Method for automatically generating high-precision map according to vehicle trajectory

Abstract

The invention relates to the field of computer geographic data processing and GIS (geographic information system) data automatic acquisition, and discloses a method for automatically generating a high-precision map according to a vehicle trajectory, which comprises the following steps of: acquiring road data by using a vehicle carrying Beidou terminal equipment, cleaning the acquired road data, and storing the cleaned data into a database; and then reading the data from the database, carrying out parallel copying on the data and generating a buffer area to obtain vector data of the traffic facilities, and adjusting and rendering different data types by utilizing a map rendering tool to obtain a high-precision map. Compared with an existing method, on the premise that the data accuracy and integrity are guaranteed, the number of hardware equipment used for road data collection is reduced, the collection cost is reduced, centimeter-level real-time dynamic motion positioning Beidou terminal equipment is utilized, the data collection error is reduced, and the data collection efficiency is improved. And the influence of factors such as weather, terrains, shelters and the like on the generation of the high-precision map is avoided to a certain extent.

Description

Technical Field

[0001] The present invention relates to the technical field of computer geographic data processing and GIS data automatic collection, and in particular to a method for automatically generating a high-precision map according to a vehicle trajectory line. Background Art

[0002] The patent document with application number CN202123437119.X discloses a multi-purpose high-precision map acquisition device, including a base, a mounting plate, a lifting assembly and a column. The mounting plate is arranged above the base, and the lifting assembly is installed between the mounting plate and the base. The column is perpendicular to the middle of the upper end surface of the mounting plate. The top of the column is equipped with an image acquisition sensor, and also includes a laser radar, which is installed on the outer wall of the column. This patent document uses a variety of complex acquisition equipment to collect high-precision maps. First, the acquisition cost is high, and second, the requirements for the external environment are also extremely high. On roads with dense populations or a large number of vehicles, the acquisition quality and efficiency will be seriously affected.

[0003] The patent document with application number CN202010846249.X discloses a high-precision map data collection method, device and system, which relates to the field of autonomous driving technology. The high-precision map data collection method includes: first obtaining the manual driving trajectory information of the target area; then controlling the autonomous driving collection device to perform autonomous driving according to the manual driving trajectory information, and obtaining high-precision map collection data of the target area. When collecting map collection data, automatic collection is performed through the autonomous driving collection device. This patent document uses a variety of collection equipment and other electronic equipment, and the collection cost is still high. At the same time, the collection results are not verified, and the collection quality cannot be verified.

[0004] In summary, the equipment, technology and manpower costs required for collecting high-precision map information using existing technologies are relatively high, involving the purchase and maintenance costs of high-precision GPS equipment, laser radar, cameras and other professional equipment, as well as the cost of regular map updates. In addition, the accuracy and completeness of high-precision map information collection are easily affected by factors such as weather, terrain, and obstructions. Summary of the invention

[0005] The present invention aims to provide a method for automatically generating a high-precision map according to a vehicle trajectory line, the method comprising the following steps: step S1 preparation: adopting a centimeter-level, real-time dynamic motion positioning Beidou terminal device, and setting parameters of the device to complete road data collection; step S2 data cleaning: cleaning the collected road data and removing abnormal data; step S3 data storage: storing the cleaned data in a PostGres database, and reading and processing it; step S4 data adjustment: comparing the processed data with a satellite map, adjusting the intersection morphology and traffic facilities, and rendering the data to generate a high-precision map.

[0006] As a further solution of the present invention, the S1 preparation work specifically includes:

[0007] S11: Place the selected centimeter-level, real-time dynamic motion positioning Beidou terminal device in the collection vehicle;

[0008] S12: Turn on the Beidou terminal device, and after the Beidou terminal device locks on at least four satellites, set the collection parameters of the Beidou terminal device, and the collection vehicle travels at a constant speed on the collected road surface, and each time a road section is passed, the attributes of the road section are edited;

[0009] S13: Define the traffic facility codes, count the traffic facilities on the road and classify them.

[0010] As a further solution of the invention, the S2 data cleaning specifically includes:

[0011] S21: When the collection is completed, the Beidou terminal device is turned off and the data is exported from the device;

[0012] S22: converting the data row coordinates to ensure data consistency;

[0013] S23: cleaning up the collected abnormal longitude and latitude data;

[0014] S24: Thinning and smoothing the data, connecting the longitude and latitude data in series through the timestamp, and obtaining the trajectory line vector data of the collected road section.

[0015] As a further solution of the invention, the S3 data storage specifically includes:

[0016] S31: Storing the data in a PostGres database, using Geotools to read the trace vector data of the collected road section, performing parallel line processing according to the attribute data of the collected road section, obtaining lane center line data with the same number of actual lanes, and encoding each piece of data according to the actual traffic equipment according to the traffic facility coding of the preparation S13;

[0017] S32: based on the lane center line data described in S31, copy the center line data of all lane center lines, central barriers, and vehicle and non-vehicle barriers parallel to the lane center line;

[0018] S33: the middle line data in S32 are then parallelized according to the collected attributes, and each piece of data is encoded according to the traffic facility code defined in S13 and the actual traffic equipment to obtain the marking facility data, and the marking facility data is stored in the PostGres database;

[0019] S34: The lane center line data described in S32 is then processed in a buffer zone, and the coded marking facility data in S33 is also brought into the buffer zone to obtain surface traffic facilities, and the generated surface traffic facility data is then stored in a PostGres database.

[0020] As a further solution of the present invention, the S4 data adjustment specifically includes:

[0021] S41: Use a GIS desktop tool to open the data stored in the PostGres database in S34, compare the data with the latest satellite base map, and adjust the intersection shape and traffic facilities;

[0022] S42: Use Geoserver to connect to the PostGres database in S41, read the traffic marking data and surface traffic facility data in the PostGres database, render in different styles according to different traffic facility coding types, and then publish to obtain a high-precision map.

[0023] Compared with the prior art, the beneficial effects of the present invention are: under the premise of ensuring data accuracy and completeness, the number of hardware devices used for road data collection is reduced, the collection cost is reduced, and the centimeter-level, real-time dynamic motion positioning Beidou terminal equipment is used to reduce data collection errors, and to a certain extent, the influence of factors such as weather, terrain, and obstructions on the generation of high-precision maps is avoided. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a flow chart of the main steps of the present invention;

[0025] Figure 2 The traffic facility coding diagram (I) of the present invention;

[0026] Figure 3 The traffic facility coding diagram (II) of the present invention;

[0027] Figure 4 A trajectory vector data graph generated for the present invention;

[0028] Figure 5 A road intersection morphology diagram generated by the present invention;

[0029] Figure 6 The renderings of the high-precision map of the present invention are released. DETAILED DESCRIPTION

[0030] A preferred embodiment of the present invention will be described in detail below in conjunction with the accompanying drawings. In the present invention, Beidou terminal equipment refers to various receiving devices developed based on the Beidou Satellite Navigation System (BDS), which can receive and process Beidou satellite signals for various applications such as positioning, navigation, timing and communication. It is widely used in scenarios such as autonomous driving, engineering surveying and mapping, precision agriculture, and drone navigation. In situations where high-precision positioning is required, RTK (Real-Time Kinematic) technology is used to achieve centimeter-level or even millimeter-level positioning accuracy, and it can be combined with a ground-based augmentation system to perform accurate real-time dynamic positioning.

[0031] In the present invention, PostgreSQL is an open source database management system that provides support for SQL and full support for transactions. PostgreSQL provides a large number of extended functions, including complex queries, foreign keys, triggers, views, transaction integrity, multi-version concurrency control, etc. PostGIS is a spatial database extension that adds geographic object support to PostgreSQL. PostGIS can create, edit, query and analyze geographic spatial data, providing powerful support for various applications.

[0032] In the present invention, GeoTools is an open source Java repository, an open source project based on the LGPL license, suitable for different operating systems, and used for data processing and operation of geographic information systems (GIS). It provides many tools and libraries for processing various geographic information system (GIS) tasks, including reading, writing, converting, analyzing and visualizing spatial data, and is suitable for various GIS application development and geospatial data processing tasks.

[0033] In the present invention, GeoServer is a Java-based software server that allows users to view and edit geospatial data. GeoServer has great flexibility in map creation and data sharing. It implements the Web Map Service (WMS) standard and can create maps in various output formats.

[0034] The method of automatically generating a high-precision map based on a vehicle trajectory in the present invention is as follows: Figure 1 The following steps are shown:

[0035] 1. Preparation

[0036] (1) Prepare equipment: correctly install the Beidou terminal equipment on the collection vehicle and ensure that the Beidou terminal equipment has obtained signals from no less than four satellites and performed positioning;

[0037] (2) Set the collection target: determine the specific range and location of the middle longitude and latitude of the lane to be collected, and clarify the collection target;

[0038] (3) Select the collection method: Select the appropriate collection method according to the specific situation, set the collection parameters, set the collection frequency to 1-10Hz, let the collection vehicle drive at a constant speed on the collected road surface, and edit the attributes of each road section after passing through it, such as the current lane, the total number of lanes, whether there are green belts, non-motorized vehicle lanes, etc.;

[0039] (4) Define traffic facility codes: Count the traffic facilities on the road and classify them, such as Figure 2 and Figure 3 As shown, it is divided into major category coding and minor category coding, and a complete code number is defined for each road traffic facility;

[0040] (5) Start collection: Start the Beidou terminal device to start collecting the longitude and latitude data of the middle of the lane. The device will record the vehicle location information, including longitude and latitude, and also record the current road information (such as road name, number of lanes, central isolation type, whether there is a non-motorized lane, etc.);

[0041] (6) Data storage and management: Properly store the collected data to ensure data security and integrity.

[0042] 2. Data Cleaning

[0043] (1) Data format check: Ensure that the data is stored in the correct format, that the latitude and longitude values ​​are expressed in the correct units, and that there are no format errors or missing values;

[0044] (2) Remove duplicate values: Check whether there are duplicate longitude and latitude points in the data. If there are duplicates, remove them to avoid data redundancy;

[0045] (3) Outlier processing: Detect and process abnormal latitude and longitude data (such as positioning drift, signal loss, values ​​beyond the lane range or obviously wrong values) through GIS desktop tools (such as Google Maps, ArcGIS, QGIS, etc.), especially RTK (Real-Time Kinematic) positioning data, and process it through base station data and differential technology to improve positioning accuracy. Outlier processing can be done through manual inspection or automatic algorithm identification and processing;

[0046] (4) Data accuracy check: Check the accuracy of the data, including the number of decimal places of longitude and latitude, to ensure that the accuracy meets the requirements;

[0047] (5) Coordinate conversion: If the data uses different coordinate systems, coordinate conversion is required to ensure data consistency, such as converting NMEA (National Marine Electronics Association) format to WGS84 coordinates;

[0048] (6) Filling missing values: For missing longitude and latitude data, interpolation or other techniques can be used to fill these missing values ​​to maintain data integrity.

[0049] (7) Data verification: Verify the cleaned data to ensure that the data meets expectations and has no errors or anomalies;

[0050] (8) Data standardization: standardize the format and unit of data to ensure data consistency and comparability;

[0051] (9) Data visualization: Visualize the cleaned data and view the distribution and trend of the data through maps or other charts to further identify potential problems.

[0052] 3. Data storage

[0053] (1) Data sorting: Sort the acquired latitude and longitude data by timestamp;

[0054] (2) Data processing: Connect the longitude and latitude data of each road section arranged in chronological order one by one to form a line segment. By calculating the distance between two adjacent longitude and latitude points, determine whether the two points can be connected into a line segment based on the distance threshold;

[0055] (3) Line segment generation: For continuous longitude and latitude points, the starting and ending points of the line segments are specified according to the intersections, so as to connect multiple longitude and latitude points into line segments;

[0056] (4) Line segment smoothing: Smoothing the generated line segments by using curve fitting or filtering technology to make the line segments smoother and consistent with the actual path;

[0057] (5) High-precision data production: Use GeoTools to read the obtained lane middle trajectory vector line segment, define the coordinate system, and use the geometry processing function of GeoTools to copy and generate the lane, further including the following steps:

[0058] a. Create a GeometryFactory object;

[0059] b. Read the original lane geometry data through Geometry;

[0060] c. Determine the lane width and road surface information based on the data obtained in step 5 of the "preparation work" stage, replicate the lane center line in parallel, and assign the traffic facility code defined in step 4 of the "preparation work" stage to the replicated data;

[0061] d. Generate a buffer (Buffer) using the parallel copied data through the BufferParameters method. The width of the buffer is the lane width, and then assign the traffic facility code defined in step 4 of the "Preparation" stage;

[0062] e. The final data is as follows Figure 4 As shown, it is stored in the PostGres database.

[0063] 4. Data Adjustment

[0064] (1) Use GIS desktop processing tools (such as Google Maps, ArcGIS, QGIS, etc.) to load the generated data, overlay the satellite map, use the satellite map as a reference, modify the vector data that has errors with the actual road surface, adjust the intersection shape and required traffic facilities (such as bus stops, safety islands, zebra crossings, etc.), and edit the geometric data of intersections and zebra crossings and other traffic facilities, such as Figure 5 As shown;

[0065] (2) Use Geoserver to connect to the PostGres database. In the GeoServer management interface, add a data store through the Stores menu and select the appropriate data store type.

[0066] (3) In GeoServer, create a map style through the Styles menu and design a map style that meets your needs, such as defining line color, fill color, etc.

[0067] (4) Finally, publish the map service in GeoServer, select the appropriate publishing method, such as Web Map Service (WMS) or Tile Map Service (WMTS), configure the service parameters and publish the map. Figure 6 shown.

Claims

1. A method for automatically generating a high-precision map based on a vehicle trajectory, characterized in that: The steps include: Step S1 preparation: use centimeter-level, real-time dynamic motion positioning Beidou terminal equipment, and set parameters for the equipment to complete road data collection; Step S2: data cleaning: cleaning the collected road data and removing abnormal data; Step S3: Data storage: The cleaned data is stored in the PostGres database and read and processed; Step S4: Data adjustment: Compare the processed data with the satellite map, adjust the intersection shape and traffic facilities, and render the data to generate a high-precision map.

2. The method for automatically generating a high-precision map according to a vehicle trajectory according to claim 1, characterized in that: The S1 preparation step further includes: S11: Place the selected centimeter-level, real-time dynamic motion positioning Beidou terminal device in the collection vehicle; S12: Turn on the Beidou terminal device, and after the Beidou terminal device locks on at least four satellites, set the collection parameters of the Beidou terminal device, and the collection vehicle travels at a constant speed on the collected road surface, and each time a road section is passed, the attributes of the road section are edited; S13: Define the traffic facility codes, count the traffic facilities on the road and classify them.

3. The method for automatically generating a high-precision map according to a vehicle trajectory according to claim 1, characterized in that: The S2 data cleaning includes the following steps: S21: When the collection is completed, the Beidou terminal device is turned off and the data is exported from the device; S22: converting the data row coordinates to ensure data consistency; S23: cleaning up the collected abnormal longitude and latitude data; S24: Thinning and smoothing the data, connecting the longitude and latitude data in series through the timestamp, and obtaining the trajectory line vector data of the collected road section.

4. The method for automatically generating a high-precision map according to a vehicle trajectory according to claim 1, characterized in that: The S3 data storage includes the following steps: S31: Storing the data in a PostGres database, using Geotools to read the trace vector data of the collected road section, performing parallel line processing according to the attribute data of the collected road section, obtaining lane center line data with the same number of actual lanes, and encoding each piece of data according to the actual traffic equipment according to the traffic facility coding of the preparation S13; S32: based on the lane center line data described in S31, copy the center line data of all lane center lines, central barriers, and vehicle and non-vehicle barriers parallel to the lane center line; S33: the middle line data in S32 are then parallelized according to the collected attributes, and each piece of data is encoded according to the traffic facility code defined in S13 and the actual traffic equipment to obtain the marking facility data, and the marking facility data PostGres is stored in the database; S34: The lane center line data described in S32 is then processed in a buffer zone, and the coded marking facility data in S33 is also brought into the buffer zone to obtain surface traffic facilities, and the generated surface traffic facility data is then stored in a PostGres database.

5. The method for automatically generating a high-precision map according to a vehicle trajectory according to claim 1, characterized in that: The S4 data adjustment includes the following steps: S41: Use a GIS desktop tool to open the data stored in the PostGres database in S34, compare the data with the latest satellite map, and adjust the intersection shape and traffic facilities; S42: Use Geoserver to connect to the database described in S41, read the traffic marking data and surface traffic facility data in the PostGres database, render them in different styles according to different traffic facility coding types, and then publish them to obtain a high-precision map.

Images