Road generation method and device

Abstract

The invention provides a road generation method and device, and the method comprises the steps: obtaining a position point forming a center line of a road; thinning the position points to obtain the position points representing the line shape of the center line as feature points; segmenting the position points according to the feature points; the position points are smoothed according to the line shape of the curve formed by the position points in the same segment, target position points in the segments are obtained, and the line shape of the curve formed by the target position points meets the line shape standard; and according to the target position point and a preset road width, generating a sideline position point of the road. The line shape of the curve formed by the target position points obtained through the smoothing processing accords with the line shape standard, namely, the line shape of the center line obtained through the smoothing processing accords with the line shape standard, so that the side line shape formed by the generated side line position points of the road based on the target position points is also standard; therefore, the data quality of the road with the width formed by the sideline position points is ensured to meet the requirement.

Description

Technical Field

[0001] This disclosure relates to the field of map technology, and in particular to a road generation method and apparatus. Background Technology

[0002] Map data records information about various entities in the real world, such as roads, buildings, and shops. In map data, roads are typically represented by line elements, and in most cases, the road is represented by a point on its centerline. Since lines have no width, this does not align with the fact that real-world roads have width. Therefore, existing technologies generate two side lines of the road based on its centerline to create a road with width.

[0003] However, the inventors of this disclosure have discovered that existing technologies generate road edge point positions by offsetting the angle bisector of the position points on the road's centerline by a certain distance to both sides of the centerline. Since the road's centerline itself is not smooth enough, directly generating edge point positions based on the centerline position points results in even more uneven road edges, leading to data quality that does not meet requirements for the width of the road formed by these edge point positions. Therefore, a technical solution is needed to generate width-bound roads with satisfactory data quality. Summary of the Invention

[0004] In view of this, the present disclosure provides a road generation method and apparatus, the main purpose of which is to smooth the position points of the center line of the road to obtain the target position points of the smooth curve, and then generate the edge position points of the road based on the target position points, so that the edge lines of the generated road are smooth as a whole, and a road with width that meets the data quality requirements is obtained.

[0005] To address the aforementioned problems, this disclosure primarily provides the following technical solutions:

[0006] In a first aspect, this disclosure provides a road generation method, comprising: obtaining the location points constituting the centerline of the road; thinning the location points to obtain location points representing the linear shape of the centerline as feature points; segmenting the location points according to the feature points; smoothing the location points according to the linear shape of the curve formed by the location points in the same segment to obtain target location points in the segment, wherein the linear shape of the curve formed by the target location points conforms to the linear shape standard; and generating the edge line location points of the road according to the target location points and a preset road width.

[0007] Secondly, this disclosure also provides a road generation apparatus, comprising: a data acquisition unit for acquiring the location points constituting the centerline of the road; a feature point determination unit for thinning the location points to obtain location points representing the linear shape of the centerline as feature points; a segmentation unit for segmenting the location points according to the feature points; a first processing unit for smoothing the location points according to the linear shape of the curve formed by the location points in the same segment to obtain target location points in the segment, wherein the linear shape of the curve formed by the target location points conforms to the linear shape standard; and a second processing unit for generating the edge line location points of the road according to the target location points and a preset road width.

[0008] Thirdly, this disclosure also provides a storage medium including a stored program, wherein the program, when running, controls the device where the storage medium is located to execute the road generation method as described above.

[0009] Fourthly, this disclosure also provides a processor for running a program, wherein the program executes the road generation method as described above during runtime.

[0010] Fifthly, this disclosure also provides a terminal device, including at least one processor; the storage medium is used to store a program executed by the processor, and data required by the processor during the execution of the program;

[0011] The processor is configured to run a program, wherein the program executes the road generation method as described above during runtime.

[0012] The road generation method and apparatus disclosed herein involve: acquiring the location points constituting the centerline of the road; thinning the location points to obtain location points representing the linear shape of the centerline as feature points; segmenting the location points according to the feature points; smoothing the location points according to the linear shape of the curve formed by the location points in the same segment to obtain target location points in the segment, wherein the linear shape of the curve formed by the target location points conforms to the linear shape standard; and generating the edge line location points of the road based on the target location points and a preset road width. Since the linear shape of the curve formed by the target location points obtained through smoothing conforms to the linear shape standard, that is, the linear shape of the smoothed centerline conforms to the linear shape standard, the edge line linear shape formed by the generated road edge line location points based on the target location points is also standard, thereby ensuring that the data quality of the road with width formed by these edge line location points meets the requirements. Attached Figure Description

[0013] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this disclosure. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0014] Figure 1 A schematic diagram of the edge lines generated by existing technology is shown;

[0015] Figure 2 A flowchart of a road generation method provided in an embodiment of this disclosure is shown;

[0016] Figure 3 This illustration shows a schematic diagram of the location of a feature point in a centerline according to an embodiment of the present disclosure;

[0017] Figure 4 This illustration shows a schematic diagram of another location point of a feature point in the center line provided by an embodiment of the present disclosure;

[0018] Figure 5 This illustration shows a schematic diagram of determining a first sampling point and initial feature points according to an embodiment of the present disclosure;

[0019] Figure 6 This illustration shows a schematic diagram of determining a second sampling point and a feature point according to an embodiment of the present disclosure;

[0020] Figure 7 This illustration shows a schematic diagram of calculating the maximum distance from a location point to a connecting line in a segment, according to an embodiment of this disclosure.

[0021] Figure 8 This illustration shows a schematic diagram of two position points located at both ends of a segment as target position points in the segment, according to an embodiment of the present disclosure.

[0022] Figure 9 This illustration shows a schematic diagram of determining the fitting position point of a segment based on optimal Bezier fitting, according to an embodiment of the present disclosure.

[0023] Figure 10 This illustration shows a schematic diagram of obtaining target position points based on Gaussian smoothing for segmentation, according to an embodiment of this disclosure.

[0024] Figure 11 This illustration shows a schematic diagram of the determination and deletion process of edge position points that cause edge overlap, according to an embodiment of the present disclosure.

[0025] Figure 12 This illustration shows a schematic diagram of a method for smoothly connecting curves formed by target position points in two adjacent segments, according to an embodiment of this disclosure.

[0026] Figure 13 This illustration shows a schematic diagram of a curve formed by the location points of the center line and the target location points, according to an embodiment of the present disclosure.

[0027] Figure 14 This illustration shows a road width that is uneven due to misalignment between the left and right lane position points, as provided in an embodiment of this disclosure.

[0028] Figure 15 This illustration shows a road width that is uniform when the left and right lane position points are aligned, according to an embodiment of the present disclosure.

[0029] Figure 16 A structural diagram of a road generation device provided in an embodiment of this disclosure is shown;

[0030] Figure 17 A structural diagram of another road generation apparatus provided in an embodiment of this disclosure is shown;

[0031] Figure 18 A structural diagram of yet another road generation device provided in an embodiment of this disclosure is shown;

[0032] Figure 19 A structural diagram of another road generation device provided in an embodiment of this disclosure is shown. Detailed Implementation

[0033] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0034] In related technologies, such as Figure 1 As shown, because the centerline of a road itself is not smooth, the road edges generated based on this uneven centerline will be even more uneven. For example, if the angle bisector of a point on the centerline is offset to either side of the centerline by a certain distance to generate the edge points, the resulting edge lines will be even more uneven, exhibiting burrs, protruding sharp corners, or obvious concavities. More seriously, in some cases, this can lead to… Figure 1 The problem shown is that the edge line formed by the edge line position points overlaps with itself at a certain position. However, the edge lines of real-world roads do not have this problem. Therefore, the data quality of the wide roads formed by the edge line position points does not meet the requirements.

[0035] Based on this, embodiments of this disclosure provide a road generation method and apparatus. The method includes: acquiring the location points constituting the centerline of the road; thinning the location points to obtain location points representing the linear shape of the centerline as feature points; segmenting the location points according to the feature points; smoothing the location points according to the linear shape of the curve formed by the location points in the same segment to obtain target location points in the segment, wherein the linear shape of the curve formed by the target location points conforms to the linear shape standard; and generating the edge line location points of the road based on the target location points and a preset road width. Since the linear shape of the curve formed by the target location points obtained by the smoothing process conforms to the linear shape standard, that is, the linear shape of the centerline after smoothing conforms to the linear shape standard, the edge line linear shape formed by the generated road edge line location points based on the target location points is also standard, thereby ensuring that the data quality of the road with width formed by these edge line location points meets the requirements.

[0036] like Figure 2 As shown, this disclosure provides a road generation method, which includes:

[0037] S1: Obtain the location of the centerline that forms the road.

[0038] Understandably, in standard map data (ordinary map data), roads in the real world are abstracted as directional lines, and these lines are represented by a series of location points, typically located along the road's centerline. Therefore, to expand a road with width, it's necessary to obtain the location points constituting the road's centerline; this data can come from pre-made standard map data. For example... Figure 1 As shown, the solid black circles represent the locations of the center lines that form the road.

[0039] S2: Thin out the position points to obtain the position points that represent the linear shape of the center line as feature points.

[0040] Understandably, since there are a large number of points that make up the centerline of a road, thinning out these points to determine those that can represent the shape of the centerline can both ensure that the shape of the centerline after smoothing retains its characteristics and improve the efficiency of smoothing.

[0041] For example, if a segment of the centerline is a straight line, the feature points are preferably located at two points at both ends of that straight line. If a segment of the centerline is an arc, the feature points are preferably located at two points at both ends of that arc, and at a point between the ends that can characterize the curvature of the arc.

[0042] In a specific implementation, the process of thinning out the position points to obtain the position points representing the linear shape of the center line as feature points can be implemented using the following example:

[0043] Example 1: Using a preset thinning distance threshold (thinning threshold), Douglas thinning is performed on the location points to obtain the location points representing the linear shape of the center line as feature points. Those skilled in the art can refer to existing Douglas thinning algorithms to implement the scheme of this disclosure; this disclosure will not elaborate on the Douglas thinning algorithm.

[0044] In the process of implementing Example 1, the inventors of this disclosure discovered that if the thinning distance threshold is large, the curve formed by the location points of the feature points tends to deviate from the center line formed by the location points before thinning, such as... Figure 3 The curve formed by the feature points circled in dashed lines deviates significantly from the center line formed by the points before thinning. If a smaller thinning distance threshold is used, the number of feature points is large, and the curve formed by these feature points retains features that the center line does not need to be preserved. For example... Figure 4 The part circled in dashed lines has a center line that is clearly a straight line, but the curve formed by the thinned feature points shows a straight line-arc-straight line-arc-straight line shape, which does not conform to the actual situation of the line shape.

[0045] Therefore, in practical applications, if the accuracy requirements of the line shape formed by the feature points are not high in the business scenario, Example 1 can be adopted. To address the problems existing in Example 1 during practice, this disclosure further provides Example 2, as follows:

[0046] Example 2: The location points are subjected to Douglas thinning twice to obtain the location points that represent the linear shape of the center line as feature points. The first distance threshold used in the first Douglas thinning is less than the second distance threshold used in the second Douglas thinning.

[0047] That is, based on the first distance threshold, all the position points constituting the center line are subjected to the first Douglas thinning to obtain the retained position points; then, for the retained position points, a second Douglas thinning is performed based on the second distance threshold to obtain the position points that represent the linear shape of the center line as feature points.

[0048] In this embodiment of the disclosure, by performing Douglas thinning on the location points twice, feature points that accurately represent the shape of the center line can be obtained, so that the curve formed by the feature points conforms to the actual shape of the line and avoids the curve formed by the thinned feature points deviating from the original center line.

[0049] In some embodiments, the centerline location points are arranged in order of road direction, that is, the centerline location points are arranged in an ordered manner. In this case, the two Douglas thinning operations on the location points to obtain the feature points characterizing the line shape of the centerline can be implemented as follows:

[0050] Based on the first distance threshold, the position points of the center line are subjected to the first Douglas thinning to obtain the position points as the first sampling points.

[0051] Obtain the distance between two adjacent first sampling points, divide the consecutive first sampling points with a distance less than the preset third distance threshold into a segment, and mark the first sampling point at the beginning and end of the segment as the initial feature point;

[0052] Based on the second distance threshold, the location point used as the first sampling point is subjected to a second Douglas thinning to obtain the location point used as the second sampling point.

[0053] The initial feature points located between two adjacent second sampling points and ranked first and last are used as feature points representing the linear shape of the center line.

[0054] In some embodiments, a fixed distance value (e.g., 5 meters or 10 meters) can be set, and the product of the centerline length and a set coefficient can be obtained. The set coefficient is a positive number less than 1, such as 0.3, 0.25, or 0.2. Then, the smaller value between the fixed distance value and the product value is selected as the third distance threshold. Of course, the third distance threshold can also be set by a technician based on experience, without affecting the implementation of this disclosure.

[0055] The following is in conjunction with the appendix Figure 5 and attached Figure 6 The specific process of the two Douglas thinning procedures will be explained.

[0056] like Figure 5 As shown, the centerline locations are arranged in order of road direction. Based on a first distance threshold, the locations constituting the centerline undergo a first Douglas thinning process, and the retained locations are used as the first sampling points. The first sampling points are as follows: Figure 5 The black dots in the left image; for the first sampling point, the distance between two adjacent first sampling points is obtained, and consecutive first sampling points with a distance less than a preset third distance threshold are grouped into a segment, such as... Figure 5 In the left image, the black dots within the same dashed circle represent sampling points belonging to the same segment; the first and last sampling points in each segment are marked as initial feature points. Figure 5 The gray dot in the right image.

[0057] After determining the first sampling point and the initial feature points, as follows: Figure 6 As shown, based on the second distance threshold, the location point used as the first sampling point is subjected to a second Douglas thinning to obtain the location point used as the second sampling point. Figure 6 The gray dots in the middle left figure), and the initial feature points located between two adjacent second sampling points and ranked first and last are used as feature points representing the linear shape of the center line ( Figure 6 The black dot circled in the right image is... Figure 5 (The gray dots shown in the right figure) Specifically:

[0058] See Figure 6 In the left image, the first and second black circles contain gray dots representing two adjacent second sampling points. The point in the first black circle corresponds to... Figure 5 The first point in the right figure was also marked as an initial feature point during the first thinning process. When determining feature points, this point was located at... Figure 6 The initial feature point located between the second black circle and the second black circle in the left image, and which is the first in the sequence, is therefore selected as the feature point. Furthermore, combining... Figure 5 As shown in the right figure, it is located Figure 6 The initial feature point between the first and second black circles in the left image, and which is the last one in the sequence, is... Figure 5 The sixth gray dot from the top in the right image, which is... Figure 6 The second black circle in the right image shows the point.

[0059] It should be noted that the values ​​of the first distance threshold and the second distance threshold can be set as needed, as long as the first distance threshold is less than the second distance threshold. This disclosure does not impose any specific restrictions on this.

[0060] Please continue reading Figure 2 S3 is executed after S2.

[0061] S3: Segment the location points according to feature points. That is, use the location points as feature points to segment the location points that constitute the centerline of the road.

[0062] For example, there are 10 points that make up the center line of a road. According to the order, the 3rd and 7th points are feature points. Then, points 1-3 are divided into one segment, points 4-7 into another segment, and points 8-10 into yet another segment.

[0063] As mentioned above, this disclosure provides two implementation methods for thinning feature points. In Implementation 1, a preset thinning distance threshold is used to perform a single Douglas thinning operation. However, if the thinning distance threshold is large, the curve formed by the feature point locations tends to deviate from the centerline formed by the locations before thinning. Furthermore, when segmenting the location points based on the feature points, if… Figure 3 As shown, the ends of the resulting segments will shift relative to the centerline, and there is a tendency for the segments to warp at the ends. Conversely, if the thinning distance threshold is small, more location points will be thinned and identified as feature points. The curve formed by these feature points retains features that the centerline does not need to retain. However, when segmenting location points based on feature points, there is a problem of over-segmentation, such as... Figure 4As shown by the dashed circle, what was originally a straight line segment will be divided into three segments due to retaining too many position points as feature points.

[0064] In this embodiment, by performing Douglas thinning twice on the location points, feature points that accurately represent the linear shape of the center line can be obtained. This makes the segmentation of the location points according to the feature points more accurate and avoids... Figure 3 The segment ends are raised and Figure 4 The problem of excessive segmentation is shown.

[0065] S4: Smooth the position points according to the curve shape formed by the position points in the same segment to obtain the target position points in the segment. The curve shape formed by the target position points conforms to the line shape standard.

[0066] In the real world, common road shapes include straight roads and curved roads. This disclosure abstracts real-world roads as lines for processing. Since the abstracted lines are also made based on the real-world conditions of the roads, for the embodiments of this disclosure, the curve formed by the target location points usually includes straight lines (straight roads) and arcs (curves). The curve formed by the target location points conforms to the line shape standard, which can be understood as: the curve formed by the target location points conforms to the straight line line shape standard or the arc line shape standard.

[0067] In this embodiment of the disclosure, the position points are smoothed according to the shape of the curve formed by the position points in the same segment. If the shape is a straight line, the position points are smoothed to obtain the target position point in the segment, and the shape of the curve formed by the target position points conforms to the straight line shape standard. If the shape is an arc, the position points are smoothed to obtain the target position point in the segment, and the shape of the curve formed by the target position points conforms to the arc shape standard.

[0068] In some embodiments, the position points are smoothed according to the curve formed by the position points in the same segment to obtain the target position points in the segment. The curve formed by the target position points conforms to the line shape standard. A specific embodiment can be as follows:

[0069] Determine the line connecting two points located at opposite ends of the same segment;

[0070] Determine the distances from other points in the segment to the connecting line. If the largest distance is less than the segment distance threshold, then the curve formed by the points in the segment is determined to be approximately a straight line. In this case, the two points at the two ends of the segment are retained as the target points in the segment, and the other points in the segment are deleted.

[0071] If the largest distance is greater than the segment distance threshold, the curve formed by the position points in the segment is determined to be arc-shaped. Then, based on the curve fitting algorithm, the position points in the segment are smoothed to obtain the fitted position points. If the fitting error of the curve formed by the fitted position points is less than the error threshold, the fitted position points are taken as the target position points in the segment.

[0072] In this embodiment of the disclosure, the same segment distance threshold can be generated for all segments, or different segment distance thresholds can be generated for different segments. For example, the corresponding segment distance threshold can be generated based on the length of the line connecting the two endpoints of the segment. When generating the segment distance threshold, it is also necessary to consider whether the threshold can accurately distinguish between straight lines and arcs. Therefore, in specific implementation, a segment distance threshold that can distinguish between straight lines and arcs can also be set based on experience. When generating the final segment distance threshold, a more reasonable value can be selected from the empirically set threshold and the calculated threshold as the final segment distance threshold.

[0073] The following describes, with reference to the accompanying drawings, a specific embodiment of smoothing the curve formed by the position points in the same segment to obtain the target position points in the segment, so that the curve formed by the obtained target position points conforms to the line shape standard.

[0074] In this embodiment of the disclosure, such as Figure 7 As shown, the dashed line indicates a segment. The dashed line is the line connecting two points at either end of this segment. By calculating the distances from other points in the segment to this line, the distance with the largest distance can be determined. Figure 7 (as shown, maximum distance), if Figure 7 If the maximum distance shown is less than the segment distance threshold corresponding to the segment, then the curve formed by the position points in the segment is determined to be a straight line. The two position points at the two ends of the segment can be retained as the target position points in the segment, and the other position points in the segment can be deleted. Figure 8 As shown, the smoothed segments conform to the straight line shape standard.

[0075] In this embodiment of the disclosure, when it is determined that the curve formed by the position points in the segment is a straight line, the two position points at both ends of the segment are retained as the target position points in the segment, and the other position points in the segment are deleted. Since the two position points can accurately represent a straight line, the segment formed by the two position points retained as target position points will not have a problem of being unsmooth, and the curve formed by the target position points of the segment conforms to the straight line standard.

[0076] Furthermore, if the largest distance in the segment is greater than the segment distance threshold, the curve formed by the position points in that segment is determined to be an arc. Based on a curve fitting algorithm, the position points in the segment can be smoothed to obtain fitted position points. If the fitting error of the curve formed by the fitted position points is less than the error threshold, the fitted position point is used as the target position point in the segment. For example, the error threshold can be set to 1.0 or any other arbitrary value.

[0077] In this embodiment of the disclosure, based on a curve fitting algorithm, the position points in the segment are smoothed to obtain fitted position points. This smoothing can be based on a Bézier curve or an optimal Bézier curve. Specifically, if the segment is divided into a first position point and a second position point as endpoints, then based on the direction of the segment, a first tangent direction is determined for the first position point, and a second tangent direction is determined for the second position point. Then, based on the first position point, the second position point, the first tangent direction, and the second tangent direction, the position points in the segment are smoothed using a Bézier curve or an optimal Bézier curve to obtain the fitted position points.

[0078] like Figure 9 As shown, based on the optimal Bézier curve, for Figure 9 The position points in the segment shown in the left figure are smoothed to obtain the fitted position points. If Figure 9 If the fitting error of the curve formed by the fitting points shown in the right figure is less than 1.0, then... Figure 9 The fitted position point shown in the right figure is used as the target position point in the segmentation.

[0079] In cases where the fitting error of the curve formed by the fitted position points is greater than the error threshold, in some embodiments, if the fitting error of the curve formed by the fitted position points is not less than the error threshold, a Gaussian smoothing algorithm is used to smooth the position points in the segment to obtain the target position points in the segment.

[0080] like Figure 10 As shown in the left figure, the fitting error of the curve formed by the fitted position points is significantly greater than the error threshold. Therefore, a Gaussian smoothing algorithm is used to smooth the position points in the segment, resulting in the following: Figure 10 The target location point in the segment shown in the right figure.

[0081] In this embodiment, when the curve formed by the position points in the segment is determined to be an arc, the position points in the segment are smoothed using a curve fitting algorithm to obtain fitted position points. If the fitting error of the curve formed by the fitted position points is less than an error threshold, the fitted position points are used as target position points in the segment. Since the target position points in the segment are determined based on the curve fitting algorithm and the fitting error is less than the threshold, the curve formed by the target position points will not have any unsmoothness issues, and the curve shape conforms to the arc shape standard. Furthermore, when the fitting error is greater than the error threshold, this disclosure uses a Gaussian smoothing algorithm to smooth the position points in the segment to obtain target position points in the segment. Since the target position points in the segment are further determined based on the Gaussian smoothing algorithm, the curve formed by the target position points will also not have any unsmoothness issues, and the curve shape also conforms to the arc shape standard.

[0082] Please continue reading Figure 2 S5 is executed after S4.

[0083] S5: Generate the edge point of the road based on the target location point and the preset road width.

[0084] In some embodiments, the preset road width is the actual road width, which can be determined based on the number of lanes and the lane width. Since the preset road width is the actual road width, after generating the edge point locations of the road, the edge point locations constitute the two edge points of the road. The road with width formed by these two edge points can be stitched and fitted with the roads in the high-precision map data.

[0085] In this embodiment of the disclosure, after determining the target location point and the preset road width, the edge line location points of the road can be determined using the angle bisector method based on the target location point and the preset road width. For example, the target location point can be offset by half of the preset road width along the directions of the two angle bisectors of the curve formed by the target location point. It is understood that the edge line location points include the left edge line location point and the right edge line location point, wherein the left edge line location point constitutes the left edge line and the right edge line location point constitutes the right edge line.

[0086] In practical applications, the road edge formed by the edge point may have the following characteristics: Figure 1 The edge line (reverse) folding issue shown needs to be detected and resolved. It should be noted that edge line (reverse) folding can also be referred to as edge line overlap, reverse overlap, etc.

[0087] To ensure that the edge lines do not fold in reverse, in this embodiment of the disclosure, after generating the edge line location points of the road, the following steps are also included:

[0088] Check if there are any edge point locations on the same edge that would cause the edge to overlap; if so, delete the edge point location that causes the edge to overlap from the edge point location of the road.

[0089] Under normal circumstances, if there is no overlap between the edges, the intersection points of the edges can be calculated pairwise, and there will be at most one intersection point. If there are more than one intersection point when calculating the intersection points of the edges, such as two intersection points, then it can be determined that the two edges overlap.

[0090] In some embodiments, the intersection points of the line segments formed by the edge point locations can be calculated pairwise to detect the edge point locations that cause the edges to overlap, but the time complexity of this method is O(n^2). 2 ), where n is the number of edge position points. If the number of edge position points is very large, the time complexity is high.

[0091] To reduce time complexity, this disclosure provides a method for detecting whether there are edge position points on the same edge that would cause the edge to overlap, comprising:

[0092] Divide the edge points arranged in sequence along the same edge into segments;

[0093] For a given segment, determine the line connecting the two edge points located at the two ends of the segment;

[0094] Get the distance from the edge position point in the segment to the projection point of the connecting line, and the distance from the projection point to the specified edge position point, where the specified edge position point is any one of the two edge position points located at the two ends of the segment;

[0095] Based on the sorting of the edge position points in the segment, the distances from the projection points of the edge position points to the specified edge position points are compared. If the distance from the projection point of the current edge position point to the specified edge position point is less than the largest distance among the previously compared edge position points, then the current edge position point is determined as the edge position point that caused the edge to overlap.

[0096] In this embodiment of the disclosure, the edge line position points arranged in sequence on the edge line are segmented. The edge line position points within a preset step length are sequentially assigned to the same segment according to their arrangement order. The preset step length is a step length coefficient multiplied by a preset road width. For example, if the step length coefficient is 4, then the preset step length is 4 times the preset road width.

[0097] The following is in conjunction with the appendix Figure 11 The specific process of detecting and deleting edge points that cause edge overlap is explained.

[0098] like Figure 11 As shown in the left figure, the edge lines formed by the edge point overlap at the bend, such as... Figure 11 As shown in the middle figure, the edge points arranged sequentially on the same edge are segmented using a preset step size, such as... Figure 11 As shown in the middle diagram, the edge line is divided into two segments, and the line connecting the two edge line points at the two ends of each segment is... Figure 11 The unidirectional arrow line shown in the middle diagram obtains the projection point of the line from the edge position point in the segment, and the distance from the projection point to the specified edge position point, such as... Figure 11 As shown in the middle diagram, the edge points pointed to by the red arrows are ordered before the edge points pointed to by the green arrows. When traversing to the edge point pointed to by the red arrow, the distance from the projection point of that edge point to the baseline (one-way arrow line) to the specified edge point is the largest. However, when traversing to the edge points pointed to by the green arrows, as... Figure 11 As shown in the middle diagram, the distance is significantly smaller than the distance to the edge point indicated by the red arrow. Therefore, it can be determined that this segment of the edge has an overlapping problem.

[0099] In this embodiment of the disclosure, after determining the edge point that causes the edge overlap, the edge point causing the edge overlap is deleted from the road edge point list, thus solving the problem of edge overlap (or reverse folding). Figure 11 As shown in the right figure, this makes the overall edge lines of the final road smooth, improves the road's aesthetics, and ensures that the data quality of the road with width formed by these edge line locations meets the requirements.

[0100] Understandably, the aforementioned method processes the location points into segments to obtain the target location points. However, whether the curves formed by the target location points in adjacent segments can smoothly connect at the junctions still needs to be checked. Figure 12 As shown in the left figure, if the curves formed by the target positions in adjacent segments do not connect smoothly at the junction, further processing is required to make them connect smoothly (e.g., Figure 12 (As shown in the figure on the right).

[0101] Based on this, in this embodiment of the disclosure, before generating the edge position points of the road based on the target position point and the preset road width, the method further includes:

[0102] The algorithm checks whether the curves formed by the target positions in two adjacent segments are smoothly connected. If not, some target positions are deleted from the two segments. Based on the remaining target positions after deleting some target positions from the two adjacent segments, a curve fitting algorithm is used to generate target positions for smoothly connecting the two segments. The curve fitting algorithm can be a Bézier curve or an optimal Bézier curve.

[0103] In some embodiments, detecting whether the curves formed by target position points in two adjacent segments are smoothly connected includes: for two adjacent segments, obtaining a first straight line formed by the last two target position points in the previous segment, and obtaining a second straight line formed by the first two target position points in the next segment; determining the angle formed by the first straight line and the second straight line, and determining whether the angle is less than a preset angle threshold; and determining whether the curves formed by target position points in two adjacent segments are smoothly connected based on whether the angle is less than the preset angle threshold.

[0104] It is understandable that the intersection of the first and second straight lines will generate four included angles. Among these four included angles, the angle with the opening facing the inside of the curve formed by the target position points in the two adjacent segments is determined, and it is judged whether this included angle is less than a preset included angle threshold.

[0105] Of course, if the first line is perpendicular to the second line, and all four included angles are 90°, then the included angle formed by the first line and the second line can be any one of the four included angles.

[0106] In some embodiments, the included angle threshold is 170°, 175°, or 180°, etc.

[0107] In this embodiment of the disclosure, by determining whether the included angle between two adjacent segments at the splicing point is less than a preset included angle threshold, it can be determined whether the curve formed by the target position points in the two adjacent segments is smoothly connected, so as to process the two adjacent segments that are not smoothly connected and achieve smooth connection.

[0108] In this embodiment of the disclosure, the target location points to be deleted from two segments can be deleted from each segment within a preset length from the connection point of the two segments. In some embodiments, the preset length is 5 meters, 4 meters, or 1 meter, etc.

[0109] In some embodiments, the preset length is related to the length of the shortest of two adjacent segments, for example, the preset length is one-tenth of the length of the shortest of two adjacent segments.

[0110] In some embodiments, the preset length is related to the degree of unevenness between two adjacent segments. For example, the preset length is related to the angle formed by the joint between two adjacent segments. Specifically, the preset length is related to the angle (referred to as the first angle) formed by the opening of the opening towards the inside of the curve formed by the target position point in the two adjacent segments among the four angles formed by the first straight line and the second straight line.

[0111] For example, the preset length is inversely proportional to the first angle; the larger the first angle, the smaller the preset length, and vice versa. Alternatively, the preset length is directly proportional to the first angle; the larger the first angle, the larger the preset length, and vice versa.

[0112] In this embodiment of the disclosure, a method for generating target position points for smoothly connecting two segments using a Bézier curve or an optimal Bézier curve is provided, based on the remaining target position points after deleting a portion of the target position points in two adjacent segments. The method includes: determining, among the remaining target position points after deleting a portion of the target position points in two adjacent segments, the first target position point in the previous segment that is closest to the connection point of the two segments, the second target position point in the subsequent segment that is closest to the connection point of the two segments, the first target direction from the tangent of the previous segment at the first target position point to the subsequent segment, and the second target direction from the tangent of the subsequent segment at the second target position point to the previous segment; and then generating target position points for smoothly connecting two segments based on the first target position point, the second target position point, the first target direction, and the second target direction, using a Bézier curve or an optimal Bézier curve.

[0113] In this embodiment of the disclosure, by detecting whether the curve formed by the target position points in two adjacent segments is smoothly connected, the two adjacent segments that are not smoothly connected are processed to generate target position points for smoothly connecting the two segments, so that the curve formed by the target position points of different segments conforms to the linear standard, and the target position points for smoothly connecting the two adjacent segments are obtained. Thus, the edge line of the road formed by the edge position points obtained from the target position points can achieve overall smoothness and improve the aesthetics of the road.

[0114] To facilitate understanding of the embodiments of this disclosure, the following is combined with Figure 13 The road generation method provided in the embodiments of this disclosure will be illustrated by example.

[0115] like Figure 13As shown in the left figure, the location points constituting the centerline of the road are obtained. These points are then divided into segments 1, 2, and 3. The location points are smoothed according to the curve formed by the location points within the same segment to obtain the target location points within each segment. Furthermore, the smoothness of the curve formed by the target location points in adjacent segments is checked. If the curves are not smooth, some target location points are deleted from the two segments. Then, based on the remaining target location points after deleting some from adjacent segments, a curve fitting algorithm is used to generate target location points for smoothly connecting the two segments. This ensures that the curve formed by the final target location points conforms to the standard curve shape. Figure 13 The right-hand diagram shows smoothing of segment 1, segment 2, and segment 3.

[0116] It should be noted that, Figure 13 For detailed steps in the example, please refer to the description of the relevant steps above, which will not be repeated here.

[0117] In this embodiment of the disclosure, road edge position points are generated based on the target position point and the preset road width. The edge position points include left edge position points and right edge position points. The left edge position points constitute the left edge, and the right edge position points constitute the right edge. The left edge and right edge are located on both sides of the center line, and the left edge position points and right edge position points are located on both sides of the target position point on the center line. The number of left edge position points and right edge position points is the same, and they are symmetrical with respect to the target position point. Therefore, when determining the road width, it can be determined based on the distance between any mutually symmetrical left edge position points and right edge position points.

[0118] However, if there are overlapping edge points on the same edge line, deleting these overlapping edge points from the road edge point list will result in an inconsistency in the number of edge points on the left and right edges. This makes it impossible to calculate the road width based on the distance between any symmetrical left and right edge points. Therefore, it is necessary to align the left and right edge points to facilitate subsequent road width calculations.

[0119] An alignment method, such as Figure 14 As shown, by fixing one edge line and determining the corresponding edge line position on another edge line based on the position points on that edge line, the alignment points can be proportionally aligned. However, this alignment method can lead to uneven road width, such as... Figure 14 As shown, the width of the road determined by the aligned edge line positions is uneven. It can be seen that the width of the road is uneven after the road turns.

[0120] Based on this, the present disclosure also provides a method for aligning edge position points, including:

[0121] Based on the feature points of the center line, determine the edge feature points on the left and right lines respectively, and the distance from the edge feature point to the feature point is the shortest.

[0122] The feature points of the centerline can be the feature points obtained in step S2, or they can be the feature points obtained by thinning the target position points obtained in step S4. This does not affect the implementation of this disclosure. The target position points obtained in step S4 can be thinned once or twice using Douglas thinning. For the specific thinning process, please refer to the relevant sections above, which will not be repeated here.

[0123] The edge position points on the left side line are segmented using the edge feature points of the left side line;

[0124] The right edge feature points are used to segment the edge position points on the right edge.

[0125] For edge feature points determined by the same two feature points, segments are formed. For the segments corresponding to the left and right lines, edge position points in the segments corresponding to the left or right lines are sampled according to the principle of equal distance, so that the number of edge position points in the segments corresponding to the left and right lines is the same, and the number is equal to the total number of edge position points in the segment with more edge position points.

[0126] In some embodiments, the point where the perpendicular line from the center line at the feature point intersects the left line is determined as the edge feature point on the left line, and the point where the perpendicular line from the center line at the feature point intersects the right line is determined as the edge feature point on the right line.

[0127] The following is in conjunction with the appendix Figure 15 The specific process of the edge position point alignment method provided in the embodiments of this disclosure will be described.

[0128] like Figure 15 As shown, the feature points of the centerline are determined, along with the perpendicular lines drawn from these feature points to the centerline. The intersections of these perpendicular lines with the left and right side lines are defined as edge feature points. The edge position points between edge feature points determined by two adjacent feature points are divided into the same segment, as shown below. Figure 15 Determine segment 1 and segment 2 of the left line, and segment 1 and segment 2 of the right line.

[0129] Specifically, for segment 1 of the left and right lines, edge point sampling is performed on the corresponding segments of the left and right lines according to the principle of proportional distance. This ensures that the number of edge point locations in the corresponding segments of the left and right lines is the same, equal to the total number of edge point locations in the segment with more edge point locations. For example, if segment 1 of the left line has 4 edge point locations and segment 1 of the right line has 2 edge point locations, then based on the 4 edge point locations in segment 1 of the left line, in segment 1 of the right line, following the principle of proportional distance (i.e., based on the proportion of distance between the 4 edge point locations in segment 1 of the left line), 4 edge point locations are sampled proportionally in segment 1 of the right line, ensuring that segment 1 of the left and right lines also includes 4 edge point locations.

[0130] For example, if there are 10 edge point locations in segment 2 of the left line and 4 edge point locations in segment 1 of the right line, then based on the 10 edge point locations in segment 1 of the left line, in segment 1 of the right line, according to the principle of proportional distance, that is, based on the proportion of distance of the 10 edge point locations in segment 1 of the left line to the distance in segment 1 of the left line, 10 edge point locations will be sampled in the same proportion in segment 1 of the right line, so that segment 1 of the left line and segment 1 of the right line also include 10 edge point locations.

[0131] In summary, it can achieve alignment of the left and right edge line positions, and ensure that the determined road width is uniform when the road width is determined based on the aligned edge line positions.

[0132] In this embodiment of the disclosure, the left and right sideline positions are sampled in the same number and at equal distances according to different segments, so that the left and right sideline positions can be aligned. When determining the road width based on the aligned sideline positions, the determined road width is guaranteed to be uniform.

[0133] Based on the above method embodiments, this disclosure also provides a road generation device 1, such as... Figure 16 As shown, the device includes: a data acquisition unit 11, a feature point determination unit 12, a segmentation unit 13, a first processing unit 14, and a second processing unit 15.

[0134] The data acquisition unit 11 is used to acquire the location points that constitute the centerline of the road.

[0135] The feature point determination unit 12 is used to thin out the position points to obtain the position points that represent the linear shape of the center line as feature points.

[0136] Segmentation unit 13 is used to segment the position points according to feature points.

[0137] The first processing unit 14 is used to smooth the position points according to the curve formed by the position points in the same segment, so as to obtain the target position points in the segment, and the curve formed by the target position points conforms to the line shape standard.

[0138] The second processing unit 15 is used to generate the edge position points of the road based on the target position point and the preset road width.

[0139] In some embodiments, the feature point determination unit 12 is specifically used to: perform Douglas thinning on the location points twice to obtain the location points representing the linear shape of the center line as feature points, wherein the first distance threshold used in the first Douglas thinning is less than the second distance threshold used in the second Douglas thinning.

[0140] In some embodiments, the position points of the centerline are arranged in the order of road direction. The feature point determination unit 12 is specifically used to perform a first Douglas thinning on the position points of the centerline based on a first distance threshold to obtain position points as first sampling points; obtain the distance between two adjacent first sampling points, divide the consecutive first sampling points with a distance less than a preset third distance threshold into a segment, and mark the first sampling points at the first and last positions in the segment as initial feature points; perform a second Douglas thinning on the position points as first sampling points based on a second distance threshold to obtain position points as second sampling points; and use the initial feature points located between two adjacent second sampling points and ranked at the first and last positions as feature points characterizing the linear shape of the centerline.

[0141] In some embodiments, the first processing unit 14 is specifically used to determine the line connecting two position points located at both ends of the same segment; determine the distance from other position points in the segment to the connecting line; if the largest distance is less than the segment distance threshold, then the curve formed by the position points in the segment is determined to be approximately a straight line, and the two position points located at both ends of the segment are retained as target position points in the segment, while other position points in the segment are deleted; if the largest distance is greater than the segment distance threshold, then the curve formed by the position points in the segment is determined to be an arc, and the position points in the segment are smoothed based on a curve fitting algorithm to obtain fitted position points; if the fitting error of the curve formed by the fitted position points is less than the error threshold, then the fitted position points are used as target position points in the segment.

[0142] In some embodiments, the first processing unit 14 is further configured to use a Gaussian smoothing algorithm to smooth the position points in the segment if the fitting error of the curve formed by the fitted position points is not less than the error threshold, so as to obtain the target position point in the segment.

[0143] In some embodiments, such as Figure 17As shown, the device also includes an overlap detection unit 16 and an overlap processing unit 17.

[0144] The overlap detection unit 16 is used to detect whether there are any edge position points on the same edge line that would cause the edge lines to overlap.

[0145] The overlap processing unit 17 is used to delete the edge point that causes the edge to overlap from the edge point of the road if there is an edge point that causes the edge to overlap.

[0146] In some embodiments, the overlap detection unit 16 is specifically used to segment the edge position points arranged in sequence on the same edge line; for a segment, determine the line connecting the two edge position points located at both ends of the segment; obtain the projection point from the edge position point in the segment to the line connecting the line, and the distance from the projection point to a specified edge position point, wherein the specified edge position point is any one of the two edge position points located at both ends of the segment; compare the distances from the projection points of the edge position points to the specified edge position points according to the sorting of the edge position points in the segment; if the distance from the projection point of the current edge position point to the specified edge position point is less than the largest of the distances from the projection points of the previously compared edge position points to the specified edge position point, then determine the current edge position point as the edge position point that causes the edge line to overlap.

[0147] In some embodiments, such as Figure 18 As shown, the device also includes a smoothing detection unit 18 and a smoothing processing unit 19.

[0148] The smoothing detection unit 18 is used to detect whether the curve formed by the target position points in two adjacent segments is smoothly connected.

[0149] The smoothing unit 19 is used to delete some target position points from the two segments if the curve formed by the target position points in two adjacent segments is not smoothly connected, and to generate target position points for smooth connection of the two segments based on the remaining target position points after deleting some target position points in the two adjacent segments using a curve fitting algorithm.

[0150] In some embodiments, the smoothing detection unit 18 is specifically used to, for two adjacent segments, obtain a first straight line formed by the last two target position points in the previous segment, obtain a second straight line formed by the first two target position points in the next segment; determine the angle formed by the first straight line and the second straight line, and determine whether the angle is less than a preset angle threshold; and determine whether the curve formed by the target position points in the two adjacent segments is smoothly connected based on whether the angle is less than the preset angle threshold.

[0151] In some embodiments, the edge line position points include a left edge line position point and a right edge line position point, with the left edge line position point constituting the left edge and the right edge line position point constituting the right edge. Figure 19 As shown, the device also includes: a sideline feature point determination unit 21, a first segmentation unit 22, a second segmentation unit 23, and a sideline position point processing unit 24.

[0152] The edge feature point determination unit 21 is used to determine the edge feature points on the left and right sides of the center line based on the feature points of the center line, with the shortest distance between the edge feature points and the feature points.

[0153] The first segmentation unit 22 is used to segment the edge position points on the left side line using the edge feature points of the left side line.

[0154] The second segmentation unit 23 is used to segment the edge position points on the right side line using the edge feature points of the right side line.

[0155] The edge line position point processing unit 24 is used to segment the edge line feature points determined by the same two feature points. The segment corresponding to the left line and the segment corresponding to the right line are sampled according to the principle of equal distance ratio. This makes the number of edge line position points in the segments corresponding to the left line and the right line the same, and the number is equal to the total number of edge line position points in the segment with more edge line position points.

[0156] It should be noted that the descriptions of the functional units and modules of the road generation device in this embodiment can be found in the descriptions in the method embodiments, and will not be repeated here.

[0157] In this embodiment, the beneficial effects achieved by the road generation device are the same as those achieved by the road generation method. For details, please refer to the relevant description of the embodiments of the road generation method, which will not be repeated here.

[0158] This disclosure also provides a storage medium including a stored program, wherein the program, when running, controls the device where the storage medium is located to perform the method described above.

[0159] This disclosure also provides a processor for running a program, wherein the program executes the method described above when it runs.

[0160] This disclosure also provides a terminal device, including at least one processor; the storage medium is used to store a program executed by the processor, and data required by the processor during the execution of the program;

[0161] The processor is used to run a program, wherein the program executes the method described above when it runs.

[0162] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, systems, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0163] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0164] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0165] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0166] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0167] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0168] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0169] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0170] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, systems, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0171] The above are merely embodiments of this disclosure and are not intended to limit the scope of this disclosure. Various modifications and variations can be made to this disclosure by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of the claims of this disclosure.

Claims

1. A road generation method, characterized in that, include: Obtain the location points that constitute the centerline of the road; The location points are thinned out to obtain the location points that represent the linear shape of the center line as feature points; The location points are segmented according to the feature points; Based on the curve formed by the position points in the same segment, the position points are smoothed to obtain the target position points in the segment, and the curve formed by the target position points conforms to the line shape standard. The edge position points of the road are generated based on the target location point and the preset road width.

2. The method according to claim 1, characterized in that, The process of thinning out the location points to obtain the location points representing the linear shape of the center line as feature points specifically includes: The location point is subjected to two Douglas thinning operations to obtain the location point representing the linear shape of the center line as the feature point. The first distance threshold used in the first Douglas thinning is less than the second distance threshold used in the second Douglas thinning.

3. The method according to claim 2, characterized in that, The location points of the centerline are arranged in order of road direction. The process of performing Douglas thinning twice on these location points yields feature points characterizing the shape of the centerline, specifically including: Based on the first distance threshold, the position points of the center line are subjected to the first Douglas thinning to obtain the position points as the first sampling points. Obtain the distance between two adjacent first sampling points, divide the consecutive first sampling points with a distance less than the preset third distance threshold into a segment, and mark the first sampling point at the beginning and end of the segment as the initial feature point; Based on the second distance threshold, the location point used as the first sampling point is subjected to a second Douglas thinning to obtain the location point used as the second sampling point; The initial feature points located between two adjacent second sampling points and ordered as the first and last points are used as feature points characterizing the shape of the center line.

4. The method according to any one of claims 1 to 3, characterized in that, The step of smoothing the position points according to the curve formed by the position points in the same segment to obtain the target position point in the segment includes: Determine the line connecting two points located at opposite ends of the same segment; Determine the distances from other position points in the segment to the connecting line. If the largest distance is less than the segment distance threshold, then the curve formed by the position points in the segment is determined to be approximately a straight line. In this case, the two position points at the two ends of the segment are retained as the target position points in the segment, and the other position points in the segment are deleted. If the largest of the distances is greater than the segment distance threshold, the curve formed by the position points in the segment is determined to be arc-shaped. Then, based on the curve fitting algorithm, the position points in the segment are smoothed to obtain the fitted position points. If the fitting error of the curve formed by the fitted position points is less than the error threshold, the fitted position points are taken as the target position points in the segment.

5. The method according to claim 4, characterized in that, The method further includes: If the fitting error of the curve formed by the fitted position points is not less than the error threshold, then the Gaussian smoothing algorithm is used to smooth the position points in the segment to obtain the target position points in the segment.

6. The method according to any one of claims 1 to 3, wherein, After generating the edge location points of the road, the method further includes: Detect whether there are any edge point locations on the same edge line that would cause the edge lines to overlap; if so, delete the edge point location location that causes the edge lines to overlap from the edge point location locations of the road.

7. The method according to claim 6, characterized in that, The detection of whether there are edge point locations on the same edge that would cause the edge to overlap includes: Divide the edge points arranged in sequence along the same edge into segments; For a given segment, determine the line connecting the two edge points located at the two ends of the segment; Obtain the projection point of the connecting line from the edge position point in the segment, and the distance from the projection point to the specified edge position point, wherein the specified edge position point is any one of the two edge position points located at both ends of the segment; Based on the sorting of the edge position points in the segment, the distances from the projection points of the current edge position points to the specified edge position points are compared. If the distance from the projection point of the current edge position point to the specified edge position point is less than the largest distance among the previously compared edge position points, then the current edge position point is determined as the edge position point that caused the edge to overlap.

8. The method according to any one of claims 1 to 3, characterized in that, Before generating the edge location points of the road, the method further includes: If the curve formed by the target position points in two adjacent segments is not smoothly connected, then some target position points are deleted from the two segments. Based on the remaining target position points after deleting some target position points from the two adjacent segments, a curve fitting algorithm is used to generate target position points for smoothly connecting the two segments.

9. The method according to claim 8, characterized in that, The detection of whether the curve formed by the target position points in two adjacent segments is smoothly connected includes: For two adjacent segments, obtain the first straight line formed by the last two target position points in the previous segment, and obtain the second straight line formed by the first two target position points in the next segment. Determine the angle formed by the first straight line and the second straight line, and determine whether the angle is less than a preset angle threshold, so as to determine whether the curve formed by the target position points in the two adjacent segments is smoothly connected.

10. The method according to any one of claims 1 to 3, characterized in that, The edge line position points include a left edge line position point and a right edge line position point, with the left edge line position point forming the left edge line and the right edge line position point forming the right edge line. The method further includes: Based on the feature points of the center line, determine the edge feature points on the left and right lines respectively, where the distance from the edge feature point to the feature point is the shortest. The edge position points on the left side line are segmented using the edge feature points of the left side line; The right edge line position points are segmented using the edge line feature points of the right edge line. According to the principle of equal distance, the edge position points in the segments corresponding to the left or right line are sampled so that the number of edge position points in the segments corresponding to the left and right lines is the same, and the number is equal to the total number of edge position points in the segment with more edge position points.

11. A road generation device, characterized in that, include: The data acquisition unit is used to acquire the location points that constitute the centerline of the road. The feature point determination unit is used to thin out the position points to obtain position points that characterize the linear shape of the center line as feature points. Segmentation unit, used to segment the position point according to the feature point; The first processing unit is used to smooth the position points according to the line shape of the curve formed by the position points in the same segment, so as to obtain the target position points in the segment, wherein the line shape of the curve formed by the target position points conforms to the line shape standard. The second processing unit is used to generate the edge position points of the road based on the target position point and the preset road width.

Images