A point cloud data matching method, system, electronic device and storage medium

Abstract

The application discloses a point cloud data matching method and system, electronic equipment and a storage medium, wherein the point cloud data matching method is applied to laser radar point cloud data matching, and comprises the following steps: acquiring scene point cloud of a to-be-matched scene and corresponding laser radar point cloud; acquiring a contribution degree reduction ratio of each point cloud in the laser radar point cloud based on a ranging distance of the laser radar point cloud; determining the reduction contribution degree of each point cloud according to the contribution degree reduction ratio; and performing point cloud matching on the laser radar point cloud in the front row of the reduction contribution degree in size order and the scene point cloud. That is, the application can determine the reduction contribution degree of the point cloud through the ranging distance of the laser radar point cloud, so as to reduce the influence of the ranging error on the point cloud matching precision.

Description

Technical Field

[0001] This application relates to the field of 3D scene modeling technology, and in particular to a point cloud data matching method, system, electronic device and storage medium. Background Technology

[0002] With the increasing maturity and widespread use of computer technology, in order to achieve LiDAR positioning, point cloud matching is usually performed using LiDAR point cloud and prior point cloud information or map information.

[0003] In the process of researching and practicing the current technology, the inventors of this application discovered that in the point cloud matching process, as long as the lidar point cloud coincides with the prior map information, it means that the point cloud matching is successful; however, during the measurement process of lidar point clouds, due to different distances, there will be ranging errors of lidar point clouds, which reduces the measurement accuracy and thus affects the point cloud matching accuracy. Summary of the Invention

[0004] The main technical problem addressed in this application is to provide a point cloud data matching method, system electronic equipment, and storage medium that can determine the reduction contribution of point cloud through the ranging distance of lidar point cloud, thereby reducing the impact of ranging error on point cloud matching accuracy.

[0005] To address the aforementioned technical problems, this application provides a point cloud data matching method applied to lidar point cloud data matching, comprising: acquiring scene point clouds of a scene to be matched, and corresponding lidar point clouds; based on the ranging distance of the lidar point clouds, acquiring the contribution reduction ratio of each point cloud in the lidar point clouds; determining the reduction contribution of each point cloud according to the contribution reduction ratio; sorting the reduction contribution by size, and matching the lidar point clouds at the top of the sorted list with the scene point clouds.

[0006] In one embodiment of this application, the contribution reduction ratio includes a first contribution reduction ratio; obtaining the contribution reduction ratio of each point cloud in the lidar point cloud based on the ranging distance of the lidar point cloud includes: obtaining the ranging distance of each point cloud in the lidar point cloud, and obtaining the maximum ranging distance of the lidar; obtaining a preset ranging error coefficient corresponding to the ranging distance; wherein the preset ranging error coefficient is positively correlated with the magnitude of the ranging distance; and determining the first contribution reduction ratio of each point cloud in the lidar point cloud based on the ranging distance, the preset ranging error coefficient, and the maximum ranging distance.

[0007] In one embodiment of this application, determining the reduction contribution of each point cloud based on the contribution reduction ratio includes: obtaining the initial contribution of each point cloud in the lidar point cloud, wherein the initial contribution of each point cloud in the lidar point cloud is the same; and determining the reduction contribution of each point cloud based on the initial contribution and the first contribution reduction ratio.

[0008] In one embodiment of this application, after obtaining the first contribution reduction ratio, the point cloud data matching method further includes: obtaining the distance change value of the ranging distance of each point cloud in the lidar point cloud; determining the second contribution reduction ratio of each point cloud in the lidar point cloud based on the distance change value, the ranging distance, and a preset suppression parameter; and determining the reduction contribution of each point cloud based on the contribution reduction ratio includes: determining the reduction contribution of each point cloud based on the first contribution reduction ratio and the second contribution reduction ratio.

[0009] In one embodiment of this application, obtaining the distance change value of the ranging distance of each point cloud in the lidar point cloud includes: obtaining the lateral point cloud change and the longitudinal point cloud change of each point cloud in the lidar point cloud within a preset angle range; and determining the distance change value based on the lateral point cloud change and the longitudinal point cloud change.

[0010] In one embodiment of this application, determining the reduction contribution of each point cloud based on the first contribution reduction ratio and the second contribution reduction ratio includes: obtaining the initial contribution of each point cloud in the lidar point cloud; and determining the reduction contribution of each point cloud based on the initial contribution, the first contribution reduction ratio, and the second contribution reduction ratio.

[0011] In one embodiment of this application, obtaining the scene point cloud of the scene to be matched and the corresponding lidar point cloud includes: obtaining prior map information of the scene to be matched; obtaining the scene point cloud of the scene to be matched based on the prior map information; and obtaining the lidar point cloud of the scene to be matched based on the prior map information; wherein, the scene point cloud is a standard point cloud.

[0012] To address the aforementioned technical problems, another technical solution adopted in this application is: providing a point cloud data matching system, including a point cloud acquisition module, a contribution reduction ratio acquisition module, a reduction contribution determination module, and a point cloud matching module. The point cloud acquisition module acquires the scene point cloud of the scene to be matched, and the corresponding LiDAR point cloud; the contribution reduction ratio acquisition module acquires the contribution reduction ratio of each point cloud in the LiDAR point cloud based on the ranging distance of the LiDAR point cloud; the reduction contribution determination module determines the reduction contribution of each point cloud according to the contribution reduction ratio; and the point cloud matching module sorts the reduction contribution values ​​by size, and matches the top-ranked LiDAR point clouds with the scene point cloud.

[0013] To solve the above-mentioned technical problems, another technical solution adopted in this application is to provide an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned point cloud data matching method.

[0014] To solve the above-mentioned technical problems, another technical solution adopted in this application is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the point cloud data matching method as described above.

[0015] Unlike current technologies, the point cloud data matching method provided in this application is applied to LiDAR point cloud data matching. It includes: acquiring scene point clouds of the scene to be matched, and the corresponding LiDAR point clouds; obtaining the contribution reduction ratio of each point cloud in the LiDAR point cloud based on the ranging distance of the LiDAR point clouds; determining the reduction contribution of each point cloud according to the contribution reduction ratio; sorting the reduction contributions by size, and matching the top-ranked LiDAR point clouds with the scene point cloud. In other words, this application can determine the reduction contribution of each point cloud through the ranging distance of the LiDAR point cloud, and based on the reduction contribution, reduce and suppress the contribution of each point cloud in point cloud matching, thereby suppressing the negative impact of ranging errors on the point cloud matching contribution, reducing the impact of ranging errors on point cloud matching accuracy, and improving point cloud matching accuracy. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the first embodiment of the point cloud data matching method in this application;

[0017] Figure 2 This is a flowchart illustrating an embodiment of step S12 in this application;

[0018] Figure 3 This is a flowchart illustrating an embodiment of step S13 in this application;

[0019] Figure 4 This is a flowchart illustrating an embodiment of step S14 in this application;

[0020] Figure 5 This is a flowchart illustrating the second embodiment of the point cloud data matching method in this application;

[0021] Figure 6 This is a flowchart illustrating step S23 in the second embodiment of this application.

[0022] Figure 7 This is a flowchart illustrating an embodiment of the point cloud data matching system of this application;

[0023] Figure 8 This is a schematic diagram of the structure of an embodiment of the electronic device of this application;

[0024] Figure 9 This is a schematic diagram of the structure of an embodiment of the computer-readable storage medium of this application. Detailed Implementation

[0025] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be particularly noted that the following embodiments are for illustrative purposes only and do not limit the scope of the invention. Similarly, the following embodiments are only some, not all, embodiments of the present invention, and all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0027] Current point cloud data matching methods, especially for LiDAR localization, primarily involve matching the LiDAR point cloud with prior point cloud or map information (point cloud matching) to achieve LiDAR localization. However, during the measurement process, LiDAR point cloud measurement errors can occur due to varying distances, affecting the final localization accuracy. Most current technical solutions address this error by using other sensors for auxiliary suppression, which can easily introduce new error effects.

[0028] Therefore, a point cloud data matching method is proposed, which can determine the reduction contribution of the point cloud by the ranging distance of the lidar point cloud, thereby reducing the impact of ranging error on the point cloud matching accuracy.

[0029] Please see Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the point cloud data matching method of this application; it should be noted that if substantially the same result is obtained, the method of this invention is not necessarily identical. Figure 1 The process sequence shown is limited to the following: Figure 1 As shown, the method includes the following steps:

[0030] S11. Obtain the scene point cloud of the scene to be matched, and the corresponding LiDAR point cloud.

[0031] The scene to be registered is the actual scene in reality. The scene point cloud is standard point cloud data, which is a set of vectors for the actual scene in a three-dimensional coordinate system, containing geometric position information. The lidar point cloud is a set of scan points. The lidar system scans the ground to obtain the three-dimensional coordinates of the ground reflection points. Each ground reflection point is distributed in three-dimensional space as a point according to the three-dimensional coordinates, and is called a scan point. Each point in the lidar point cloud contains three-dimensional coordinate information, and sometimes also contains color information, reflection intensity information, echo number information, etc.

[0032] Specifically, a 3D modeling tool is used to create a 3D model of the actual scene, thereby obtaining the scene point cloud corresponding to the actual scene; within the range of the obtained scene point cloud, a LiDAR system is used to scan the area to obtain the LiDAR point cloud within that area.

[0033] S12. Based on the ranging distance of the lidar point cloud, obtain the contribution reduction ratio of each point cloud in the lidar point cloud.

[0034] In the process of ranging from a LiDAR point cloud, the distance between each frame of the LiDAR point cloud and another frame within the same area may vary. Therefore, different frames of LiDAR point clouds have different errors compared to the actual scene point cloud. Thus, based on the change in ranging distance, a reduction limit can be applied to the contribution of each point cloud in the LiDAR point cloud to point cloud matching, thereby reducing the impact of the LiDAR point cloud ranging error on the point cloud matching accuracy.

[0035] Specifically, in the process of acquiring the LiDAR point cloud corresponding to the scene to be matched, the ranging distance of each point cloud in each frame is acquired. Based on the error between the ranging distance of each point cloud in the LiDAR point cloud and the scene point cloud of the real scene, the contribution reduction ratio of each point cloud in the LiDAR point cloud is determined; where the larger the ranging distance, the smaller the corresponding contribution reduction ratio.

[0036] S13. Determine the reduction contribution of each point cloud based on the contribution reduction ratio.

[0037] Among them, the contribution is related to the matching confidence in the point cloud matching process. That is, the greater the contribution, the higher the matching confidence in the point cloud matching process. The contribution reduction ratio refers to a reduction limit on the contribution, which limits the contribution of point clouds at different distances to different degrees, and makes the point cloud with the larger error have a smaller contribution to the point cloud matching.

[0038] Specifically, after obtaining the ranging distance of the lidar point cloud and the contribution reduction ratio of each point cloud in the lidar point cloud, the contribution of each point cloud in the lidar point cloud is reduced and limited by the corresponding contribution reduction ratio, so as to obtain the reduction contribution corresponding to each point cloud in the lidar point cloud.

[0039] S14. Sort the reduction contribution by size, and match the top-ranked LiDAR point clouds with the scene point clouds.

[0040] The sorting process involves ranking the points based on their reduction contribution from largest to smallest. The top-ranked points are those with the largest reduction contribution in different frames. This process aims to obtain the LiDAR point cloud that best matches the scene point cloud and has the smallest error. In other words, it obtains the reduction contribution corresponding to each pose, and the reduction contribution corresponds to the matching confidence. The optimal pose is determined based on the best matching confidence, which means obtaining the pose that best matches the LiDAR point cloud and the scene point cloud.

[0041] Specifically, after obtaining the reduction contribution of each point cloud in the LiDAR point cloud, the reduction contribution of the pose corresponding to the point cloud is calculated based on the reduction contribution of each point cloud. The reduction contribution of each pose is obtained, and then the target poses are sorted from largest to smallest according to the reduction contribution of each pose to obtain the top target poses. Then, the LiDAR point cloud corresponding to the target pose is matched with the scene point cloud.

[0042] Point cloud matching refers to the matching between point cloud data obtained by LiDAR scanning the surrounding environment and a grid map, or between point cloud maps themselves. Specifically, it involves finding a pose point in the map such that as many LiDAR data points as possible fall into the corresponding grid cells in the grid map. This pose point is the result of point cloud matching, representing the LiDAR's pose in the map.

[0043] For example, a LiDAR system might have six laser point clouds. Point cloud matching involves calculating and counting the number of point clouds overlapping with the black grid at different poses of the LiDAR on the map. A higher number of overlapping point clouds indicates a higher confidence level for the corresponding pose. The pose with the highest number of overlapping point clouds is the point cloud matching result. If only the number of overlapping point clouds is counted to filter and determine the matching result, each point cloud contributes equally to the entire matching process. In this case, if the LiDAR has the same number of overlapping point clouds with the black grid at two different poses, a valid result cannot be selected. Due to the influence of the ranging error coefficient, each point cloud has a different error—some have large errors, some have small errors. Larger errors will result in larger errors in the point cloud matching result. Therefore, it is necessary to differentiate the contribution of point clouds during matching based on their ranging errors; that is, point clouds with larger errors contribute less, and point clouds with smaller errors contribute more.

[0044] If 5 points fall within the corresponding grid, and neglecting error, the matching confidence score = (O 5) / 6 100%, considering error, the matching confidence level = (O1 r1+O2 r2+O3 r3+O4 r4+O5 r5) / 6 100%, Oi is the original confidence level, which is usually 1, ri is the reduction ratio, which is related to the error ranging coefficient, and the matching confidence level is related to the reduction contribution.

[0045] In this embodiment, the scene point cloud of the scene to be matched and the corresponding LiDAR point cloud are acquired; based on the ranging distance of the LiDAR point cloud, the contribution reduction ratio of each point cloud in the LiDAR point cloud is obtained; according to the contribution reduction ratio, the reduction contribution of each point cloud is determined; the reduction contribution is sorted by size, and the LiDAR point clouds at the top of the ranking are matched with the scene point cloud; that is, this embodiment reduces the contribution of point cloud matching based on the ranging distance of each point cloud in the LiDAR point cloud to obtain the reduction contribution of each point cloud, and then, based on the reduction contribution, the LiDAR point clouds at the top of the reduction contribution are selected to be matched with the scene point cloud, thereby reducing the impact of the ranging error of the point cloud on the matching accuracy in point cloud matching.

[0046] See Figure 2 , Figure 2 This is a flowchart illustrating an embodiment of step S12 in this application;

[0047] The contribution reduction ratio includes the first contribution reduction ratio;

[0048] like Figure 2 As shown, it includes the following steps:

[0049] S121. Obtain the ranging distance of each point cloud in the lidar point cloud, and obtain the maximum ranging distance of the lidar.

[0050] Among them, the ranging distance refers to the distance between the ranging device and the measured point as measured by the lidar, and the maximum ranging distance refers to the maximum distance that the lidar can measure.

[0051] Specifically, in the process of acquiring the lidar point cloud, the ranging distance of each point in the lidar point cloud is obtained, and the maximum ranging distance corresponding to the lidar is obtained according to the preset parameters or test data of the lidar.

[0052] S122. Obtain the preset ranging error coefficient corresponding to the ranging distance; wherein, the preset ranging error coefficient is positively correlated with the magnitude of the ranging distance;

[0053] Among them, the preset ranging error coefficient is related to the ranging distance; it is also related to the material of the reflecting material when the material of the reflecting material is different; and it is also related to the environmental factors when the environmental factors are different.

[0054] Specifically, after obtaining the ranging distance of each point cloud in the lidar point cloud, the corresponding preset ranging error coefficient is obtained from the preset ranging error database based on the current test environment factors, the material of the reflective material, and the ranging distance.

[0055] For example, when the reflective material is black, under normal temperature conditions (10℃~30℃, humidity ≤93%, no condensation), and without reliability testing, the absolute error is ≤10mm, the relative average deviation is ≤8%, and the range is ≤13mm when the distance range is 0.15~0.3m; the absolute error is ≤20mm when the distance range is 0.3~1m, the relative average deviation is ≤8%, and the range is ≤25mm; the absolute error is ≤4% when the distance range is 1~2m, the relative average deviation is ≤0.7%, and the range is ≤4.5%; and the absolute error is ≤5% when the distance range is 2~4m, the relative average deviation is ≤1.2%, and the range is ≤5.5%. Outside of normal temperature conditions (-10℃~10℃, 30℃~40℃, humidity ≤93%, no condensation), or after reliability testing, the absolute error is ≤13mm, the relative average deviation is ≤8%, and the range is ≤15mm when the distance range is 0.15~0.3m; the absolute error is ≤30mm, the relative average deviation is ≤8%, and the range is ≤35mm when the distance range is 1~2m; the absolute error is ≤4.5%, the relative average deviation is ≤0.7%, and the range is ≤5% when the distance range is 2~4m; the absolute error is ≤6%, the relative average deviation is ≤1.2%, and the range is ≤6.5%.

[0056] When the reflective material is white, under normal temperature conditions (10℃~30℃, humidity ≤93%, no condensation), and without reliability testing, the absolute error is ≤10mm, the relative average deviation is ≤8%, and the range is ≤10mm at a distance range of 0.15~0.3m; the absolute error is ≤20mm, the relative average deviation is ≤0.5%, and the range is ≤25mm at a distance range of 1~2m; the absolute error is ≤2.5%, the relative average deviation is ≤0.5%, and the range is ≤3.5% at a distance range of 2~4m. Outside of normal temperature conditions (-10℃~10℃, 30℃~40℃, humidity ≤93%, no condensation), or after reliability testing, the absolute error is ≤10mm, the relative average deviation is ≤8%, and the range is ≤13mm when the distance range is 0.15~0.3m; the absolute error is ≤25mm, the relative average deviation is ≤0.6%, and the range is ≤30mm when the distance range is 1~2m; the absolute error is ≤3.5%, the relative average deviation is ≤0.6%, and the range is ≤4.5% when the distance range is 2~4m; and the absolute error is ≤5%, the relative average deviation is ≤0.6%, and the range is ≤5.5% when the distance range is 2~4m.

[0057] S123. Based on the ranging distance, the preset ranging error coefficient, and the maximum ranging distance, determine the first contribution reduction ratio of each point cloud in the lidar point cloud.

[0058] The contribution reduction ratio includes the first contribution reduction ratio;

[0059] Specifically, after obtaining the ranging distance, preset ranging error coefficient, and corresponding maximum ranging distance for each point cloud in the lidar point cloud, the first contribution reduction ratio for each point cloud in the lidar point cloud is determined by the contribution reduction formula, as shown below:

[0060] ,

[0061] in, The reduction ratio for the first contribution The distance measured for each point in the lidar point cloud. This is the ranging error coefficient. This represents the maximum distance to be measured.

[0062] In this embodiment, by predicting the ranging error coefficient and the corresponding maximum ranging distance for each point cloud in the lidar point cloud, the first contribution reduction ratio for each point cloud in the lidar point cloud can be determined. Then, based on the first contribution reduction ratio, the reduction contribution of each point cloud is determined, which reduces the impact of ranging error on matching accuracy for subsequent point cloud matching.

[0063] See Figure 3 , Figure 3 This is a flowchart illustrating an embodiment of step S13 in this application;

[0064] like Figure 3 As shown, it includes the following steps:

[0065] S131. Obtain the initial contribution of each point cloud in the lidar point cloud, wherein the initial contribution of each point cloud in the lidar point cloud is the same.

[0066] The initial contribution refers to the contribution of each point cloud to the overall matching rate after a successful point cloud match. In the current point cloud matching, the initial contribution of each point cloud is the same.

[0067] Specifically, for each point cloud in the lidar point cloud that can be matched, the corresponding initial contribution is obtained. It can be understood that the initial contribution is the same here.

[0068] S132. Determine the reduction contribution of each point cloud based on the initial contribution and the first contribution reduction ratio.

[0069] Specifically, after obtaining the initial contribution of each point cloud in the lidar point cloud, the initial contribution is calculated according to the first contribution reduction ratio to determine the reduction contribution corresponding to each point cloud; the calculation is as follows:

[0070] ,

[0071] in, To reduce contribution, The reduction ratio for the first contribution This is the initial contribution level.

[0072] See Figure 4 , Figure 4 This is a flowchart illustrating an embodiment of step S14 in this application;

[0073] like Figure 4 As shown, it includes the following steps:

[0074] S141. Obtain the prior map information of the scene to be matched;

[0075] The scene to be matched is an actual scene in the real world, and the prior map information can refer to the three-dimensional model information corresponding to the scene to be matched obtained through three-dimensional modeling.

[0076] S142. Obtain the scene point cloud of the scene to be matched based on prior map information, and obtain the lidar point cloud of the scene to be matched based on prior map information.

[0077] The scene point cloud is a standard point cloud. The lidar point cloud is the point cloud data obtained by lidar scanning of the scene to be matched.

[0078] Specifically, a 3D model of the scene to be matched is performed to obtain the 3D model corresponding to the scene to be matched, and then the scene point cloud corresponding to the scene to be matched is obtained based on the 3D model; the lidar scans the scene to be matched to obtain the lidar point cloud corresponding to the scene to be matched.

[0079] In this embodiment, before acquiring the corresponding point cloud data, the scene to be matched is determined first, and then the corresponding point cloud data is acquired in the scene to be matched. This can reduce the measurement error caused by environmental changes and improve the accuracy of subsequent point cloud matching.

[0080] See Figure 5 , Figure 5 This is a flowchart illustrating the second embodiment of the point cloud data matching method in this application;

[0081] like Figure 5 As shown, it includes the following steps:

[0082] S21. Obtain the scene point cloud of the scene to be matched, and the corresponding LiDAR point cloud;

[0083] S22. Based on the ranging distance, preset ranging error coefficient, and maximum ranging distance of the lidar point cloud, determine the first contribution reduction ratio of each point cloud in the lidar point cloud.

[0084] S23. Based on the distance change value, ranging distance, and preset suppression parameters, determine the second contribution reduction ratio of each point cloud in the lidar point cloud;

[0085] The distance change value is determined based on the ranging distance of each point cloud in the lidar point cloud. The second contribution reduction ratio refers to the contribution reduction ratio formed by the changes in the point cloud in different directions such as the horizontal and vertical directions.

[0086] Specifically, the distance change value between each point cloud in the LiDAR point cloud and the scene point cloud is obtained, and a corresponding preset suppression parameter is set for each distance change value. Based on the distance change value, the ranging distance, and the preset suppression parameter, the second contribution reduction ratio of each point cloud in the LiDAR point cloud is calculated and determined as follows:

[0087] ,

[0088] in, The second contribution reduction ratio, The standard deviation of the distance variation values. For measuring distance, This is the suppression parameter.

[0089] S24. Determine the reduction contribution of each point cloud based on the first contribution reduction ratio and the second contribution reduction ratio.

[0090] in,

[0091] Specifically, after obtaining the first contribution reduction ratio and the second contribution reduction ratio, the reduction contribution of each point cloud in the lidar point cloud can be determined based on the first contribution reduction ratio, the second contribution reduction ratio, and the initial contribution, as shown below:

[0092] ,

[0093] in, The standard deviation of the distance variation values. For measuring distance, To suppress parameters, This is the ranging error coefficient. This represents the maximum distance to be measured.

[0094] S25. Sort the reduction contribution by size, and match the top-ranked LiDAR point clouds with the scene point clouds.

[0095] Sort by size, which means sorting according to the size of the reduction contribution.

[0096] Specifically, after obtaining the reduction contribution of each point cloud in the LiDAR point cloud, the reduction contribution of the pose corresponding to the point cloud is calculated based on the reduction contribution of each point cloud. The reduction contribution of each pose is obtained, and then the target poses are sorted from largest to smallest according to the reduction contribution of each pose to obtain the top target poses. Then, the LiDAR point cloud corresponding to the target pose is matched with the scene point cloud.

[0097] In this embodiment, by obtaining the first contribution reduction ratio, the impact of ranging error on matching accuracy can be effectively reduced. In order to further improve the matching accuracy, the second contribution reduction ratio corresponding to the distance change value based on the point cloud is also obtained, which can effectively reduce the impact of the change in ranging distance on the matching accuracy of point cloud matching.

[0098] See Figure 6 , Figure 6 This is a flowchart illustrating an embodiment of step S23 in this application;

[0099] like Figure 6 As shown, it includes the following steps:

[0100] S231. Obtain the lateral and longitudinal changes of each point cloud in the lidar point cloud within a preset angle range.

[0101] In order to reduce interference caused by scene changes, a preset angle range is set, and within this preset angle range, the lateral and vertical point cloud changes of each point cloud in the lidar point cloud are acquired.

[0102] Specifically, lateral point cloud change refers to the degree of change of the point cloud in the lidar point cloud based on the scene point cloud in the horizontal direction, while longitudinal point cloud change refers to the degree of change of the point cloud in the lidar point cloud based on the scene point cloud in the vertical direction.

[0103] In some embodiments, the degree of change in point cloud can also be measured in ways other than standard deviation, such as variance, covariance, etc.

[0104] S232. Determine the distance change value based on the changes in the horizontal and vertical point clouds.

[0105] Specifically, after acquiring the lateral and longitudinal point cloud changes in the lidar point cloud, the distance change value is determined based on the lateral and longitudinal point cloud changes, and then the standard deviation of the corresponding distance change value is obtained based on the distance change value.

[0106] In this embodiment, by measuring the lateral and longitudinal changes of the point cloud, the degree of change of each point cloud in the lidar point cloud within a preset angle range can be obtained. Then, based on the degree of change, the contribution of the corresponding point cloud is increased, thereby improving the contribution of distant point clouds, so as to ensure the longitudinal positioning accuracy and corresponding matching accuracy in point cloud matching.

[0107] In some embodiments, point clouds in complex environments contribute more to the matching than point clouds in simple environments; for example, point clouds in polyline environments contribute more to the matching than point clouds in straight-line environments.

[0108] In some embodiments, it is also necessary to obtain the initial contribution of each point cloud in the LiDAR point cloud, wherein the initial contribution of each point cloud in the LiDAR point cloud is the same; the initial contribution refers to the contribution of each point cloud to the overall matching rate after successful point cloud matching, and in the current point cloud matching, the initial contribution of each point cloud is the same. Specifically, for each point cloud in the LiDAR point cloud that can be matched, the corresponding initial contribution is obtained, and it can be understood that the initial contribution here is the same.

[0109] See Figure 7 , Figure 7 This is a schematic diagram of the structure of an embodiment of the point cloud data matching system in this application;

[0110] like Figure 7 As shown, the point cloud data matching system 300 includes: a point cloud acquisition module 310, a contribution reduction ratio acquisition module 320, a reduction contribution determination module 330, and a point cloud matching module 340. The point cloud acquisition module 310 acquires the scene point cloud of the scene to be matched, as well as the corresponding LiDAR point cloud. The contribution reduction ratio acquisition module 320 acquires the contribution reduction ratio of each point cloud in the LiDAR point cloud based on the ranging distance of the LiDAR point cloud. The reduction contribution determination module 330 determines the reduction contribution of each point cloud based on the contribution reduction ratio. The point cloud matching module 340 sorts the reduction contribution by size and matches the top-ranked LiDAR point clouds with the scene point cloud.

[0111] See Figure 8 , Figure 8 This is a schematic diagram of an embodiment of the electronic device in this application. This electronic device can perform the point cloud data matching step in the above method.

[0112] The electronic device 400 includes a memory 420, a processor 410, and a computer program stored in the memory 420 and executable on the processor 410. When the processor 410 executes the computer program, it can perform the point cloud data matching execution steps in the above method. For related details, please refer to the detailed description in the above method, which will not be repeated here.

[0113] Please see Figure 9 , Figure 9 This is a schematic diagram of an embodiment of the computer-readable storage medium 500 in this application. The computer-readable storage medium 500 stores a computer program 510, which, when executed by a processor, implements the point cloud data matching execution step in the above-described method. For related details, please refer to the detailed description in the above-described method; it will not be repeated here.

[0114] The above solution acquires the scene point cloud and the corresponding LiDAR point cloud of the scene to be matched; based on the ranging distance of the LiDAR point cloud, it obtains the contribution reduction ratio of each point cloud in the LiDAR point cloud; based on the contribution reduction ratio, it determines the reduction contribution of each point cloud; it sorts the reduction contributions by size, and uses the top-ranked LiDAR point clouds to perform point cloud matching with the scene point cloud; that is, this embodiment reduces the contribution of point cloud matching based on the ranging distance of each point cloud in the LiDAR point cloud to obtain the reduction contribution of each point cloud, and then, based on the reduction contribution, selects the top-ranked LiDAR point clouds in the reduction contribution to perform point cloud matching with the scene point cloud, thereby reducing the impact of point cloud ranging errors on matching accuracy.

[0115] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection of apparatuses or units, and may be electrical, mechanical, or other forms.

[0116] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0117] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0118] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0119] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A point cloud data matching method, characterized in that, Applications include: LiDAR point cloud data matching Obtain the scene point cloud of the scene to be matched, and the corresponding LiDAR point cloud; Based on the ranging distance of the lidar point cloud, the contribution reduction ratio of each point cloud in the lidar point cloud is obtained, wherein the contribution reduction ratio includes a first contribution reduction ratio; the ranging distance of each point cloud in the lidar point cloud is obtained, and the maximum ranging distance of the lidar is obtained; a preset ranging error coefficient corresponding to the ranging distance is obtained; wherein the preset ranging error coefficient is positively correlated with the magnitude of the ranging distance; based on the ranging distance, the preset ranging error coefficient, and the maximum ranging distance, the first contribution reduction ratio of each point cloud in the lidar point cloud is determined; the contribution reduction ratio is a reduction limit for reducing the contribution; and the distance change value of the ranging distance of each point cloud in the lidar point cloud is obtained; based on the distance change value, the ranging distance, and a preset suppression parameter, a second contribution reduction ratio of each point cloud in the lidar point cloud is determined. The reduction contribution of each point cloud is determined according to the reduction ratio, wherein the reduction contribution of each point cloud is determined according to the first reduction ratio and the second reduction ratio. The reduction contribution is sorted by size, and the top-ranked LiDAR point clouds are matched with the scene point clouds.

2. The point cloud data matching method according to claim 1, characterized in that, The step of determining the reduction contribution of each point cloud based on the reduction ratio includes: The initial contribution of each point cloud in the lidar point cloud is obtained, wherein the initial contribution of each point cloud in the lidar point cloud is the same. The reduction contribution of each point cloud is determined based on the initial contribution and the first contribution reduction ratio.

3. The point cloud data matching method according to claim 1, characterized in that, The step of obtaining the distance change value of the ranging distance for each point cloud in the lidar point cloud includes: Acquire the lateral and longitudinal changes of each point cloud in the lidar point cloud within a preset angle range; The distance change value is determined based on the horizontal point cloud changes and the vertical point cloud changes.

4. The point cloud data matching method according to claim 3, characterized in that, The step of determining the reduction contribution of each point cloud based on the first contribution reduction ratio and the second contribution reduction ratio includes: Obtain the initial contribution of each point cloud in the lidar point cloud; The reduction contribution of each point cloud is determined based on the initial contribution, the first contribution reduction ratio, and the second contribution reduction ratio.

5. The point cloud data matching method according to claim 1, characterized in that, The acquisition of the scene point cloud of the scene to be matched, and the corresponding LiDAR point cloud, includes: Obtain prior map information for the scene to be matched; The scene point cloud of the scene to be matched is obtained based on the prior map information, and the lidar point cloud of the scene to be matched is obtained based on the prior map information; wherein, the scene point cloud is a standard point cloud.

6. A point cloud data matching system, characterized in that, include: The point cloud acquisition module is used to acquire the scene point cloud of the scene to be matched, as well as the corresponding LiDAR point cloud; The contribution reduction ratio acquisition module acquires the contribution reduction ratio of each point cloud in the lidar point cloud based on the ranging distance of the lidar point cloud, wherein the contribution reduction ratio includes a first contribution reduction ratio; acquires the ranging distance of each point cloud in the lidar point cloud and the maximum ranging distance of the lidar; acquires a preset ranging error coefficient corresponding to the ranging distance; wherein the preset ranging error coefficient is positively correlated with the magnitude of the ranging distance; determines the first contribution reduction ratio of each point cloud in the lidar point cloud based on the ranging distance, the preset ranging error coefficient, and the maximum ranging distance; the contribution reduction ratio is a reduction limit for reducing the contribution; and acquires the distance change value of the ranging distance of each point cloud in the lidar point cloud; and determines a second contribution reduction ratio of each point cloud in the lidar point cloud based on the distance change value, the ranging distance, and a preset suppression parameter. The contribution reduction determination module determines the reduction contribution of each point cloud according to the contribution reduction ratio, wherein the reduction contribution of each point cloud is determined according to the first contribution reduction ratio and the second contribution reduction ratio. The point cloud matching module is used to sort the reduction contribution by size, and then perform point cloud matching between the top-ranked lidar point clouds and the scene point clouds.

7. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the point cloud data matching method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the point cloud data matching method as described in any one of claims 1 to 5.

Images