Three-dimensional road center line generation method and device, equipment and storage medium

Abstract

The embodiment of the invention provides a three-dimensional road center line generation method and device, equipment and a storage medium, and relates to the technical field of vehicles. The method comprises the following steps: performing analysis processing on obtained curve rule grid CRG data of a target road to obtain road related data of the target road; establishing a three-dimensional mapping relational expression according to the road related data; determining an initial center line of the target road according to the transverse deviation range of the target road; and according to the three-dimensional mapping relational expression, performing iterative updating processing on at least part of the initial center points included in the initial center line to obtain updated three-dimensional center points, and according to the updated three-dimensional center points, determining the three-dimensional center line of the target road. According to the method, the elevation data can be added in the iterative calculation process of the center line, and the accuracy of the determined three-dimensional center line is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicles, in particular to a three-dimensional road centerline generation method, device, equipment and storage medium. BACKGROUND

[0002] CRG (Curved Regular Grid) is a data format for describing complex road surface models. It represents the road surface as a grid with fixed sampling intervals (u, v), where each grid point contains three-dimensional coordinate (X, Y, Z) information. This allows CRG to accurately represent the planar shape and elevation changes of the road surface.

[0003] In related technologies, when determining the road centerline based on CRG data, the centerline is generally determined in the two-dimensional plane (X, Y), ignoring the Z-axis data (i.e., elevation data).

[0004] This results in a centerline that cannot accurately reflect the geometric characteristics of the road in three-dimensional space, affecting the accuracy of the road centerline, and reducing the accuracy and reliability of the high-definition map generated based on the determined centerline, and affecting the accuracy of autonomous driving decisions. SUMMARY

[0005] The embodiments of the present application provide a three-dimensional road centerline generation method, device, equipment and storage medium, which can add elevation data in the iterative calculation process of the centerline, improving the accuracy of the determined three-dimensional centerline.

[0006] In a first aspect, the embodiments of the present application provide a three-dimensional road centerline generation method, comprising:

[0007] parsing and processing the obtained CRG (Curved Regular Grid) data of the target road to obtain road-related data of the target road;

[0008] establishing a three-dimensional mapping relationship based on the road-related data, wherein the three-dimensional mapping relationship is used to map the road-related data from a two-dimensional parameter domain to a three-dimensional space domain;

[0009] determining an initial centerline of the target road based on the lateral offset range of the target road, wherein the initial centerline includes an initial center point under each longitudinal interface of the target road;

[0010] iteratively updating at least part of the initial center points included in the initial centerline based on the three-dimensional mapping relationship to obtain updated three-dimensional center points, and determining a three-dimensional centerline of the target road based on the updated three-dimensional center points.

[0011] In a possible implementation, the longitudinal interface under the lateral offset range comprises a first boundary point and a second boundary point; and the iterative updating of at least part of the initial center points comprised in the initial center line according to the three-dimensional mapping relationship to obtain updated three-dimensional center points comprises:

[0012] determining a longitudinal interface to be updated, and repeatedly performing the following process for each initial center point under the longitudinal interface to be updated: determining a unit tangent vector at a position of the initial center point according to the three-dimensional mapping relationship; determining a first projection parameter of the first boundary point under the unit tangent vector and a second projection parameter of the second boundary point under the unit tangent vector; if it is determined that a projection difference between the first projection parameter and the second projection parameter does not meet an iterative convergence requirement, updating the initial center point according to the unit tangent vector and the three-dimensional mapping relationship to obtain an updated initial center point;

[0013] if it is determined that the projection difference meets the iterative convergence requirement, determining the initial center point as the updated three-dimensional center point.

[0014] In a possible implementation, the initial center point comprises lateral data under the longitudinal interface; and the updating of the initial center point according to the unit tangent vector and the three-dimensional mapping relationship to obtain an updated initial center point comprises:

[0015] determining new lateral data according to the unit tangent vector and the initial center point;

[0016] determining an original three-dimensional coordinate point according to the initial center point and the three-dimensional mapping relationship, and determining a new three-dimensional coordinate point according to the new lateral data and the three-dimensional mapping relationship;

[0017] determining the updated initial center point according to a difference between the new three-dimensional coordinate point and the original three-dimensional coordinate point.

[0018] In a possible implementation, the determination of the longitudinal interface to be updated comprises:

[0019] determining road slope information at a position of an initial center point in the initial center line;

[0020] determining a center point iterative step length according to the road slope information, and determining the longitudinal interface to be updated according to the center point iterative step length.

[0021] In a possible implementation, the road-related data at least comprises longitudinal data, lateral data, elevation data, heading angle data, and reference line data; and the establishment of the three-dimensional mapping relationship according to the road-related data comprises:

[0022] determining first coordinate data according to the heading angle data, the lateral data, and a projection of the longitudinal data on a first coordinate axis of the reference line data;

[0023] determining second coordinate data according to the heading angle data, the lateral data, and a projection of the longitudinal data on a second coordinate axis of the reference line data;

[0024] determining third coordinate data according to the elevation data and a first interpolation manner;

[0025] determining the three-dimensional mapping relationship according to the first coordinate data, the second coordinate data, and the third coordinate data.

[0026] In a possible implementation, before determining the third coordinate data according to the elevation data and the first interpolation manner, the method further includes:

[0027] determining data distribution information of the elevation data;

[0028] determining abnormal data in the elevation data according to the data distribution information;

[0029] updating the abnormal data according to a second interpolation manner to obtain processed elevation data.

[0030] In a possible implementation, the method further includes:

[0031] determining a heading angle integral error of the reference line data according to a start position and an end position in the reference line data;

[0032] if it is determined that the heading angle integral error meets heading angle correction requirements, determining a correction coefficient according to the heading angle integral error;

[0033] correcting the first coordinate data and the second coordinate data according to the correction coefficient.

[0034] In a second aspect, an embodiment of the present application provides a three-dimensional road center line generation device, including:

[0035] a parsing unit configured to parse CRG data of a target road to obtain road-related data of the target road;

[0036] a mapping unit configured to establish a three-dimensional mapping relationship according to the road-related data; the three-dimensional mapping relationship is used to map the road-related data from a two-dimensional parameter domain to a three-dimensional space domain.

[0037] determining unit configured to determine an initial center line of the target road according to a lateral offset range of the target road, wherein the initial center line comprises an initial center point under each longitudinal interface of the target road;

[0038] iterating unit configured to perform iterative updating processing on at least part of the initial center points comprised in the initial center line according to the three-dimensional mapping relationship, to obtain updated three-dimensional center points, and determine a three-dimensional center line of the target road according to the updated three-dimensional center points.

[0039] In a possible implementation, the longitudinal interface comprises a first boundary point and a second boundary point under the lateral offset range; at this time, the iterating unit is configured to:

[0040] determine a longitudinal interface to be updated, and repeatedly perform the following process on the initial center point under each longitudinal interface to be updated: determine a unit tangent vector at a position of the initial center point according to the three-dimensional mapping relationship; determine a first projection parameter of the first boundary point under the unit tangent vector, and determine a second projection parameter of the second boundary point under the unit tangent vector; if it is determined that a projection difference between the first projection parameter and the second projection parameter does not meet an iterative convergence requirement, update the initial center point according to the unit tangent vector and the three-dimensional mapping relationship, to obtain an updated initial center point;

[0041] if it is determined that the projection difference meets the iterative convergence requirement, determine the initial center point as the updated three-dimensional center point.

[0042] In a possible implementation, the initial center point comprises lateral data under the longitudinal interface; at this time, the iterating unit is configured to:

[0043] determine new lateral data according to the unit tangent vector and the initial center point;

[0044] determine an original three-dimensional coordinate point according to the initial center point and the three-dimensional mapping relationship, and determine a new three-dimensional coordinate point according to the new lateral data and the three-dimensional mapping relationship;

[0045] determine the updated initial center point according to a difference between the new three-dimensional coordinate point and the original three-dimensional coordinate point.

[0046] In a possible implementation, the iterating unit is configured to:

[0047] determine road slope information at a position of an initial center point in the initial center line;

[0048] Based on the road slope information, the center point iteration step size is determined, and based on the center point iteration step size, the longitudinal interface to be updated is determined.

[0049] In one possible implementation, the road-related data includes at least longitudinal data, lateral data, elevation data, heading angle data, and reference line data; in this case, the mapping unit is used for:

[0050] The first coordinate data is determined based on the heading angle data, the lateral data, and the projection of the longitudinal data onto the first coordinate axis of the reference line data;

[0051] The second coordinate data is determined based on the heading angle data, the lateral data, and the projection of the longitudinal data onto the second coordinate axis of the reference line data;

[0052] Based on the elevation data and the first interpolation method, determine the third coordinate data;

[0053] The three-dimensional mapping relationship is determined based on the first coordinate data, the second coordinate data, and the third coordinate data.

[0054] In one possible implementation, before determining the third coordinate data based on the elevation data and the first interpolation method, the device is further configured to:

[0055] Determine the data distribution information of the elevation data;

[0056] Based on the data distribution information, abnormal data in the elevation data are identified;

[0057] The abnormal data is updated according to the second interpolation method to obtain the processed elevation data.

[0058] In one possible implementation, the device is also used for:

[0059] The heading angle integral error of the reference line data is determined based on the start and end positions in the reference line data.

[0060] If it is determined that the integral error of the heading angle meets the heading angle correction requirements, then the correction coefficient is determined based on the integral error of the heading angle.

[0061] The first coordinate data and the second coordinate data are corrected according to the correction coefficient.

[0062] Thirdly, embodiments of this application provide a computer device, including: a memory and a processor;

[0063] The memory stores computer-executed instructions;

[0064] The processor executes computer execution instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0065] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0066] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0067] The three-dimensional road centerline generation method, apparatus, device, and storage medium provided in this application can determine the road-related data of the target road based on the curve rule grid (CRG) data of the target road, and establish a three-dimensional mapping relationship based on the road-related data. This allows the road-related data to be mapped from a two-dimensional parameter domain to a three-dimensional spatial domain, providing a three-dimensional data foundation for subsequent three-dimensional centerline generation and ensuring that the elevation data of the target road can participate in the calculation process of the three-dimensional centerline generation. Specifically, the initial centerline of the target road can be determined based on the lateral offset range of the target road. Then, according to the three-dimensional mapping relationship, at least a portion of the initial center points included in the initial centerline are iteratively updated to obtain updated three-dimensional center points. This implementation ensures that the elevation values ​​of the determined three-dimensional center points are consistent with the actual road, thereby improving the accuracy of the determined three-dimensional center points. After determining the three-dimensional centerline of the target road based on the updated three-dimensional center points, the three-dimensional geometric accuracy of the three-dimensional centerline can be improved. Furthermore, when generating a high-precision map based on the three-dimensional centerline obtained according to the above embodiments, the reliability of the high-precision map can be improved; or, when making autonomous driving decisions based on the three-dimensional centerline, the accuracy of autonomous driving control decisions can be improved, thereby improving vehicle performance. Attached Figure Description

[0068] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0069] Figure 1 A flowchart illustrating a method for generating a three-dimensional road centerline provided in this application embodiment;

[0070] Figure 2 A flowchart illustrating another method for generating a three-dimensional road centerline provided in this application embodiment;

[0071] Figure 3 A schematic diagram illustrating the implementation process of a three-dimensional road centerline generation method provided in this application embodiment;

[0072] Figure 4 A schematic diagram of a three-dimensional road centerline generation device provided in this application embodiment;

[0073] Figure 5 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application.

[0074] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0075] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0076] With the rapid development of autonomous driving technology, the reliance on high-definition maps (HD Maps) is increasing. High-definition maps not only need to provide precise lane-level planar location, but also accurate three-dimensional geometric information such as elevation, slope, and curvature, so that vehicles can perform precise positioning and path planning.

[0077] In related technologies, when determining the road centerline based on CRG data, the centerline is generally determined on a two-dimensional plane (X, Y), ignoring the Z-axis data (i.e., elevation data). In practice, the three-dimensional boundary of the CRG pavement can first be projected onto a two-dimensional plane, and the two-dimensional centerline can be determined on this plane using a centerline extraction algorithm. Then, the obtained two-dimensional centerline is remapped back into three-dimensional space; that is, after interpolating the elevation data, the elevation value of the two-dimensional centerline is determined, thereby determining the three-dimensional centerline.

[0078] In this implementation, elevation information is not directly involved in the centerline calculation process, resulting in the extracted centerline failing to accurately reflect the road's geometric characteristics in three-dimensional space. For example, in areas with significant slope variations (e.g., bridge approach, slope top, slope bottom), the elevation value of the two-dimensional centerline deviates from the actual elevation of the road's center point. This makes it impossible to accurately determine key information such as the centerline's three-dimensional slope and inclination angle, thereby reducing the reliability of the generated high-precision map and affecting the accuracy of the autonomous vehicle's control decisions.

[0079] Therefore, a new method for determining the three-dimensional centerline is urgently needed.

[0080] The method for generating three-dimensional road centerlines provided in this application can establish a three-dimensional mapping relationship based on road-related data, and perform iterative processing based on the three-dimensional mapping relationship to obtain the three-dimensional centerline. This allows elevation data to directly participate in the iterative calculation process of the three-dimensional centerline, thereby making the generated three-dimensional centerline more closely match the actual road surface and improving the accuracy of the determined three-dimensional centerline.

[0081] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0082] Figure 1 This is a flowchart illustrating a method for generating a three-dimensional road centerline according to an embodiment of this application, as shown below. Figure 1 As shown, the method includes:

[0083] S101. The obtained curve regular grid (CRG) data of the target road is parsed and processed to obtain the road-related data of the target road.

[0084] In one example, road-related data for a target road can be obtained by parsing the header information of the curve rule raster (CRG) data.

[0085] Optionally, the road surface data block in the curve regular raster (CRG) data may include road reference line data, which may include, but is not limited to: heading angle data, lateral raster values, longitudinal raster values, elevation data, and parametric equation coefficients of the road reference line.

[0086] Here, the road reference line can be understood as the road centerline obtained through preliminary fitting. At this point, the coefficients of the parametric equations are also the parameters of the corresponding fitted curve. Subsequently, based on the curve rule grid (CRG) data, the three-dimensional road centerline generation method provided in this application embodiment can be used to determine the three-dimensional centerline with high precision for scenarios requiring high precision.

[0087] S102. Based on road-related data, establish a three-dimensional mapping relationship; wherein, the three-dimensional mapping relationship is used to map road-related data from the two-dimensional parameter domain to the three-dimensional spatial domain.

[0088] Optionally, based on road-related data, a mapping relationship can be established from the two-dimensional parameter domain (u, v) to the three-dimensional space (X, Y, Z), resulting in a three-dimensional mapping formula. Here, u represents the long side parameter along the target road direction, i.e., the longitudinal grid value (each longitudinal grid value corresponds to a longitudinal interface); v represents the short side parameter perpendicular to the target road line, i.e., the transverse grid value.

[0089] S103. Determine the initial centerline of the target road based on the lateral offset range of the target road; wherein, the initial centerline includes the initial center point under each longitudinal interface of the target road.

[0090] In one example, the lateral offset range of the target road can be understood as the range determined by the minimum and maximum lateral positions.

[0091] At this point, the initial centerline of the target road can be determined based on the midpoint within the lateral offset range.

[0092] S104. According to the three-dimensional mapping formula, at least some of the initial center points included in the initial center line are iteratively updated to obtain updated three-dimensional center points, and the three-dimensional center line of the target road is determined based on the updated three-dimensional center points.

[0093] In one example, the iteration can begin with the initial center point included by the initial center line as the initial value in the vertical interface corresponding to the initial center line that needs to be iterated, thus enabling the finding of a point in that corresponding vertical interface. , making the point The intersection of the tangent plane at that point and the transverse boundary line below the corresponding longitudinal interface, and the distance from point [missing information]. The lengths are equal, thus determining the three-dimensional center point under the corresponding longitudinal interface.

[0094] As described above, this embodiment of the application can determine the road-related data of the target road based on the curve regularity grid (CRG) data of the target road, and establish a three-dimensional mapping relationship based on the road-related data. This allows the road-related data to be mapped from the two-dimensional parameter domain to the three-dimensional spatial domain, providing a three-dimensional data foundation for subsequent three-dimensional centerline generation and ensuring that the elevation data of the target road can participate in the calculation process of the three-dimensional centerline generation. Specifically, the initial centerline of the target road can be determined based on the lateral offset range of the target road. Then, according to the three-dimensional mapping relationship, at least a portion of the initial center points included in the initial centerline are iteratively updated to obtain updated three-dimensional center points. This implementation method ensures that the elevation values ​​of the determined three-dimensional center points are consistent with the actual road, thereby improving the accuracy of the determined three-dimensional center points. After determining the three-dimensional centerline of the target road based on the updated three-dimensional center points, the three-dimensional geometric accuracy of the three-dimensional centerline can be improved. Furthermore, when generating a high-precision map based on the three-dimensional centerline obtained according to the above embodiments, the reliability of the high-precision map can be improved; or, when making autonomous driving decisions based on the three-dimensional centerline, the accuracy of autonomous driving control decisions can be improved, thereby improving vehicle performance.

[0095] Figure 2 A flowchart illustrating another method for generating a three-dimensional road centerline provided in this application embodiment is shown below. Figure 2 As shown, in this embodiment... Figure 1 Based on the examples, a detailed description of the method for generating three-dimensional road centerlines is provided. This method includes:

[0096] S201. The obtained curve regular grid (CRG) data of the target road is parsed and processed to obtain the road-related data of the target road.

[0097] Optionally, the road-related data may include at least longitudinal data (determined based on longitudinal grid values), lateral data (determined based on lateral grid values), elevation data, heading angle data, and reference line data (determined based on parametric equation coefficients, which can determine the reference coordinates of the reference line).

[0098] S202. Based on road-related data, establish a three-dimensional mapping relationship; wherein, the three-dimensional mapping relationship is used to map road-related data from the two-dimensional parameter domain to the three-dimensional spatial domain.

[0099] In one example, the first coordinate data can be determined based on the heading angle data, the lateral data, and the projection of the longitudinal data onto the first coordinate axis of the reference line data.

[0100] Optionally, the first coordinate data can be understood as the coordinate data of the X-axis in the three-dimensional spatial domain.

[0101] At this point, the first coordinate data can be determined using the following formula (1).

[0102] (1)

[0103] in, Representing vertical data The projection of the reference line data onto the X-axis (i.e., the first coordinate axis). Represents horizontal data. Representing vertical data The heading angle at the location.

[0104] In one example, the second coordinate data can be determined based on the projection of the heading angle data, the lateral data, and the longitudinal data onto the second coordinate axis of the reference line data.

[0105] Optionally, the second coordinate data can be understood as the Y-axis coordinate data in the three-dimensional spatial domain.

[0106] At this point, the second coordinate data can be determined using the following formula (2).

[0107] (2)

[0108] in, Representing vertical data Projection of the reference line data onto the Y-axis (i.e., the second coordinate axis).

[0109] In one example, the third coordinate data can be determined based on the elevation data and the first interpolation method.

[0110] Optionally, the third coordinate data can be understood as the coordinate data of the Z-axis in the three-dimensional spatial domain.

[0111] Optionally, the first interpolation method can be linear interpolation, bilinear interpolation, cubic spline interpolation, etc. The type of the first interpolation method is not limited here; it should be chosen based on actual needs.

[0112] Optionally, if the first interpolation method is bilinear interpolation, then the third coordinate data can be determined by the following formula (3).

[0113] (3)

[0114] in, , represents the normalized position of point u within the current grid cell in the u direction. , represents the normalized position of point v in the current grid cell in the v direction, j and k represent the grid cell index where the parameter (u,v) is located, and Z represents the elevation data.

[0115] when When, it indicates that the point is close to the left boundary of the grid cell. ;

[0116] when When, it indicates that the point is close to the right boundary of the grid cell. ;

[0117] when When the time is specified, it indicates that the point is inside a grid cell.

[0118] when When, it indicates that the point is close to the left boundary of the grid cell. ;

[0119] when When, it indicates that the point is close to the right boundary of the grid cell. ;

[0120] when When the time is specified, it indicates that the point is inside a grid cell.

[0121] Finally, based on the first coordinate data, the second coordinate data, and the third coordinate data, the three-dimensional mapping relationship is determined.

[0122] Optionally, the three-dimensional mapping relationship can be written as: At this point, the three-dimensional mapping relationship can be seen in the following formula (4).

[0123] (4)

[0124] In the above implementation, the accuracy of the determined coordinate data can be improved by using the heading angle that changes dynamically with the longitudinal data, combined with the projection of the longitudinal data onto each coordinate axis, thereby improving the accuracy of the three-dimensional mapping relationship.

[0125] In one possible implementation, to further ensure the accuracy of the determined elevation data, this embodiment of the application may determine the data distribution information of the elevation data before determining the third coordinate data based on the elevation data and the first interpolation method, and identify abnormal data in the elevation data based on the data distribution information. Then, the abnormal data is updated according to the second interpolation method to obtain the processed elevation data.

[0126] Optionally, data that does not conform to a normal distribution in the elevation data can be identified as outliers. For example, this can be determined based on "3..." The principle is to remove abnormal data from the elevation data.

[0127] Optionally, the second interpolation method can indicate the interpolation method at the location of the supplementary abnormal data. For example, the second interpolation method can be bilinear interpolation, inverse distance weighted interpolation, cubic spline interpolation, etc. The type of the second interpolation method is not limited here, and the actual needs shall prevail.

[0128] Optionally, the first interpolation method and the second interpolation method can be the same or different; this is not limited here.

[0129] This implementation method can eliminate abnormal data in curve regular raster (CRG) data, thereby obtaining more accurate elevation data and improving the accuracy of the three-dimensional centerline determined based on the elevation data.

[0130] In one alternative implementation, since the process of determining the projection of longitudinal data onto the corresponding coordinate axis requires integration along the longitudinal data using the heading angle, and the heading angle data has limited accuracy, long-distance integration can lead to offset and insufficient heading angle accuracy.

[0131] Based on this, embodiments of this application can also determine the heading angle integral error of the reference line data according to the start and end positions in the reference line data. If it is determined that the heading angle integral error meets the heading angle correction requirements, then a correction coefficient is determined based on the heading angle integral error; and the first coordinate data and the second coordinate data are corrected according to the correction coefficient.

[0132] Optionally, when determining the heading angle integration error, one can first start from the starting position of the target road and perform forward integration. At this point, the ending position under forward integration can be compared with the ending position in the curve regular grid (CRG) data to obtain the heading angle forward integration error. Then, starting from the ending position of the target road, reverse integration can be performed to obtain the heading angle reverse integration error. Finally, the heading angle forward integration error and the heading angle reverse integration error can be fused to obtain the final heading angle integration error.

[0133] Optionally, the heading angle correction requirement can indicate a heading angle integral error threshold. In this case, if it is determined that the heading angle integral error is greater than the heading angle integral error threshold indicated by the heading angle correction requirement, then the heading angle correction requirement is determined to be met; otherwise, the heading angle correction requirement is determined not to be met.

[0134] Optionally, if the heading angle correction requirement is met, the heading angle integral error can be evenly distributed across each longitudinal data point to obtain correction coefficients. Then, the first and second coordinate data can be corrected based on these correction coefficients.

[0135] In one possible implementation, a segmented integration method can be used to correct the heading angle integral error. That is, the target road can be divided into multiple independent segments, calculated segment by segment, and then spliced ​​together. Compared with the single-segment integration method, this reduces the accumulation of integration error. The method for correcting the heading angle integral error is not limited here, but is based on what is feasible.

[0136] In the above embodiments, the accuracy of coordinate transformation can be improved by correcting the heading angle integral error, thereby improving the accuracy of the three-dimensional mapping relationship, and further improving the accuracy of the three-dimensional centerline determined according to the three-dimensional mapping relationship.

[0137] Optionally, after determining the accurate three-dimensional mapping relationship according to the above process, the initial three-dimensional center point can be determined, and after iterative updating, the updated three-dimensional center point can be determined, thereby determining the three-dimensional centerline.

[0138] S203. Determine the initial centerline of the target road based on the lateral offset range of the target road; wherein the initial centerline includes the initial center point under each longitudinal interface of the target road.

[0139] Optionally, it is assumed that the lateral offset range includes the minimum lateral data. and the largest horizontal data Therefore, the set of boundary points for the lateral offset range of the target road can be determined as follows: , .

[0140] in, That is, the first boundary point. This is the second boundary point.

[0141] At this point, assume that the initial centerline of the target road is at... Therefore, the lateral data of the initial center point under each longitudinal interface included in the initial center line can be expressed as shown in the following formula (5).

[0142] (5)

[0143] Then, based on each longitudinal interface included by the initial centerline... The initial center point at the location is iteratively updated to obtain the updated three-dimensional centerline, as described in the process below.

[0144] S204. Determine the road slope information based on the location of the initial center point in the initial centerline.

[0145] Optionally, road slope information can be determined based on the ratio of changes in elevation data to longitudinal data.

[0146] S205. Based on the road slope information, determine the iteration step size of the center point, and based on the iteration step size of the center point, determine the longitudinal interface to be updated.

[0147] In one implementation, multiple slope intervals can be pre-divided, and a corresponding center point iteration step size can be pre-set for each slope interval. Then, the corresponding center point iteration step size can be determined based on the slope interval to which the road slope information at the initial center point location belongs.

[0148] In another implementation, the initial center point iteration step size can be set to a default value (e.g., 1). Then, the step size change is determined based on the rate of change of the road slope information, and a new center point iteration step size is determined based on the step size change.

[0149] The method for determining the iteration step size of the center point is not limited here; the method that can be implemented is the standard.

[0150] Optionally, after determining the iteration step size of the center point, a new vertical interface to be updated can be determined, and the initial center point under the new vertical interface to be updated can be iteratively updated.

[0151] This implementation method can determine a smaller center point iteration step size when the road slope information is large, and a larger center point iteration step size when the road slope information is small, thus achieving both computational accuracy and computational efficiency.

[0152] Optionally, after determining the vertical interface to be updated, when iteratively updating the initial center point of the vertical interface to be updated according to the three-dimensional mapping relationship, the process described below can be referred to.

[0153] S206. For each initial center point under the vertical interface to be updated, repeat the following process: Determine the unit tangent vector at the location of the initial center point according to the three-dimensional mapping relationship; determine the first projection parameter of the first boundary point under the unit tangent vector, and determine the second projection parameter of the second boundary point under the unit tangent vector; if the projection difference between the first projection parameter and the second projection parameter does not meet the iterative convergence requirement, then update the initial center point according to the unit tangent vector and the three-dimensional mapping relationship to obtain the updated initial center point.

[0154] In one example, when determining the unit tangent vector at the location of the initial center point, see the following formula (6).

[0155] (6)

[0156] in, For example, a pre-set lateral offset. It can be 2 centimeters.

[0157] Optionally, when determining the first projection parameters of the boundary point (taking the first boundary point, i.e. the left boundary point, as an example) under the unit tangent vector, you can refer to the following formula (7).

[0158] (7)

[0159] in, The initial center point is determined based on the three-dimensional mapping formula, the horizontal data, and the vertical data of the initial center point.

[0160] In one example, the iterative convergence requirement can indicate the iterative convergence threshold. In this case, if the projection difference between the first projection parameter and the second projection parameter is less than the iterative convergence threshold indicated by the iterative convergence requirement, it means that the iterative convergence requirement is met; otherwise, it means that the iterative convergence requirement is not met.

[0161] At this point, based on the three-dimensional mapping relationship, the unit tangent vector of the initial center point can be determined under a pre-set lateral offset. The initial center point can then be verified and calibrated based on the unit tangent vector, thereby obtaining a center point with equal projection distances between the left and right boundary points. This achieves dynamic determination of the three-dimensional center point and improves the accuracy of the determined three-dimensional center point.

[0162] Optionally, since the initial center point includes the horizontal data under the vertical interface, if the projection difference does not meet the iterative convergence requirement, the new horizontal data can be determined first based on the unit tangent vector and the initial center point; then, the original three-dimensional coordinate points can be determined based on the initial center point and the three-dimensional mapping relationship, and the new three-dimensional coordinate points can be determined based on the new horizontal data and the three-dimensional mapping relationship; finally, the updated initial center point can be determined based on the difference between the new three-dimensional coordinate points and the original three-dimensional coordinate points.

[0163] For details, please refer to formula (8) below.

[0164] (8)

[0165] in, This represents the vertical data of the current vertical interface that needs to be updated.

[0166] At this point, the initial center point can be updated based on the unit tangent vector, thus taking into account elevation data to update the initial center point and make the updated initial center point more closely match the actual road.

[0167] S207. If it is determined that the projection difference meets the iterative convergence requirement, then the initial center point is determined as the updated three-dimensional center point.

[0168] Optionally, the updated three-dimensional center point can be obtained by substituting the vertical data corresponding to the vertical interface (i.e., the corresponding vertical interface to be updated) and the updated horizontal data into the three-dimensional mapping formula.

[0169] S208. Based on the updated three-dimensional center point, determine the three-dimensional centerline of the target road.

[0170] Figure 3 This is a schematic diagram illustrating the implementation process of a three-dimensional road centerline generation method provided in an embodiment of this application, as shown below. Figure 3 As shown, in the specific implementation process, after obtaining the curve regular grid (CRG) data of the target road, the curve regular grid (CRG) data can be parsed first to obtain the road-related data of the target road, and a three-dimensional mapping relationship can be established based on the road-related data.

[0171] Next, the horizontal offset range and the vertical distance range can be determined. At this point, the vertical data of the vertical interface to be updated (i.e., the vertical data of the vertical interface to be updated) within the vertical distance range... ), performs the process of determining the updated center point of the three dimensions.

[0172] Specifically, starting from the initial center point of the initial centerline, first determine the unit tangent vector of the initial center point, and then determine the projection parameters of the first and second boundary points under the unit tangent vector, i.e., the first projection parameter and the second projection parameter. Next, determine the projection difference between the first and second projection parameters. Then, determine whether the projection difference is less than the iterative convergence threshold indicated by the iterative convergence requirement.

[0173] If so, it indicates that the iterative convergence requirement is met. At this point, the updated 3D center point is obtained, and the iterative update of the initial center point under the next vertical interface to be updated continues. The process ends when the next vertical interface to be updated exceeds the vertical distance range.

[0174] If not, update the initial center point to obtain the updated initial center point, and continue to determine whether the projection difference is less than the iterative convergence threshold indicated by the iterative convergence requirement under the updated initial center point.

[0175] Figure 4 This is a schematic diagram of the structure of a three-dimensional road centerline generation device provided in an embodiment of this application, as shown below. Figure 4 As shown, the three-dimensional road centerline generation device 40 provided in this embodiment includes:

[0176] The parsing unit 401 is used to parse and process the acquired curve regular grid (CRG) data of the target road to obtain the road-related data of the target road.

[0177] The mapping unit 402 is used to establish a three-dimensional mapping relationship based on road-related data; wherein, the three-dimensional mapping relationship is used to map the road-related data from the two-dimensional parameter domain to the three-dimensional spatial domain.

[0178] The determining unit 403 is used to determine the initial centerline of the target road based on the lateral offset range of the target road; wherein the initial centerline includes the initial center point under each longitudinal interface of the target road.

[0179] The iteration unit 404 is used to iteratively update at least a portion of the initial center points included in the initial center line according to the three-dimensional mapping relationship, to obtain updated three-dimensional center points, and to determine the three-dimensional center line of the target road based on the updated three-dimensional center points.

[0180] In one possible implementation, the vertical interface includes a first boundary point and a second boundary point within the horizontal offset range; in this case, the iteration unit 404 is used to:

[0181] The vertical interface to be updated is determined, and the following process is repeated for the initial center point under each vertical interface to be updated: According to the three-dimensional mapping relationship, the unit tangent vector at the location of the initial center point is determined; the first projection parameter of the first boundary point under the unit tangent vector is determined, and the second projection parameter of the second boundary point under the unit tangent vector is determined; if it is determined that the projection difference between the first projection parameter and the second projection parameter does not meet the iterative convergence requirement, the initial center point is updated according to the unit tangent vector and the three-dimensional mapping relationship to obtain the updated initial center point.

[0182] If it is determined that the projection difference meets the iterative convergence requirement, then the initial center point is determined as the updated three-dimensional center point.

[0183] In one possible implementation, the initial center point includes lateral data below the vertical interface; in this case, the iteration unit 404 is used to:

[0184] Determine the new lateral data based on the unit tangent vector and the initial center point;

[0185] Based on the initial center point and the three-dimensional mapping formula, the original three-dimensional coordinate points are determined, and based on the new horizontal data and the three-dimensional mapping formula, the new three-dimensional coordinate points are determined.

[0186] The updated initial center point is determined based on the difference between the new 3D coordinate point and the original 3D coordinate point.

[0187] In one possible implementation, the iteration unit 404 is used for:

[0188] Determine the road slope information based on the location of the initial center point in the initial centerline;

[0189] Based on the road slope information, determine the iteration step size of the center point, and based on the iteration step size of the center point, determine the longitudinal interface to be updated.

[0190] In one possible implementation, the road-related data includes at least longitudinal data, lateral data, elevation data, heading angle data, and reference line data; in this case, the mapping unit 402 is used for:

[0191] The first coordinate data is determined based on the projection of the heading angle data, lateral data, and longitudinal data onto the first coordinate axis of the reference line data;

[0192] The second coordinate data is determined based on the projection of the heading angle data, lateral data, and longitudinal data onto the second coordinate axis of the reference line data;

[0193] The third coordinate data is determined based on the elevation data and the first interpolation method;

[0194] Based on the first coordinate data, the second coordinate data, and the third coordinate data, determine the three-dimensional mapping relationship.

[0195] In one possible implementation, before determining the third coordinate data based on the elevation data and the first interpolation method, the device is further configured to:

[0196] Determine the data distribution information of elevation data;

[0197] Based on the data distribution information, abnormal data in the elevation data were identified;

[0198] The abnormal data is updated according to the second interpolation method to obtain the processed elevation data.

[0199] In one possible implementation, the device is also used for:

[0200] Determine the heading angle integral error of the reference line data based on the start and end positions in the reference line data;

[0201] If it is determined that the integral error of the heading angle meets the heading angle correction requirements, then the correction coefficient is determined based on the integral error of the heading angle.

[0202] The first and second coordinate data are corrected according to the correction coefficients.

[0203] The three-dimensional road centerline generation device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0204] Figure 5 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application.Figure 5 As shown, the computer device 50 provided in this embodiment includes at least one processor 501 and a memory 502. Optionally, the computer device 50 further includes a communication component 503. The processor 501, memory 502, and communication component 503 are connected via a bus 504.

[0205] In a specific implementation, at least one processor 501 executes computer execution instructions stored in memory 502, causing at least one processor 501 to perform the above-described method.

[0206] The specific implementation process of processor 501 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0207] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0208] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0209] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0210] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0211] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0212] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0213] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0214] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0215] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0216] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0217] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0218] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0219] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for generating a three-dimensional road centerline, characterized in that, include: The acquired curve regularity grid (CRG) data of the target road is parsed and processed to obtain the road-related data of the target road; Based on the road-related data, a three-dimensional mapping relationship is established; wherein, the three-dimensional mapping relationship is used to map the road-related data from the two-dimensional parameter domain to the three-dimensional spatial domain; Based on the lateral offset range of the target road, the initial centerline of the target road is determined; wherein, the initial centerline includes the initial center point under each longitudinal interface of the target road; According to the three-dimensional mapping formula, at least some of the initial center points included in the initial center line are iteratively updated to obtain updated three-dimensional center points, and the three-dimensional center line of the target road is determined based on the updated three-dimensional center points.

2. The method according to claim 1, characterized in that, The vertical interface includes a first boundary point and a second boundary point within the horizontal offset range; the iterative update process, based on the three-dimensional mapping formula, of at least a portion of the initial center points included in the initial centerline to obtain updated three-dimensional center points includes: The vertical interface to be updated is determined, and the following process is repeated for the initial center point under each vertical interface to be updated: Based on the three-dimensional mapping formula, the unit tangent vector at the location of the initial center point is determined; the first projection parameter of the first boundary point under the unit tangent vector is determined, and the second projection parameter of the second boundary point under the unit tangent vector is determined; if the projection difference between the first projection parameter and the second projection parameter does not meet the iterative convergence requirement, the initial center point is updated based on the unit tangent vector and the three-dimensional mapping formula to obtain the updated initial center point. If it is determined that the projection difference satisfies the iterative convergence requirement, then the initial center point is determined as the updated three-dimensional center point.

3. The method according to claim 2, characterized in that, The initial center point includes the horizontal data under the vertical interface; the initial center point is updated according to the unit tangent vector and the three-dimensional mapping relationship to obtain the updated initial center point, including: Based on the unit tangent vector and the initial center point, determine the new lateral data; Based on the initial center point and the three-dimensional mapping formula, the original three-dimensional coordinate point is determined, and based on the new horizontal data and the three-dimensional mapping formula, the new three-dimensional coordinate point is determined. The updated initial center point is determined based on the difference between the new three-dimensional coordinate point and the original three-dimensional coordinate point.

4. The method according to claim 2, characterized in that, Determine the vertical interface to be updated, including: Determine the road slope information based on the location of the initial center point in the initial centerline; Based on the road slope information, the center point iteration step size is determined, and based on the center point iteration step size, the longitudinal interface to be updated is determined.

5. The method according to any one of claims 1-4, characterized in that, The road-related data includes at least longitudinal data, lateral data, elevation data, heading angle data, and reference line data; the step of establishing a three-dimensional mapping relationship based on the road-related data includes: The first coordinate data is determined based on the heading angle data, the lateral data, and the projection of the longitudinal data onto the first coordinate axis of the reference line data; The second coordinate data is determined based on the heading angle data, the lateral data, and the projection of the longitudinal data onto the second coordinate axis of the reference line data; Based on the elevation data and the first interpolation method, determine the third coordinate data; The three-dimensional mapping relationship is determined based on the first coordinate data, the second coordinate data, and the third coordinate data.

6. The method according to claim 5, characterized in that, Before determining the third coordinate data based on the elevation data and the first interpolation method, the method further includes: Determine the data distribution information of the elevation data; Based on the data distribution information, abnormal data in the elevation data are identified; The abnormal data is updated according to the second interpolation method to obtain the processed elevation data.

7. The method according to claim 5, characterized in that, The method further includes: The heading angle integral error of the reference line data is determined based on the start and end positions in the reference line data. If it is determined that the integral error of the heading angle meets the heading angle correction requirements, then the correction coefficient is determined based on the integral error of the heading angle. The first coordinate data and the second coordinate data are corrected according to the correction coefficient.

8. A three-dimensional road centerline generation device, characterized in that, include: The parsing unit is used to parse and process the acquired curve regular grid (CRG) data of the target road to obtain the road-related data of the target road. A mapping unit is used to establish a three-dimensional mapping relationship based on the road-related data; wherein the three-dimensional mapping relationship is used to map the road-related data from a two-dimensional parameter domain to a three-dimensional spatial domain. A determining unit is configured to determine the initial centerline of the target road based on the lateral offset range of the target road; wherein the initial centerline includes the initial center point under each longitudinal interface of the target road; An iterative unit is used to iteratively update at least a portion of the initial center points included in the initial center line according to the three-dimensional mapping formula, to obtain updated three-dimensional center points, and to determine the three-dimensional center line of the target road according to the updated three-dimensional center points.

9. A computer device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-7.

Images