Road plane determination method, apparatus, device, medium, and vehicle

Abstract

The embodiments of the present disclosure provide a road plane determination method, device, equipment, medium and vehicle. The road plane determination method comprises: performing grid division on a region of interest to obtain a plurality of division grids; selecting a first grid in a near grid, selecting a second grid in a far grid, and randomly selecting a plurality of third grids in an intermediate grid; determining point coordinates based on point cloud coordinates included in each of the first grid, the second grid and the randomly selected third grid, respectively, and using three point coordinates to form a candidate triangle; selecting a candidate triangle meeting a preset constraint condition as a target triangle, and determining a plane of the target triangle as a plane where the region of interest is located. The embodiments of the present disclosure limit the selection of point coordinates based on the idea of the random sample consensus algorithm, so that the obtained point coordinates possibly represent the plane characteristics of the target road region, and the fitting degree of the plane where the target road is located is improved.

Description

Technical Field

[0001] This disclosure relates to the field of point cloud data processing, specifically to a method, apparatus, equipment, medium, and vehicle for determining a road plane. Background Technology

[0002] Because point cloud data collected by LiDAR can perceive the environment more accurately, LiDAR is now widely used in the field of vehicle assisted driving. In vehicle assisted driving scenarios, environmental point cloud data collected by LiDAR can be used not only to identify obstacles in the road environment, but also to determine the road's smoothness and slope. The core of determining road smoothness and slope is to determine the road plane based on the point cloud data of the target road area.

[0003] To reduce the amount of data processed, existing methods for determining road planes based on point cloud data of a target road area are based on the random sample consensus algorithm. This algorithm randomly samples the point cloud data of the target road area and uses the coordinates of the randomly sampled points to fit the road plane. However, due to the inherent randomness of the random sample consensus algorithm, the positional characteristics and representativeness of the random point coordinates obtained by this method are not ideal, resulting in a poor fit between the plane obtained based on the random point cloud coordinates and the real plane. Summary of the Invention

[0004] To address the aforementioned technical problems, embodiments of this disclosure provide a method, apparatus, device, medium, and vehicle for determining road planes.

[0005] In a first aspect, embodiments of this disclosure provide a method for determining a road plane, comprising:

[0006] Obtain the point cloud coordinates within the region of interest of the target road;

[0007] The region of interest is divided into grids according to preset rules to obtain multiple grids and point cloud coordinates within the grids.

[0008] Multiple first grids are selected from the nearby grids, the second grid farthest from the first grids is selected from the distant grids, and multiple third grids are randomly selected from the intermediate grids. The nearby grids are the grids in the divided grids whose longitudinal distance to the point cloud acquisition point is less than the first longitudinal distance, the distant grids are the grids in the divided grids whose longitudinal distance to the point cloud acquisition point is greater than the second longitudinal distance, the second longitudinal distance is greater than the first longitudinal distance, and the intermediate grids are the grids in the divided grids other than the nearby grids and the distant grids.

[0009] Based on the point cloud coordinates of each first grid, the corresponding second grid, and a randomly selected third grid, a point coordinate is determined, and the three point coordinates are used to form a candidate triangle;

[0010] Select a candidate triangle that meets the preset constraints as the target triangle, and use the plane determined by the target triangle as the plane where the region of interest is located.

[0011] Optionally, selecting candidate triangles that meet preset constraints as target triangles includes:

[0012] Evaluation scores are calculated based on the shape and area of ​​the candidate triangles. The evaluation scores are positively correlated with the similarity of the shapes of the candidate triangles and the ideal isosceles triangles, and positively correlated with the area of ​​the candidate triangles. The ideal isosceles triangle is the isosceles triangle with the hypotenuse of the target road area as its base and the largest area.

[0013] The plane determined by the candidate triangle with the best evaluation score is selected as the plane where the target road area is located.

[0014] Optionally, calculating the evaluation score based on the shape and area of ​​the candidate triangle includes:

[0015] Calculate the angle difference between each angle of the candidate triangle and the angle of the ideal isosceles triangle;

[0016] The angle evaluation score of the candidate triangle is determined based on the mean square error of the three angle differences.

[0017] The ratio of the area of ​​the candidate triangle to the area of ​​the target road region is calculated and used as the area evaluation score;

[0018] Calculate the weighted sum of the angle evaluation score and the area evaluation score, and use the weighted sum as the evaluation score of the candidate triangle.

[0019] Optionally, the grid obtained by dividing the region of interest into grids according to preset rules includes at least three columns;

[0020] The selection of multiple first grids in the nearby grid includes:

[0021] If there are unselected grids in the outer column, the first grid is randomly selected from the unselected grids;

[0022] If there are no unselected grids in the outer column, select the first grid in the inner column.

[0023] Optionally, before dividing the region of interest into a grid according to preset rules, the method further includes:

[0024] Traverse the point cloud coordinates within the region of interest to determine the maximum and minimum height coordinates;

[0025] Calculate the height difference between the maximum height coordinate and the minimum height coordinate;

[0026] If the height difference is less than the height threshold, the step of dividing the region of interest into grids according to preset rules is performed.

[0027] Optionally, before dividing the region of interest into a grid according to preset rules, the method further includes:

[0028] Statistical analysis is performed on the height of the point cloud coordinates within the region of interest to determine the height distribution characteristics of the point cloud.

[0029] If the point cloud height distribution features meet the preset features, the step of dividing the region of interest into grids according to the preset rules is performed.

[0030] Optionally, it also includes: deleting the farthest sub-region in the region of interest when the height difference is greater than the height threshold or the point cloud height distribution features do not meet the preset features;

[0031] Repeat the steps of determining the maximum and minimum heights or performing statistical analysis on the height of the point cloud coordinates within the region of interest.

[0032] Optionally, the point coordinates are determined based on the point cloud coordinates of each first grid, the corresponding second grid, and a randomly selected third grid, including:

[0033] The point cloud coordinates of each of the first grid, the second grid, and the third grid are downsampled to obtain the corresponding point coordinates. The downsampling method includes one of the following: calculating the mean value of the road surface point cloud coordinates in the corresponding grid, calculating the median value of the road surface point cloud coordinates in the corresponding grid, and randomly selecting a road surface point cloud coordinate in the corresponding grid.

[0034] Secondly, embodiments of this disclosure provide a road plane determination device, comprising:

[0035] The point cloud coordinate acquisition unit is used to acquire the point cloud coordinates within the region of interest of the target road;

[0036] A grid division unit is used to divide the region of interest into grids according to preset rules, thereby obtaining multiple grids and point cloud coordinates within the grids.

[0037] A grid selection unit is used to select multiple first grids in a nearby grid, select the second grid farthest from the first grids in a distant grid, and randomly select multiple third grids in an intermediate grid. The nearby grids are grids in the divided grid whose longitudinal distance to the point cloud acquisition point is less than a first longitudinal distance. The distant grids are grids in the divided grid whose longitudinal distance to the point cloud acquisition point is greater than a second longitudinal distance, where the second longitudinal distance is greater than the first longitudinal distance. The intermediate grids are grids in the divided grid other than the nearby grids and the distant grids.

[0038] The triangle building unit is used to determine a point coordinate based on the point cloud coordinates of each first grid, the corresponding second grid, and a randomly selected third grid, and to form a candidate triangle using the three point coordinates;

[0039] The plane determination unit is used to select candidate triangles that meet preset constraints as target triangles, and to use the plane determined by the target triangles as the plane where the region of interest is located.

[0040] Thirdly, embodiments of this disclosure provide an electronic device, including a processor and a memory, the memory being used to store a computer program; when the computer program is loaded by the processor, it causes the processor to execute the road plane determination method as described above.

[0041] Fourthly, embodiments of this disclosure provide a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to implement the aforementioned method for determining the road surface.

[0042] Fifthly, embodiments of this disclosure provide a vehicle including an onboard control chip and a lidar. The onboard control chip processes the point cloud coordinates of a region of interest on a target road collected by the lidar and determines the plane where the region of interest is located using the road plane determination method described above.

[0043] The technical solution provided in this disclosure has the following advantages compared with the prior art:

[0044] The solution provided in this disclosure divides the target road area into grids. After determining the grids for the target road area, a first grid is randomly selected from the nearby grids, a second grid farthest from the first grid is selected from the distant grids, and a third grid is randomly selected from the intermediate grids. Then, candidate triangles are constructed using the point cloud coordinates determined from the first, second, and third grids. The plane determined by the candidate triangle with the highest evaluation score is selected as the plane containing the target road area. The solution provided in this disclosure employs a random sampling consensus algorithm to select point coordinates and also restricts the selection rules of point coordinates through pre-defined rules, ensuring that the obtained point coordinates optimally represent the planar characteristics of the target road area, thereby improving the fitting degree of the determined target road plane. Attached Figure Description

[0045] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0046] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without any creative effort, wherein:

[0047] Figure 1 This is a flowchart of a method for determining a road surface according to an embodiment of this disclosure;

[0048] Figure 2 This is a schematic diagram illustrating the determination of a target triangle within a grid area of ​​a target road region according to an embodiment of this disclosure;

[0049] Figure 3 This is a schematic diagram of the structure of the road plane determination device provided in an embodiment of this disclosure;

[0050] Figure 4 This is a schematic diagram of the structure of an electronic device provided in some embodiments of this disclosure. Detailed Implementation

[0051] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0052] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below. It should be noted that the concepts of "first", "second", etc., used in this disclosure are only used to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0053] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0054] This disclosure provides a method for determining a road plane, which modifies the method for determining point coordinates based on a random sampling consensus algorithm, thereby obtaining a fitted plane that better fits the actual road plane.

[0055] Figure 1 This is a flowchart of a method for determining a road surface according to an embodiment of this disclosure. Figure 1 As shown, the method provided in this embodiment includes S110-S150.

[0056] It should be noted that the determination of the road plane provided in this embodiment is performed by an electronic device. The aforementioned electronic device may be a vehicle infotainment system, or a remote server or other device connected to the vehicle infotainment system and capable of acquiring point cloud data collected by LiDAR.

[0057] S110: Obtain the point cloud coordinates within the region of interest of the target road.

[0058] The target road area is a pre-determined road area. In practical applications, after acquiring the point cloud coordinates collected by the lidar, the electronic device can preprocess the point cloud coordinates to exclude point cloud coordinates determined by lasers reflected from roadside curbs, vegetation, or other non-road surface obstacles, thereby determining the point cloud coordinates within the area of ​​interest on the road surface.

[0059] In practice, electronic devices can use point cloud semantic segmentation models to process the acquired point cloud coordinates, determine the attribute type of the point cloud coordinates, that is, determine whether the point cloud coordinates are reflected by curbs, green plants or other road obstacles, or reflected by the road surface or by road participants, and determine the target road area including the road surface point cloud coordinates based on the attribute type of the point cloud coordinates.

[0060] S120: Divide the region of interest into grids according to preset rules to obtain multiple grids and point cloud coordinates within the grids.

[0061] After determining the region of interest (ROI) of the target road, the electronic device can then divide the ROI into multiple meshes according to preset rules. The specific rules and how the electronic device divides the ROI according to these rules will be analyzed later. After obtaining each mesh, the point cloud coordinates within that mesh can be determined based on its coordinate range.

[0062] S130: Select multiple first grids in the nearby grid, select the second grid farthest from the first grid in the distant grid, and randomly select multiple third grids in the middle grid.

[0063] The nearby grid consists of multiple grids whose longitudinal distance to the point cloud acquisition point is less than the first longitudinal distance. The aforementioned point cloud acquisition point is the location of the LiDAR, which is the coordinate reference point determined by the LiDAR. The aforementioned longitudinal distance refers to the distance in the direction the vehicle is pointing. In other words, the nearby grid is a set of grids whose distance to the LiDAR is less than the first longitudinal distance.

[0064] The distant grid consists of grids whose longitudinal distance to the point cloud acquisition point is greater than the second longitudinal distance among multiple grid divisions. It should be noted that the aforementioned second longitudinal distance is greater than the first longitudinal distance. The intermediate grid consists of all grids other than the nearby and distant grids.

[0065] In practice, the specific determination of multiple first grids, second grids, and third grids depends on the number of grids to be divided and the confidence level to be achieved in random sampling; no specific limitations are made here.

[0066] It should also be noted that, in this embodiment of the present disclosure, the electronic device may randomly select a nearby grid in succession, and if it is determined that this nearby grid is not selected, it shall be used as the first grid. Similarly, the electronic device may randomly select an intermediate grid in succession, and if it is determined that this intermediate grid is not selected, it shall be used as the third grid.

[0067] After determining each first grid, the electronic device can determine the second grid that is furthest from the first grid based on the positional relationship and size of the grids. It should be noted that multiple first grids may correspond to the same second grid.

[0068] S140: Determine a point coordinate based on the point cloud coordinates of each first grid, the corresponding second grid, and a randomly selected third grid, and use the three point coordinates to form a candidate triangle.

[0069] After determining the first, second, and third grids, the electronic device randomly selects one of the first grids, the corresponding second grid, and a randomly selected third grid as a grid pair. Based on the point cloud coordinates of each of the three grids in the pair, it determines the coordinates of a point. Each point coordinate includes coordinates in three dimensions (x, y, and z directions).

[0070] Figure 2 This is a schematic diagram illustrating the determination of a target triangle within a grid region of a target road area, according to an embodiment of this disclosure. For example... Figure 2 As shown, in this embodiment of the present disclosure, after selecting the first grid, the second grid, and the third grid, the coordinates of the points in the three grids are connected to obtain candidate triangles.

[0071] According to the principles of triangle formation, the coordinates of three points can form a triangle, that is, a candidate triangle. The candidate triangle mentioned here is not a real triangle, but a hypothetical triangle constructed using the coordinates of three points.

[0072] Depending on the specific circumstances, the aforementioned candidate triangles could be any of the following: right triangles, acute triangles, or obtuse triangles. Given the coordinates of the three points, the area of ​​the candidate triangle can be determined through calculation.

[0073] It should be noted that the point cloud coordinates within the aforementioned grid are road surface point cloud coordinates, that is, coordinates representing the road surface. In a specific implementation of this disclosure, the electronic device can use the point cloud semantic segmentation model described above to determine road obstacles located in the target road area, exclude the point cloud data representing the road obstacles, and use the remaining point cloud coordinates as the point cloud coordinates of the target road region of interest.

[0074] It should also be noted that due to the previous processing, some first, second, and third grids do not have corresponding point cloud coordinates. In this case, S140 cannot be executed to form candidate triangles, so such first, second, and third grids can be deleted.

[0075] S150: Select a candidate triangle that meets the preset constraints as the target triangle, and use the plane determined by the target triangle as the plane where the region of interest is located.

[0076] Using the method provided in this disclosure, the electronic device divides the target road area into grids. After determining the grids for the target road area, it selects the first grid from the nearest grids, the second grid from the farthest grids, and a third grid randomly selected from the middle grids. Then, it constructs candidate triangles using the point cloud coordinates determined from the first, second, and third grids, and selects the plane determined by the candidate triangle with the highest evaluation score as the plane containing the target road area.

[0077] The solution provided in this embodiment adopts the idea of ​​random sampling consensus algorithm to select point coordinates, and restricts the selection rules of point coordinates through pre-set rules, so that the obtained point coordinates can optimally represent the planar characteristics of the target road area, thereby improving the fitting degree of the determined target road plane.

[0078] In some embodiments, the target triangle is determined based on the following preset constraints: (1) the target triangle should be a triangle that covers the target road region of interest as much as possible, and its area should theoretically be as large as possible; (2) the target triangle should be an isosceles right triangle with the diagonal of the region of interest as its base as much as possible.

[0079] In the corresponding S150, candidate triangles that meet the preset constraints are selected as target triangles, including S151-S152.

[0080] S151: Calculate the evaluation score based on the shape and area of ​​the candidate triangle.

[0081] In this embodiment of the disclosure, the evaluation score of the candidate triangle is determined with reference to the shape and area of ​​an ideal isosceles triangle. An ideal isosceles triangle is an isosceles triangle with the diagonal of the target road area as its base and the largest possible area.

[0082] The closer the area and shape of each candidate triangle are to the ideal candidate triangle, the higher the corresponding evaluation score. That is to say: (1) The evaluation score of the candidate triangle is related to the size of the candidate triangle area. The larger the area of ​​the candidate triangle, the higher the corresponding evaluation score; (2) The evaluation score of the candidate triangle is positively correlated with the similarity of the candidate triangle to the ideal isosceles triangle shape. The closer the two shapes are, the higher the corresponding evaluation score.

[0083] Therefore, in this embodiment of the disclosure, the corresponding evaluation score can be calculated based on the shape and area of ​​the target triangle, with reference to the shape and area of ​​an ideal isosceles triangle.

[0084] S152: Select the plane determined by the candidate triangle with the best evaluation score as the plane where the target road area is located.

[0085] After calculating the evaluation score for each candidate triangle, the electronic device can then select the candidate triangle with the highest evaluation score and use the plane defined by this triangle as the plane where the target road area is located.

[0086] As previously stated, the electronic device performs S151 to calculate an evaluation score based on the shape and area of ​​each candidate triangle. In some embodiments of this disclosure, S51 may include S1511-S1514.

[0087] S1511: Calculate the angle difference between each angle of the candidate triangle and the angle of the ideal isosceles triangle.

[0088] S1512: Determine the angle evaluation score of the candidate triangle based on the mean square error of the three angle differences.

[0089] The angle differences between each angle of the candidate triangle and the corresponding angle of the ideal isosceles triangle are obtained by subtracting each angle from the corresponding angle of the ideal isosceles triangle. The angle differences can be positive, negative, or zero. After determining the three angle differences, the mean square error (MSE) is calculated based on these differences to obtain the angle evaluation score of the candidate triangle.

[0090] In this embodiment of the disclosure, the higher the angle evaluation score, the more similar the candidate triangle is to the ideal isosceles triangle. Therefore, the angle evaluation score is inversely correlated with the root mean square error. To this end, in some embodiments of this disclosure, the electronic device can calculate the evaluation score using the following method. Where δ is the mean square error of the angle differences corresponding to the candidate triangles, δ worst Let δ be the root mean square error of the angle difference between a hypothetical triangle with an angle of 180 degrees and an ideal isosceles triangle. best It is 0.

[0091] S1513: Calculate the ratio of the area of ​​the candidate triangle to the area of ​​the target road area, and use it as the area evaluation score.

[0092] The area evaluation score is calculated using a formula. The calculation yields s, where s triangle Let s be the area of ​​the candidate triangle. target It is the area of ​​the ideal isosceles triangle, which is one-quarter of the area of ​​the target road area.

[0093] S1514: Calculate the weighted sum of the angle evaluation score and the area evaluation score, and use the weighted sum as the evaluation score of the candidate triangle.

[0094] After obtaining the angle evaluation score and the area evaluation score, the two scores can be weighted and summed to obtain the evaluation score of the candidate triangle. Evaluation score c = wa ×a+w b ×b, where w a As the weight of the angle evaluation score, w b The weighting of the area evaluation score.

[0095] Of course, other comparison methods can also be used to determine the evaluation score of the candidate triangle in this embodiment. For example, the difference between the side length of the candidate triangle and the side length of the ideal triangle can be used to reflect the shape and area of ​​the candidate triangle, and then the evaluation score can be calculated.

[0096] As previously stated, in S120, the target road area is divided into grids according to preset rules, resulting in multiple grids. In specific implementations, the grid shape in the preset grid division rules can be rectangular, triangular, rhomboid, etc., and this embodiment does not impose any particular limitation. A more preferred grid shape is rectangular or rhomboid.

[0097] In some embodiments, the grid obtained by dividing the target road area into grids according to preset rules includes at least three rows and at least three columns, preferably at least four rows and four columns, and the more rows and columns, the better, provided that computing resources are sufficient. For example, in a specific application, the grid obtained by dividing the target road area into grids includes 10 rows and 4 columns. The three rows closest to the point cloud acquisition point are designated as the near grid, and the three rows farthest from the point cloud acquisition point are designated as the far grid.

[0098] In some embodiments of this disclosure, S130 selects a plurality of first grids in the nearby grid, including S131-S133.

[0099] S131: Determine if there are any unselected grids in the outer grid; if so, execute S132; otherwise, execute S133.

[0100] S132: Randomly select the first grid in the outer column.

[0101] S133: Randomly select the first grid in the inner column.

[0102] In this embodiment, the outer columns are the columns on the outside of the grid. For example, in a 10x4 grid, columns 1 and 4 are the outer columns, and the corresponding columns 2 and 3 are the inner columns. Of course, in this embodiment, the outer and inner columns are relative. When the target road area has more columns, the outer columns and corresponding inner columns can be determined according to the specific circumstances.

[0103] In this embodiment, selecting the first grid from the outer grid can be achieved by first selecting the grid closest to the point cloud acquisition point within the outer grid, and then selecting the grid farthest from the point cloud acquisition point within the outer grid. Corresponding to the aforementioned selection of the first grid, the second grid is highly likely to be a grid located in the diagonal region.

[0104] By employing the aforementioned S131-S133, the base of the constructed candidate triangle can be made as large as possible, thereby maximizing the area of ​​the candidate triangle.

[0105] Optionally, in some embodiments of this disclosure, the electronic device may also execute S160-S170 before executing S110.

[0106] S160: Traverse the point cloud coordinates of the region of interest of the target road, determine the maximum and minimum height coordinates, and calculate the height difference between the maximum and minimum height coordinates.

[0107] S170: Determine if the height difference is less than the height threshold; if so, execute S110.

[0108] By traversing the point cloud coordinates within the region of interest (ROI) of the target road, the maximum and minimum height coordinates are determined. The difference between these two values ​​is calculated, and it is determined whether the difference is less than a height threshold. This allows us to determine if the points within the ROI of the target road are likely to lie on a plane. If the height difference is less than the height threshold, the point cloud coordinates represent a plane. If the height difference is greater than the height threshold, the height fluctuation of the point cloud coordinates is too large, and it cannot be considered a plane point cloud.

[0109] Optionally, in some embodiments of this disclosure, the electronic device may also execute S180-S190 before executing S110.

[0110] S180: Perform statistical analysis on the height of point cloud coordinates within the region of interest of the target road to determine the height distribution characteristics of the point cloud.

[0111] S190: Determine whether the point cloud distribution features meet the preset features; if so, execute S110.

[0112] To determine the height distribution characteristics of point cloud coordinates within the region of interest of the target road, a height statistical analysis is performed. This involves calculating the average and standard deviation of the point cloud coordinate heights and using these values ​​to determine the distribution range of the point cloud heights. The aforementioned preset characteristic is a specific distribution range, such as [μ-3δ, μ+3δ], where μ is the mean of the point cloud heights and δ is the root mean square deviation of the point cloud heights.

[0113] If the point cloud height distribution characteristics meet the preset characteristics, it can be determined that the point cloud height meets the expected distribution, and therefore the point cloud can be determined to be in a plane.

[0114] Optionally, in some embodiments of this disclosure, if the electronic device determines in S170 that the height difference is greater than or equal to the height threshold, or if it determines in S190 that the point cloud height distribution features do not meet the preset features, the electronic device may also execute S200.

[0115] S200: Delete the furthest sub-region in the region of interest of the target road, and re-execute S160 or 180.

[0116] In practical applications, situations that cause the point cloud height distribution of the target road's region of interest to deviate from the preset rules may include: (1) sloping areas in the road; and (2) numerous potholes in the road, meaning the road is severely uneven. In normal roads, the probability of situation (2) occurring is much lower than the probability of situation (1). In other words, the situation causing the point cloud coordinate height distribution of the target road's region of interest to deviate from the preset rules is most likely due to sloping areas in the road. It should be noted that the aforementioned sloping areas are relative to the road areas near the point cloud acquisition point in the target road's region of interest; sloping areas refer to road areas far from the nearby road areas.

[0117] Deleting the furthest sub-region within the target road region of interest involves deleting the region of the target road region of interest that is furthest from the point cloud acquisition point by a preset width, resulting in a new target road region of interest.

[0118] As described above, in S140, a point coordinate is determined based on the road surface point cloud coordinates included in each first grid, the corresponding second grid, and a randomly selected third grid. Optionally, in some embodiments of this disclosure, S130 specifically involves: downsampling the point cloud coordinates of each first grid, second grid, and third grid to obtain the corresponding point coordinates.

[0119] The downsampling method is used to obtain the corresponding point coordinates. This involves calculating a position coordinate based on the point cloud coordinates in each grid and using this position coordinate as the corresponding point coordinate. In specific implementation, the downsampling method can be one of the following: calculating the mean value of the road surface point cloud coordinates in the corresponding grid, calculating the median value of the road surface point cloud coordinates in the corresponding grid, and randomly selecting a road surface point cloud coordinate in the corresponding grid.

[0120] In practice, downsampling of the road surface point cloud corresponding to the grid can be performed only on the height coordinates. The horizontal and vertical coordinates of the points in the corresponding grid are then used as the horizontal and vertical coordinates of the point coordinates.

[0121] In addition to providing the aforementioned method for determining the road surface, this disclosure also provides a road surface determination device 300.

[0122] Figure 3 This is a schematic diagram of the structure of the road plane determination device provided in an embodiment of this disclosure. Figure 3 As shown, the road plane determination device 300 includes a point cloud coordinate acquisition unit 301, a grid division unit 302, a grid selection unit 303, a triangle construction unit 304, and a plane determination unit 305.

[0123] The point cloud coordinate acquisition unit 301 is used to acquire the point cloud coordinates of the region of interest of the target road.

[0124] The grid division unit 302 is used to divide the region of interest into grids according to preset rules, so as to obtain multiple grids and point cloud coordinates within the grids.

[0125] The grid selection unit 303 is used to select multiple first grids in the nearby grid, select the second grid farthest from the first grid in the distant grid, and randomly select multiple third grids in the intermediate grid. The nearby grid is the grid in the grid that has a longitudinal distance from the point cloud acquisition point that is less than the first longitudinal distance. The distant grid is the grid in the grid that has a longitudinal distance from the point cloud acquisition point that is greater than the second longitudinal distance. The second longitudinal distance is greater than the first longitudinal distance. The intermediate grid is the grid in the grid that is excluding the nearby grid and the distant grid.

[0126] The triangle building unit 304 is used to determine a point coordinate based on the point cloud coordinates of each first grid, the corresponding second grid, and a randomly selected third grid, and to form a candidate triangle using the three point coordinates.

[0127] The plane determination unit 305 is used to select candidate triangles that meet preset constraints as target triangles, and to use the plane determined by the target triangle as the plane where the region of interest is located.

[0128] In some embodiments of this disclosure, the plane determination unit 305 calculates an evaluation score based on the shape and area of ​​the candidate triangles, and selects the plane determined by the candidate triangle with the best evaluation score as the plane where the target road area is located. The evaluation score is positively correlated with the shape similarity between the candidate triangle and the ideal isosceles triangle, and positively correlated with the area of ​​the candidate triangle. The ideal isosceles triangle has the base of the hypotenuse of the target road area and is the isosceles triangle with the largest area.

[0129] In some embodiments, the plane determination unit 305 includes an angle difference calculation subunit, an angle scoring subunit, an area scoring subunit, and a score subunit. The angle difference calculation subunit calculates the angle difference between each angle of the candidate triangle and the angle of the ideal isosceles triangle. The angle scoring subunit determines the angle evaluation score of the candidate triangle based on the mean square error of the three angle differences. The area scoring subunit calculates the ratio of the area of ​​the candidate triangle to the area of ​​the target road region, using this ratio as the area evaluation score. The score subunit calculates the weighted sum of the angle evaluation score and the area evaluation score, using this weighted sum as the evaluation score of the target triangle.

[0130] In some embodiments of this disclosure, the grid obtained by dividing the target road area into grids according to a pre-set division rule includes at least three rows and at least three columns. Correspondingly, if there are unselected grids in the outer columns, the grid selection unit 303 randomly selects the first grid from the unselected grids; if there are no unselected grids in the outer columns, it selects the first grid from the inner columns.

[0131] In some embodiments of this disclosure, the road plane determination device 300 further includes a traversal unit, a difference calculation unit, and a first judgment unit. The traversal unit is used to traverse the point cloud coordinates within the region of interest and determine the maximum and minimum height coordinates. The difference calculation unit is used to calculate the height difference between the maximum and minimum height coordinates. The first judgment unit is used to determine whether the height difference is less than a height threshold. If the height difference is less than the height threshold, the mesh division unit 302 performs the step of meshing the region of interest according to preset rules.

[0132] In some embodiments of this disclosure, the road plane determination device 300 further includes a statistical feature determination unit and a second judgment unit. The statistical feature determination unit is used to perform statistical analysis on the height of point cloud coordinates within the region of interest to determine the point cloud height distribution characteristics. The second judgment unit is used to determine whether the point cloud height distribution characteristics meet preset characteristics. If the point cloud height distribution characteristics meet the preset characteristics, the mesh division unit 302 performs the step of meshing the region of interest according to preset rules.

[0133] In some embodiments of this disclosure, the road plane determination device 300 further includes a deletion unit. The deletion unit is used to delete the furthest sub-region within the region of interest when the height difference exceeds a height threshold or the point cloud height distribution characteristics do not meet preset features. Correspondingly, the traversal unit re-executes the steps of determining the maximum and minimum heights, and the statistical feature determination unit re-executes the steps of statistically analyzing the height of the point cloud coordinates within the region of interest.

[0134] In some embodiments of this disclosure, the triangle determination unit 303 performs downsampling processing on the point cloud coordinates of each first grid, second grid and third grid to obtain the corresponding point coordinates. The downsampling method includes one of the following: calculating the mean value of the road surface point cloud coordinates in the corresponding grid, calculating the median value of the road surface point cloud coordinates in the corresponding grid, and randomly selecting a road surface point cloud coordinate in the corresponding grid.

[0135] Figure 4 This is a schematic diagram of the structure of an electronic device provided in some embodiments of this disclosure. See below for details. Figure 4 It shows a schematic diagram of a structure suitable for implementing the electronic device 400 in the embodiments of this disclosure. Figure 4 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0136] like Figure 4 As shown, the electronic device 400 may include a processing unit 401 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in the read-only memory ROM 402 or a program loaded from the storage device 408 into the random access memory RAM 403. The RAM 403 also stores various programs and data required for the operation of the electronic device 400. The processing unit 401, ROM 402, and RAM 403 are interconnected via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0137] Typically, the following devices can be connected to I / O interface 405: input devices 406 including, for example, touchscreens, touchpads, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 407 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 408 including, for example, magnetic tapes, hard disks, etc.; and communication devices 409. Communication device 409 allows electronic device 400 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 4 An electronic device 400 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0138] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 409, or installed from storage device 408, or installed from ROM 402. When the computer program is executed by processing device 401, it performs the functions defined in the methods of embodiments of this disclosure.

[0139] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0140] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0141] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0142] The aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to: acquire point cloud coordinates within a region of interest (ROI) of a target road; divide the ROI into multiple grids according to preset rules, obtaining multiple grids and point cloud coordinates within each grid; select multiple first grids from nearby grids, select the second grid furthest from the first grids from distant grids, and randomly select multiple third grids from intermediate grids, wherein the nearby grids are those whose longitudinal distance to the point cloud acquisition point is less than a first longitudinal distance, the distant grids are those whose longitudinal distance to the point cloud acquisition point is greater than a second longitudinal distance, the second longitudinal distance being greater than the first longitudinal distance, and the intermediate grids are those other than the nearby and distant grids; determine a point coordinate based on the point cloud coordinates of each first grid, the corresponding second grid, and a randomly selected third grid, and form a candidate triangle using the three point coordinates; select a candidate triangle that meets preset constraints as the target triangle, and use the plane determined by the target triangle as the plane containing the ROI.

[0143] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0144] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0145] The units described in the embodiments of this disclosure can be implemented in software or hardware. The names of the units are not, in some cases, intended to limit the specific unit.

[0146] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0147] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include, based on electrical connections of one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0148] This disclosure also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it can implement the methods of any of the above method embodiments. The execution method and beneficial effects are similar, and will not be described again here.

[0149] Furthermore, in a fifth aspect, embodiments of this disclosure provide a vehicle including an onboard control chip and a lidar. The onboard control chip processes the point cloud coordinates of a target road area collected by the lidar and uses the road plane determination method described above to determine the plane in which the target road area is located.

[0150] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, 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 a 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 the element.

[0151] The above description is merely a specific embodiment of this disclosure, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to the embodiments described herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for determining a road plane, characterized in that The method comprises the following steps: acquiring point cloud coordinates in a target road region of interest; dividing the region of interest into grids according to a preset rule to obtain a plurality of division grids and point cloud coordinates in the division grids; selecting a plurality of first grids in near grids, selecting a second grid farthest from the first grids in far grids, and randomly selecting a plurality of third grids in intermediate grids, the near grids being grids in the division grids with a longitudinal distance to a point cloud collection point less than a first longitudinal distance, the far grids being grids in the division grids with a longitudinal distance to the point cloud collection point greater than a second longitudinal distance, the second longitudinal distance being greater than the first longitudinal distance, and the intermediate grids being grids in the division grids other than the near grids and the far grids; determining a point coordinate for each of the first grids, the corresponding second grid, and a randomly selected third grid based on the point cloud coordinates included in each of the first grids, the corresponding second grid, and the randomly selected third grid, and using the three point coordinates to form a candidate triangle; selecting a candidate triangle meeting a preset constraint condition as a target triangle, and determining a plane of the target triangle as a plane on which the region of interest is located; wherein the step of selecting a candidate triangle meeting a preset constraint condition as a target triangle comprises: calculating an evaluation score according to the shape and area of the candidate triangle, the evaluation score being positively correlated with the shape similarity degree between the candidate triangle and an ideal isosceles triangle, and the evaluation score being positively correlated with the area of the candidate triangle, the ideal isosceles triangle being an isosceles triangle with a base equal to the diagonal of the region of interest and a maximum area; selecting a plane determined by a candidate triangle with an optimal evaluation score as the plane on which the region of interest is located.

2. The method of claim 1, wherein, The step of calculating an evaluation score according to the shape and area of the candidate triangle comprises: calculating an angle difference value between each angle of the candidate triangle and an angle of an ideal isosceles triangle; determining an angle evaluation score of the candidate triangle based on the mean square deviation of the three angle difference values; calculating a ratio of the area of the candidate triangle to the area of the region of interest as an area evaluation score; calculating a weighted sum value of the angle evaluation score and the area evaluation score as the evaluation score of the candidate triangle.

3. The method of claim 1, wherein, The grids obtained by dividing the region of interest into grids according to a preset rule comprise at least three columns; The step of selecting a plurality of first grids in near grids comprises: in the case that there is an unselected grid in the outer column, randomly selecting the first grid in the unselected grid; in the case that there is no unselected grid in the outer column, selecting the first grid in the inner column.

4. The method of claim 1, wherein, Before the step of dividing the region of interest into grids according to a preset rule, the method further comprises: traversing the point cloud coordinates in the region of interest to determine a maximum height coordinate and a minimum height coordinate; calculating a height difference value of the maximum height coordinate and the minimum height coordinate; in the case that the height difference value is less than a height threshold, performing the step of dividing the region of interest into grids according to a preset rule. In the case that the height difference is greater than the height threshold, a farthest sub-region in the region of interest is deleted; the steps of determining the maximum height coordinate and the minimum height coordinate are re-executed.

5. The method of claim 1, wherein, Before the region of interest is divided into grids according to the preset rule, the method further comprises: statistical analysis is performed on the height of the point cloud coordinates in the region of interest to determine the point cloud height distribution characteristics; in the case that the point cloud height distribution characteristics meet the preset characteristics, the step of dividing the region of interest into grids according to the preset rule is executed.

6. The method of claim 5, wherein, Further comprising: in the case that the point cloud height distribution characteristics do not meet the preset characteristics, a farthest sub-region in the region of interest is deleted; the step of performing statistical analysis on the height of the point cloud coordinates in the region of interest is re-executed.

7. The method according to any one of claims 1 to 6, characterized in that, The point cloud coordinates included in each of the first grid, the corresponding second grid and a randomly selected third grid each determine a point coordinate, comprising: respectively, the point cloud coordinates of each of the first grid, the second grid and the third grid are down-sampled to obtain corresponding point coordinates, and the down-sampling method comprises one of the following: calculating the mean value of the road point cloud coordinates in the corresponding grid, calculating the median value of the road point cloud coordinates in the corresponding grid, and randomly selecting one road point cloud coordinate in the corresponding grid.

8. A device for determining a road plane, characterized in that Comprising: a point cloud coordinate acquisition unit configured to acquire point cloud coordinates in a region of interest of a target road; a grid division unit configured to divide the region of interest into grids according to a preset rule to obtain a plurality of division grids and point cloud coordinates in the division grids; a grid selection unit configured to select a plurality of first grids in near grids, a second grid farthest from the first grids in far grids, and a plurality of third grids randomly in intermediate grids, the near grids being grids in the division grids having a longitudinal distance to a point cloud collection point less than a first longitudinal distance, the far grids being grids in the division grids having a longitudinal distance to the point cloud collection point greater than a second longitudinal distance, the second longitudinal distance being greater than the first longitudinal distance, and the intermediate grids being grids in the division grids other than the near grids and the far grids; a triangle construction unit configured to respectively determine a point coordinate based on the point cloud coordinates included in each of the first grid, the corresponding second grid and a randomly selected third grid, and to form a candidate triangle by using the three point coordinates; The plane determination unit is configured to select a candidate triangle meeting a preset constraint condition as a target triangle, and determine a plane of the target triangle as a plane where the region of interest is located; wherein the selection of the candidate triangle meeting the preset constraint condition as the target triangle comprises: calculating an evaluation score according to a shape and an area of the candidate triangle, the evaluation score being in a positive correlation with a shape similarity degree between the candidate triangle and an ideal isosceles triangle, and the evaluation score being in a positive correlation with the area of the candidate triangle, the ideal isosceles triangle being an isosceles triangle with a base being a diagonal of the region of interest and having a largest area; and selecting a plane determined by a candidate triangle with an optimal evaluation score as the plane where the region of interest is located.

9. An electronic device, comprising: The computer program is loaded by the processor, and the processor executes the method for determining a road plane according to any one of claims 1-7. The storage medium stores a computer program, and when the computer program is executed by the processor, the processor implements the method for determining a road plane according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The vehicle-mounted control chip processes point cloud coordinates of a target road region of interest collected by the laser radar, and determines a plane where the target road region of interest is located by using the method for determining a road plane according to any one of claims 1-7.

11. A vehicle characterized by comprising: ​

Images