A method for automatically identifying and repairing coordinate errors of a road portal device

Abstract

The application discloses a kind of road portal equipment coordinate error automatic identification and repair method, comprising: road network modeling, bus route matching through road section, bus vehicle GPS track preprocessing, bus vehicle through portal passing record matching, bus vehicle through portal passing record data preprocessing, bus vehicle GPS track matching bus route through road section, bus vehicle track through portal coordinate calculation, bus vehicle correction portal coordinate result exception handling.This method can be based on the spatiotemporal fusion of bus vehicle GPS track and portal passing record data, realize portal coordinate position anomaly automatic identification and repair, significantly improve portal coordinate correction efficiency and improve coordinate precision, provide guarantee for subsequent portal data restore vehicle track, traffic sign index calculation and other applications.

Description

Technical Field

[0001] This invention belongs to the field of traffic data governance technology, specifically relating to a method for automatic identification and repair of coordinate errors in road checkpoint equipment. Background Technology

[0002] Vehicle information collected by road checkpoints is the most important data source for traffic management. After nearly 20 years of construction, the number of road checkpoint devices in large cities generally reaches hundreds of thousands.

[0003] However, in practical applications, the coordinates of road checkpoint devices exhibit numerous anomalies, leading to errors in vehicle location identification and severely impacting the accuracy of applications such as vehicle trajectory reconstruction, intersection traffic flow calculation, and traffic signal control. Currently, the identification of coordinate errors in road checkpoint devices is mostly done manually, one by one. With hundreds of thousands of devices, there is a lack of methods for batch anomaly identification, resulting in a large workload, low efficiency, and a lack of automatic repair methods.

[0004] In summary, there is an urgent need for an automatic identification and repair method for coordinate errors in road checkpoint equipment, which would solve the problem that the identification and repair of coordinate errors in existing road checkpoint equipment relies on manual verification, and improve the quality of road checkpoint data. Summary of the Invention

[0005] The purpose of this invention is to automatically identify coordinate recognition problems in hundreds of thousands of sets of equipment in batches, and to automatically repair erroneous coordinates of equipment, thereby reducing manual labor and improving the quality of road checkpoint data.

[0006] To achieve the above-mentioned objectives, the present invention provides a method for automatic identification and repair of coordinate errors in road checkpoint equipment, comprising the following steps:

[0007] Based on the connectivity of road network nodes and vehicle traffic rules, establish road segment, node, basic segment, connecting segment, and vehicle-road network modeling rules and generate corresponding map elements. Based on the relationship between road segments and nodes and the connectivity of connecting segments between upstream and downstream road segments of nodes, establish the upstream and downstream road network topology of nodes. Based on the upstream and downstream road network topology, achieve accurate matching of bus routes and segments in different directions by filtering the direction angle of road segments and the distance between road segments and bus routes. Based on the travel time difference and distance of adjacent time trajectory points of buses, calculate the interval travel speed, identify and remove low-speed drifting GPS trajectory points according to the speed threshold, and extract the valid coordinates from the bus trajectory points. Based on the effective coordinates in the bus trajectory points and the fusion matching of the deduplicated records of vehicles passing through the checkpoint, the coordinates are obtained by automatically acquiring the accurate matching and correction of the road segment location collected by the checkpoint equipment through data spatiotemporal fusion. By correcting the upstream and downstream relationships of the road segment and intersection where the checkpoint coordinates are located, as well as the deviation distance from the original checkpoint coordinates, abnormal values ​​of the bus trajectory checkpoint coordinates are filtered out.

[0008] Preferably, the filtering of road segment direction angles and distances between road segments and bus routes includes: matching road segments traversed by bus routes; generating a buffer zone based on the bus route coordinate sequence to filter ground road segments around the bus routes; for road segments within the buffer zone, drawing perpendicular lines from the start and end points of the road segments to find the perpendicular foot of the bus routes; and matching road segments traversed by bus routes based on the minimum absolute value of the difference between the road segment direction angles at the start and end points and the perpendicular lines from the bus routes, thereby achieving accurate matching of the directional bus routes and road segments.

[0009] Preferably, after filtering the direction angle of the road segment and the distance between the road segment and the bus route, the method further includes: matching the bus vehicle with the bus route, deduplicating the GPS trajectory of the bus vehicle according to the vehicle and the bus route, so as to achieve accurate matching of the bus routes and road segments in different directions.

[0010] Preferably, the distance is calculated using Hein's formula, as follows:

[0011]

[0012] in: For the Earth's radius, Latitude Longitude Due to latitude difference, This is due to the difference in longitude.

[0013] Preferably, after identifying and eliminating low-speed drift GPS trajectory points based on the vehicle speed threshold, the method further includes: Extract the bus vehicle list and travel time period, deduplicate the bus GPS trajectory by vehicle to obtain the bus vehicle list, the bus route ID, and the minimum and maximum travel time.

[0014] The system deduplicates bus passage data by recording the passage time of each bus at the checkpoint. For the same bus, the data is sorted by time. A threshold is set based on the time difference between two adjacent records to filter duplicate passage data, retaining the earliest recorded data. This process extracts the valid coordinates from the bus's trajectory points.

[0015] This invention provides a method for automatic identification and repair of coordinate errors at road checkpoint equipment, including: road network modeling, matching bus routes to road segments, preprocessing bus GPS trajectories, matching bus passage records at checkpoints, preprocessing bus passage record data, matching bus GPS trajectories to road segments along bus routes, calculating bus coordinates at checkpoints, and handling anomalies in corrected bus checkpoint coordinate results. This method, based on the spatiotemporal fusion of bus GPS trajectories and checkpoint passage record data, enables automatic identification and repair of checkpoint coordinate anomalies, significantly improving checkpoint coordinate correction efficiency and accuracy, and providing a guarantee for subsequent applications such as checkpoint data reconstruction of vehicle trajectories and calculation of traffic indicators. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the overall process for correcting the coordinates of checkpoint equipment based on the fusion of GPS trajectory of public transport vehicles and vehicle passage records at checkpoints; Figure 2 A schematic diagram of the turning topology of the city's road network nodes; Figure 3 Distribution of city bus routes; Figure 4 The results of matching bus routes with road segments; Figure 5 This refers to the matching relationship between public transport vehicles and routes. Figure 6 For public transport vehicles GPS trajectory anomaly identification results for 52093D; Figure 7 For public transport vehicles The GPS trajectory and checkpoint vehicle passage records of 52093D were matched. Figure 8 For public transport vehicles 52093D vehicle passing through checkpoint record duplicate detection and identification results; Figure 9 Buses Example of trajectory interval matching correction checkpoint in 52093D; Figure 10 Statistics on the number of road sections for which GPG trajectory checkpoint coordinates of public transport vehicles have been corrected; Figure 11 The result is the identification of the reverse road segment at the intersection. Detailed Implementation

[0017] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be understood that after reading the teachings of this invention, those skilled in the art can make various alterations or modifications to the invention, and these equivalent forms also fall within the scope defined by the appended claims.

[0018] This invention provides a method for automatic identification and repair of coordinate errors in road checkpoint equipment, comprising the following steps: Step 1: Based on road network modeling and bus route matching, establish road segments, nodes, basic segments, connecting segments, and vehicle-road network modeling rules through road network node connectivity and vehicle traffic rules, and generate corresponding map elements. Based on the relationship between road segments and nodes, and the connectivity of upstream and downstream road segments connecting segments of nodes, establish the upstream and downstream road network topology of nodes, including nodes, upstream entrance road segments of nodes, downstream exit road segments of nodes, turning relationships between upstream and downstream road segments, and downstream road segment length. By filtering the direction angle of road segments and the distance between road segments and bus routes, accurate matching of bus routes and road segments in different directions is achieved.

[0019] The selection of road segment direction angles and distances between road segments and bus routes includes: matching road segments traversed by bus routes; generating a buffer zone based on the bus route coordinate sequence to select ground road segments around the bus routes; for road segments within the buffer zone, drawing perpendicular lines from the start and end points of the road segments to find the perpendicular foot of the bus routes; and matching road segments traversed by bus routes based on the minimum absolute value of the difference between the road segment direction angles at the start and end points and the perpendicular foot of the bus routes, thereby achieving accurate matching of the bus routes and road segments in different directions.

[0020] After filtering the direction angle of the road segment and the distance between the road segment and the bus route, the method further includes: matching the bus vehicle with the bus route, deduplicating the GPS trajectory of the bus vehicle according to the vehicle and the bus route, so as to achieve accurate matching of the bus routes and road segments in different directions.

[0021] Step 2: Data preprocessing. By using adjacent time trajectory points of buses and the travel time difference, low-speed deviation trajectory points (low-speed drift trajectory points of bus GPS trajectory points) are identified and removed based on the vehicle speed during the interval journey. Valid coordinates are then extracted from the bus trajectory points. Specifically: The identification of abnormal points in bus GPS trajectories involves sorting bus GPS trajectory points by time. Two trajectory points from the same bus with adjacent times form a time interval. The time difference and distance between the two trajectory points are calculated based on their acquisition time and coordinates. The distance is then divided by the time difference to obtain the interval's travel speed. A speed threshold is set to identify and remove low-speed drifting GPS trajectory points. The interval length (distance) is calculated using the Heinrich's formula, as follows:

[0022]

[0023] in: The radius of the Earth is 6,378,137 meters. Latitude (radians) Longitude (in radians) Due to latitude difference, This is due to the difference in longitude.

[0024] After identifying and eliminating low-speed drifting GPS trajectory points, the process also includes: Extract the bus vehicle list and travel time period, deduplicate the bus GPS trajectory by vehicle to obtain the bus vehicle list, the bus route ID, and the minimum and maximum travel time.

[0025] The system deduplicates bus passage data by recording the passage time of each bus at the checkpoint. For the same bus, the data is sorted by time. A threshold is set based on the time difference between two adjacent records to filter duplicate passage data, retaining the earliest recorded data. This process extracts the valid coordinates from the bus's trajectory points.

[0026] Step 3: After GPS preprocessing of the bus vehicle trajectory and deduplication of vehicle passage records at checkpoints, coordinates are obtained through fusion and matching. The coordinates of the road sections collected by the checkpoint equipment are automatically matched and corrected through spatiotemporal data fusion. Details are as follows: Obtain bus GPS trajectory records of passing through checkpoints. Match the vehicle license plate number and time period in the bus GPS trajectory records with the checkpoint passing records to obtain the checkpoint passing information of the bus GPS trajectory data, including the checkpoint equipment number, collection time, and vehicle license plate number.

[0027] The bus checkpoint passage records are matched with adjacent time GPS trajectory points of the bus. The matched bus GPS trajectory point sequence is sorted by time according to the bus license plate number. Two trajectory points with adjacent times form a trajectory interval. The trajectory intervals of the bus checkpoint passage time that fall before and after the GPS trajectory interval are obtained.

[0028] The location and coordinates of the checkpoint equipment are obtained by matching the bus's GPS trajectory interval and the road segment traversed by the bus route during the bus passage time. The GPS coordinates of the bus's GPS trajectory interval at two points before and after are obtained. For each GPS coordinate point, two points with the closest perpendicular distance are found within the matched road segment of the bus route. The two perpendiculars of the trajectory at the previous time point are denoted as... and The two perpendicular feet of the trajectory at the next time point are denoted as and The perpendicular distance between the two points in time allows the Cartesian set to form four combinations, namely: , , , If two perpendicular points exist in the same road segment R among the four combinations, and the latter perpendicular point is downstream of the direction of travel of the vehicle in road segment R, then the two trajectory points are matched to road segment R. If no upstream or downstream perpendicular points exist in the same road segment, then based on the road segments where the two perpendicular points are located, the shortest path is searched for using the road network topology and the shortest path algorithm. The combination with the smallest sum of road segment lengths traversed by the shortest path is selected as the upstream and downstream road segments for perpendicular point combination matching. After the two trajectory points are matched to road segments, the time difference between the time the bus passes through the checkpoint and the time the two trajectory points pass through the checkpoint is calculated, along with the time difference between the time the bus passes through the checkpoint and the time the two trajectory points pass through the checkpoint. The position and coordinates of the checkpoint equipment in the interval between the two perpendicular points in the bus route segment (passing road segment sequence) are obtained by interpolation according to the time difference ratio.

[0029] Step 4: Identification and processing of abnormal results for bus GPS checkpoint correction coordinates. This involves filtering out abnormal bus trajectory checkpoint coordinate values ​​based on the upstream and downstream relationships of the road segment where the checkpoint correction coordinates are located and the deviation distance from the original checkpoint coordinates. Specifically: Based on the summary of bus vehicle checkpoint coordinate records, the number of road segments after checkpoint correction is counted. For road segments with more than 3 correction results, the two results with the largest number of correction records are selected.

[0030] Classify and process according to the number of road segments: The bus trajectory correction checkpoint has only one road segment. The coordinates of the correction record with the smallest distance between the checkpoint correction coordinates and the checkpoint coordinate distance are selected.

[0031] For bus GPS-matched checkpoint coordinate correction, two road segments are categorized according to their relationship with the intersection. If the two road segments are connected by a U-turn from the entrance road segment to the exit road segment of the same intersection, both road segments are retained, and the coordinates of each segment are recorded based on the minimum distance between the checkpoint correction coordinates and the checkpoint coordinate distance. If the two road segments are connected by a U-turn from the entrance road segment to the exit road segment of different intersections, the road segment and coordinates of the correction record with the minimum distance between the checkpoint correction coordinates and the checkpoint coordinate distance are retained.

[0032] Example 1 like Figure 1As shown, this paper provides an implementation route for a method to correct checkpoint device coordinates based on the fusion of bus GPS trajectory and checkpoint vehicle passage records. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be understood that after reading the teachings of this invention, those skilled in the art can make various alterations or modifications to the invention, and these equivalent forms also fall within the scope defined by the appended claims.

[0033] Phase 1: Road network modeling and bus route matching Step 1: Obtain the layer elements of road segments, nodes, basic segments, lanes, and connecting segments in the calibration checkpoint coordinate area, and create node turning topology relationships. Taking the correction of a city checkpoint as an example, obtain the layer elements of city road segments, nodes, basic segments, lanes, and connecting segments. Based on the relationship between road segments and upstream and downstream nodes, and whether there are connecting segments connecting road segments, create road segment node topology turning relationships, totaling 317,737 turning relationships. See the example below. Figure 2 .

[0034] Step 2: Obtain the bus route layer for the calibration checkpoint coordinate area. There are a total of 1719 bus routes in the city. See [link / details]. Figure 3 .

[0035] Step 3: Matching Bus Routes to Road Segments. Set a 50-meter buffer zone for bus routes and a 30-meter vertical distance threshold between the start / end point of a road segment and the bus route segment. First, filter road segments within the buffer zone. Then, further filter bus route segments within the range based on the vertical distance threshold between the start / end point of the road segment and the bus route segment. Also, extract the perpendicular interval between the start / end points of the bus route segments passing through the road segment's start / end point. Calculate the direction angle of this perpendicular interval. Select the road segment with the smallest absolute value of the difference between the direction angle of the perpendicular interval and the direction angle of the road segment as the matching road segment. The matching results are shown below. Figure 4 .

[0036] Step 4: Matching buses with bus routes. The bus trajectories are deduplicated based on the buses and routes to obtain the matching relationships between buses and routes. See [link / details]. Figure 5 .

[0037] Phase Two: Data Preprocessing Step 1: Preprocess anomalies in the GPS vehicle trajectories of buses. (Taking bus route 1151 as an example.) Taking bus 52093D as an example, the anomaly preprocessing process is explained. First, the data is obtained... The 52093D bus GPS trajectory recorder sorts bus trajectories by GPS elapsed time. Two trajectory points with adjacent times form a range. The time difference and distance are calculated based on the elapsed time and coordinates of the two trajectory points within each range. The range speed is then calculated based on the range distance and time difference, with a speed threshold of 5 km / h. Abnormal ranges are identified, and the previous trajectory point is retained for each abnormal range. See the calculation example below. Figure 6 .

[0038] Step 2: Match bus vehicle trajectory with checkpoint vehicle passing records. (Based on bus vehicle...) Taking bus 52093D as an example, based on the bus's trajectory date of 2025-11-30, filter the records of vehicles passing through checkpoints, see... Figure 7 .

[0039] Step 3: Preprocessing of bus passage records at checkpoints. (Based on bus vehicles...) Taking vehicle 52093D as an example at a checkpoint, a 2-minute threshold is set for repeated detection. Duplicate records are detected. If two adjacent detection records for the same vehicle at the same checkpoint occur within 2 minutes, they are considered duplicates. The duplicate detection record is marked with a 1, while the duplicate detection record is marked with a 0. Duplicate records are filtered out. (See [link to relevant documentation]). Figure 8 .

[0040] The third stage involves fusing and correcting the checkpoint coordinates by integrating the pre-processed bus GPS trajectory and checkpoint vehicle passage records.

[0041] Public buses Taking 52093D as an example, the time it takes for a bus to pass through the checkpoint is recorded as follows: , The time is 23:45:41; the time the bus passes through the checkpoint and the most recent time preceding the bus's trajectory are recorded as 23:45:41. , The time is 23:45:30; the time the bus passes through the checkpoint and the most recent time after the bus's trajectory are recorded as 23:45:30. , The time is 23:45:50. Using a trajectory interval matching algorithm, the road segment containing the two trajectory points before and after the trajectory interval is determined to be 23585. Based on the time of the two trajectory points before and after the bus... The time difference is 20 seconds. The time it takes for the bus to pass through the checkpoint and the time it takes to reach the previous trajectory point is obtained. The time difference is 11 seconds. According to Heinrich's formula, the distance between the two trajectory points of the bus is 180 meters. Time difference and Time difference percentage estimation The distance is 99 meters. Based on this distance, the corrected position and coordinates of the checkpoint are obtained. See Figure 9 .

[0042] Phase Four: Handling Anomalies in Bus GPS Checkpoint Coordinate Correction. Based on the summarized results of bus GPS trajectory correction checkpoint coordinates, the number of road sections requiring bus trajectory correction at checkpoints is calculated. (See attached table.) Figure 10 The results are then adjusted based on the number of corrected road segments. Based on the statistical results of corrected road segments, same-direction and opposite-direction road segments at the same intersection are assessed and corrected. See [link / reference]. Figure 11 .

Claims

1. A method for automatic identification and repair of coordinate errors in road checkpoint equipment, characterized in that, Includes the following steps: Based on the connectivity of road network nodes and vehicle traffic rules, establish road segment, node, basic segment, connecting segment, and vehicle-road network modeling rules and generate corresponding map elements. Based on the relationship between road segments and nodes and the connectivity of connecting segments between upstream and downstream road segments of nodes, establish the upstream and downstream road network topology of nodes. Based on the upstream and downstream road network topology, achieve accurate matching of bus routes and segments in different directions by filtering the direction angle of road segments and the distance between road segments and bus routes. Based on the travel time difference and distance of adjacent time trajectory points of buses, calculate the interval travel speed, identify and remove low-speed drifting GPS trajectory points according to the speed threshold, and extract the valid coordinates from the bus trajectory points. Based on the effective coordinates in the bus trajectory points and the fusion matching of the deduplicated records of vehicles passing through the checkpoint, the coordinates are obtained by automatically acquiring the accurate matching and correction of the road segment location collected by the checkpoint equipment through data spatiotemporal fusion. By correcting the upstream and downstream relationships of the road segment and intersection where the checkpoint coordinates are located, as well as the deviation distance from the original checkpoint coordinates, abnormal values ​​of the bus trajectory checkpoint coordinates are filtered out.

2. The method for automatic identification and repair of coordinate errors in road checkpoint equipment as described in claim 1, characterized in that, The selection of road segment direction angles and distances between road segments and bus routes includes: matching road segments traversed by bus routes; generating a buffer zone based on the bus route coordinate sequence to select ground road segments around the bus routes; for road segments within the buffer zone, drawing perpendicular lines from the start and end points of the road segments to find the perpendicular foot of the bus routes; and matching road segments traversed by bus routes based on the minimum absolute value of the difference between the road segment direction angles at the start and end points and the perpendicular foot of the bus routes, thereby achieving accurate matching of the bus routes and road segments in different directions.

3. The method for automatic identification and repair of coordinate errors in road checkpoint equipment as described in claim 1, characterized in that, After filtering the direction angle of the road segment and the distance between the road segment and the bus route, the method further includes: matching the bus vehicle with the bus route, deduplicating the GPS trajectory of the bus vehicle according to the vehicle and the bus route, so as to achieve accurate matching of the bus routes and road segments in different directions.

4. The method for automatic identification and repair of coordinate errors in road checkpoint equipment as described in claim 1, characterized in that, The distance is calculated using Hein's formula, as follows: in: For the Earth's radius, Latitude Longitude Due to latitude difference, This is due to the difference in longitude.

5. The method for automatic identification and repair of coordinate errors in road checkpoint equipment as described in claim 1, characterized in that, After identifying and eliminating low-speed drifting GPS trajectory points based on vehicle speed thresholds, the process further includes: Extract the bus vehicle list and travel time period, deduplicate the bus GPS trajectory by vehicle to obtain the bus vehicle list, the bus route ID, and the minimum and maximum travel time. The system deduplicates bus passage data by recording the passage time of each bus at the checkpoint. For the same bus, the data is sorted by time. A threshold is set based on the time difference between two adjacent records to filter duplicate passage data, retaining the earliest recorded data. This process extracts the valid coordinates from the bus's trajectory points.

Images