Traffic data fusion method and system based on multi-source perception

Abstract

The invention relates to the technical field of artificial intelligence data processing, and provides a traffic data fusion method and system based on multi-source perception, which collects traffic conditions and demand data, improves the data processing efficiency through position labels and grouping, improves the data consistency through space-time matching optimization, and improves the data fusion efficiency. According to the multi-source traffic data fusion method, the data quality is guaranteed through anomaly detection and correction, systematization is constructed to improve the intelligent level of traffic management, the problem of space-time difference in multi-source traffic data fusion is effectively solved, and the efficiency of traffic management and the scientificity of decision making are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of artificial intelligence data processing, and provides a traffic data fusion method and system based on multi-source perception. Background Art

[0002] Multi-source perception refers to the process of obtaining traffic data from multiple angles and levels through different types of sensors, devices or systems. In the field of transportation, traffic monitoring cameras are an important means of obtaining real-time traffic data. Through video monitoring, road conditions, vehicle driving conditions, pedestrian behavior, etc. can be observed in real time, providing a basis for traffic management and law enforcement. Sensors such as radar, infrared, and laser can sense various physical quantities in the traffic environment, such as vehicle speed, distance, and flow. These data are of great significance for traffic flow analysis, speed management, and accident warning. Satellite positioning systems such as GPS and Beidou and vehicle navigation systems can provide accurate location information of vehicles and provide support for traffic planning, navigation, and rescue. Through the Internet of Things technology, traffic facilities, vehicles, pedestrians, etc. can be connected to realize real-time data collection and transmission. This helps to build an intelligent transportation system and improve traffic efficiency and management level.

[0003] In the actual use of multi-source perception in traffic data fusion, there are the following technical pain points: the data formats collected by different perception devices are different, and the data collected by different perception devices have temporal and spatial differences, such as different sampling frequencies and asynchronous data update times. The temporal and spatial differences will affect the accuracy of traffic data fusion and reduce the efficiency of traffic management. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a traffic data fusion method and system based on multi-source perception to solve the problem that spatiotemporal differences will affect the accuracy of traffic data fusion and reduce the efficiency of traffic management.

[0005] In order to solve the above technical problems, the specific technical solutions of the present invention are as follows:

[0006] In a first aspect, the present invention provides a traffic data fusion method based on multi-source perception, comprising:

[0007] Step S101, obtaining traffic condition data and traffic demand data, the traffic condition data including traffic flow, road vehicle speed, road vehicle density and road condition data, the road condition data including road flatness, road damage and road construction image data, the traffic demand data including residents' travel purpose, residents' travel time, residents' travel mode and residents' travel distance;

[0008] Step S102, extracting location information from the traffic condition data to obtain location information corresponding to the traffic condition data, creating location tags for the traffic condition data according to the location information corresponding to the traffic condition data, obtaining location-tagged traffic condition data, extracting location information from the traffic demand data to obtain location information corresponding to the traffic demand data, creating location tags for the traffic demand data according to the location information corresponding to the traffic demand data, obtaining location-tagged traffic demand data;

[0009] Step S103, substituting the position-marked traffic condition data and the position-marked traffic demand data into a preset data position grouping model to obtain traffic condition data grouping data and traffic demand data grouping data;

[0010] Step S104, matching the traffic condition data group data and the traffic demand data group data according to the time information and the location information to obtain a traffic data matching result;

[0011] Step S105, substitute the traffic data matching result into the preset traffic data simulation model. If the traffic data simulation model operates normally, the traffic data matching result is qualified. If the traffic data simulation model operates abnormally, the traffic data matching result is unqualified. When the traffic data matching result is unqualified, the traffic condition data and traffic demand data are corrected to obtain the corrected traffic condition data and traffic demand data.

[0012] Furthermore, in the traffic data fusion method based on multi-source perception provided by the present invention, step S103 includes:

[0013] Substituting the traffic condition data and traffic demand data after the position marking into a preset data position grouping model respectively, the preset data position grouping model groups the substituted data according to a preset grouping rule or algorithm to generate traffic condition data grouping data and traffic demand data grouping data;

[0014] The traffic condition data group data and the traffic demand data group data are substituted into the preset data integrity detection model to obtain the data integrity detection result. If the data integrity detection result is qualified, the traffic condition data group data and the traffic demand data group data are valid. If the data integrity detection result is unqualified, a group data error warning message is generated.

[0015] Furthermore, in the traffic data fusion method based on multi-source perception provided by the present invention, step S104 includes:

[0016] Determine data matching rules based on actual needs of traffic management;

[0017] The matching rules are based on the proximity of time information and the proximity of location information. The determined matching rules are used to execute the matching rules on the traffic condition data grouping data and the traffic demand data grouping data. The time-space weighted method is used to optimize the matching results and generate traffic data matching results. The traffic data matching results include traffic condition data and traffic demand data that meet the matching rules, as well as the correlation information between the traffic condition data and the traffic demand data.

[0018] Furthermore, in the traffic data fusion method based on multi-source perception provided by the present invention, step S105 includes:

[0019] Substituting the traffic data matching results into the preset traffic data simulation model, marking the identified abnormal data, wherein the abnormal data marking includes abnormal data features, abnormal data time, abnormal data location and abnormal data value;

[0020] Based on traffic condition data, traffic demand data, time and location, check whether there is data missing or redundant, and match correction methods according to the causes of abnormal data. Correction methods include data interpolation, smoothing and outlier removal;

[0021] Use the selected correction method to correct the abnormal data, update the corrected data to the original data set, and re-substitute the corrected data into the preset traffic data simulation model to verify whether the corrected data is qualified. If the corrected data is still unqualified, repeat the above steps and iterate repeatedly until all data pass the verification.

[0022] In a second aspect, a traffic data fusion system based on multi-source perception is provided, which applies the traffic data fusion method based on multi-source perception, and comprises:

[0023] A multi-source data acquisition unit is used to acquire traffic condition data and traffic demand data. The traffic condition data includes traffic flow, road vehicle speed, road vehicle density and road condition data. The road condition data includes road flatness, road damage and road construction image data. The traffic demand data includes residents' travel purpose, residents' travel time, residents' travel mode and residents' travel distance.

[0024] a data source location determination unit, configured to extract location information from the traffic condition data, obtain location information corresponding to the traffic condition data, establish location tags for the traffic condition data according to the location information corresponding to the traffic condition data, obtain location-tagged traffic condition data, extract location information from the traffic demand data, obtain location information corresponding to the traffic demand data, establish location tags for the traffic demand data according to the location information corresponding to the traffic demand data, and obtain location-tagged traffic demand data;

[0025] A traffic data grouping unit, used for substituting the traffic condition data after the position marking and the traffic demand data after the position marking into a preset data position grouping model to obtain traffic condition data grouping data and traffic demand data grouping data;

[0026] A data processing unit, used for matching the traffic condition data grouping data and the traffic demand data grouping data according to the time information and the location information to obtain a traffic data matching result;

[0027] The traffic data consistency repair unit is used to substitute the traffic data matching result into the preset traffic data simulation model. If the traffic data simulation model runs normally, the traffic data matching result is qualified. If the traffic data simulation model runs abnormally, the traffic data matching result is unqualified. When the traffic data matching result is unqualified, the traffic condition data and traffic demand data are corrected to obtain the corrected traffic condition data and traffic demand data.

[0028] Furthermore, in the traffic data fusion system based on multi-source perception provided by the present invention, the traffic data grouping unit in the step is also used for:

[0029] Substituting the traffic condition data and traffic demand data after the position marking into a preset data position grouping model respectively, the preset data position grouping model groups the substituted data according to a preset grouping rule or algorithm to generate traffic condition data grouping data and traffic demand data grouping data;

[0030] The traffic condition data group data and the traffic demand data group data are substituted into the preset data integrity detection model to obtain the data integrity detection result. If the data integrity detection result is qualified, the traffic condition data group data and the traffic demand data group data are valid. If the data integrity detection result is unqualified, a group data error warning message is generated.

[0031] Furthermore, in the traffic data fusion system based on multi-source perception provided by the present invention, the data processing unit is also used for:

[0032] Determine data matching rules based on actual needs of traffic management;

[0033] The matching rules are based on the proximity of time information and the proximity of location information. The determined matching rules are used to execute the matching rules on the traffic condition data grouping data and the traffic demand data grouping data. The time-space weighted method is used to optimize the matching results and generate traffic data matching results. The traffic data matching results include traffic condition data and traffic demand data that meet the matching rules, as well as the correlation information between the traffic condition data and the traffic demand data.

[0034] Furthermore, in the traffic data fusion system based on multi-source perception provided by the present invention, the traffic data consistency repair unit is further used to:

[0035] Substituting the traffic data matching results into the preset traffic data simulation model, marking the identified abnormal data, wherein the abnormal data marking includes abnormal data features, abnormal data time, abnormal data location and abnormal data value;

[0036] Based on traffic condition data, traffic demand data, time and location, check whether there is data missing or redundant, and match correction methods according to the causes of abnormal data. Correction methods include data interpolation, smoothing and outlier removal;

[0037] Use the selected correction method to correct the abnormal data, update the corrected data to the original data set, and re-substitute the corrected data into the preset traffic data simulation model to verify whether the corrected data is qualified. If the corrected data is still unqualified, repeat the above steps and iterate repeatedly until all data pass the verification.

[0038] The beneficial effects of the present invention are mainly reflected in the following aspects:

[0039] The present invention significantly improves the efficiency and accuracy of traffic management through multi-source perception and data fusion technology. The present invention obtains traffic data and demand data from multiple sources, covering multiple dimensions such as traffic flow, vehicle speed, road conditions, and residents' travel purposes, thereby improving the comprehensiveness and diversity of data. These data are established through location tags, so that each piece of data has clear location information, reducing the temporal and spatial differences caused by different data sources.

[0040] In the data grouping and matching process, the present invention adopts a preset data position grouping model and a spatiotemporal weighting method to accurately match the position and time information, optimizes the matching results, and significantly improves the accuracy of traffic data fusion.

[0041] Through multi-source data acquisition and fusion, the present invention realizes real-time collection and comprehensive analysis of traffic data, enabling traffic managers to timely understand traffic conditions and demand changes, and quickly respond to traffic incidents and congestion. The introduction of data grouping and integrity detection mechanisms reduces the complexity of data processing, improves data processing efficiency, and enables large-scale traffic data to be effectively processed and analyzed in a short time.

[0042] The present invention uses a preset traffic data simulation model to verify the matching results, can identify and mark abnormal data, and correct the data through data interpolation, smoothing and outlier removal, thereby improving the consistency and accuracy of the data. Accurate data provides a reliable basis for decision-making for traffic managers, making traffic planning, management and control more scientific and reasonable, helping to optimize the allocation of traffic resources and improve the overall operating efficiency of the traffic system.

[0043] The implementation of this invention provides strong technical support for the construction of intelligent transportation systems. Through multi-source perception and data fusion technology, it realizes the comprehensive collection and real-time processing of traffic information, laying a solid foundation for the construction and development of intelligent transportation systems. The widespread application of intelligent transportation systems will further enhance the level of intelligent transportation management, promote the digital transformation of the transportation industry, and improve the safety, reliability and convenience of transportation systems.

[0044] In summary, the present invention effectively solves the impact of temporal and spatial differences on the accuracy of traffic data fusion through multi-source perception and data fusion technology, improves the efficiency and accuracy of traffic management, and injects new vitality into the development of intelligent transportation systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0046] Figure 1 A schematic diagram of a traffic data fusion method based on multi-source perception provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. The technical solutions provided by the embodiments of the present invention are described in detail below in conjunction with the drawings.

[0048] In order to better understand the purpose of the present invention, the present invention is described in further detail below.

[0049] In a first aspect, the present invention provides a traffic data fusion method based on multi-source perception, comprising:

[0050] Step S101, obtaining traffic condition data and traffic demand data, the traffic condition data including traffic flow, road vehicle speed, road vehicle density and road condition data, the road condition data including road flatness, road damage and road construction image data, the traffic demand data including residents' travel purpose, residents' travel time, residents' travel mode and residents' travel distance;

[0051] Step S101 is to obtain traffic condition data and traffic demand data. The specific contents are as follows:

[0052] Traffic Flow: Collecting data about the flow of vehicles on the road helps understand the congestion and traffic load on the road.

[0053] Road vehicle speed: Record the speed of vehicles on the road to assess the impact of traffic flow and road conditions on vehicle speed.

[0054] Road vehicle density: measures the density of vehicles on the road and reflects the actual situation of traffic congestion.

[0055] Road condition data: including road roughness, road damage, and road construction image data. These data provide the physical condition of the road and are essential for analyzing traffic conditions and predicting potential problems.

[0056] Residents’ travel purposes: Understanding the purpose of residents’ travel, such as going to work, shopping, entertainment, etc., helps analyze the distribution and pattern of transportation demand.

[0057] Residents’ travel time: Record residents’ travel time to analyze the time distribution and peak hours of traffic demand.

[0058] Residents’ travel modes: Collecting the modes of transportation chosen by residents when traveling, such as walking, cycling, public transportation, private cars, etc., helps to understand the usage and demand of various modes of transportation.

[0059] Resident travel distance: Measure the distance residents travel to analyze the spatial distribution and scope of transportation demand.

[0060] By acquiring these detailed traffic status data and traffic demand data, the present invention can provide a solid foundation for subsequent traffic data fusion, analysis and traffic management.

[0061] Step S102, extracting location information from the traffic condition data to obtain location information corresponding to the traffic condition data, creating location tags for the traffic condition data according to the location information corresponding to the traffic condition data, obtaining location-tagged traffic condition data, extracting location information from the traffic demand data to obtain location information corresponding to the traffic demand data, creating location tags for the traffic demand data according to the location information corresponding to the traffic demand data, obtaining location-tagged traffic demand data;

[0062] Step S102 is a step of extracting location information from traffic condition data and traffic demand data and establishing location tags. The specific process is as follows:

[0063] Extract location-related information from traffic condition data, such as longitude, latitude, road name, intersection location, etc. Location information can accurately identify the geographical location where the traffic condition data occurs.

[0064] Based on the extracted location information, a location tag is created for the traffic condition data. The location tag associates the traffic condition data with a specific geographic location, so that subsequent data processing and analysis can be performed based on the geographic location.

[0065] Similarly, location-related information is extracted from the traffic demand data, such as the starting point, end point, and roads of residents' travel. Based on the extracted location information, location tags are established for the traffic demand data. This allows the traffic demand data to be associated with specific geographic locations, facilitating subsequent data fusion and analysis.

[0066] Through the processing of step S102, both the traffic condition data and the traffic demand data are given accurate location tags, which provides a solid foundation for subsequent data processing, matching and fusion.

[0067] Step S103, substituting the position-marked traffic condition data and the position-marked traffic demand data into a preset data position grouping model to obtain traffic condition data grouping data and traffic demand data grouping data;

[0068] Step S103 is a key step of substituting the location-marked traffic condition data and traffic demand data into a preset data location grouping model to obtain grouped data. The specific process is as follows: the traffic condition data and traffic demand data are both location-marked, that is, each piece of data is associated with a specific geographic location.

[0069] The traffic condition data and traffic demand data with location tags are substituted into a preset data location grouping model. This model is built based on a geographic information system (GIS) or spatial data analysis method, and can group the data according to its geographical location.

[0070] In the data location grouping model, traffic condition data and traffic demand data are grouped according to the data's geographical location, spatial relationship or preset grouping rules. The purpose of grouping is to group adjacent or similar data together to facilitate subsequent data processing and analysis.

[0071] After the grouping operation, traffic condition data grouping data and traffic demand data grouping data are obtained. These grouping data will be used for subsequent data matching, fusion and analysis.

[0072] Through the processing of step S103, the traffic condition data and the traffic demand data are effectively grouped, which facilitates the subsequent data processing and analysis, and helps to improve the accuracy and efficiency of traffic data fusion, thereby improving the effect of traffic management.

[0073] Step S104, matching the traffic condition data group data and the traffic demand data group data according to the time information and the location information to obtain a traffic data matching result;

[0074] Step S104 is a key step of matching the traffic condition data group data and the traffic demand data group data according to the time information and the location information to obtain the traffic data matching result. The specific process is as follows:

[0075] The traffic condition data group data and the traffic demand data group data are both prepared and contain accurate time information and location information. The traffic condition data group data are matched with the traffic demand data group data according to the time information and location information. The purpose of matching is to find the traffic condition and demand data at the same time, the same location, or similar time and location.

[0076] In the matching process, the influence of time and space factors should be fully considered. Since traffic conditions and demand change with time and space, it is necessary to improve the consistency or similarity of matching data in time and space.

[0077] After the matching operation, traffic data matching results are obtained. These matching results will include traffic conditions and demand data at the same time, the same location, or similar time and location, providing a basis for subsequent traffic data analysis and traffic management.

[0078] Through the processing of step S104, the traffic condition data group data and the traffic demand data group data are effectively matched, and a traffic data matching result containing time and space information is obtained, which helps to improve the accuracy and reliability of traffic data fusion and provide more accurate information support for traffic management.

[0079] Step S105, substitute the traffic data matching result into the preset traffic data simulation model. If the traffic data simulation model operates normally, the traffic data matching result is qualified. If the traffic data simulation model operates abnormally, the traffic data matching result is unqualified. When the traffic data matching result is unqualified, the traffic condition data and traffic demand data are corrected to obtain the corrected traffic condition data and traffic demand data.

[0080] Step S105 is to verify the traffic data matching result and correct it. The specific process is as follows:

[0081] The traffic data matching result obtained in step S104 is substituted into a preset traffic data simulation model. This model is constructed based on traffic flow theory, traffic simulation or machine learning, and can simulate the operation status of the traffic system.

[0082] Run the substituted data in the traffic data simulation model and observe the output results and operation status of the model. If the model runs normally, it means that the traffic data matching results are reasonable; if the model runs abnormally, it means that there are problems with the traffic data matching results.

[0083] Based on the operation results of the traffic data simulation model, it is judged whether the traffic data matching results are qualified. The qualified standard is whether the indicators such as traffic flow parameters, traffic congestion index, traffic accident rate, etc. output by the model are consistent with the expected or historical data.

[0084] If the traffic data matching result is unqualified, the traffic condition data and traffic demand data need to be corrected. The correction methods include data cleaning, data interpolation, outlier processing, time and space weighting, etc., to improve the accuracy, consistency and completeness of the data.

[0085] After correction, new traffic status data and traffic demand data are obtained. These data should be more in line with the actual situation and can more accurately reflect the operation status and demand of the traffic system.

[0086] If necessary, the corrected data can be substituted into the traffic data simulation model again for verification to improve the accuracy and reliability of the data.

[0087] Through the processing of step S105, the accuracy and reliability of the traffic data matching results can be improved, providing a solid foundation for subsequent traffic data analysis and traffic management. At the same time, by correcting the traffic status data and traffic demand data, the quality and availability of the data can be improved, further improving the effect of traffic management.

[0088] Specifically, the traffic data fusion method based on multi-source perception provided by the present invention, step S103 includes:

[0089] Substituting the traffic condition data and traffic demand data after the position marking into a preset data position grouping model respectively, the preset data position grouping model groups the substituted data according to a preset grouping rule or algorithm to generate traffic condition data grouping data and traffic demand data grouping data;

[0090] The traffic condition data group data and the traffic demand data group data are substituted into the preset data integrity detection model to obtain the data integrity detection result. If the data integrity detection result is qualified, the traffic condition data group data and the traffic demand data group data are valid. If the data integrity detection result is unqualified, a group data error warning message is generated.

[0091] Specifically, the traffic data fusion method based on multi-source perception provided by the present invention, step S104 includes:

[0092] Determine data matching rules based on actual needs of traffic management;

[0093] The matching rules are based on the proximity of time information and the proximity of location information. The determined matching rules are used to execute the matching rules on the traffic condition data grouping data and the traffic demand data grouping data. The time-space weighted method is used to optimize the matching results and generate traffic data matching results. The traffic data matching results include traffic condition data and traffic demand data that meet the matching rules, as well as the correlation information between the traffic condition data and the traffic demand data.

[0094] Specifically, the traffic data fusion method based on multi-source perception provided by the present invention, step S105 includes:

[0095] Substituting the traffic data matching results into the preset traffic data simulation model, marking the identified abnormal data, wherein the abnormal data marking includes abnormal data features, abnormal data time, abnormal data location and abnormal data value;

[0096] Based on traffic condition data, traffic demand data, time and location, check whether there is data missing or redundant, and match correction methods according to the causes of abnormal data. Correction methods include data interpolation, smoothing and outlier removal;

[0097] Use the selected correction method to correct the abnormal data, update the corrected data to the original data set, and re-substitute the corrected data into the preset traffic data simulation model to verify whether the corrected data is qualified. If the corrected data is still unqualified, repeat the above steps and iterate repeatedly until all data pass the verification.

[0098] In a second aspect, a traffic data fusion system based on multi-source perception is provided, which applies the traffic data fusion method based on multi-source perception, and comprises:

[0099] A multi-source data acquisition unit is used to acquire traffic condition data and traffic demand data. The traffic condition data includes traffic flow, road vehicle speed, road vehicle density and road condition data. The road condition data includes road flatness, road damage and road construction image data. The traffic demand data includes residents' travel purpose, residents' travel time, residents' travel mode and residents' travel distance.

[0100] a data source location determination unit, configured to extract location information from the traffic condition data, obtain location information corresponding to the traffic condition data, establish location tags for the traffic condition data according to the location information corresponding to the traffic condition data, obtain location-tagged traffic condition data, extract location information from the traffic demand data, obtain location information corresponding to the traffic demand data, establish location tags for the traffic demand data according to the location information corresponding to the traffic demand data, and obtain location-tagged traffic demand data;

[0101] A traffic data grouping unit, used for substituting the traffic condition data after the position marking and the traffic demand data after the position marking into a preset data position grouping model to obtain traffic condition data grouping data and traffic demand data grouping data;

[0102] A data processing unit, used for matching the traffic condition data grouping data and the traffic demand data grouping data according to the time information and the location information to obtain a traffic data matching result;

[0103] The traffic data consistency repair unit is used to substitute the traffic data matching result into the preset traffic data simulation model. If the traffic data simulation model runs normally, the traffic data matching result is qualified. If the traffic data simulation model runs abnormally, the traffic data matching result is unqualified. When the traffic data matching result is unqualified, the traffic condition data and traffic demand data are corrected to obtain the corrected traffic condition data and traffic demand data.

[0104] Specifically, the present invention provides a traffic data fusion system based on multi-source perception, wherein the traffic data grouping unit is further used to:

[0105] Substituting the traffic condition data and traffic demand data after the position marking into a preset data position grouping model respectively, the preset data position grouping model groups the substituted data according to a preset grouping rule or algorithm to generate traffic condition data grouping data and traffic demand data grouping data;

[0106] The traffic condition data group data and the traffic demand data group data are substituted into the preset data integrity detection model to obtain the data integrity detection result. If the data integrity detection result is qualified, the traffic condition data group data and the traffic demand data group data are valid. If the data integrity detection result is unqualified, a group data error warning message is generated.

[0107] Specifically, the present invention provides a traffic data fusion system based on multi-source perception, wherein the data processing unit is further used for:

[0108] Determine data matching rules based on actual needs of traffic management;

[0109] The matching rules are based on the proximity of time information and the proximity of location information. The determined matching rules are used to execute the matching rules on the traffic condition data grouping data and the traffic demand data grouping data. The time-space weighted method is used to optimize the matching results and generate traffic data matching results. The traffic data matching results include traffic condition data and traffic demand data that meet the matching rules, as well as the correlation information between the traffic condition data and the traffic demand data.

[0110] Specifically, the present invention provides a traffic data fusion system based on multi-source perception, wherein the traffic data consistency repair unit is further used to:

[0111] Substituting the traffic data matching results into the preset traffic data simulation model, marking the identified abnormal data, wherein the abnormal data marking includes abnormal data features, abnormal data time, abnormal data location and abnormal data value;

[0112] Based on traffic condition data, traffic demand data, time and location, check whether there is data missing or redundant, and match correction methods according to the causes of abnormal data. Correction methods include data interpolation, smoothing and outlier removal;

[0113] Use the selected correction method to correct the abnormal data, update the corrected data to the original data set, and re-substitute the corrected data into the preset traffic data simulation model to verify whether the corrected data is qualified. If the corrected data is still unqualified, repeat the above steps and iterate repeatedly until all data pass the verification.

[0114] The present invention provides a traffic data fusion method and system based on multi-source perception, and aims to solve the problems that spatiotemporal differences affect the accuracy of traffic data fusion and reduce traffic management efficiency. The following specific technical solutions are provided to solve these problems:

[0115] Obtain traffic status data and traffic demand data, covering multiple dimensions of traffic information, such as traffic flow, vehicle speed, road conditions, and residents' travel purposes, etc. Extract location information from the acquired data and establish location tags so that each piece of data has clear location information, which is convenient for subsequent processing and matching.

[0116] Substitute the position-marked data into the preset data position grouping model for grouping processing to reduce the complexity of data processing and improve processing efficiency. Perform integrity checks on the grouped data to improve the integrity and effectiveness of the data, and generate warning information in a timely manner for incomplete data.

[0117] According to the time information and location information, the traffic condition data group and the traffic demand data group are matched, and the time-space weighted method is used to optimize the matching results to generate traffic data matching results containing related information. This step enables data from different data sources to be accurately matched, providing strong support for subsequent traffic management and decision-making.

[0118] Substitute the matched data into the traffic data simulation model for verification, identify and mark abnormal data. For abnormal data, match the corresponding correction method (such as data interpolation, smoothing, outlier removal, etc.) based on its time, location, and characteristics, and correct the data. Substitute the corrected data back into the model for verification until all data pass the verification to improve the accuracy and consistency of the data.

[0119] The present invention designs and implements a traffic data fusion system based on multi-source perception, which includes modules such as a multi-source data acquisition unit, a data source location determination unit, a traffic data grouping unit, a data processing unit, and a traffic data consistency repair unit. Through the actual operation and verification of the system, it is proved that the technical solution can effectively solve the impact of spatiotemporal differences on the accuracy of traffic data fusion and improve the efficiency of traffic management.

[0120] These technical solutions effectively solve the impact of temporal and spatial differences on the accuracy of traffic data fusion through refined data processing processes, effective data matching and consistency repair mechanisms, and systematic implementation methods, thereby improving the efficiency of traffic management and the scientific nature of decision-making.

[0121] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations, and the above-described embodiments of the present invention do not constitute a limitation on the scope of protection of the present invention.

Claims

1. A traffic data fusion method based on multi-source perception, characterized in that: include: Step S101, obtaining traffic condition data and traffic demand data, the traffic condition data including traffic flow, road vehicle speed, road vehicle density and road condition data, the road condition data including road flatness, road damage and road construction image data, the traffic demand data including residents' travel purpose, residents' travel time, residents' travel mode and residents' travel distance; Step S102, extracting location information from the traffic condition data to obtain location information corresponding to the traffic condition data, creating location tags for the traffic condition data according to the location information corresponding to the traffic condition data, obtaining location-tagged traffic condition data, extracting location information from the traffic demand data to obtain location information corresponding to the traffic demand data, creating location tags for the traffic demand data according to the location information corresponding to the traffic demand data, obtaining location-tagged traffic demand data; Step S103, substituting the position-marked traffic condition data and the position-marked traffic demand data into a preset data position grouping model to obtain traffic condition data grouping data and traffic demand data grouping data; Step S104, matching the traffic condition data group data and the traffic demand data group data according to the time information and the location information to obtain a traffic data matching result; Step S105, substitute the traffic data matching result into the preset traffic data simulation model. If the traffic data simulation model operates normally, the traffic data matching result is qualified. If the traffic data simulation model operates abnormally, the traffic data matching result is unqualified. When the traffic data matching result is unqualified, the traffic condition data and traffic demand data are corrected to obtain the corrected traffic condition data and traffic demand data.

2. The traffic data fusion method based on multi-source perception as claimed in claim 1, characterized in that: The step S103 includes: Substituting the traffic condition data and traffic demand data after the position marking into a preset data position grouping model respectively, the preset data position grouping model groups the substituted data according to a preset grouping rule or algorithm to generate traffic condition data grouping data and traffic demand data grouping data; The traffic condition data group data and the traffic demand data group data are substituted into the preset data integrity detection model to obtain the data integrity detection result. If the data integrity detection result is qualified, the traffic condition data group data and the traffic demand data group data are valid. If the data integrity detection result is unqualified, a group data error warning message is generated.

3. The traffic data fusion method based on multi-source perception as claimed in claim 1, characterized in that: The step S104 includes: Determine data matching rules based on actual needs of traffic management; The matching rules are based on the proximity of time information and the proximity of location information. The determined matching rules are used to execute the matching rules on the traffic condition data grouping data and the traffic demand data grouping data. The time-space weighted method is used to optimize the matching results and generate traffic data matching results. The traffic data matching results include traffic condition data and traffic demand data that meet the matching rules, as well as the correlation information between the traffic condition data and the traffic demand data.

4. The traffic data fusion method based on multi-source perception as claimed in claim 1, characterized in that: The step S105 includes: Substituting the traffic data matching results into the preset traffic data simulation model, marking the identified abnormal data, wherein the abnormal data marking includes abnormal data features, abnormal data time, abnormal data location and abnormal data value; Based on traffic condition data, traffic demand data, time and location, check whether there is data missing or redundant, and match correction methods according to the causes of abnormal data. Correction methods include data interpolation, smoothing and outlier removal; Use the selected correction method to correct the abnormal data, update the corrected data to the original data set, and re-substitute the corrected data into the preset traffic data simulation model to verify whether the corrected data is qualified. If the corrected data is still unqualified, repeat the above steps and iterate repeatedly until all data pass the verification.

5. A traffic data fusion system based on multi-source perception, using the traffic data fusion method based on multi-source perception as claimed in any one of claims 1 to 4, characterized in that: include: A multi-source data acquisition unit is used to acquire traffic condition data and traffic demand data. The traffic condition data includes traffic flow, road vehicle speed, road vehicle density and road condition data. The road condition data includes road flatness, road damage and road construction image data. The traffic demand data includes residents' travel purpose, residents' travel time, residents' travel mode and residents' travel distance. a data source location determination unit, configured to extract location information from the traffic condition data, obtain location information corresponding to the traffic condition data, establish location tags for the traffic condition data according to the location information corresponding to the traffic condition data, obtain location-tagged traffic condition data, extract location information from the traffic demand data, obtain location information corresponding to the traffic demand data, establish location tags for the traffic demand data according to the location information corresponding to the traffic demand data, and obtain location-tagged traffic demand data; A traffic data grouping unit, used for substituting the traffic condition data after the position marking and the traffic demand data after the position marking into a preset data position grouping model to obtain traffic condition data grouping data and traffic demand data grouping data; A data processing unit, used for matching the traffic condition data grouping data and the traffic demand data grouping data according to the time information and the location information to obtain a traffic data matching result; The traffic data consistency repair unit is used to substitute the traffic data matching result into the preset traffic data simulation model. If the traffic data simulation model runs normally, the traffic data matching result is qualified. If the traffic data simulation model runs abnormally, the traffic data matching result is unqualified. When the traffic data matching result is unqualified, the traffic condition data and traffic demand data are corrected to obtain the corrected traffic condition data and traffic demand data.

6. A traffic data fusion system based on multi-source perception as claimed in claim 5, characterized in that: The traffic data grouping unit in the step is also used for: Substituting the traffic condition data and traffic demand data after the position marking into a preset data position grouping model respectively, the preset data position grouping model groups the substituted data according to a preset grouping rule or algorithm to generate traffic condition data grouping data and traffic demand data grouping data; The traffic condition data group data and the traffic demand data group data are substituted into the preset data integrity detection model to obtain the data integrity detection result. If the data integrity detection result is qualified, the traffic condition data group data and the traffic demand data group data are valid. If the data integrity detection result is unqualified, a group data error warning message is generated.

7. The traffic data fusion system based on multi-source perception as claimed in claim 5, characterized in that: The data processing unit is further used for: Determine data matching rules based on actual needs of traffic management; The matching rules are based on the proximity of time information and the proximity of location information. The determined matching rules are used to execute the matching rules on the traffic condition data grouping data and the traffic demand data grouping data. The time-space weighted method is used to optimize the matching results and generate traffic data matching results. The traffic data matching results include traffic condition data and traffic demand data that meet the matching rules, as well as the correlation information between the traffic condition data and the traffic demand data.

8. The traffic data fusion system based on multi-source perception as claimed in claim 5, characterized in that: The traffic data consistency repair unit in the step is also used to: Substituting the traffic data matching results into the preset traffic data simulation model, marking the identified abnormal data, wherein the abnormal data marking includes abnormal data features, abnormal data time, abnormal data location and abnormal data value; Based on traffic condition data, traffic demand data, time and location, check whether there is data missing or redundant, and match correction methods according to the causes of abnormal data. Correction methods include data interpolation, smoothing and outlier removal; Use the selected correction method to correct the abnormal data, update the corrected data to the original data set, and re-substitute the corrected data into the preset traffic data simulation model to verify whether the corrected data is qualified. If the corrected data is still unqualified, repeat the above steps and iterate repeatedly until all data pass the verification.

Images