A road boundary processing method, device and equipment

Abstract

The present invention discloses a method, device and equipment for processing road boundaries. The method includes: taking the road boundary included in the target road as the object to be processed, obtaining a series of road boundary points on the road boundary; taking the road boundary points as reference points to match road markings within a preset spatial range to associate the matched road markings with the road boundary; determining the inner and outer sides of the target road based on the geometric position relationship between each road boundary and the associated road markings; and performing movement processing on the road boundary based on the position of the road boundary on the inner and outer sides of the target road. The method associates the road boundary with the road markings, and then determines the inner and outer sides of the target road based on their geometric position relationship, so as to perform movement processing on the road boundary. The method greatly reduces the number of road boundaries appearing on the inner side of the road, solves the technical problem of fuzzy distinction between adjacent road boundaries, and improves the accuracy of the produced high-precision map.

Description

Technical Field

[0001] The present invention relates to the field of geographic information technology, in particular to the field of high-precision map production technology, and in particular to a method, device and equipment for processing road boundaries. Background Art

[0002] High-precision maps can be used for autonomous driving, and the identification and production of road signs are important components of generating high-precision maps. Road signs are generally identified and extracted through road image information collected by cameras mounted on vehicles. The applicant has previously proposed a number of patents for producing high-precision maps using multi-source crowdsourcing data. In the process of producing high-precision maps, the road signs (road marking data, road boundary data, etc.) included in the multi-source crowdsourcing data are calibrated, optimized, clustered and aggregated, laying the foundation for the later production of high-precision maps. Summary of the invention

[0003] In order to solve the abnormal location of road boundary data during the aggregation processing of multi-source crowdsourcing data, and thus prepare accurate high-precision maps, while enriching technical routes and increasing selection space, a road boundary processing method, device and equipment are provided in the embodiments of the present invention.

[0004] In a first aspect, an embodiment of the present invention provides a method for processing a road boundary, wherein the road boundary is a road boundary of a high-precision map produced based on multi-source crowdsourcing data, and the processing method may include:

[0005] Taking the road boundary included in the target road as the object to be processed, obtaining a series of road boundary points on the road boundary;

[0006] Performing road marking matching within a preset spatial range with the road boundary point as a reference point, so as to associate the matched road marking with the road boundary;

[0007] Determining the inner and outer sides of the target road based on the geometric position relationship between each road boundary and the associated road markings;

[0008] The road boundary is moved based on the position of the road boundary inside and outside the target road.

[0009] In one embodiment, determining the inner and outer sides of the target road based on the geometric position relationship between each road boundary and the associated road marking may include:

[0010] Determine the data types of the road boundary and road markings included in the target road;

[0011] If the data types include original road boundary data and original road marking data, statistics are performed on the geometric position relationship between each group of original road boundaries and the associated original road markings, so as to determine the inner and outer sides of the target road based on the proportion of the original road boundaries located at the positions of the original road markings in the statistical results;

[0012] If the data type does not include original road boundary data and original road marking data, determine whether the road boundary and the road marking are from the same batch of data; if they are from the same batch of data, determine the inner and outer sides of the target road based on the position of the road boundary; otherwise, determine the inner and outer sides of the target road based on the road marking associated with the road boundary and whether there are other road boundaries or road markings within the spatial range of the road marking.

[0013] In one embodiment, the determining the inner and outer sides of the target road based on the road markings associated with the road boundary and whether there are other road boundaries or road markings within the spatial range of the road markings may include:

[0014] Making a judgment based on the road markings associated with the road boundary;

[0015] If there is another road boundary that is not associated with the road marking within the first width threshold of the road marking, and the height of the road boundary is the same as the height of the road marking, then the side where the road boundary is located is the outer side of the target road; wherein the first width threshold is determined based on the width of the emergency lane in the target road;

[0016] If there are other road markings with the same height and opposite direction within the second width threshold of the road marking, the side where the other road markings are located is the outer side of the target road; wherein the second width threshold is determined based on the width of the lane in the target road;

[0017] If there are other road markings with the same height and direction within the third width threshold range of the road marking, the side where the other road markings are located is the inner side of the target road; wherein the third width threshold is determined based on the width of the lane in the target road, and the third width threshold is greater than the second width threshold.

[0018] In one embodiment, if it is determined that the data type includes original road boundary data and original road marking data, then before statistics are performed on the geometric position relationship between each group of original road boundaries and associated original road markings, it also includes: based on the timestamp attributes in the original road boundary data and the original road marking data, verifying the association relationship between the original road boundaries and the associated original road markings; if the timestamp attributes are the same, statistics are performed.

[0019] In one embodiment, if the inner and outer sides of the target road are not determined based on the data types of the road boundary and road markings included in the target road, the method further includes: judging whether the target road is a herringbone road based on the attribute information of the road boundary;

[0020] If it is a herringbone road, there is another herringbone road marking within the fourth width threshold range on one side of the road marking, and there are other road markings with the same height and direction within the fifth width threshold range on the other side of the road marking, then the side where the other herringbone road marking is located is the outer side of the target road;

[0021] or,

[0022] If it is a herringbone road, there is another herringbone road marking within the fourth width threshold range on one side of the road marking, and there is another road boundary with the same height and direction and no correlation with the road marking within the fifth width threshold range on the other side of the road marking, then the side where the other herringbone road marking is located is the outer side of the target road;

[0023] The fourth width threshold is determined based on the width of a lane in the target road, and the fifth width threshold is determined based on the widths of two lanes in the target road.

[0024] In one embodiment, the moving process of the road boundary based on the position of the road boundary inside and outside the target road may include:

[0025] If the road boundary is inside the target road, a series of road boundary points on the road boundary are moved to the outside of the target road by a preset moving length threshold, and the moved series of road boundary points are connected to achieve moving processing of the road boundary;

[0026] If the road boundary is outside the target road, and the road marking associated with the road boundary has other road markings with the same height and opposite direction within the preset sixth width threshold, a series of the road boundary points are moved to the inside of the target road by a preset moving length threshold, and the series of road boundary points after the movement are connected to achieve movement processing of the road boundary;

[0027] If the road boundary is outside the target road, and there is a road boundary point of a road boundary associated with other road markings within the preset seventh width threshold range of the road boundary point, then the straight-line distances between a series of the road boundary points and the associated road markings are calculated, and the series of road boundary points are moved to the inside of the target road by the difference between the straight-line distance and the seventh width threshold, and the series of road boundary points after the movement are connected to achieve movement processing of the road boundary;

[0028] The sixth width threshold is greater than the seventh width threshold.

[0029] In one embodiment, if the difference between the straight-line distance and the seventh width threshold is less than the moving length threshold, a series of the road boundary points are moved toward the inside of the target road by a preset moving length threshold, and the series of road boundary points after the movement are connected to achieve movement processing of the road boundary.

[0030] In one embodiment, acquiring a series of road boundary points on the road boundary may include:

[0031] Performing line data thinning processing on the road boundary to obtain a series of road boundary points representing the linear shape of the road boundary; or,

[0032] Points are taken on the road boundary with a preset first distance threshold, and a series of road boundary points are obtained on the road boundary; wherein the first distance threshold is determined according to the smoothness of the line data of the target road and / or the road boundary.

[0033] In one embodiment, the performing road marking matching within a preset spatial range using the road boundary point as a reference point to associate the matched road marking with the road boundary may include:

[0034] Taking the road boundary point as a reference point and a second distance threshold as a search range, a road marking line with the same height, same direction, same left and right sides of the road and closest to the road boundary point within the spatial range of the second distance threshold is taken as a matched road marking line;

[0035] Binding the matched road markings to the road boundary to associate the matched road markings with the road boundary;

[0036] The second distance threshold is determined according to the width of the emergency lane in the target road.

[0037] In one embodiment, the processing method may also include: judging whether there are multiple road boundaries within a preset spatial range on one side of the target road based on the geometric line type of the road boundary; if so, performing redundant processing on the road boundaries included on one side of the target road to retain a road boundary closest to the target road.

[0038] In a second aspect, an embodiment of the present invention provides a device for processing road boundaries, wherein the road boundaries are road boundaries of a high-precision map produced based on multi-source crowdsourcing data, and the processing device may include:

[0039] An acquisition module, used for taking a road boundary included in a target road as an object to be processed, and acquiring a series of road boundary points on the road boundary;

[0040] An association module, used to match road markings within a preset spatial range using the road boundary point as a reference point, so as to associate the matched road markings with the road boundary;

[0041] A determination module, configured to determine the inner and outer sides of the target road based on a geometric positional relationship between each of the road boundaries and associated road markings;

[0042] The processing module is used to perform movement processing on the road boundary based on the position of the road boundary inside and outside the target road.

[0043] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for processing road boundaries as described in the first aspect.

[0044] In a fourth aspect, an embodiment of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for processing road boundaries as described in the first aspect is implemented.

[0045] The beneficial effects of the above technical solution provided by the embodiment of the present invention include at least:

[0046] The embodiment of the present invention provides a method, device and equipment for processing road boundaries. The processing method associates the road boundary with the road marking, and then determines the inner and outer sides of the target road based on their geometric position relationship, and moves the road boundary according to the position of the road boundary inside and outside the target road. The method greatly reduces the number of road boundaries that appear on the inside of the road, solves the technical problem of fuzzy distinction between adjacent road boundaries, and improves the accuracy of the produced high-precision map.

[0047] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0048] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0050] Figure 1 A flowchart of a method for processing road boundaries provided in an embodiment of the present invention;

[0051] Figure 2 A flowchart of a detailed road boundary processing method provided in an embodiment of the present invention;

[0052] Figure 3 This is a specific flow chart of step S26;

[0053] Figure 4 This is an example 1 of determining the inside and outside of a target road provided in an embodiment of the present invention;

[0054] Figure 5 This is a second example of determining the inside and outside of a target road provided in an embodiment of the present invention;

[0055] Figure 6 This is Example 3 of judging the inside and outside of a herringbone road provided in an embodiment of the present invention;

[0056] Figure 7 for Figure 4 Example 1 shows the effect of moving the road boundary;

[0057] Figure 8 for Figure 5 Example 2 shows the effect of moving the road boundary;

[0058] Fig. 9 This is the third road boundary movement effect diagram provided in an embodiment of the present invention;

[0059] Fig.10 It is a schematic diagram of the structure of a road boundary processing device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0061] When the inventor processed the road marking data in the original data, the inventor found that the road boundary position in the identified road marking data was abnormal, for example, the road boundary appeared inside the road. At the same time, the inventor also found that when the road marking data was aggregated, due to the numerous aggregation processing steps and complex logic, the road boundary position would still be abnormal after the aggregation processing, and the location of the road boundary is outside the lane, which is in principle impassable. In response to the above technical problems, the inventor tried to move the road boundary data inside the road directly to the outside of the road, but this would lead to other overall deficiencies, such as omissions in identifying the inner boundary of the road, errors in the boundary movement distance, etc., and the road boundary inside the road would need to be cleaned and deleted separately later, resulting in data loss, inconsistency with the input, and other problems. In view of the above problems, the present invention is proposed to provide a method, device and equipment for processing road boundaries that overcome the above problems or at least partially solve the above problems.

[0062] In an embodiment of the present invention, a method for processing road boundaries is provided, wherein the road boundaries are road boundaries of a high-precision map produced based on multi-source crowdsourcing data, referring to Figure 1 As shown, the road boundary processing method may include the following steps:

[0063] Step S11: Taking the road boundary included in the target road as the object to be processed, a series of road boundary points are obtained on the road boundary.

[0064] Step S12: performing road marking matching within a preset spatial range with the road boundary point as a reference point, so as to associate the matched road marking with the road boundary.

[0065] Step S13: Determine the inner and outer sides of the target road based on the geometric position relationship between each road boundary and the associated road markings.

[0066] Step S14: Based on the position of the road boundary inside and outside the target road, the road boundary is moved.

[0067] The above-mentioned road boundary processing method provided in the embodiment of the present invention associates the road boundary with the road marking, and then determines the inner and outer sides of the target road based on their geometric position relationship, and moves the road boundary according to the position of the road boundary inside and outside the target road. This method greatly reduces the number of road boundaries appearing on the inside of the road, solves the technical problem of fuzzy distinction between adjacent road boundaries, and improves the accuracy of the produced high-precision map.

[0068] In a specific embodiment, this embodiment provides a detailed road boundary processing method, referring to Figure 2 As shown, the method may include the following steps:

[0069] Step S21, obtaining target road data to be processed, wherein the target road data includes road boundaries and road markings.

[0070] The target road data acquired in this step is composed of road surface markings, which may include road boundary data and road marking data, wherein the road marking data may include lane lines, road guide lines (arrows), zebra crossings, etc. Of course, in this embodiment, the road markings can be described by the most lane lines in the road marking data, and the embodiment of the present invention does not make specific limitations on this.

[0071] In this embodiment, the target road data can be data after multi-source crowdsourcing data has been aggregated, or it can be original data that has not been aggregated, or it can be original data retained after aggregation. In the process of making a high-precision map, it is first necessary to process the original road markings to obtain a preliminary high-precision map. At the same time, since multi-source crowdsourcing data is also continuously collected, the high-precision map needs to be updated. Therefore, the target road data obtained in this step will have new original data (incremental data) that has not been aggregated. It should be noted that the road boundary method provided in this embodiment is a method for optimizing the mobile processing of road boundaries in the target road data after aggregation. Generally, in the process of making a high-precision map, the original data will be deleted or marked after the aggregation process in order to ensure the efficient processing performance of the computer equipment.

[0072] It should also be noted that the road boundaries and road markings in the target road have data attributes, such as data type (boundary type or marking type), geometry three-dimensional coordinates, is_left left and right sides of the vehicle, longitudinal_type herringbone scene, cluster_id class name (used to associate with original data), data_type data batch, etc.

[0073] Step S22, based on the geometric line type of the road boundary, determine whether there are multiple road boundaries within a preset spatial range on one side of the target road; if so, execute step S23; otherwise, execute step S24.

[0074] In this step, the road boundaries in the target road are processed redundantly. In the specific implementation, it is confirmed based on the coordinate attribute geometry of the road boundary whether there are other road boundaries with the same height, direction, and left and right sides of the road within the preset space (2m) of each boundary; if not, the road boundary does not need to be processed redundantly; if so, execute the following step S23.

[0075] In this embodiment, the distance of the preset spatial range can be half the lane width determined according to the lane width of the target road, that is, 2m. Of course, the preset spatial range can also be set by other empirical values, and the embodiment of the present invention does not make specific limitations on this.

[0076] Step S23, redundancy processing is performed on the road boundaries included in one side of the target road to retain a road boundary closest to the target road. In this step, if there are multiple road boundaries, the distance between each road boundary and the target road is calculated respectively, and then the road boundary closest to the target road is retained. In this embodiment, when the distance between each road boundary and the target road is calculated respectively, each target road is taken as a surface data, and the vertical distance between each road boundary and the edge of the surface data is taken as the distance between the road boundary and the target road.

[0077] Step S24: Taking the road boundary included in the target road as the object to be processed, a series of road boundary points are obtained on the road boundary.

[0078] In step S24 of this embodiment, a series of road boundary points are obtained (extracted) on the road boundary to be processed, so that the road boundary points can be used for matching, association and movement in subsequent steps. In the embodiment of the present invention, there are two ways to obtain a series of road boundary points on the road boundary, which are as follows:

[0079] <Method 1>

[0080] The line data of the road boundary is thinned to obtain a series of road boundary points that characterize the linear shape of the road boundary. Since the road boundary data is vector data, the method 1 in this embodiment can thin the road boundary through the existing thinning algorithm to obtain a series of road boundary points. These road boundary points characterize the direction of the road boundary, laying a foundation for matching association and movement in subsequent steps.

[0081] It should be noted here that when obtaining a series of road boundary points through a thinning algorithm in this method 1, a series of road boundary points with a quantity or density that is suitable for actual needs can be obtained by adjusting parameters such as the thinning threshold, step size and filter value. The embodiment of the present invention is not specifically limited here.

[0082] <Method 2>

[0083] Points are taken on the road boundary with a preset first distance threshold, and a series of road boundary points are obtained on the road boundary; wherein the first distance threshold is determined according to the smoothness of the line data of the target road and / or the road boundary. Method 2 in this embodiment is to directly take points on the road boundary with a set distance threshold, and the intervals between the obtained road boundary points are constant. Since the first threshold in this application is determined according to the smoothness of the line data of the target road and / or the road boundary, the series of road boundary points obtained based on the first distance threshold can also characterize the actual direction of the road boundary. For example, in an embodiment of the present invention, a first distance threshold of 0.8m is used to take points on the road boundary, and the smoothness of the corners of adjacent road boundary points is related to the distance of 0.8m.

[0084] Step S25: performing road marking matching within a preset spatial range with the road boundary point as a reference point, so as to associate the matched road marking with the road boundary.

[0085] This step is to establish the association relationship between the road boundary and the road marking. In the specific implementation, firstly, the road boundary point is used as the reference point, and the second distance threshold is used as the search range. The road markings with the same height, the same direction, the same left and right sides of the road and the closest distance to the road boundary point within the spatial range of the second distance threshold are used as matching road markings; then, the matching road markings are bound to the road boundary to associate the matching road markings with the road boundary; wherein the second distance threshold is determined according to the width of the emergency lane in the target road. In step S25 of this embodiment, the value of the second distance threshold is determined according to the width of the emergency lane in the target road, and its value can be 2.5m.

[0086] It should be noted that in this step, if there is a missing road marking or no matching road marking is found in the search within the second distance threshold range, the association relationship between the road boundary and the road marking is no longer established, and the road boundary cannot be realized based on the method provided in the embodiment of the present invention. Of course, if such a situation occurs, the map data is incrementally updated to obtain the aggregated road marking again, and then the step can be continued; or, the unassociated road boundary can be optimized according to the connected front and rear road boundaries, which is not specifically limited in the embodiment of the present invention.

[0087] Step S26: Determine the inner and outer sides of the target road based on the geometric position relationship between each road boundary and the associated road markings.

[0088] Since the road boundary is located outside the road marking in the actual road, and the road boundary is located at the edge of the entire target road, after constructing the association between the road boundary and the road marking, it is necessary to focus on each road marking to confirm its left and right sides, that is, which side of the road marking is the outside of the road and which side is the inside of the road. In this step, the starting point of the road marking vector direction can be defined as the axial direction, the entire road marking as the arm, the counterclockwise direction of the marking as the left side, and the clockwise direction as the right side, so that it is necessary to determine whether the road boundary is on the left or right side of the lane marking.

[0089] In the specific implementation of step S26 of this embodiment, reference may be made to Figure 3 , which may specifically include the following steps:

[0090] Step S261, determine the data type of the road boundary and road marking included in the target road. If the data type includes original road boundary data and original road marking data, execute step S262; if the data type does not include original road boundary data and original road marking data, execute step S263.

[0091] In this embodiment, since the target road data to be processed obtained in step S21 includes data after aggregation processing, original data that has not been aggregated, and original data retained after aggregation processing, the method used to determine the inside and outside of the target road in step S26 is different for different data types.

[0092] Step S263: Count the geometric position relationship between each group of original road boundaries and the associated original road markings, and determine the inner and outer sides of the target road based on the proportion of the original road boundaries located at the original road marking positions in the statistical results.

[0093] In the specific implementation of this step, step S263 of this embodiment takes the side where the road boundary is located as the outer side of the target road, referring to Figure 4 As shown, Figure 4The orange-yellow dotted line on the left represents the road boundary, the purple solid line represents the road marking, the orange-yellow solid line on the right represents the original data of the road boundary before aggregation, and the purple surface data on the right represents the original data of the road marking before aggregation. First, find the original data corresponding to each pair of road boundaries and road markings in the original data according to the class name cluster_id of the road boundary and the road marking; then, for each group of original data, the geometric position relationship (coordinate geometry) between the original road boundary and the associated original road marking is determined respectively on which side (left or right) of the original road marking the original road boundary in each group of original data is located, and output the result (in this embodiment, it can be determined based on the position relationship between the centroid of the road marking and the road boundary); then count all the left and right side judgment results, and determine whether the original road boundary is on the left or right side of the original road marking based on the majority; the side with the result that the original road boundary is on one side of the original road marking is determined as the outer side of the target road. It should be noted here that if the number of output results of the original road boundary being located on the left side of the original road marking and the original road boundary being located on the right side of the original road marking is the same, the current position of the road marking to be processed after aggregation is taken as the outer side of the target road.

[0094] In the embodiment of the present invention, the original road boundary data and the original road marking data are used for statistical judgment, instead of simply relying on the positional relationship between the road boundary and the road marking that has been clustered to judge the inner and outer sides of the target road. The purpose of this is to avoid the position error between the final road boundary and the road marking outputted during the processing of the original data due to a series of processing such as calibration, posture optimization, clustering and aggregation of the original data. The occurrence of this error can be effectively avoided by statistics on the positional relationship of the original data.

[0095] In an optional embodiment, before executing step S263, the following judgment needs to be performed, namely, step S262, based on the timestamp attributes in the original road boundary data and the original road marking data, the association relationship between the original road boundary and the associated original road marking is verified; if the timestamp attributes are the same, statistics are performed, that is, step S263 is executed. In this embodiment, if the timestamps are the same, it means that the original road boundary data and the original road marking data are captured and collected at the same time. By verifying the timestamp attributes of each group of original road boundary data and original road marking data, it can effectively avoid that the original road boundary and the associated original road marking are not in a real association relationship, which further improves the accuracy of the judgment of the inside and outside of the target road.

[0096] Step S264: determine whether the road boundary and the road marking are from the same batch of data; if they are from the same batch of data, execute step S265; otherwise, execute step S266. In this step, if the data type does not include the original road boundary data and the original road marking data, it means that the road boundary is no longer moved for the first time, that is, there may be incremental data and scenes after aggregate movement processing.

[0097] Step S265, determine the inner and outer sides of the target road based on the position of the road boundary. If this step is the same batch of data, the data_type attribute of the road boundary and the road marking is the same, then it is considered that the current road boundary is already outside the road, and no additional judgment is made. In this embodiment, two values ​​in the data_type attribute are used, namely, old data: 2; new data: None (empty). In this way, it is determined that the road boundary data and the road marking data are from the same batch of data.

[0098] Step S266: The road markings associated with the road boundary are used as a reference, and the inner and outer sides of the target road are determined based on whether there are other road boundaries or road markings within the road marking space. In this step, if it is determined that the road boundary and the road markings are not from the same batch of data, the inner and outer sides of the target road can be determined based on the current road boundary and road markings and the road markings as a reference. If any of the following scenarios is met, the inner and outer sides of the target road can be determined.

[0099] Specifically, the judgment is made based on the road markings associated with the road boundary. Figure 5 As shown, it can be divided into the following three scenarios. Figure 5 As shown, Figure 5 The medium to dark green solid line with an upward arrow is the associated road sign, and the dark red dotted line with an upward arrow is the road boundary.

[0100] Scenario 1: If there are other road boundaries that are not associated with the road marking within the first width threshold of the road marking, and the height of the road boundary is the same as the height of the road marking, then the side where the road boundary is located is the outer side of the target road; wherein, the first width threshold is determined based on the width of the emergency lane in the target road. In this embodiment, the first width threshold is determined based on the width of the emergency lane, and based on the actual road, the first width threshold is 2.5 meters. Within the range of 2.5 meters, there are other road boundaries ( Figure 5 If the other road boundary is at the same height as the road marking, the side where the other road boundary is located is the outer side of the target road.

[0101] Scenario 2: If there are other road markings with the same height and opposite directions within the second width threshold of the road marking, the side where the other road markings are located is the outer side of the target road; wherein the second width threshold is determined based on the width of the lane in the target road. In this embodiment, the second width threshold is determined based on the width of the lane in the target road, and the second width threshold is 3 meters based on the actual road. Within the range of 3 meters, there are other road markings with the same height and opposite directions as the road marking (dark green solid line with an upward arrow) ( Figure 5 If it is a solid sky blue line with a downward arrow), the side where the other road marking is located is the outer side of the target road.

[0102] Scenario 3: If there are other road markings with the same height and direction within the third width threshold of the road marking, the side where the other road markings are located is the inner side of the target road; wherein the third width threshold is determined based on the width of the lane in the target road, and the third width threshold is greater than the second width threshold. In this embodiment, the third width threshold is also determined based on the width of the lane in the target road. The third width threshold is determined to be 5.5 meters based on the width of one and a half lanes in the actual road. If there are other road markings with the same height and direction as the road marking (dark green solid line with an upward arrow) within a range of 5.5 meters ( Figure 5 If the other road line is a solid line with an upward arrow in light green, the side where the other road line is located is the inner side of the target road, and the other side is the outer side of the road.

[0103] In another optional embodiment, if the inner and outer sides of the target road cannot be determined by the above method, the target road in this embodiment may be a non-straight road, most likely a herringbone road. Figure 3 As shown, the above step S26 may further include the following steps when implemented: step S267, judging whether the target road is a herringbone road based on the attribute information of the road boundary; step S268, if it is a herringbone road, there is another herringbone road marking within the fourth width threshold range on one side of the road marking, and there are other road markings with the same height and direction within the fifth width threshold range on the other side of the road marking, then the side where the other herringbone road marking is located is the outer side of the target road; step S269, if it is a herringbone road, there is another herringbone road marking within the fourth width threshold range on one side of the road marking, and there are other road boundaries with the same height and direction and no correlation with the road marking within the fifth width threshold range on the other side of the road marking, then the side where the other herringbone road marking is located is the outer side of the target road; wherein the fourth width threshold is determined based on the width of the lane in the target road, and the fifth width threshold is determined based on the widths of two lanes in the target road.

[0104] Reference Figure 6 As shown, Figure 6The dark green solid line with an arrow pointing to the right is the associated road sign, and the dark red dotted line with an arrow pointing to the right is the road boundary. In this embodiment, the fourth width threshold is determined based on the lane width of the target road, and can be 3 meters in this embodiment. The fifth width threshold is determined based on the width of two lanes in the target road, and can be 7.5 meters in this embodiment. Figure 6 As shown, if the road markings ( Figure 6 There is another herringbone road marking within 3 meters of the dark green solid line with an arrow pointing right. Figure 6 If the two road markings are at the same height and in the same direction, then the other herringbone road marking ( Figure 6 The side with the middle sky blue solid line with an upper right arrow) is the outer side of the target road; or, refer to Figure 6 As shown, if it is a herringbone road, the road marking ( Figure 6 There is another herringbone road marking within 3 meters of the dark green solid line with an arrow pointing right. Figure 6 Sky blue solid line with an upper right arrow), and road markings ( Figure 6 There is a road marking with the same height and direction within 7.5 meters on the other side of the dark green solid line with an arrow pointing right. Figure 6 Other road boundaries that have no associated relationship (medium dark green solid line with arrow pointing right) Figure 6 light green dashed line with a right-pointing arrow), then the other herringbone road marking ( Figure 6 The side where the center blue solid line with an upper right arrow is located is the outer side of the target road.

[0105] In this embodiment, the herringbone road is distinguished to solve the actual scenario that cannot be distinguished based on the above data type, thereby improving the accuracy of the inner and outer side discrimination of the target road in the embodiment of the present invention.

[0106] Step S27: Based on the position of the road boundary inside and outside the target road, move the road boundary.

[0107] When this step is implemented, it is specifically divided into moving inside the road and moving outside the road, as follows;

[0108] If the road boundary is inside the target road, a series of road boundary points on the road boundary are moved to the outside of the target road by a preset moving length threshold, and the series of road boundary points after the movement are connected to achieve the movement of the road boundary. Figure 8 As shown, the preset moving length threshold in this embodiment is 0.1m (empirical value), the road boundary point located on the inner side of the target road is directly assigned a value of 0.1m, and the road boundary point is moved 0.1m to the outer side of the target road and then connected to realize the movement processing of the road boundary.

[0109] If the road boundary is outside the target road, and the road markings associated with the road boundary have other road markings with the same height and opposite direction within the preset sixth width threshold, a series of road boundary points are moved to the inside of the target road by a preset moving length threshold, and the series of road boundary points after the movement are connected to realize the movement processing of the road boundary; refer to Figure 7 And reference Fig. 9 As shown in the upper and lower figures on the left side, in this embodiment, the sixth threshold width is 1m. If there are other road markings with the same height and opposite direction within 1m of the road markings associated with the road boundary, the road boundary points located outside the target road are directly assigned a value of 0.1m, and the road boundary points are moved 0.1m to the inside of the target road and then connected to realize the movement processing of the road boundary.

[0110] If the road boundary is outside the target road, and there are road boundary points of other road boundaries associated with road markings within the preset seventh width threshold, then the straight-line distances between a series of road boundary points and the associated road markings are calculated, and a series of road boundary points are moved to the inside of the target road by the straight-line distance and the seventh width threshold difference, and the moved series of road boundary points are connected to achieve the movement of the road boundary; wherein the sixth width threshold is greater than the seventh width threshold. Fig. 9 As shown, in this embodiment, the seventh width threshold is 0.25m. If there are road boundary points of roads associated with other road markings within the range of 0.25m, that is, the distance between the two types of road boundary points is too close, the calculated moving straight-line distance will be subtracted from the seventh width threshold (0.25m), and then the difference will be used to awaken the movement.

[0111] In another optional embodiment, in order to better distinguish two types of road boundaries that are too close to each other, in this embodiment, if the difference between the straight-line distance and the seventh width threshold is less than the moving length threshold (0.1m), a series of road boundary points are moved to the inside of the target road by a preset moving length threshold, and the series of road boundary points after the movement are connected to achieve the movement processing of the road boundary. By moving the two types of road boundaries that are too close to each other by a large distance (0.1m), the two types of road boundaries can be clearly distinguished.

[0112] It should be noted that, refer to Fig. 9 As shown, the road boundary can be smoothed after moving, that is, Fig. 9 The road boundary after the movement is straightened and smoothed, etc., and this embodiment of the present invention does not make any specific limitation to this.

[0113] The above-mentioned road boundary processing method provided in the embodiment of the present invention optimizes and upgrades the road boundaries in the process of making high-precision maps, greatly reduces the problem of road boundaries appearing on the inside of the road, introduces original data to participate in the judgment of the inside and outside of the road, and significantly improves the persuasiveness and accuracy of the boundary movement plan; and special optimization processing is also done for complex road scenes such as multi-lane and herringbone, which not only solves the problem of adjacent boundary intersection, but also improves the universality and completeness of the entire high-precision map production process.

[0114] Based on the same inventive concept, an embodiment of the present invention further provides a road boundary processing device, wherein the road boundary is a road boundary of a high-precision map produced based on multi-source crowdsourcing data, referring to Fig.10 As shown, the processing device may include: an acquisition module 11, an association module 12, a determination module 13 and a processing module 14, and its working principle is as follows:

[0115] The acquisition module 11 is used to take the road boundary included in the target road as the object to be processed, and acquire a series of road boundary points on the road boundary;

[0116] The association module 12 is used to match the road markings within a preset spatial range using the road boundary point as a reference point, so as to associate the matched road markings with the road boundary;

[0117] The determination module 13 is used to determine the inner and outer sides of the target road based on the geometric position relationship between each road boundary and the associated road markings;

[0118] The processing module 14 is used for moving the road boundary based on the position of the road boundary inside and outside the target road.

[0119] In one embodiment, the determination module 13 is specifically configured to:

[0120] Determine the data types of the road boundary and road markings included in the target road;

[0121] If the data types include original road boundary data and original road marking data, statistics are performed on the geometric position relationship between each group of original road boundaries and the associated original road markings, so as to determine the inner and outer sides of the target road based on the proportion of the original road boundaries located at the positions of the original road markings in the statistical results;

[0122] If the data type does not include original road boundary data and original road marking data, determine whether the road boundary and the road marking are from the same batch of data; if they are from the same batch of data, determine the inner and outer sides of the target road based on the position of the road boundary; otherwise, determine the inner and outer sides of the target road based on the road marking associated with the road boundary and whether there are other road boundaries or road markings within the spatial range of the road marking.

[0123] In another optional embodiment, the determination module 13 is further configured to:

[0124] Making a judgment based on the road markings associated with the road boundary;

[0125] If there is another road boundary that is not associated with the road marking within the first width threshold of the road marking, and the height of the road boundary is the same as the height of the road marking, then the side where the road boundary is located is the outer side of the target road; wherein the first width threshold is determined based on the width of the emergency lane in the target road;

[0126] If there are other road markings with the same height and opposite direction within the second width threshold of the road marking, the side where the other road markings are located is the outer side of the target road; wherein the second width threshold is determined based on the width of the lane in the target road;

[0127] If there are other road markings with the same height and direction within the third width threshold range of the road marking, the side where the other road markings are located is the inner side of the target road; wherein the third width threshold is determined based on the width of the lane in the target road, and the third width threshold is greater than the second width threshold.

[0128] In another optional embodiment, if it is determined that the data type includes original road boundary data and original road marking data, then before statistics are performed on the geometric position relationship between each group of original road boundaries and associated original road markings, the determination module 13 is specifically used to verify the association relationship between the original road boundaries and the associated original road markings based on the timestamp attributes in the original road boundary data and the original road marking data; if the timestamp attributes are the same, statistics are performed.

[0129] In another optional embodiment, if the inner and outer sides of the target road are not determined based on the data types of the road boundary and road markings included in the target road, the determination module 13 is further configured to:

[0130] Determining whether the target road is a herringbone road based on the attribute information of the road boundary;

[0131] If it is a herringbone road, there is another herringbone road marking within the fourth width threshold range on one side of the road marking, and there are other road markings with the same height and direction within the fifth width threshold range on the other side of the road marking, then the side where the other herringbone road marking is located is the outer side of the target road;

[0132] or,

[0133] If it is a herringbone road, there is another herringbone road marking within the fourth width threshold range on one side of the road marking, and there is another road boundary with the same height and direction and no correlation with the road marking within the fifth width threshold range on the other side of the road marking, then the side where the other herringbone road marking is located is the outer side of the target road;

[0134] The fourth width threshold is determined based on the width of a lane in the target road, and the fifth width threshold is determined based on the widths of two lanes in the target road.

[0135] In another optional embodiment, the processing module 14 is specifically used for:

[0136] If the road boundary is inside the target road, a series of road boundary points on the road boundary are moved to the outside of the target road by a preset moving length threshold, and the moved series of road boundary points are connected to achieve moving processing of the road boundary;

[0137] If the road boundary is outside the target road, and the road marking associated with the road boundary has other road markings with the same height and opposite direction within the preset sixth width threshold, a series of the road boundary points are moved to the inside of the target road by a preset moving length threshold, and the series of road boundary points after the movement are connected to achieve movement processing of the road boundary;

[0138] If the road boundary is outside the target road, and there is a road boundary point of a road boundary associated with other road markings within the preset seventh width threshold range of the road boundary point, then the straight-line distances between a series of the road boundary points and the associated road markings are calculated, and the series of road boundary points are moved to the inside of the target road by the difference between the straight-line distance and the seventh width threshold, and the series of road boundary points after the movement are connected to achieve movement processing of the road boundary;

[0139] The sixth width threshold is greater than the seventh width threshold.

[0140] In another optional embodiment, the processing module 14 can also be specifically used for: if the difference between the straight-line distance and the seventh width threshold is less than the moving length threshold, a series of the road boundary points are moved toward the inside of the target road by a preset moving length threshold, and the series of road boundary points after the movement are connected to realize the movement processing of the road boundary.

[0141] In another optional embodiment, the acquisition module 11 is specifically used for:

[0142] Performing line data thinning processing on the road boundary to obtain a series of road boundary points representing the linear shape of the road boundary; or,

[0143] Points are taken on the road boundary with a preset first distance threshold, and a series of road boundary points are obtained on the road boundary; wherein the first distance threshold is determined according to the smoothness of the line data of the target road and / or the road boundary.

[0144] In another optional embodiment, the association module 12 is specifically used for:

[0145] Taking the road boundary point as a reference point and a second distance threshold as a search range, a road marking line with the same height, same direction, same left and right sides of the road and closest to the road boundary point within the spatial range of the second distance threshold is taken as a matched road marking line;

[0146] Binding the matched road markings to the road boundary to associate the matched road markings with the road boundary;

[0147] The second distance threshold is determined according to the width of the emergency lane in the target road.

[0148] Based on the same inventive concept, an embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the above-mentioned road boundary processing method is implemented.

[0149] Based on the same inventive concept, a computer device is also provided in an embodiment of the present invention, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the above-mentioned road boundary processing method when executing the program.

[0150] The principles of the problems solved by the above-mentioned devices, media, and related equipment in the embodiments of the present invention are similar to the aforementioned road boundary processing method, so their implementation can refer to the implementation of the aforementioned road boundary processing method, and the repeated parts will not be repeated.

[0151] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.

[0152] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0153] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0154] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0155] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for processing road boundaries, wherein the road boundaries are road boundaries of a high-precision map produced based on multi-source crowdsourcing data, characterized in that: include: Taking the road boundary included in the target road as the object to be processed, obtaining a series of road boundary points on the road boundary; Performing road marking matching within a preset spatial range with the road boundary point as a reference point, so as to associate the matched road marking with the road boundary; Determining the inner and outer sides of the target road based on the geometric position relationship between each road boundary and the associated road markings; The road boundary is moved based on the position of the road boundary inside and outside the target road.

2. The processing method according to claim 1, characterized in that: The step of determining the inner and outer sides of the target road based on the geometric position relationship between each road boundary and the associated road markings includes: Determine the data types of the road boundary and road markings included in the target road; If the data types include original road boundary data and original road marking data, statistics are performed on the geometric position relationship between each group of original road boundaries and the associated original road markings, so as to determine the inner and outer sides of the target road based on the proportion of the original road boundaries located at the positions of the original road markings in the statistical results; If the data type does not include original road boundary data and original road marking data, determine whether the road boundary and the road marking are from the same batch of data; if they are from the same batch of data, determine the inner and outer sides of the target road based on the position of the road boundary; otherwise, determine the inner and outer sides of the target road based on the road marking associated with the road boundary and whether there are other road boundaries or road markings within the spatial range of the road marking.

3. The processing method according to claim 2, characterized in that: The determining of the inner and outer sides of the target road based on the road markings associated with the road boundary as a reference and whether there are other road boundaries or road markings within the spatial range of the road markings includes: Making a judgment based on the road markings associated with the road boundary; If there is another road boundary that is not associated with the road marking within the first width threshold of the road marking, and the height of the road boundary is the same as the height of the road marking, then the side where the road boundary is located is the outer side of the target road; wherein the first width threshold is determined based on the width of the emergency lane in the target road; If there are other road markings with the same height and opposite direction within the second width threshold of the road marking, the side where the other road markings are located is the outer side of the target road; wherein the second width threshold is determined based on the width of the lane in the target road; If there are other road markings with the same height and direction within the third width threshold range of the road marking, the side where the other road markings are located is the inner side of the target road; wherein the third width threshold is determined based on the width of the lane in the target road, and the third width threshold is greater than the second width threshold.

4. The processing method according to claim 2, characterized in that: If it is determined that the data type includes original road boundary data and original road marking data, then before statistics are performed on the geometric position relationship between each group of original road boundaries and associated original road markings, the method also includes: verifying the association relationship between the original road boundaries and the associated original road markings based on the timestamp attributes in the original road boundary data and the original road marking data; and performing statistics if the timestamp attributes are the same.

5. The processing method according to claim 2, characterized in that: If the inner and outer sides of the target road are not determined based on the data types of the road boundary and road markings included in the target road, the method further includes: judging whether the target road is a herringbone road based on the attribute information of the road boundary; If it is a herringbone road, there is another herringbone road marking within the fourth width threshold range on one side of the road marking, and there are other road markings with the same height and direction within the fifth width threshold range on the other side of the road marking, then the side where the other herringbone road marking is located is the outer side of the target road; or, If it is a herringbone road, there is another herringbone road marking within the fourth width threshold range on one side of the road marking, and there is another road boundary with the same height and direction and no correlation with the road marking within the fifth width threshold range on the other side of the road marking, then the side where the other herringbone road marking is located is the outer side of the target road; The fourth width threshold is determined based on the width of a lane in the target road, and the fifth width threshold is determined based on the widths of two lanes in the target road.

6. The processing method according to claim 1, characterized in that: The moving process of the road boundary based on the position of the road boundary inside and outside the target road includes: If the road boundary is inside the target road, a series of road boundary points on the road boundary are moved to the outside of the target road by a preset moving length threshold, and the moved series of road boundary points are connected to achieve moving processing of the road boundary; If the road boundary is outside the target road, and the road marking associated with the road boundary has other road markings with the same height and opposite direction within the preset sixth width threshold, a series of the road boundary points are moved to the inside of the target road by a preset moving length threshold, and the series of road boundary points after the movement are connected to achieve movement processing of the road boundary; If the road boundary is outside the target road, and there is a road boundary point of a road boundary associated with other road markings within the preset seventh width threshold range of the road boundary point, then the straight-line distances between a series of the road boundary points and the associated road markings are calculated, and the series of road boundary points are moved to the inside of the target road by the difference between the straight-line distance and the seventh width threshold, and the series of road boundary points after the movement are connected to achieve movement processing of the road boundary; The sixth width threshold is greater than the seventh width threshold.

7. The processing method according to claim 6, characterized in that: If the difference between the straight-line distance and the seventh width threshold is less than the moving length threshold, a series of the road boundary points are moved toward the inside of the target road by a preset moving length threshold, and the series of road boundary points after the movement are connected to achieve movement processing of the road boundary.

8. The processing method according to claim 1, characterized in that: The step of acquiring a series of road boundary points on the road boundary comprises: Performing line data thinning processing on the road boundary to obtain a series of road boundary points representing the linear shape of the road boundary; or, Points are taken on the road boundary with a preset first distance threshold, and a series of road boundary points are obtained on the road boundary; wherein the first distance threshold is determined according to the smoothness of the line data of the target road and / or the road boundary.

9. The processing method according to claim 1, characterized in that: The performing road marking matching within a preset spatial range with the road boundary point as a reference point to associate the matched road marking with the road boundary includes: Taking the road boundary point as a reference point and a second distance threshold as a search range, a road marking line with the same height, same direction, same left and right sides of the road and closest to the road boundary point within the spatial range of the second distance threshold is taken as a matched road marking line; Binding the matched road markings to the road boundary to associate the matched road markings with the road boundary; The second distance threshold is determined according to the width of the emergency lane in the target road.

10. The processing method according to any one of claims 1 to 9, characterized in that: Also includes: Based on the geometric line type of the road boundary, it is determined whether there are multiple road boundaries within a preset spatial range on one side of the target road; if so, the road boundaries included in one side of the target road are redundantly processed to retain a road boundary closest to the target road.

11. A device for processing road boundaries, wherein the road boundaries are road boundaries of a high-precision map produced based on multi-source crowdsourcing data, characterized in that: include: An acquisition module, used for taking a road boundary included in a target road as an object to be processed, and acquiring a series of road boundary points on the road boundary; An association module, used to match road markings within a preset spatial range using the road boundary point as a reference point, so as to associate the matched road markings with the road boundary; A determination module, configured to determine the inner and outer sides of the target road based on a geometric positional relationship between each of the road boundaries and associated road markings; The processing module is used to perform movement processing on the road boundary based on the position of the road boundary inside and outside the target road.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the road boundary processing method according to any one of claims 1 to 10 is realized.

13. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the road boundary processing method according to any one of claims 1 to 10 is implemented.

Images