantity information and the like.
The classification characteristic data may also include real-time information of the monitored road segments in the road side 30 in the area, for example: (1) Time-space related data including traffic accident data, motor vehicle travel data, non-motor vehicle and pedestrian travel data, and current road congestion level; (2) Time-related data, including weather data and environmental data; (3) Spatially related data, including point of interest data, road design data, and the like.
Based on the above, the invention constructs the data monitoring system of intelligent traffic, optimizes the transmission path of the source node and correlates the related data, and performs feature screening processing and updating based on the vehicle monitoring screening data in the vehicle driving area after preprocessing, thereby being applicable to the information transmission requirement in the traffic operation of a large area, having the preprocessing correlation classification effect, simplifying the data transmission, being beneficial to improving the data processing effect of different channels, further improving the subsequent screening classification updating effect, reducing the scale of classification data, reducing the complexity of classification calculation, simultaneously reducing the cost of monitoring the traffic data of the large area and improving the operation effect of intelligent traffic.
Referring to fig. 3, in combination with the system of the above embodiment, the present embodiment further provides a method for screening and updating vehicle driving data of intelligent traffic, which includes the following steps:
step S101, based on source node data acquired by a road side end 30, an intelligent scheduling management platform 10 optimizes and correlates a transmission path of the source node acquired by the road side end 30 to form vehicle monitoring data in a vehicle driving area;
step S102, data preprocessing is carried out on vehicle monitoring data in a vehicle driving area;
step S103, performing feature screening processing based on the preprocessed vehicle monitoring data in the vehicle driving area, and screening out preset vehicle driving update data meeting preset conditions as classification feature data;
step S104, inputting the classification characteristic data into a preset classification model to perform classification model training and setting a punishment function, and obtaining a target data screening model when a loss function between a classification result output by the preset classification model and a real label corresponding to the classification characteristic data meets a preset convergence condition.
In summary, the intelligent traffic data in a large area is monitored in a low cost, high stability and optimized pre-classification mode, and based on the normalized classification characteristic data, the accuracy of the classification data is further improved through differential analysis, so that the intelligent traffic data screening and updating monitoring system is constructed in real time.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present application, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
Any process or method description in a flowchart or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process. And the scope of the preferred embodiments of the present application includes additional implementations in which functions may be performed in a substantially simultaneous manner or in an opposite order from that shown or discussed, including in accordance with the functions that are involved.
Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions.
It is to be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. All or part of the steps of the methods of the embodiments described above may be performed by a program that, when executed, comprises one or a combination of the steps of the method embodiments, instructs the associated hardware to perform the method.
In addition, each functional unit in each embodiment of the present application may be integrated in one processing module, or each unit may exist alone physically, or two or more units may be integrated in one module. The integrated modules may be implemented in hardware or in software functional modules. The integrated modules described above, if implemented in the form of software functional modules and sold or used as a stand-alone product, may also be stored in a computer-readable storage medium. The storage medium may be a read-only memory, a magnetic or optical disk, or the like.
The foregoing is merely specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily think of various changes or substitutions within the technical scope of the present application, and these should be covered in the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (5)
1. The vehicle driving data screening and updating system based on the track big data is characterized by comprising an intelligent scheduling management platform (10) for screening and updating the driving data, a vehicle-mounted end (20) for monitoring vehicle internal information and a road side end (30) for monitoring vehicle external environment; the intelligent scheduling management platform (10) comprises a path optimization module, an association processing module, a data acquisition module, a data preprocessing module, a feature screening module and a model building module; the vehicle-mounted end (20) is provided with vehicle-mounted OBU equipment, and the vehicle-mounted OBU equipment is used for acquiring data of a vehicle loaded with the vehicle-mounted OBU equipment in a vehicle driving area; the road side end (30) is provided with RSU road side equipment for road vehicle monitoring and platform monitoring equipment for road station monitoring, the RSU road side equipment and the platform monitoring equipment respectively acquire corresponding road monitoring data in a vehicle driving area, wherein the RSU road side equipment is connected with the platform monitoring equipment, and the RSU road side equipment is connected with vehicle-mounted OBU equipment;
the intelligent scheduling management platform (10) optimizes the transmission path of the source node in the acquired vehicle driving area through a path optimization module, and associates the acquired road monitoring data in the vehicle driving area through an association processing module to form vehicle monitoring screening data in the vehicle driving area; the intelligent scheduling management platform (10) optimizes the transmission path of the acquired vehicle driving area source node through a path optimization module, and comprises a path optimization module in the intelligent scheduling management platform (10), wherein the priority selection between the monitoring nodes of the relay forwarding node on RSU road side equipment and platform monitoring equipment of the road side end (30) and the corresponding sink node is specifically as follows:
acquiring source nodes of a road side end (30) in a vehicle driving area, wherein the source nodes comprise nodes of RSU road side equipment and platform monitoring equipment in road monitoring, namely all data transmission paths for transmitting road monitoring data to corresponding sink nodes;
calculating a preferred value for each data transmission path, the preferred value being calculated as shown in formula (3):
（3）；
wherein,data transmission path for indicating source node to transmit monitoring data of RSU road side equipment and station monitoring equipment to corresponding sink node>Is a preferred value of (2); />Is a weight coefficient; />Representing data transmission path->The number of nodes is monitored by the RSU road side equipment and the station monitoring equipment; />Representing data transmission path->Middle (f)iThe number of neighbor nodes of the data transmission node; />Representing data transmission path->Middle (f)iPacket loss rate of data transmission node and previous monitoring node; />The method comprises the steps of setting a preset packet loss rate threshold value;
based on the preferred value of the data transmission path, sequencing and selecting the relay forwarding node as a new source node;
the method comprises the steps of obtaining a transmission path of an internal node in a vehicle driving area to optimize, and calling an association processing module in the intelligent scheduling management platform (10), wherein the association classification transmission between a monitoring node of a relay forwarding node on RSU road side equipment and platform monitoring equipment of a road side end (30) and a corresponding sink node is specifically as follows:
acquiring a plurality of source nodes of a road side end (30) in a vehicle driving area, wherein the source nodes comprise nodes of RSU road side equipment and nodes of platform monitoring equipment in road monitoring, namely all data transmission paths for transmitting road monitoring data to corresponding sink nodes;
obtaining a node data set according to a preset area transmission path, and finding out a double frequent set from the node data set;
calculating the support degree and the confidence degree of each double item set in the double item frequent set, and constructing an association structure diagram among the aggregation nodes of the vehicle driving area according to the support degree and the confidence degree of the double item set;
obtaining important feature vectors and secondary feature vectors of all sink nodes in the association structure diagram, and carrying out feature combination on the important feature vectors and the secondary feature vectors to obtain node feature vectors;
inputting the association structure diagram and the node feature vector into a GCN model for convolution classification, outputting classification result data, and transmitting the classification result data to an intelligent scheduling management platform (10);
performing data preprocessing on the vehicle monitoring data in the vehicle driving area, namely performing outlier rejection and missing value processing on the classification result data after the classification result data are transmitted to an intelligent scheduling management platform (10), performing feature screening processing on the vehicle monitoring data in the vehicle driving area in different areas, screening preset vehicle driving updating data meeting preset conditions according to a target data screening model, and updating traffic index data in different areas;
data preprocessing is carried out on the vehicle monitoring screening data in the vehicle driving area; and (3) carrying out feature screening processing on the vehicle monitoring screening data in the vehicle driving area after data preprocessing, screening out preset vehicle driving update data meeting preset conditions, taking the preset vehicle driving update data as classification feature data, carrying out discretization processing on the vehicle monitoring data in the history vehicle driving area after preprocessing to enable traffic data in different areas to be discrete data, and carrying out normalization processing on the discrete data, wherein the normalization processing is shown in a formula (1):
（1） ；
wherein,indicate->Dimension data->；/>Indicate->First->Sample values; />And->Respectively represent +.>Maximum and minimum values of samples in the dimensional data; />Indicating>Dimension data itemSample value->Represents normalized->Dimension data->Sample values;
screening traffic data of different areas after normalization processing, and screening out preset vehicle running update data meeting preset conditions as classification characteristic data; inputting the classification characteristic data into a preset classification model to perform classification model training and setting a punishment function, and obtaining a target data screening model when a loss function between a classification result output by the preset classification model and a real label corresponding to the classification characteristic data meets a preset convergence condition;
the loss function adopts a mean square error as an estimated cost function MSE, and the difference between a predicted value and a true value is output by a measurement target data screening model, as shown in a formula (2):
（2）；
wherein,a single predictor representing the ith sample, < +.>The true value of the i-th sample is represented, and n is the sequence number.
2. The vehicle travel data screening and updating system based on trajectory big data according to claim 1, wherein the outlier rejection includes:
the preset classification result data are stored in a database of the intelligent scheduling management platform (10), and repeated transmission preset classification result data of the same sink node are deleted; and/or the number of the groups of groups,
setting a similarity threshold of the preset classification result data, and deleting the preset classification result data in the same sink node data similarity threshold; and/or the number of the groups of groups,
deleting the data with one or more missing preset classification result data at the tail of the same sink node data.
3. The system for screening and updating vehicle running data based on big track data according to claim 1, wherein the missing value processing comprises calculating the mode of data corresponding to the data abnormal item or the data missing item in different sink node data preset item data according to the preset classification result data with the data abnormal item or the data missing item in the head or the middle part of the same sink node data, and replacing the mode with the corresponding abnormal or missing data.
4. The vehicle driving data screening and updating system based on the track big data according to claim 1, wherein the classification characteristic data comprises operation information of a monitored vehicle at a vehicle-mounted terminal (20), and specifically comprises: scheduling information, GPS information for debugging, attendance information, station entering and exiting information, overtime information, shift starting information, operation vehicle statistics information, IC card data import information, real-time departure information, operation vehicle inquiry information, peak interval inquiry information, first station report rate information, time period line average operation speed, time period personnel single completion rate information and time period personnel single completion personnel quantity information;
the classification characteristic data also comprises real-time information of the monitored road section of the road side end (30), and specifically comprises the following steps: spatio-temporal related data, time related data, and spatial related data.
5. A method for a vehicle travel data screening and updating system applied to track big data, comprising the steps of:
based on source node data acquired by a road side end (30), an intelligent scheduling management platform (10) optimizes and correlates a transmission path of the source node acquired by the road side end (30) to form vehicle monitoring data in a vehicle driving area; the intelligent scheduling management platform (10) optimizes the transmission path of the acquired vehicle driving area source node through a path optimization module, and comprises a path optimization module in the intelligent scheduling management platform (10), wherein the priority selection between the monitoring nodes of the relay forwarding node on RSU road side equipment and platform monitoring equipment of the road side end (30) and the corresponding sink node is specifically as follows:
acquiring source nodes of a road side end (30) in a vehicle driving area, wherein the source nodes comprise nodes of RSU road side equipment and platform monitoring equipment in road monitoring, namely all data transmission paths for transmitting road monitoring data to corresponding sink nodes;
calculating a preferred value for each data transmission path, the preferred value being calculated as shown in formula (3):
（3）；
wherein,data transmission path for indicating source node to transmit monitoring data of RSU road side equipment and station monitoring equipment to corresponding sink node>Is a preferred value of (2); />Is a weight coefficient; />Representing data transmission path->The number of nodes is monitored by the RSU road side equipment and the station monitoring equipment; />Representing data transmission path->Middle (f)iThe number of neighbor nodes of the data transmission node; />Representing data transmission path->Middle (f)iPacket loss rate of data transmission node and previous monitoring node; />The method comprises the steps of setting a preset packet loss rate threshold value;
based on the preferred value of the data transmission path, sequencing and selecting the relay forwarding node as a new source node;
the method comprises the steps of obtaining a transmission path of an internal node in a vehicle driving area to optimize, and calling an association processing module in the intelligent scheduling management platform (10), wherein the association classification transmission between a monitoring node of a relay forwarding node on RSU road side equipment and platform monitoring equipment of a road side end (30) and a corresponding sink node is specifically as follows:
acquiring a plurality of source nodes of a road side end (30) in a vehicle driving area, wherein the source nodes comprise nodes of RSU road side equipment and nodes of platform monitoring equipment in road monitoring, namely all data transmission paths for transmitting road monitoring data to corresponding sink nodes;
obtaining a node data set according to a preset area transmission path, and finding out a double frequent set from the node data set;
calculating the support degree and the confidence degree of each double item set in the double item frequent set, and constructing an association structure diagram among the aggregation nodes of the vehicle driving area according to the support degree and the confidence degree of the double item set;
obtaining important feature vectors and secondary feature vectors of all sink nodes in the association structure diagram, and carrying out feature combination on the important feature vectors and the secondary feature vectors to obtain node feature vectors;
inputting the association structure diagram and the node feature vector into a GCN model for convolution classification, outputting classification result data, and transmitting the classification result data to an intelligent scheduling management platform (10);
performing data preprocessing on vehicle monitoring data in the vehicle driving area;
performing feature screening processing based on the preprocessed vehicle monitoring data in the vehicle driving area, and screening out preset vehicle driving update data meeting preset conditions as classification feature data;
and inputting the classification characteristic data into a preset classification model to perform classification model training and setting a punishment function, and obtaining a target data screening model when a loss function between a classification result output by the preset classification model and a real label corresponding to the classification characteristic data meets a preset convergence condition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211420559.0A CN115909734B (en) | 2022-11-14 | 2022-11-14 | Vehicle driving data screening and updating system and method based on track big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211420559.0A CN115909734B (en) | 2022-11-14 | 2022-11-14 | Vehicle driving data screening and updating system and method based on track big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115909734A CN115909734A (en) | 2023-04-04 |
| CN115909734B true CN115909734B (en) | 2024-01-09 |
Family
ID=86493326
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211420559.0A Active CN115909734B (en) | 2022-11-14 | 2022-11-14 | Vehicle driving data screening and updating system and method based on track big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115909734B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117648213A (en) * | 2024-01-30 | 2024-03-05 | 杭银消费金融股份有限公司 | Data transmission path model building method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6480783B1 (en) * | 2000-03-17 | 2002-11-12 | Makor Issues And Rights Ltd. | Real time vehicle guidance and forecasting system under traffic jam conditions |
| CN107507428A (en) * | 2017-09-11 | 2017-12-22 | 安徽实运信息科技有限责任公司 | One kind is used for road traffic congestion degree detecting system |
| CN114187763A (en) * | 2021-12-07 | 2022-03-15 | 罗鑫 | Vehicle driving data screening method and system for intelligent traffic |
| CN114743401A (en) * | 2022-03-23 | 2022-07-12 | 华录智达科技股份有限公司 | Data visualization bus dispatching management platform based on bus digital transformation |
| CN115273480A (en) * | 2022-09-28 | 2022-11-01 | 南京智慧交通信息股份有限公司 | Traffic early warning system and method based on big data analysis |
-
2022
- 2022-11-14 CN CN202211420559.0A patent/CN115909734B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6480783B1 (en) * | 2000-03-17 | 2002-11-12 | Makor Issues And Rights Ltd. | Real time vehicle guidance and forecasting system under traffic jam conditions |
| CN107507428A (en) * | 2017-09-11 | 2017-12-22 | 安徽实运信息科技有限责任公司 | One kind is used for road traffic congestion degree detecting system |
| CN114187763A (en) * | 2021-12-07 | 2022-03-15 | 罗鑫 | Vehicle driving data screening method and system for intelligent traffic |
| CN114743401A (en) * | 2022-03-23 | 2022-07-12 | 华录智达科技股份有限公司 | Data visualization bus dispatching management platform based on bus digital transformation |
| CN115273480A (en) * | 2022-09-28 | 2022-11-01 | 南京智慧交通信息股份有限公司 | Traffic early warning system and method based on big data analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115909734A (en) | 2023-04-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109000668B (en) | Real-time intelligent navigation method based on Internet of vehicles | |
| Chen et al. | A short-term traffic prediction model in the vehicular cyber–physical systems | |
| US8606727B2 (en) | Method and system for traffic prediction based on space-time relation | |
| CN103678917B (en) | A kind of real-time arrival time Forecasting Methodology of public transport based on simulated annealing | |
| CN102708698B (en) | Vehicle optimal-path navigation method based on vehicle internet | |
| CN102622879B (en) | Traffic information providing apparatus | |
| CN101599219A (en) | Traffic signal control system | |
| CN110836675B (en) | Decision tree-based automatic driving search decision method | |
| CN108444486B (en) | Navigation route sorting method and device | |
| CN115909734B (en) | Vehicle driving data screening and updating system and method based on track big data | |
| CN104794895A (en) | Multisource traffic information fusion method for expressways | |
| CN111932915A (en) | Bus right transfer and ride-sharing signal priority method based on arrival time stepping prediction | |
| CN112530158B (en) | Road network supplementing method based on historical track | |
| JP5378002B2 (en) | Vehicle motion estimation device, vehicle motion estimation method, and vehicle motion estimation program | |
| JP3402263B2 (en) | Traffic information management device and traffic information management method | |
| CN115331425A (en) | Traffic early warning method, device and system | |
| CN110954119B (en) | Urban road motor vehicle optimal route induction method based on big data | |
| CN116853272A (en) | Automatic driving vehicle behavior prediction method and system integrating complex network and graph converter | |
| CN105806355A (en) | Green vehicle path navigation system and method | |
| CN115164922A (en) | Path planning method, system, equipment and storage medium | |
| CN115060278A (en) | Intelligent vehicle battery replacement navigation method and system based on multi-target genetic algorithm | |
| CN117546218A (en) | Distributed multitasking machine learning for traffic prediction | |
| CN114596709A (en) | Data processing method, device, equipment and storage medium | |
| CN113362628A (en) | Bidirectional dynamic shortest path display system in intelligent large traffic dispatching system | |
| KR19990068262A (en) | Automobile traffic information providing system_ |
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 |