Intelligent road network road condition calculation method and system based on multi-source data fusion

Abstract

The invention discloses an intelligent road network road condition calculation method and system based on multi-source data fusion, and relates to the technical field of intelligent traffic, and the method comprises the steps: creating a PostgreSQL spatial database and an extension table, and carrying out the structural storage of the obtained multi-source data, the multi-source data comprises OSM road network data, Internet road condition data and electronic police checkpoint data; matching the OSM road network data with the internet road condition data through the similarity between the spatial position and the attribute; through vehicle track correlation and space projection, calculating the average speed of a bayonet road section, combining data quality dynamic weighting and fusing multi-source data to generate road conditions, dividing road condition grades, and switching road condition priorities according to weather states; and based on the Kafka message queue, triggering periodic updating of the road condition data. Through accurate road network matching, multi-source complementarity fusion, dynamic weight adjustment and visual display, the defects of coverage limitation, environment sensitivity and poor real-time performance existing in existing single-source data are overcome.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to an intelligent road network condition calculation method and system based on multi-source data fusion. Background Art

[0002] With the acceleration of urbanization, the number of motor vehicles continues to rise, and urban traffic congestion is becoming more and more serious. Accurately grasping real-time road conditions information is crucial to improving traffic efficiency, alleviating congestion, and optimizing traffic management.

[0003] Traditional road condition monitoring relies primarily on single data sources such as traffic cameras and induction coils. Specifically, traffic cameras use image recognition to determine road conditions, such as the number of vehicles and queue lengths. However, these cameras are significantly affected by weather conditions such as heavy rain and fog, as well as by lighting conditions. Loss of image clarity can easily lead to misjudgments of road conditions. Induction coils, buried beneath the road, are highly accurate in detecting traffic flow on specific sections of road. However, due to their limited coverage, they can only detect vehicles at the installation location and fail to reflect the overall state of the road network. Furthermore, some road condition monitoring methods that rely on user data suffer from a lack of real-time performance. Delays in data collection and processing make it difficult to cope with rapidly changing traffic conditions. For example, during short periods of congestion caused by sudden traffic accidents, travelers and traffic management departments cannot be provided with the latest road conditions in a timely manner. Furthermore, the single data source provides limited road condition information, making it difficult to comprehensively consider the impact of multiple factors on traffic and accurately assess road capacity.

[0004] It can be seen that due to the shortcomings of existing technologies such as limited coverage, data accuracy being greatly affected by environmental interference, lack of real-time performance and single information, it is difficult to accurately, comprehensively and in real time obtain and display road network conditions information, and it is unable to provide a reliable basis for traffic decision-making. Summary of the Invention

[0005] In a first aspect of the present invention, in order to solve the above technical problems, a method for calculating intelligent road network conditions based on multi-source data fusion is provided, comprising the steps of: Create a PostgreSQL spatial database and extended tables to perform structured storage of the acquired multi-source data, including OSM road network data, Internet traffic data, and police checkpoint data. Matching the OSM road network data with the Internet traffic condition data based on spatial location and attribute similarity; By correlating vehicle trajectories and spatial projection, the average speed of the checkpoint section is calculated. The road condition is generated by dynamically weighting and fusing multi-source data based on data quality, and the road condition level is divided. At the same time, the road condition priority is switched according to weather conditions. Trigger periodic updates of traffic condition data based on the Kafka message queue.

[0006] Furthermore, the method for structured storage of multi-source data includes: Storing the OSM road network data in the PostgreSQL spatial database, and constructing a road network basic table including road nodes, road sections and topological relationships; Obtaining dynamic traffic data from an Internet map platform in real time and storing it in a basic traffic table. The Internet traffic data includes road names, road section geometry information, center point coordinates and angles; The electronic police checkpoint data and vehicle pass records are obtained from the traffic management system and stored in the device information table.

[0007] Furthermore, the method for matching the OSM road network data with the Internet traffic condition data includes: Use the ST_DWithin function to filter OSM road segments whose geometric distance to Internet road conditions is ≤ 100 meters; Calculate the closest point of the road condition geometry point to the OSM road segment using the ST_ClosestPoint function, and determine the relative position of the current road condition geometry point to the OSM road segment using the ST_LineLocatePoint function, filtering out matches where the relative position deviates outside a threshold. The matching accuracy of the OSM road network data and the Internet traffic data is ensured by comparing the road name, grade and road angle attributes.

[0008] Furthermore, the matching method of the OSM road network data and the Internet traffic condition data is also provided with a manual correction interface, which marks abnormal matching points through an interactive interface and manually adjusts the binding relationship between the OSM road network data and the Internet traffic condition data to correct abnormal matching results.

[0009] Furthermore, the method for calculating the average speed of the checkpoint section includes: Correlate vehicle traffic records at adjacent checkpoints by license plate number, calculate time differences, and filter out abnormal records with time differences less than 5 seconds; Use the ST_LineInterpolatePoint function to obtain the projection points of the vehicle on adjacent checkpoint road segments, and use the ST_Distance function to calculate the distance between the two points; The average speed of the checkpoint section is calculated according to the distance and the time difference.

[0010] Furthermore, the step of dynamically weighting and fusing multi-source data to generate road conditions includes: The data quality of Internet traffic conditions and checkpoint data was evaluated using the number of sampling points as a weighting factor; According to the formula Output fusion speed, where Represents the weight of the bayonet data; represents the weight of the Internet traffic condition, and ; 、 They represent the speed of bayonet fusion and the speed of Internet fusion respectively; According to the ratio of the fused speed to the road speed limit, the road condition level is divided into smooth, slow, congested and severely congested.

[0011] Furthermore, the weight of the checkpoint data is determined by the sampling density and the health of the device, which satisfies the expression:

[0012] Where, is the actual number of samples; is the standard sampling number; The number of normal bayonet slots; The total number of bayonet slots.

[0013] Furthermore, the priority strategy for the road condition classification is: In normal weather conditions, the bayonet fusion speed is preferred; In case of bad weather or abnormal checkpoint data, switch to Internet traffic data.

[0014] Furthermore, the data source of the multi-source data can be expanded to at least one of floating vehicle GPS data, bus smart card data, and mobile phone signaling data.

[0015] A second aspect of the present invention provides an intelligent road network condition calculation system based on multi-source data fusion for implementing the method, comprising: Data acquisition module, used to obtain OSM road network data, Internet traffic data and electronic police checkpoint data in real time; The road network matching module is configured with a PostGIS spatial database to perform the binding of the OSM road network and Internet road conditions and manual correction; a fusion calculation module configured to run a bayonet velocity calculation model and a dynamic weighted fusion algorithm; A real-time update module, which triggers periodic traffic updates through a message queue and stores the fusion results in a database; and The visualization module is used to bind the fused road conditions to the electronic map and render the output in real time.

[0016] Furthermore, the fusion calculation module includes a data quality assessment unit for counting the number of Internet traffic condition sampling points to generate Internet traffic condition weights, and verifying the checkpoint device status and sampling density to generate checkpoint data weights.

[0017] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: The present invention solves the shortcomings of existing single-source data, such as limited coverage, environmental sensitivity, and poor real-time performance, through precise road network matching, multi-source complementary fusion, dynamic weight adjustment, and intuitive visual display. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 This is a flowchart of the overall process disclosed in the embodiment of the present invention; Figure 2 This is a flowchart of binding multi-source data to a road network according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0021] The present invention aims to provide an intelligent road network traffic condition calculation method based on multi-source data fusion to achieve full coverage, high precision and strong real-time traffic status perception at the road network level.

[0022] Example 1 See also Figure 1-2 , the method is described in detail below.

[0023] S1. Create a PostgreSQL spatial database and extended tables to perform structured storage on the acquired multi-source data, including OSM road network data, Internet traffic data, and electronic police checkpoint data.

[0024] In this embodiment, the method for structured storage of multi-source data includes: S11. Download the OSM map data for the target area; use Navicat Premium or DBeaver to create a PostgreSQL spatial database and enable the PostGIS spatial extension.

[0025] S12. Store the OSM map data in the PostgreSQL spatial database and construct a road network basic table containing road nodes, road sections, and topological relationships.

[0026] There are three types of OSM map data elements: nodes, ways, and relations. OSM map data is written into eight basic tables: nodes, node_tags, ways, way_tags, way_nodes, relations, relation_tags, and relation_members.

[0027] S13. Obtain dynamic traffic data from the Internet map platform in real time and store it in the traffic basic table. The Internet traffic data includes road name, road section geometry information, center point coordinates and angle.

[0028] This embodiment calls the Amap / Baidu map API to obtain basic traffic condition data and dynamic traffic condition data; creates a traffic condition basic table t_b_amap_original_data, which contains the primary key id, traffic_id, road name traffic_name, road section longitude and latitude string geom, road section center point traffic_lonlat, and road section angle traffic_angle; and imports Internet traffic conditions into the traffic condition basic table.

[0029] S14. Obtain the electronic police checkpoint data and vehicle traffic records from the traffic management system and store them in the device information table. The device information table is the device_info table, which contains the device ID, device name, device location, and device binding information.

[0030] S2. Match OSM road network data with Internet traffic data based on spatial location and attribute similarity.

[0031] The matching method between OSM road network data and Internet traffic condition data specifically includes: S21. Use the ST_DWithin function to filter OSM road segments whose geometric distance to Internet road conditions is ≤ 100 meters.

[0032] S22. Calculate the closest point of the road geometry point to the OSM road segment using the ST_ClosestPoint function, and use the ST_LineLocatePoint function to determine the relative position (between 0 and 1) of the current road geometry point to the OSM road segment, and filter out matches whose relative positions deviate from the threshold.

[0033] S23. Ensure the matching accuracy between OSM road network data and Internet traffic data by comparing road name, grade, and road angle attributes; and store the matching results in the t_r_internet table.

[0034] Those skilled in the art will further explain that ST_DWithin(A,B,D) takes as input geometry A, geometry B, and a distance threshold (meters) and outputs a Boolean value (true / false) to determine whether the distance between A and B is ≤ D. ST_ClosestPoint(line,point) takes as input line geometry and point geometry and outputs the point geometry (the closest point on the line) to calculate the shortest projection point from the point to the line segment. ST_LineLocatePoint(line,point) takes as input line geometry and point geometry and outputs a floating-point number (0-1) to calculate the scaled position of the projection point on the line segment.

[0035] In this embodiment, the matching method of OSM road network data and Internet traffic condition data is also provided with a manual correction interface, which marks abnormal matching points through an interactive interface and manually adjusts the binding relationship between OSM road network data and Internet traffic condition data to correct abnormal matching results.

[0036] S3. Calculate the average speed of the checkpoint section through vehicle trajectory correlation and spatial projection. Combined with data quality, dynamically weighted fusion of multi-source data is used to generate road conditions, which are then graded and prioritized based on weather conditions.

[0037] S31. The calculation method for the average speed at the checkpoint section includes: First, the vehicle traffic records of adjacent checkpoints are linked by license plate number, the time difference is calculated, and abnormal records with a time difference of less than 5 seconds are filtered out. Then, the two projection points of the vehicle on the adjacent checkpoint road sections are obtained using the ST_LineInterpolatePoint function. The distance between the two points is calculated using the ST_Distance function (multiplied by 111000 to convert to meters). Finally, the average speed (km / h) of the checkpoint section is calculated based on the distance and time difference.

[0038] S32. Dynamically weighting and fusing multi-source data to generate traffic conditions, i.e., combining Internet traffic conditions and checkpoint speeds, and generating final traffic condition results using dynamic weights, specifically including the following steps: The number of sampling points is used as a weighting factor to evaluate the data quality of Internet traffic data and checkpoint data. In this embodiment, if Internet traffic data has more than 5 sampling points, the weight is 1; if checkpoint data has more than 20 sampling points, the weight is 1.

[0039] According to the formula Output fusion speed, where Indicates the weight of the checkpoint data; represents the weight of Internet traffic conditions, and ; 、 They represent the speed of card fusion and the speed of Internet fusion respectively. Furthermore, the weight of card data is determined by the sampling density and device health, which satisfies the expression:

[0040] Where, is the actual number of samples; is the standard sampling number; The number of normal bayonet slots; The total number of bayonet slots.

[0041] Based on the ratio of the above-mentioned fusion speed to the road speed limit, the road condition level is divided into smooth, slow, congested and severely congested, as shown in the following table:

[0042] S33. The priority strategy for road condition classification is: in normal weather, the checkpoint fusion speed is given priority, that is, the Internet traffic conditions are used as the calculation basis, and the road conditions such as the electronic police checkpoint are calculated as the high priority scheme; in severe weather or when the checkpoint data is abnormal, switch to the Internet traffic condition data.

[0043] In this embodiment, the conditions for determining the dominance of the checkpoint data include: a. The target road section must have ≥2 checkpoints, and the distance between adjacent checkpoints must be ≤5 km; b. The sampling frequency of the checkpoint data is ≥100 vehicles / hour, which means the data density meets the standard; c. The card access equipment is in normal condition, with no fault alarm records; d. The weather data is judged to be normal, that is, there is no severe weather such as heavy rain or heavy snow.

[0044] When all the above conditions are met, the bayonet weight ≥0.7.

[0045] S4. Trigger periodic updates of traffic condition data based on the Kafka message queue.

[0046] This example encapsulates the stored procedure, specifically the fusion computing logic, into the t_r_internet stored procedure, which includes data clearing, recalculation, and logging. It uses a Kafka queue to receive internet traffic and police checkpoint data for related calculations, pushing data to the t_r_internet table every two minutes.

[0047] Example 2 It should be noted that the data source of the multi-source data obtained in this embodiment can be expanded to at least one of floating vehicle GPS data, bus smart card data, and mobile phone signaling data.

[0048] For example, mobile phone signaling data can be used to supplement or replace some data sources in Example 1. Specifically, communication operators can obtain the spatial distribution and movement patterns of mobile phone terminals through mobile phone signaling, and analyze them in combination with existing data sources in the transportation industry to obtain real-time traffic status of the road network.

[0049] In addition, it is also possible to consider introducing vehicle-to-everything (V2X) technology, which enables communication between vehicles (V2V) and vehicles and infrastructure (V2I) to obtain richer vehicle driving information.

[0050] Example 3 In addition to the GIS-based road network matching algorithm disclosed in Example 1, the semantic segmentation technology in deep learning can also be used to process images or data containing road information, directly identify the road section where the vehicle is located, and achieve matching of road condition information with the road network.

[0051] Example 4 In the data fusion solution disclosed in Example 1, in addition to weighted fusion algorithms and machine learning-based fusion methods, Bayesian inference algorithms can also be used for data fusion. Specifically, based on the probability distribution of data from different data sources, the probability of fused road condition information is calculated using the Bayesian formula to determine the road condition. Data fusion can also be attempted using DS evidence theory to address data source uncertainty and conflicts, thereby improving the reliability of the fusion results.

[0052] The present invention also provides an intelligent road network condition calculation system based on multi-source data fusion for implementing the above method, which mainly includes a data acquisition module, a road network matching module, a fusion calculation module, a real-time update module and a visualization module, wherein the data acquisition module is used to obtain OSM road network data, Internet road condition data and electronic police checkpoint data in real time; the road network matching module is configured to carry a PostGIS spatial database to perform binding of the OSM road network and Internet road conditions and manual correction; the fusion calculation module is configured to run a checkpoint speed calculation model and a dynamic weighted fusion algorithm; the real-time update module is used to trigger periodic road condition updates through a message queue and store the fusion results in a database; the visualization module is used to bind the fused road conditions to an electronic map and render and output them in real time.

[0053] Among them, the fusion calculation module includes a data quality assessment unit, which is used to count the number of Internet traffic condition sampling points to generate Internet traffic condition weights, and to verify the checkpoint equipment status and sampling density to generate checkpoint data weights.

[0054] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent road network condition calculation method based on multi-source data fusion, characterized in that: Including steps: Create a PostgreSQL spatial database and extended tables to perform structured storage of the acquired multi-source data, including OSM road network data, Internet traffic data, and police checkpoint data. Matching the OSM road network data with the Internet traffic condition data based on spatial location and attribute similarity; By correlating vehicle trajectories and spatial projection, the average speed of the checkpoint section is calculated. The road condition is generated by dynamically weighting and fusing multi-source data based on data quality, and the road condition level is divided. At the same time, the road condition priority is switched according to weather conditions. Trigger periodic updates of traffic condition data based on the Kafka message queue.

2. The intelligent road network traffic condition calculation method based on multi-source data fusion according to claim 1 is characterized in that: The method for structured storage of multi-source data includes: Storing the OSM road network data in the PostgreSQL spatial database, and constructing a road network basic table including road nodes, road sections and topological relationships; Obtaining real-time traffic data from an Internet map platform and storing it in a basic traffic table. The Internet traffic data includes road names, road section geometry information, center point coordinates, and angles. The electronic police checkpoint data and vehicle pass records are obtained from the traffic management system and stored in the device information table.

3. The intelligent road network traffic condition calculation method based on multi-source data fusion according to claim 1 is characterized in that: The matching method of the OSM road network data and the Internet traffic condition data includes: Use the ST_DWithin function to filter OSM road segments whose geometric distance to Internet road conditions is ≤ 100 meters; Calculate the closest point of the road condition geometry point to the OSM road segment using the ST_ClosestPoint function, and determine the relative position of the current road condition geometry point to the OSM road segment using the ST_LineLocatePoint function, filtering out matches where the relative position deviates outside a threshold. The matching accuracy of the OSM road network data and the Internet traffic data is ensured by comparing the road name, grade and road angle attributes.

4. The intelligent road network condition calculation method based on multi-source data fusion according to claim 1 or 3, characterized in that: The matching method of OSM road network data and Internet traffic condition data is also provided with a manual correction interface, which marks abnormal matching points through an interactive interface and manually adjusts the binding relationship between the OSM road network data and the Internet traffic condition data to correct abnormal matching results.

5. The intelligent road network traffic condition calculation method based on multi-source data fusion according to claim 1 is characterized in that: The calculation method of the average speed of the checkpoint section includes: Correlate vehicle traffic records at adjacent checkpoints by license plate number, calculate time differences, and filter out abnormal records with time differences less than 5 seconds; Use the ST_LineInterpolatePoint function to obtain the projection points of the vehicle on adjacent checkpoint road segments, and use the ST_Distance function to calculate the distance between the two points; The average speed of the checkpoint section is calculated according to the distance and the time difference.

6. The intelligent road network condition calculation method based on multi-source data fusion according to claim 1 is characterized in that: The step of dynamically weighting and fusing multi-source data to generate road conditions includes: The data quality of Internet traffic conditions and checkpoint data was evaluated using the number of sampling points as a weighting factor; According to the formula Output fusion speed, where Represents the weight of the bayonet data; represents the weight of the Internet traffic condition, and ; 、 They represent the speed of bayonet fusion and the speed of Internet fusion respectively; According to the ratio of the fused speed to the road speed limit, the road condition level is divided into smooth, slow, congested and severely congested.

7. The intelligent road network traffic condition calculation method based on multi-source data fusion according to claim 6 is characterized in that: The weight of the card access data is determined by the sampling density and the health of the equipment, which satisfies the expression: Where, is the actual number of samples; is the standard sampling number; The number of normal bayonet slots; The total number of bayonet slots.

8. The intelligent road network traffic condition calculation method based on multi-source data fusion according to claim 1 or 6, characterized in that: The priority strategy for the road condition classification is: In normal weather conditions, the bayonet fusion speed is preferred; In case of bad weather or abnormal checkpoint data, switch to Internet traffic data.

9. The intelligent road network condition calculation method based on multi-source data fusion according to claim 1 is characterized in that: The data source of the multi-source data can be expanded to at least one of floating vehicle GPS data, public transportation smart card data, and mobile phone signaling data.

10. An intelligent road network condition calculation system based on multi-source data fusion for implementing the method according to any one of claims 1 to 9, characterized in that: include: Data acquisition module, used to obtain OSM road network data, Internet traffic data and electronic police checkpoint data in real time; The road network matching module is configured with a PostGIS spatial database to perform the binding of the OSM road network and Internet road conditions and manual correction; a fusion calculation module configured to run a bayonet velocity calculation model and a dynamic weighted fusion algorithm; Real-time update module, used to trigger periodic traffic condition updates through message queues and store fusion results in the database; as well as The visualization module is used to bind the fused road conditions to the electronic map and render the output in real time.

11. The intelligent road network condition calculation system based on multi-source data fusion according to claim 10, characterized in that: The fusion calculation module includes a data quality assessment unit, which is used to count the number of Internet traffic condition sampling points to generate Internet traffic condition weights, and verify the checkpoint device status and sampling density to generate checkpoint data weights.

Images