High-precision map data processing method and device, storage medium and processor

Abstract

The invention discloses a high-precision map data processing method and device, a storage medium and a processor. In the scheme, on the basis of a preset spacing range and a calculated spacing between adjacent data points in original data of a to-be-processed path, downsampling is carried out on a plurality of data points in the to-be-processed path, and a data point set conforming to the preset spacing range is obtained; performing window division on the path curve based on the curvature of the path curve fitted by the data point set to obtain a plurality of to-be-processed windows; traversing and calculating the curvature of the curve segment in each to-be-processed window, comparing the curvature of the curve segment in each to-be-processed window with a preset curvature, determining the sampling frequency of each to-be-processed window, and performing down-sampling on the curve segment in each to-be-processed window according to the obtained sampling frequency to obtain a down-sampled curve segment in each to-be-processed window; and generating a processed data point set of the to-be-processed path. Compared with the prior art that the data volume of the high-precision map is too large under the condition that the data precision of the high-precision map is ensured, the method has obvious advantages.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method, device, storage medium and processor for processing high-precision map data. Background Art

[0002] HD maps contain a wealth of detailed and precise road information. Compared to traditional maps, HD maps offer higher resolution and richer data. They provide detailed information down to the centimeter level, including road geometry, lane location, traffic signs and traffic light positions, road slopes, and curve curvature. This precise data is crucial for the implementation and operation of technologies such as autonomous vehicles and advanced driver assistance systems, helping vehicles more accurately perceive their surroundings and make safer and more effective driving decisions.

[0003] To ensure rich and accurate information, HD maps require large amounts of data. Generally, a city-level HD map can contain tens or even hundreds of gigabytes (GB). The exact amount depends on the area covered by the map, the level of detail, and the required accuracy.

[0004] In the process of processing high-precision map data, downsampling is usually used to reduce the amount of data. Common downsampling methods include uniform downsampling and bandpass downsampling. Uniform downsampling performs sampling at a set frequency or step size, which directly discards some sampling points. Bandpass downsampling performs sampling by setting a cutoff frequency and only retains frequency components within a specified range. However, both sampling methods have flaws: uniform downsampling is achieved by simply reducing the number of sampling points, which directly leads to information loss. If the downsampling ratio is large, this information loss will be more obvious, affecting the accurate restoration of the signal; bandpass downsampling not only requires the selection of an appropriate filter, but may also cause spectral aliasing, which affects the quality and accuracy of the signal and leads to a decrease in signal fidelity.

[0005] Faced with the problem of excessive amount of high-precision map data, how to process high-precision map data while ensuring its accuracy is a technical problem that needs to be solved urgently. Summary of the Invention

[0006] Based on the above problems, the present application provides a high-precision map data processing method, device, storage medium and processor, the purpose of which is to reduce the amount of high-precision map data while ensuring the accuracy of high-precision map data.

[0007] The embodiments of this application disclose the following technical solutions:

[0008] In a first aspect, the present application provides a method for processing high-precision map data, the method comprising:

[0009] Obtaining raw data of a path to be processed in a high-precision map; the raw data of the path to be processed includes location information of multiple data points on the path to be processed; the location information includes latitude and longitude coordinates, heading, speed, and location acquisition time;

[0010] Based on a preset spacing range and the calculated spacing between adjacent data points in the original data of the path to be processed, downsampling multiple data points in the path to be processed to obtain a set of data points that meet the preset spacing range;

[0011] Fitting a path curve according to the data points in the data point set;

[0012] Dividing the path curve into windows based on the curvature of the path curve to obtain a plurality of windows to be processed;

[0013] The curvature of the curve segments in each window to be processed is traversed and calculated, and the curvature of the curve segments in each window to be processed is compared with a preset curvature, and a sampling frequency of each window to be processed is determined. The curve segments in each window to be processed are downsampled according to the obtained sampling frequency to generate a set of processed data points of the path to be processed; the curve segments in each window to be processed are composed of multiple data points.

[0014] Optionally, after traversing and calculating the curvature of the curve segment in each window to be processed, the method further includes:

[0015] Calculating the difference in curvature between curve segments of two adjacent windows to be processed, and determining whether the difference is within a preset range;

[0016] If the difference is within a preset range, merging the two adjacent windows to be processed into one window to be processed;

[0017] The curvature of the curve segment in the window to be processed is compared with a preset curvature to determine the sampling frequency of the window to be processed, and the curve segment in the window to be processed is downsampled according to the obtained sampling frequency.

[0018] Optionally, traversing and calculating the curvature of the curve segment in each window to be processed, comparing the curvature of the curve segment in each window to be processed with a preset curvature, and determining the sampling frequency of each window to be processed, including:

[0019] If the curvature of the curve segment in the window to be processed is lower than the first preset curvature, setting the sampling frequency of the window to be processed to the first sampling frequency;

[0020] If the curvature of the curve segment in the window to be processed is higher than or equal to the first preset curvature and lower than the second preset curvature, setting the sampling frequency of the window to be processed to the second sampling frequency; the first sampling frequency is lower than the second sampling frequency;

[0021] If the curvature of the curve segment in the window to be processed is higher than or equal to the second preset curvature, the sampling frequency of the window to be processed is set to a third sampling frequency; and the second sampling frequency is lower than the third sampling frequency.

[0022] Optionally, calculating the distance between adjacent data points in the original data of the path to be processed includes:

[0023] If the longitude and latitude coordinates of the first data point are (o i ,p i ), the latitude and longitude coordinates of the second data point are (o i-1 ,p i-1 ), then calculate the distance L between the first data point and the second data point i The specific formula is:

[0024] L i =√(o i -o i-1 ) 2 +(p i -p i-1 ) 2 ;

[0025] The first data point and the second data point are data points located adjacent to each other in the original data of the path to be processed.

[0026] A second aspect of the present application provides a high-precision map data processing device, the device comprising:

[0027] A data acquisition module is used to acquire original data of the path to be processed; the original data of the path to be processed includes position information of multiple data points;

[0028] a first downsampling processing module, configured to downsample a plurality of data points in the path to be processed based on a preset spacing range and a calculated spacing between adjacent data points in the original data of the path to be processed, to obtain a set of data points that meets the preset spacing range;

[0029] The second downsampling processing module is used to fit a path curve based on the data points in the data point set; divide the path curve into windows based on the curvature of the path curve to obtain multiple windows to be processed; traversally calculate the curvature of the curve segment in each window to be processed, compare the curvature of the curve segment in each window to be processed with a preset curvature, determine the sampling frequency of each window to be processed, downsample the curve segment in each window to be processed according to the obtained sampling frequency, and generate a processed data point set for the path to be processed; the curve segment in each window to be processed is composed of multiple data points.

[0030] In an optional implementation, the apparatus further includes: a window merging module;

[0031] The window merging module is used to calculate the difference in curvature of the curve segments of two adjacent windows to be processed, and determine whether the difference is within a preset range;

[0032] If the difference is within a preset range, the two adjacent windows to be processed are merged into one window to be processed.

[0033] In an optional implementation, the second downsampling processing module is specifically configured to:

[0034] If the curvature of the curve segment in the window to be processed is lower than the first preset curvature, setting the sampling frequency of the window to be processed to the first sampling frequency;

[0035] If the curvature of the curve segment in the window to be processed is higher than or equal to the first preset curvature and lower than the second preset curvature, setting the sampling frequency of the window to be processed to the second sampling frequency; the first sampling frequency is lower than the second sampling frequency;

[0036] If the curvature of the curve segment in the window to be processed is higher than or equal to the second preset curvature, the sampling frequency of the window to be processed is set to a third sampling frequency; and the second sampling frequency is lower than the third sampling frequency.

[0037] In an optional implementation, the first downsampling processing module is specifically configured to:

[0038] If the longitude and latitude coordinates of the first data point are (o i ,p i ), the latitude and longitude coordinates of the second data point are (o i-1 ,p i-1 ), then calculate the distance L between the first data point and the second data point i The specific formula is:

[0039]

[0040] The first data point and the second data point are data points located adjacent to each other in the original data of the path to be processed.

[0041] The third aspect of the present application provides a computer-readable storage medium, which stores a computer program. When the program is executed by a processor, the high-precision map data processing method provided in any implementation method of the first aspect is implemented.

[0042] A fourth aspect of the present application provides a processor for running a computer program, which, when running, executes the high-precision map data processing method provided in any implementation of the first aspect.

[0043] Compared with the existing technology, this application has the following beneficial effects:

[0044] The high-precision map data processing method provided in the present application performs two downsamplings on the original data of the path to be processed in the acquired high-precision map. First, based on the preset spacing range and the calculated spacing between adjacent data points in the original data of the path to be processed, multiple data points in the path to be processed are downsampled to obtain a set of data points that meet the preset spacing range. The original data is downsampled for the first time within the preset spacing range to remove data points outside the preset spacing range, effectively reducing the number of sampling points and removing noise points. While reducing the amount of data, it ensures that each sampling point truly expresses the path, and completely restores the path to be processed in the high-precision map on a spatial scale.

[0045] Then, a path curve is fitted based on the data points in the data point set; the path curve is windowed based on the curvature of the path curve to obtain multiple windows to be processed; the curvature of the curve segments in each window to be processed is calculated, and the curvature of the curve segments in each window to be processed is compared with a preset curvature to determine the sampling frequency of each window to be processed. The curve segments in each window to be processed are downsampled based on the obtained sampling frequency to generate a processed data point set for the path to be processed. The sampling frequency is set based on the adaptability of the curvature of the curve segments in each window to be processed. After the second downsampling, the amount of data at the sampling points is further reduced, while ensuring the accuracy of the path representation of the generated data point set for the path to be processed, thereby effectively solving the problems encountered in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0047] Figure 1 A flowchart of a high-precision map data processing method provided in an embodiment of the present application;

[0048] Figure 2 This is an effect diagram of a path curve window division provided in an embodiment of the present application;

[0049] Figure 3 A flowchart of another high-precision map data processing method provided in an embodiment of the present application;

[0050] Figure 4 This is a rendering of the effect of merging adjacent windows of a path curve provided in an embodiment of the present application;

[0051] Figure 5 A schematic structural diagram of a high-precision map data processing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0052] As described above, in the current HD map data processing process, the problem of excessive HD map data volume exists, and the accuracy of the HD map data cannot be guaranteed.

[0053] In view of the above problems, the inventors have proposed a high-precision map data processing method, device, storage medium and processor after research, which obtain the original data of the path to be processed in the high-precision map; based on a preset spacing range and the calculated spacing between adjacent data points in the original data of the path to be processed, downsample multiple data points in the path to be processed to obtain a set of data points that meets the preset spacing range; fit a path curve according to the data points in the data point set; divide the path curve into windows based on the curvature of the path curve to obtain multiple windows to be processed; traverse and calculate the curvature of the curve segment in each window to be processed, compare the curvature of the curve segment in each window to be processed with the preset curvature, determine the sampling frequency of each window to be processed, downsample the curve segment in each window to be processed according to the obtained sampling frequency, and generate a processed data point set for the path to be processed.

[0054] In order to help those skilled in the art better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying 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 those skilled in the art without creative work are within the scope of protection of this application.

[0055] See also Figure 1 , which is a flow chart of a high-precision map data processing method provided by an embodiment of the present application. Figure 1 As shown, the method includes the following steps:

[0056] S101. Obtain the original data of the path to be processed in the high-precision map.

[0057] The original data of the path to be processed includes location information of multiple data points on the path to be processed; the location information includes latitude and longitude coordinates, heading, speed and location acquisition time.

[0058] High-precision maps provide highly accurate positioning information and richer map content, primarily serving autonomous driving. Map data acquisition equipment uses a Global Navigation Satellite System (GNSS) board in conjunction with a Real-Time Kinematic (RTK) base station to obtain all-weather positioning information from the GNSS. This raw positioning data includes the current latitude and longitude coordinates, heading, speed, and time of acquisition.

[0059] For example, as shown in Table 1, the table shown in Table 1 contains multiple columns, each column represents a different data type, roll_imu records the roll angle of the inertial measurement unit (IMU), and all values ​​are 0; time_stamp is a timestamp, indicating the time point of data collection; gps.pos_x and gps.pos_y are the longitude and latitude coordinates of the global positioning system (GPS) position; gps.speed indicates the speed of the GPS; gps.azimuth is the azimuth of the GPS, indicating the angle of the moving direction. From the data recorded in the table, the data in the first few rows remain almost unchanged in timestamp, GPS position, speed and azimuth, until the second to last row, the GPS speed and azimuth change, which may mean that the device started to move or changed direction at a certain moment. This type of table is often used to record sensor data, such as self-driving cars, drones or other systems that require precise position and motion information.

[0060] Table 1

[0061]

[0062] S102 : Based on a preset spacing range and the calculated spacing between adjacent data points in the original data of the path to be processed, downsample multiple data points in the path to be processed to obtain a set of data points that meets the preset spacing range.

[0063] In a possible implementation, the distance between adjacent data points in the original data of the path to be processed is calculated as follows:

[0064] If the latitude and longitude coordinates of the first data point are (o i ,p i ), the latitude and longitude coordinates of the second data point are (o i-1 ,p i-1 ), then calculate the distance L between the first data point and the second data point i The specific formula is:

[0065]

[0066] The first data point and the second data point are data points located adjacent to each other in the original data of the path to be processed.

[0067] By setting a preset spacing range and performing sampling, a set of data points with a specific spacing can be obtained. The preset spacing range is defined by an upper spacing limit and a lower spacing limit. The upper spacing limit of the preset spacing range is used to remove some sudden noise and data points that deviate from the original data path trajectory, while the lower spacing limit of the preset spacing range is used to filter out redundant and repeated data points that are too dense in the path of the original data. In other words, by setting a preset spacing range, data points outside the preset spacing range can be removed, effectively reducing the number of sampling points and removing noise points. While reducing the amount of data, it ensures that each sampling point truly expresses the path, and completely restores the path to be processed in the high-precision map on a spatial scale.

[0068] S103: Fitting a path curve according to the data points in the data point set.

[0069] To further downsample the HD map data, a path curve needs to be fitted based on the data points in the data point set. In practical applications, path fitting can typically be performed using one of a variety of methods, including but not limited to linear regression, polynomial regression, spline interpolation, or Gaussian process regression.

[0070] S104 : Divide the path curve into windows based on the curvature of the path curve to obtain a plurality of windows to be processed.

[0071] In a possible implementation, each window to be processed contains the same number of data points.

[0072] The number of data points that each window should contain is determined by calculating the curvature of the curve, for example, Figure 2 As shown, for Figure 2 The displayed curve path can be set to n windows, each of which contains 8 data points representing the longitude and latitude coordinates.

[0073] S105. Traverse and calculate the curvature of the curve segments in each window to be processed, compare the curvature of the curve segments in each window to be processed with the preset curvature, determine the sampling frequency of each window to be processed, downsample the curve segments in each window to be processed according to the obtained sampling frequency, and generate a processed data point set of the path to be processed.

[0074] Wherein, the curve segment in each window to be processed is composed of multiple data points.

[0075] In a possible implementation, determining the sampling frequency of each window to be processed includes:

[0076] If the curvature of the curve segment in the window to be processed is lower than the first preset curvature, setting the sampling frequency of the window to be processed to the first sampling frequency;

[0077] If the curvature of the curve segment in the window to be processed is higher than or equal to the first preset curvature and lower than the second preset curvature, setting the sampling frequency of the window to be processed to the second sampling frequency; the first sampling frequency is lower than the second sampling frequency;

[0078] If the curvature of the curve segment in the window to be processed is higher than or equal to the second preset curvature, the sampling frequency of the window to be processed is set to a third sampling frequency; and the second sampling frequency is lower than the third sampling frequency.

[0079] The implementation method here is only an example. In actual applications, different frequency setting intervals may be set, for example, two or four.

[0080] Curvature is a crucial aspect of a path's geometric significance. On straight roads with low curvature, reducing the sampling frequency ensures that data points accurately represent the original path while significantly reducing the amount of sampled data. On complex, curved roads with high curvature, increasing the sampling frequency ensures that the path representation remains undistorted. This improves the downsampling efficiency of the overall positioning data and ensures the accuracy of the data's representation of the path.

[0081] The HD map data processing method provided in this embodiment downsamples the raw data of the path to be processed in the acquired HD map twice. The first downsampling is performed within a preset spacing range, removing data points outside the preset spacing range. This effectively reduces the number of sampling points and eliminates noise points. While reducing the data volume, it ensures that each sampling point accurately represents the path, completely restoring the HD map path at the spatial scale. A second downsampling is then performed by calculating curvature and windowing. The sampling frequency is set based on the curvature adaptability of the curve segments in each window to be processed. This second downsampling further reduces the amount of sampling point data while ensuring the accuracy of the path representation by the generated set of data points for the path to be processed, effectively resolving the challenges faced by existing technologies.

[0082] See also Figure 3 , which is a flow chart of the high-precision map data processing method provided by the embodiment of this application. Based on the previous embodiment, the high-precision map data processing method is improved by adding a step of merging the windows to be processed. Figure 3 As shown, the method includes the following steps:

[0083] S201. Obtain the original data of the path to be processed in the high-precision map.

[0084] The original data of the path to be processed includes location information of multiple data points on the path to be processed; the location information includes latitude and longitude coordinates, heading, speed and location acquisition time.

[0085] Use the GNSS board of the map data acquisition device in combination with the RTK base station to obtain all-weather positioning information provided by the global navigation satellite system and obtain the raw data of the positioning information, including the latitude and longitude coordinates of the current position, heading, speed, and position acquisition time.

[0086] S202 : Based on a preset spacing range and the calculated spacing between adjacent data points in the original data of the path to be processed, downsample multiple data points in the path to be processed to obtain a set of data points that meets the preset spacing range.

[0087] In one possible implementation, the distance between adjacent data points in the original data of the path to be processed is calculated as follows:

[0088] If the longitude and latitude coordinates of the first data point are (o i ,p i ), the latitude and longitude coordinates of the second data point are (o i-1 ,p i-1 ), then calculate the distance L between the first data point and the second data point i The specific formula is:

[0089]

[0090] The first data point and the second data point are data points located adjacent to each other in the original data of the path to be processed.

[0091] Downsampling is a data processing technique used to reduce the number of data points while preserving the characteristics of the original data as much as possible. Downsampling the original data within a preset spacing range effectively reduces the number of sampling points and removes noise. While reducing the data volume, it ensures that each sampling point accurately represents the positioning, completely recreating the path to be processed as captured by the data acquisition device at the spatial scale.

[0092] S203: Fit a path curve according to the data points in the data point set.

[0093] The set of data points obtained in the previous step that meet the preset spacing range is expressed as a curve, and a path curve is fitted according to the data points in the data point set. Path fitting can usually be performed using one of a variety of methods, including but not limited to linear regression, polynomial regression, spline interpolation, or Gaussian process regression.

[0094] S204 : Divide the path curve into windows based on the curvature of the path curve to obtain a plurality of windows to be processed.

[0095] In a possible implementation, each window to be processed contains the same number of data points.

[0096] For the path curve fitted in the previous step, n windows can be set, each window containing m data points representing longitude and latitude coordinates.

[0097] S205. Traverse and calculate the curvature of the curve segments in each window to be processed, calculate the difference in curvature of the curve segments of two adjacent windows to be processed, and determine whether the difference is within a preset range; if the difference is within the preset range, merge the two adjacent windows to be processed into one window to be processed.

[0098] like Figure 4 As shown, Figure 4 This example demonstrates the effect of merging two adjacent processing windows on a curved path. For example, merging the first and second windows before sampling effectively reduces the number of sampling points and ensures that each point accurately represents the path. Merging adjacent windows based on a preset curvature difference threshold simplifies processing and reduces data volume.

[0099] S206. Compare the curvature of the curve segments in each window to be processed with a preset curvature, determine the sampling frequency of each window to be processed, downsample the curve segments in each window to be processed according to the obtained sampling frequency, and generate a processed data point set of the path to be processed.

[0100] In a possible implementation, the curve segment in each window to be processed is composed of multiple data points.

[0101] If the curvature of the curve segment in the window to be processed is lower than the first preset curvature, setting the sampling frequency of the window to be processed to the first sampling frequency;

[0102] If the curvature of the curve segment in the window to be processed is higher than or equal to the first preset curvature and lower than the second preset curvature, setting the sampling frequency of the window to be processed to the second sampling frequency; the first sampling frequency is lower than the second sampling frequency;

[0103] If the curvature of the curve segment in the window to be processed is higher than or equal to the second preset curvature, the sampling frequency of the window to be processed is set to a third sampling frequency; and the second sampling frequency is lower than the third sampling frequency.

[0104] The implementation method here is only an example. In actual applications, different frequency setting intervals may be set, for example, two or four.

[0105] The HD map data processing method provided in this embodiment downsamples the raw data of the path to be processed in the acquired HD map twice. The first downsample is performed within a preset spacing range, removing data points outside the preset spacing range, effectively reducing the number of sampling points and eliminating noise points. While reducing the data volume, this method ensures that each sampling point accurately represents the path and fully restores the path to be processed at the spatial scale. After window division, adjacent windows are judged and merged according to preset rules, and the merged windows to be processed are then downsampled again. This method further reduces the number of data points and the amount of data at the sampling points, while ensuring the accuracy of the path representation of the generated data point set for the path to be processed.

[0106] Based on the high-precision map data processing method introduced in the previous embodiment, the present application also provides a high-precision map data processing device. Figure 5 Figure 2 is a schematic diagram of the structure of the device. Figure 5 As shown, the high-precision map data processing device includes:

[0107] The data acquisition module 501 is used to acquire the original data of the path to be processed; the original data of the path to be processed includes the position information of multiple data points;

[0108] A first downsampling processing module 502 is configured to downsample a plurality of data points in the path to be processed based on a preset spacing range and a calculated spacing between adjacent data points in the original data of the path to be processed, to obtain a set of data points that meets the preset spacing range;

[0109] The second downsampling processing module 503 is used to fit a path curve based on the data points in the data point set; divide the path curve into windows based on the curvature of the path curve to obtain multiple windows to be processed; traversally calculate the curvature of the curve segment in each window to be processed, compare the curvature of the curve segment in each window to be processed with a preset curvature, determine the sampling frequency of each window to be processed, downsample the curve segment in each window to be processed according to the obtained sampling frequency, and generate a processed data point set for the path to be processed; the curve segment in each window to be processed is composed of multiple data points.

[0110] In an optional embodiment, the apparatus further comprises: a window merging module;

[0111] The window merging module is used to calculate the difference in curvature of the curve segments of two adjacent windows to be processed, and determine whether the difference is within a preset range;

[0112] If the difference is within a preset range, the two adjacent windows to be processed are merged into one window to be processed.

[0113] In an optional implementation manner, the second downsampling processing module is specifically configured to:

[0114] If the curvature of the curve segment in the window to be processed is lower than the first preset curvature, setting the sampling frequency of the window to be processed to the first sampling frequency;

[0115] If the curvature of the curve segment in the window to be processed is higher than or equal to the first preset curvature and lower than the second preset curvature, setting the sampling frequency of the window to be processed to the second sampling frequency; the first sampling frequency is lower than the second sampling frequency;

[0116] If the curvature of the curve segment in the window to be processed is higher than or equal to the second preset curvature, the sampling frequency of the window to be processed is set to a third sampling frequency; and the second sampling frequency is lower than the third sampling frequency.

[0117] In an optional implementation manner, the first downsampling processing module is specifically configured to:

[0118] If the longitude and latitude coordinates of the first data point are (o i ,p i), the latitude and longitude coordinates of the second data point are (o i-1 ,p i-1 ), then calculate the distance L between the first data point and the second data point i The specific formula is:

[0119]

[0120] The first data point and the second data point are data points located adjacent to each other in the original data of the path to be processed.

[0121] In addition, an embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the program is executed by a processor, a high-precision map data processing method as described in any of the method embodiments is implemented.

[0122] In addition, an embodiment of the present application also provides a processor for running a computer program, which, when running, executes the high-precision map data processing method described in any implementation method of the aforementioned method embodiment.

[0123] It should be noted that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The device embodiment described above is merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components indicated as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without expending creative work.

[0124] The above is merely one specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A high-precision map data processing method, characterized in that: include: Obtain the original data of the path to be processed in the high-precision map; The original data of the path to be processed includes location information of a plurality of data points on the path to be processed; The location information includes latitude and longitude coordinates, heading, speed and location acquisition time; Based on a preset spacing range and the calculated spacing between adjacent data points in the original data of the path to be processed, downsampling a plurality of data points in the path to be processed to obtain a set of data points that meet the preset spacing range; Fitting a path curve according to the data points in the data point set; Dividing the path curve into windows based on the curvature of the path curve to obtain a plurality of windows to be processed; The curvature of the curve segment in each window to be processed is calculated by traversing, and the curvature of the curve segment in each window to be processed is compared with the preset curvature, and the sampling frequency of each window to be processed is determined. The curve segment in each window to be processed is downsampled according to the obtained sampling frequency to generate a processed data point set of the path to be processed; the curve segment in each window to be processed is composed of multiple data points.

2. The method according to claim 1, characterized in that After traversing and calculating the curvature of the curve segment in each window to be processed, it also includes: Calculating the difference in curvature of curve segments of two adjacent windows to be processed, and determining whether the difference is within a preset range; If the difference is within a preset range, the two adjacent windows to be processed are merged into one window to be processed.

3. The method according to claim 1, characterized in that The curvature of the curve segment in each window to be processed is calculated by traversing, and the curvature of the curve segment in each window to be processed is compared with the preset curvature to determine the sampling frequency of each window to be processed, including: If the curvature of the curve segment in the window to be processed is lower than a first preset curvature, setting the sampling frequency of the window to be processed to the first sampling frequency; If the curvature of the curve segment in the window to be processed is higher than or equal to the first preset curvature and lower than the second preset curvature, the sampling frequency of the window to be processed is set to the second sampling frequency; the first sampling frequency is lower than the second sampling frequency; If the curvature of the curve segment in the window to be processed is higher than or equal to the second preset curvature, the sampling frequency of the window to be processed is set to the third sampling frequency; the second sampling frequency is lower than the third sampling frequency.

4. The method according to claim 1, characterized in that: Calculating the distance between adjacent data points in the original data of the path to be processed includes: If the longitude and latitude coordinates of the first data point are (o i ,p i ), the latitude and longitude coordinates of the second data point are (o i-1 ,p i-1 ), then calculate the distance L between the first data point and the second data point i The specific formula is: The first data point and the second data point are data points that are adjacent in position in the original data of the path to be processed.

5. A high-precision map data processing device, characterized in that: include: A data acquisition module, used to acquire the original data of the path to be processed; The original data of the path to be processed includes location information of multiple data points; A first downsampling processing module, configured to downsample a plurality of data points in the path to be processed based on a preset spacing range and a calculated spacing between adjacent data points in the original data of the path to be processed, so as to obtain a set of data points that conform to the preset spacing range; A second downsampling processing module is used to fit a path curve according to the data points in the data point set; divide the path curve into windows based on the curvature of the path curve to obtain a plurality of windows to be processed; each window to be processed contains a consistent number of data points; The curvature of the curve segment in each window to be processed is calculated by traversing, and the curvature of the curve segment in each window to be processed is compared with the preset curvature, and the sampling frequency of each window to be processed is determined. The curve segment in each window to be processed is downsampled according to the obtained sampling frequency to generate a processed data point set of the path to be processed; the curve segment in each window to be processed is composed of multiple data points.

6. The device according to claim 5, characterized in that The device also includes: a window merging module; The window merging module is used to calculate the difference in curvature of the curve segments of two adjacent windows to be processed, and determine whether the difference is within a preset range; If the difference is within a preset range, the two adjacent windows to be processed are merged into one window to be processed.

7. The device according to claim 5, characterized in that The second downsampling processing module is specifically used for: If the curvature of the curve segment in the window to be processed is lower than the first preset curvature, setting the sampling frequency of the window to be processed to the first sampling frequency; If the curvature of the curve segment in the window to be processed is higher than or equal to the first preset curvature and lower than the second preset curvature, the sampling frequency of the window to be processed is set to the second sampling frequency; the first sampling frequency is lower than the second sampling frequency; If the curvature of the curve segment in the window to be processed is higher than or equal to the second preset curvature, the sampling frequency of the window to be processed is set to the third sampling frequency; the second sampling frequency is lower than the third sampling frequency.

8. The device according to claim 5, characterized in that The first downsampling processing module is specifically used for: If the longitude and latitude coordinates of the first data point are (o i ,p i ), the latitude and longitude coordinates of the second data point are (o i-1 ,p i-1 ), then calculate the distance L between the first data point and the second data point i The specific formula is: The first data point and the second data point are data points that are adjacent in position in the original data of the path to be processed.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the program is executed by the processor, the high-precision map data processing method as described in any one of claims 1 to 4 is implemented.

10. A processor, characterized in that: Used to run a computer program, which, when running, executes the high-precision map data processing method as described in any one of claims 1-4.

Images