A method, device, equipment and storage medium for updating high-precision maps

Abstract

The present application relates to a method, device, equipment and storage medium for updating a high-precision map, the method comprising obtaining a first high-precision map and a second high-precision map; parameterizing the first high-precision map and the second high-precision map to obtain each parameterized map cube; calculating the center value and covariance matrix of each map cube to obtain the eigenvalues ​​of the eigenvector of the covariance matrix, wherein there are at least three eigenvalues; sorting the eigenvalues ​​in order of size, and adjusting the order of the eigenvectors of the covariance matrix of each map cube in the sorted order; judging whether the first high-precision map needs to be updated based on the eigenvalues ​​or eigenvectors; if the judgment result meets the preset conditions, replacing the first high-precision map with the second high-precision map. The present application has the effect of updating the map as the scene changes.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to a method, device, equipment and storage medium for updating a high-precision map. Background Art

[0002] At present, autonomous driving maps are difficult to update as the scene changes, and there is a lag in the update method.

[0003] In a time-varying scenario where the scene changes over time, for the positioning thread of an autonomous vehicle, when the scene changes significantly, the map needs to be updated as the scene changes. Otherwise, not only will the real-time positioning result be unreliable, but the subsequent mapping thread will also experience pose divergence during post-processing global optimization due to excessive real-time positioning errors, affecting mapping accuracy.

[0004] With respect to the above-mentioned related technologies, the inventors found that existing autonomous driving maps are difficult to update as the scene changes, resulting in unreliable real-time positioning results and pose divergence of subsequent mapping threads during post-processing global optimization, which affects mapping accuracy. Summary of the invention

[0005] In order to update the autonomous driving map as the scene changes, the present application provides a method, device, equipment and storage medium for updating a high-precision map.

[0006] On the first aspect, the present application provides a method for updating a high-precision map, which has the characteristic of enabling the autonomous driving map to be updated as the scene changes.

[0007] This application is achieved through the following technical solutions:

[0008] A method for updating a high-precision map comprises the following steps:

[0009] Obtain a first high-precision map and a second high-precision map;

[0010] Parameterizing the first high-precision map and the second high-precision map to obtain parameterized map cubes;

[0011] Calculating the center value and the covariance matrix of each map cube to obtain eigenvalues ​​of the eigenvector of the covariance matrix, wherein the eigenvalues ​​are at least three;

[0012] Sorting the eigenvalues ​​in order of magnitude, and adjusting the order of the eigenvectors of the covariance matrix of each map cube accordingly according to the sorted order;

[0013] Based on the eigenvalue or the eigenvector, determining whether the first high-precision map needs to be updated;

[0014] If the judgment result meets the preset conditions, the second high-precision map is used to replace the first high-precision map.

[0015] In a preferred example, the present application may be further configured as follows: the step of determining whether the first high-precision map needs to be updated based on the eigenvalue or the eigenvector includes:

[0016] Counting the number of the parameterized map cubes that need to be updated in the first high-precision map;

[0017] Determining whether the total number of the parameterized map cubes that need to be updated for the first high-precision map exceeds a preset first threshold;

[0018] If the total number of the parameterized map cubes that need to be updated for the first high-precision map exceeds a preset first threshold, the first high-precision map needs to be updated.

[0019] In a preferred example, the present application may be further configured as follows: the step of determining whether the first high-precision map needs to be updated based on the eigenvalue or the eigenvector includes:

[0020] When the autonomous driving vehicle is in an accelerating state, if the eigenvalue satisfies a condition, a first angle between a eigenvector of a maximum eigenvalue mapping of an updated parameterized map cube of the second high-precision map and a eigenvector of a maximum eigenvalue mapping of an updated parameterized map cube of the corresponding first high-precision map is calculated;

[0021] If the first angle is greater than a preset second threshold, it is determined that the parameterized map cube of the corresponding first high-precision map needs to be updated.

[0022] In a preferred example, the present application may be further configured as follows: the step of determining whether the first high-precision map needs to be updated based on the eigenvalue or the eigenvector includes:

[0023] When the autonomous driving vehicle is in an accelerating state, if the eigenvalue satisfies a condition, a second angle between a eigenvector of the maximum eigenvalue mapped by the updated parameterized map cube of the second high-precision map and a eigenvector of the maximum eigenvalue mapped by the updated parameterized map cube of the corresponding first high-precision map, and a third angle between a eigenvector of the second largest eigenvalue mapped by the updated parameterized map cube of the second high-precision map and a eigenvector of the second largest eigenvalue mapped by the updated parameterized map cube of the corresponding first high-precision map are calculated;

[0024] If the second angle is greater than or equal to a preset third threshold, and the third angle is greater than or equal to the third threshold, it is determined that the parameterized map cube of the corresponding first high-precision map needs to be updated.

[0025] In a preferred example, the present application may be further configured as follows: the step of determining whether the first high-precision map needs to be updated based on the eigenvalue or the eigenvector includes:

[0026] Based on the eigenvalues, obtaining a first factor and a second factor;

[0027] Determining whether the absolute value of the first factor is less than or equal to a preset fourth threshold, and whether the absolute value of the second factor is less than or equal to the fourth threshold;

[0028] When the preset conditions are met, the angle between the feature vector of the maximum eigenvalue mapping of the updated parameterized map cube of the second high-precision map and the feature vector of the maximum eigenvalue mapping of the corresponding updated parameterized map cube of the first high-precision map is calculated to obtain a fourth angle, and the angle between the feature vector of the second largest eigenvalue mapping of the updated parameterized map cube of the second high-precision map and the feature vector of the second largest eigenvalue mapping of the corresponding updated parameterized map cube of the first high-precision map is calculated to obtain a fifth angle, and the angle between the feature vector of the minimum eigenvalue mapping of the updated parameterized map cube of the second high-precision map and the feature vector of the minimum eigenvalue mapping of the corresponding updated parameterized map cube of the first high-precision map is calculated to obtain a sixth angle;

[0029] Determining whether the fourth angle, the fifth angle, and the sixth angle are all greater than or equal to a preset fifth threshold;

[0030] When the fourth angle, the fifth angle and the sixth angle are all greater than or equal to the fifth threshold, it is determined that the corresponding parameterized map cube of the first high-precision map needs to be updated.

[0031] In a preferred example, the present application may be further configured as follows: the step of obtaining the first factor and the second factor based on the characteristic value includes:

[0032] Calculate the ratio of the smallest eigenvalue to the second largest eigenvalue and the ratio of the second largest eigenvalue to the largest eigenvalue for each updated parameterized map cube of the first high-precision map, and calculate the ratio of the smallest eigenvalue to the second largest eigenvalue and the ratio of the second largest eigenvalue to the largest eigenvalue for each updated parameterized map cube of the corresponding second high-precision map;

[0033] The first factor is obtained by subtracting the ratio of the minimum eigenvalue to the second largest eigenvalue of each updated parameterized map cube of the first high-precision map from the ratio of the minimum eigenvalue to the second largest eigenvalue of each updated parameterized map cube of the corresponding second high-precision map;

[0034] And, the second factor is obtained by subtracting the ratio of the second largest eigenvalue to the largest eigenvalue of each parameterized map cube updated in the first high-precision map from the ratio of the second largest eigenvalue to the largest eigenvalue of each parameterized map cube updated in the corresponding second high-precision map.

[0035] In a preferred example, the present application can be further configured as follows: the calculation formulas of the center value of each map cube, the covariance matrix and the eigenvalue of the eigenvector of the covariance matrix are respectively,

[0036]

[0037]

[0038] Cov*V=λ*V

[0039] In the formula, Cen represents the center value, p k Represents a point in each map cube, n represents the number of points in each map cube, Cov represents the covariance matrix, V represents the eigenvector of the covariance matrix, and λ represents the eigenvalue.

[0040] On the second aspect, the present application provides an updating device for an autonomous driving map, which has the characteristic of updating the autonomous driving map as the scene changes.

[0041] This application is achieved through the following technical solutions:

[0042] An updating device for an autonomous driving map, comprising:

[0043] A map acquisition module, used to acquire a first high-precision map and a second high-precision map;

[0044] A map parameterization module, used to parameterize the first high-precision map and the second high-precision map to obtain parameterized map cubes;

[0045] A calculation module, used for calculating the center value and covariance matrix of each map cube to obtain eigenvalues ​​of the eigenvector of the covariance matrix, wherein the eigenvalues ​​are at least three;

[0046] A sorting module, used for sorting the eigenvalues ​​in order of magnitude, and adjusting the order of the eigenvectors of the covariance matrix of each map cube according to the sorted order;

[0047] A judgment module, used for judging whether the first high-precision map needs to be updated based on the eigenvalue or the eigenvector;

[0048] The map update module is used to replace the first high-precision map with the second high-precision map when the judgment result meets the preset conditions.

[0049] On the third aspect, the present application provides a computer device having the feature of enabling the autonomous driving map to be updated as the scene changes.

[0050] This application is achieved through the following technical solutions:

[0051] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any one of the above-mentioned methods for updating a high-precision map when executing the computer program.

[0052] In a fourth aspect, the present application provides a computer-readable storage medium having the characteristic of enabling the autonomous driving map to be updated as the scene changes.

[0053] This application is achieved through the following technical solutions:

[0054] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of any one of the above-mentioned methods for updating a high-precision map.

[0055] In summary, compared with the prior art, the technical solution provided by this application has at least the following beneficial effects:

[0056] Obtain a first high-precision map, i.e., an old autonomous driving map, and obtain a second high-precision map, i.e., a new autonomous driving map reconstructed based on a preset posture, divide the first high-precision map and the second high-precision map into preset sizes respectively to obtain initial cubes, and parameterize each initial cube to obtain each parameterized map cube of the first high-precision map and each parameterized map cube of the second high-precision map, so as to serve as a data basis for judging whether each updated parameterized map cube of the first high-precision map needs to be updated compared with each updated parameterized map cube of the second high-precision map; based on the parameterized map cubes, calculate the center value and covariance matrix of each map cube to obtain at least three eigenvalues ​​of the eigenvector of the covariance matrix, so as to characterize each parameterized map cube and obtain eigenvalues ​​and eigenvectors, so as to judge whether each updated parameterized map cube of the first high-precision map needs to be updated compared with each updated parameterized map cube of the second high-precision map. data basis; based on the eigenvalue or eigenvector, determine whether each parameterized map cube updated by the first high-precision map needs to be updated compared with each parameterized map cube updated by the second high-precision map, and record the initial cubes that need to be updated of the first high-precision map to accurately detect whether the first high-precision map changes with the scene change; determine whether the total number of initial cubes that need to be updated of the first high-precision map exceeds a preset first threshold, and when the total number of initial cubes that need to be updated of the first high-precision map exceeds the first threshold, use a reconstructed autonomous driving map to replace the old autonomous driving map, so that when the scene changes significantly, the reconstructed second high-precision map is used to replace the old autonomous driving map, so that the system can transmit less map resource data while updating the second high-precision map as the scene changes, thereby improving the transmission rate, making the real-time positioning result obtained according to the autonomous driving map more reliable, and also conducive to improving the mapping accuracy of subsequent mapping threads. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 A flowchart of a method for updating a high-precision map is provided as an exemplary embodiment of the present application.

[0058] Figure 2 A structural block diagram of a device for updating an autonomous driving map provided in accordance with another exemplary embodiment of the present application. DETAILED DESCRIPTION

[0059] This specific embodiment is merely an explanation of the present application and is not a limitation of the present application. After reading this specification, those skilled in the art may make modifications to the present embodiment without any creative contribution as needed, but such modifications are protected by the patent law as long as they are within the scope of the claims of the present application.

[0060] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0061] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0062] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0063] Reference Figure 1 , an embodiment of the present application provides a method for updating a high-precision map, and the main steps of the method are described as follows.

[0064] S1: Obtain a first high-precision map and a second high-precision map;

[0065] S2: parameterizing the first high-precision map and the second high-precision map to obtain parameterized map cubes;

[0066] S3: Calculate the center value and covariance matrix of each map cube to obtain eigenvalues ​​of the eigenvector of the covariance matrix, wherein there are at least three eigenvalues;

[0067] S4: sorting the eigenvalues ​​in order of magnitude, and adjusting the order of the eigenvectors of the covariance matrix of each map cube according to the sorted order;

[0068] S5: Based on the eigenvalue or the eigenvector, determining whether the first high-precision map needs to be updated;

[0069] S6: If the judgment result meets the preset conditions, the second high-precision map is used to replace the first high-precision map.

[0070] Further, the step of determining whether each updated parameterized map cube of the first high-precision map needs to be updated compared to each updated parameterized map cube of the second high-precision map includes:

[0071] When the autonomous driving vehicle is in an accelerating state, if the maximum eigenvalue of the updated parameterized map cube of the first high-precision map is greater than 12 times the second largest eigenvalue, and the maximum eigenvalue of the updated parameterized map cube of the corresponding second high-precision map is less than or equal to 9 times the second largest eigenvalue, it is determined that the updated parameterized map cube of the corresponding first high-precision map needs to be updated.

[0072] Further, the step of determining whether each updated parameterized map cube of the first high-precision map needs to be updated compared to each updated parameterized map cube of the second high-precision map includes:

[0073] When the autonomous driving vehicle is in an accelerating state, if the maximum eigenvalue of the updated parameterized map cube of the first high-precision map is greater than 12 times the second largest eigenvalue, and the maximum eigenvalue of the corresponding updated parameterized map cube of the second high-precision map is greater than 9 times the second largest eigenvalue, calculate a first angle between a eigenvector of the maximum eigenvalue mapping of the updated parameterized map cube of the second high-precision map and a eigenvector of the maximum eigenvalue mapping of the updated parameterized map cube of the corresponding first high-precision map;

[0074] If the first angle is greater than a preset second threshold, it is determined that the updated parameterized map cube of the corresponding first high-precision map needs to be updated.

[0075] Further, the step of determining whether each updated parameterized map cube of the first high-precision map needs to be updated compared to each updated parameterized map cube of the second high-precision map includes:

[0076] When the autonomous driving vehicle is in an accelerating state, if the second largest eigenvalue of the updated parameterized map cube of the first high-precision map is greater than 12 times the minimum eigenvalue, and the second largest eigenvalue of the updated parameterized map cube of the corresponding second high-precision map is less than or equal to 9 times the minimum eigenvalue, it is determined that the updated parameterized map cube of the corresponding first high-precision map needs to be updated.

[0077] Further, the step of determining whether each updated parameterized map cube of the first high-precision map needs to be updated compared to each updated parameterized map cube of the second high-precision map includes:

[0078] When the autonomous driving vehicle is in an accelerating state, if the second largest eigenvalue of the updated parameterized map cube of the first high-precision map is greater than 12 times the smallest eigenvalue, and the second largest eigenvalue of the corresponding updated parameterized map cube of the second high-precision map is greater than 9 times the smallest eigenvalue, calculate the second angle between the eigenvector of the maximum eigenvalue mapping of the updated parameterized map cube of the second high-precision map and the eigenvector of the maximum eigenvalue mapping of the updated parameterized map cube of the corresponding first high-precision map, and the third angle between the eigenvector of the second largest eigenvalue mapping of the updated parameterized map cube of the second high-precision map and the eigenvector of the second largest eigenvalue mapping of the updated parameterized map cube of the corresponding first high-precision map;

[0079] If the second angle is greater than or equal to a preset third threshold, and the third angle is greater than or equal to the third threshold, it is determined that the updated parameterized map cube of the corresponding first high-precision map needs to be updated.

[0080] Further, the step of determining whether each updated parameterized map cube of the first high-precision map needs to be updated compared to each updated parameterized map cube of the second high-precision map includes:

[0081] Calculate the ratio of the minimum eigenvalue to the second largest eigenvalue and the ratio of the second largest eigenvalue to the largest eigenvalue of each updated parameterized map cube of the first high-precision map, and calculate the ratio of the minimum eigenvalue to the second largest eigenvalue and the ratio of the second largest eigenvalue to the largest eigenvalue of each updated parameterized map cube of the corresponding second high-precision map;

[0082] Calculate the difference between the ratio of the minimum eigenvalue to the second largest eigenvalue of each updated parameterized map cube of the first high-precision map and the ratio of the minimum eigenvalue to the second largest eigenvalue of each updated parameterized map cube of the corresponding second high-precision map to obtain a first difference, and the difference between the ratio of the second largest eigenvalue to the largest eigenvalue of each updated parameterized map cube of the first high-precision map and the ratio of the second largest eigenvalue to the largest eigenvalue of each updated parameterized map cube of the corresponding second high-precision map to obtain a second difference;

[0083] Determine whether the absolute value of the first difference is less than or equal to a preset fourth threshold, and whether the absolute value of the second difference is less than or equal to the fourth threshold;

[0084] When the absolute value of the first difference is less than or equal to the fourth threshold and the absolute value of the second difference is less than or equal to the fourth threshold, calculate the angle between the feature vector of the maximum eigenvalue mapping of the updated parameterized map cube of the second high-precision map and the feature vector of the maximum eigenvalue mapping of the corresponding updated parameterized map cube of the first high-precision map to obtain the fourth angle, and calculate the angle between the feature vector of the second largest eigenvalue mapping of the updated parameterized map cube of the second high-precision map and the feature vector of the second largest eigenvalue mapping of the corresponding updated parameterized map cube of the first high-precision map to obtain the fifth angle, and calculate the angle between the feature vector of the minimum eigenvalue mapping of the updated parameterized map cube of the second high-precision map and the feature vector of the minimum eigenvalue mapping of the corresponding updated parameterized map cube of the first high-precision map to obtain the sixth angle;

[0085] Determining whether the fourth angle, the fifth angle, and the sixth angle are all greater than or equal to a preset fifth threshold;

[0086] When the fourth angle, the fifth angle and the sixth angle are all greater than or equal to the fifth threshold, it is determined that the updated parameterized map cube of the corresponding first high-precision map needs to be updated.

[0087] Furthermore, the calculation formulas for the center value of each map cube, the covariance matrix, and the eigenvalues ​​of the eigenvector of the covariance matrix are,

[0088]

[0089]

[0090] Cov*V=λ*V

[0091] In the formula, Cen represents the center value, p k Represents a point in each map cube, n represents the number of points in each map cube, Cov represents the covariance matrix, V represents the eigenvector of the covariance matrix, and λ represents the eigenvalue.

[0092] The above embodiments are described in detail below.

[0093] In a preferred embodiment, a first high-precision map and a second high-precision map reconstructed based on a preset posture are obtained. The first high-precision map is an old autonomous driving map, and the second high-precision map is a new autonomous driving map reconstructed based on a preset posture. The preset posture can be a posture obtained by obtaining the current posture, or a posture obtained by obtaining the current posture and performing local optimization.

[0094] The first high-precision map and the second high-precision map are respectively divided into initial cubes according to a size of 4m*4m*4m, and each initial cube is parameterized according to a normal distribution to obtain parameterized map cubes.

[0095] Each point in the initial cube is represented by p k It means that according to the points in each initial cube, the center value Cen, the covariance matrix Cov and the eigenvalue λ of the eigenvector V of the covariance matrix Cov are calculated for each map cube. Among them, the center value Cen is calculated according to the points in each initial cube, and the center value Cen and the covariance matrix Cov are used to represent the points in each initial cube. Because the spatial coordinates of a radar point cloud such as (x, y, z) are three-dimensional, the dimension of the covariance matrix Cov is 3x3. Therefore, in this embodiment, there can be three corresponding eigenvalues ​​λ of the eigenvector V of the covariance matrix Cov.

[0096] The calculation formulas for the central value Cen, the covariance matrix Cov, and the eigenvalue λ of the eigenvector V of the covariance matrix Cov are,

[0097]

[0098]

[0099] Cov*V=λ*V

[0100] In the formula, Cen represents the center value, p k Represents a point in each map cube, n represents the number of points in each map cube, Cov represents the covariance matrix, V represents the eigenvector of the covariance matrix, and λ represents the eigenvalue.

[0101] Then, the three eigenvalues ​​of each parameterized map cube are sorted in order of size. In this embodiment, the three eigenvalues ​​of each parameterized map cube are sorted in order from large to small, and the following is obtained: in, represents the largest eigenvalue, represents the second largest eigenvalue, Represents the smallest eigenvalue. And adjust the order of the eigenvectors of the covariance matrix of each map cube in the sorted order, and update the parameterized map cube at the same time.

[0102] Based on the eigenvalues ​​and the eigenvectors, it is determined whether each updated parameterized map cube of the first high-precision map needs to be updated compared to each updated parameterized map cube of the second high-precision map.

[0103] The specific steps for determining whether the first high-precision map needs to be updated include the following three situations.

[0104] 1) When the autonomous driving vehicle is in an accelerating state, if the largest eigenvalue of the updated parameterized map cube of the first high-precision map is greater than 12 times the second largest eigenvalue, that is, for the first high-precision map If the largest eigenvalue of the updated parameterized map cube of the corresponding second high-precision map is less than or equal to 9 times the second largest eigenvalue, the updated parameterized map cube of the corresponding second high-precision map does not satisfy When the updated parameterized map cube of the corresponding first high-precision map needs to be updated, no subsequent judgment is required;

[0105] If the largest eigenvalue of the updated parameterized map cube of the corresponding second high-precision map is greater than 9 times the second largest eigenvalue, the updated parameterized map cube of the corresponding second high-precision map satisfies When , the eigenvector of the maximum eigenvalue mapping of the updated parameterized map cube of the second high-precision map is calculated and the eigenvector of the largest eigenvalue mapping of the updated parameterized map cube corresponding to the first high-precision map The angle size of , obtain the calculated value of the first angle;

[0106] Determine the magnitude relationship between the first angle and a preset second threshold; in this embodiment, the preset second threshold may be 6°;

[0107] If the first angle is greater than the second threshold, it is determined that the updated parameterized map cube of the corresponding first high-precision map needs to be updated; if the first angle is less than or equal to the second threshold, it is determined that the updated parameterized map cube of the corresponding first high-precision map does not need to be updated.

[0108] 2) When the autonomous driving vehicle is in an accelerating state, if the second largest eigenvalue of the updated parameterized map cube of the first high-precision map is greater than 12 times the smallest eigenvalue, that is, for the first high-precision map The updated parameterized map cube of the corresponding second high-precision map has a second largest eigenvalue less than or equal to 9 times the smallest eigenvalue, that is, the updated parameterized map cube of the corresponding second high-precision map does not satisfy When the updated parameterized map cube of the corresponding first high-precision map needs to be updated, no subsequent judgment is required;

[0109] When the second largest eigenvalue of the updated parameterized map cube of the corresponding second high-precision map is greater than 9 times the smallest eigenvalue, that is, the updated parameterized map cube of the corresponding second high-precision map satisfies When the eigenvalue mapping feature vector of the updated parameterized map cube of the second high-precision map is calculated, and the eigenvector of the largest eigenvalue mapping of the updated parameterized map cube corresponding to the first high-precision map The second angle size, and the eigenvector of the second largest eigenvalue mapping of the updated parameterized map cube of the second high-precision map and the eigenvector of the second largest eigenvalue mapping of the updated parameterized map cube corresponding to the first high-precision map The size of the third angle;

[0110] If the second angle is greater than or equal to the preset third threshold, and the third angle is greater than or equal to the third threshold, that is, the second angle and the third angle are both greater than or equal to the third threshold, it is determined that the updated parameterized map cube of the corresponding first high-precision map needs to be updated; if the second angle and the third angle are both less than the third threshold, it is determined that the updated parameterized map cube of the corresponding first high-precision map does not need to be updated. In this embodiment, the preset third threshold can be 6°.

[0111] 3) Directly calculate the ratio of the smallest eigenvalue to the second largest eigenvalue of each parameterized map cube updated in the first high-precision map The ratio of the second largest eigenvalue to the largest eigenvalue And calculate the ratio of the smallest eigenvalue to the second largest eigenvalue of each parameterized map cube updated in the corresponding second high-precision map The ratio of the second largest eigenvalue to the largest eigenvalue

[0112] The absolute value of the first difference is obtained by subtracting the ratio of the minimum eigenvalue to the second largest eigenvalue of each updated parameterized map cube of the first high-precision map from the ratio of the minimum eigenvalue to the second largest eigenvalue of each updated parameterized map cube of the corresponding second high-precision map, and the absolute value of the second difference is obtained by subtracting the ratio of the second largest eigenvalue to the largest eigenvalue of each updated parameterized map cube of the first high-precision map from the ratio of the second largest eigenvalue to the largest eigenvalue of each updated parameterized map cube of the corresponding second high-precision map;

[0113] Determine whether the absolute value of the first difference is less than or equal to a preset fourth threshold, and whether the absolute value of the second difference is less than or equal to the fourth threshold; in this embodiment, the fourth threshold may be 0.15;

[0114] When the absolute value of the first difference is less than or equal to the fourth threshold and the absolute value of the second difference is less than or equal to the fourth threshold, that is, the absolute value of the first difference and the absolute value of the second difference do not exceed the fourth threshold, the eigenvector of the maximum eigenvalue mapping of the updated parameterized map cube of the second high-precision map is calculated and the eigenvector of the largest eigenvalue mapping of the updated parameterized map cube corresponding to the first high-precision map The angle size is obtained to obtain the fourth angle, and the eigenvector of the second largest eigenvalue mapping of the updated parameterized map cube of the second high-precision map is calculated and the eigenvector of the second largest eigenvalue mapping of the updated parameterized map cube corresponding to the first high-precision map The angle size is obtained to obtain the fifth angle, and the eigenvector of the minimum eigenvalue mapping of the updated parameterized map cube of the second high-precision map is calculated and the eigenvector of the minimum eigenvalue mapping of the updated parameterized map cube corresponding to the first high-precision map The angle size of , we get the sixth angle;

[0115] Determine whether the fourth angle, the fifth angle, and the sixth angle are all greater than or equal to a preset fifth threshold; in this embodiment, the fifth threshold may be 8°;

[0116] When the fourth angle, the fifth angle and the sixth angle are all greater than or equal to the fifth threshold, it is determined that the updated parameterized map cube of the corresponding first high-precision map needs to be updated; when the fourth angle, the fifth angle and the sixth angle are all less than the fifth threshold, it is determined that the updated parameterized map cube of the corresponding first high-precision map does not need to be updated.

[0117] Record and count the number of initial cubes that need to be updated for the first high-precision map in each judgment case.

[0118] It is determined whether the total number of initial cubes that need to be updated in the first high-precision map exceeds a preset first threshold. In this embodiment, the first threshold may be 50.

[0119] When the total number of initial cubes that need to be updated in the first high-precision map exceeds a first threshold, a reconstructed second high-precision map is used to replace the old autonomous driving map.

[0120] In summary, a method for updating a high-precision map obtains a first high-precision map and a second high-precision map reconstructed based on a preset posture, divides the first high-precision map and the second high-precision map into preset sizes respectively, obtains initial cubes, and parameterizes each initial cube to obtain each parameterized map cube of the first high-precision map and each parameterized map cube of the second high-precision map, which serve as a data basis for judging whether each parameterized map cube updated by the first high-precision map needs to be updated compared with each parameterized map cube updated by the second high-precision map; based on the parameterized map cubes, three eigenvalues ​​of the center value, the covariance matrix, and the eigenvector of the covariance matrix of each map cube are calculated to characterize each parameterized map cube and obtain the eigenvalues ​​and the eigenvectors as the data basis for judging whether each parameterized map cube updated by the first high-precision map needs to be updated compared with each parameterized map cube updated by the second high-precision map. Based on the eigenvalues ​​or eigenvectors, determine whether each updated parameterized map cube of the first high-precision map needs to be updated compared to each updated parameterized map cube of the second high-precision map, and record the initial cubes of the first high-precision map that need to be updated to accurately detect whether the first high-precision map changes as the scene changes; determine whether the total number of initial cubes that need to be updated of the first high-precision map exceeds a preset first threshold, and when the total number of initial cubes that need to be updated of the first high-precision map exceeds the first threshold, use the reconstructed second high-precision map to replace the old autonomous driving map, so that when the scene changes significantly, the reconstructed autonomous driving map is used to replace the old autonomous driving map, so that the system can transmit less map resource data while updating the second high-precision map as the scene changes, thereby improving the transmission rate, making the real-time positioning result obtained according to the autonomous driving map more reliable, and also helping to improve the mapping accuracy of subsequent mapping threads.

[0121] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0122] Reference Figure 2 The embodiment of the present application also provides an updating device for an autonomous driving map, which corresponds one-to-one to the updating method for a high-precision map in the above embodiment. The updating device for an autonomous driving map includes:

[0123] A map acquisition module, used to acquire a first high-precision map and a second high-precision map;

[0124] A map parameterization module, used to parameterize the first high-precision map and the second high-precision map to obtain parameterized map cubes;

[0125] A calculation module, used for calculating the center value and covariance matrix of each map cube to obtain eigenvalues ​​of the eigenvector of the covariance matrix, wherein the eigenvalues ​​are at least three;

[0126] A sorting module, used for sorting the eigenvalues ​​in order of magnitude, and adjusting the order of the eigenvectors of the covariance matrix of each map cube according to the sorted order;

[0127] A judgment module, used for judging whether the first high-precision map needs to be updated based on the eigenvalue or the eigenvector;

[0128] The map update module is used to replace the first high-precision map with the second high-precision map when the judgment result meets the preset conditions.

[0129] The judgment module includes a first judgment module and a second judgment module.

[0130] A first judgment module is used to judge whether each updated parameterized map cube of the first high-precision map needs to be updated compared with each updated parameterized map cube of the second high-precision map based on the eigenvalue or the eigenvector, and record the initial cube of the first high-precision map that needs to be updated;

[0131] The second judgment module is used to judge whether the total number of the initial cubes that need to be updated in the first high-precision map exceeds a preset first threshold.

[0132] For the specific definition of an updating device for an autonomous driving map, please refer to the definition of an updating method for a high-precision map in the above text, which will not be repeated here. Each module in the above-mentioned updating device for an autonomous driving map can be implemented in whole or in part by software, hardware, and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0133] In one embodiment, a computer device is provided, which may be a server. The computer device includes a processor, a memory, a network interface, and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, any of the above-mentioned methods for updating a high-precision map is implemented.

[0134] In one embodiment, a computer-readable storage medium is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are implemented:

[0135] S1: Obtain a first high-precision map and a second high-precision map;

[0136] S2: parameterizing the first high-precision map and the second high-precision map to obtain parameterized map cubes;

[0137] S3: Calculate the center value and covariance matrix of each map cube to obtain eigenvalues ​​of the eigenvector of the covariance matrix, wherein there are at least three eigenvalues;

[0138] S4: sorting the eigenvalues ​​in order of magnitude, and adjusting the order of the eigenvectors of the covariance matrix of each map cube according to the sorted order;

[0139] S5: Based on the eigenvalue or the eigenvector, determining whether the first high-precision map needs to be updated;

[0140] S6: If the judgment result meets the preset conditions, the second high-precision map is used to replace the first high-precision map.

[0141] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0142] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above.

Claims

1. A method for updating a high-precision map, characterized in that: The following steps are included: Obtain a first high-precision map and a second high-precision map; Parameterizing the first high-precision map and the second high-precision map to obtain parameterized map cubes; Calculating the center value and the covariance matrix of each map cube to obtain eigenvalues ​​of the eigenvector of the covariance matrix, wherein the eigenvalues ​​are at least three; Sorting the eigenvalues ​​in order of magnitude, and adjusting the order of the eigenvectors of the covariance matrix of each map cube accordingly according to the sorted order; Based on the eigenvalue or the eigenvector, determining whether the first high-precision map needs to be updated includes: Based on the eigenvalues, obtaining the first factor and the second factor includes calculating the ratio of the minimum eigenvalue to the second largest eigenvalue and the ratio of the second largest eigenvalue to the largest eigenvalue of each updated parameterized map cube of the first high-precision map, and calculating the ratio of the minimum eigenvalue to the second largest eigenvalue and the ratio of the second largest eigenvalue to the largest eigenvalue of each updated parameterized map cube of the corresponding second high-precision map; obtaining the first factor by subtracting the ratio of the minimum eigenvalue to the second largest eigenvalue of each updated parameterized map cube of the first high-precision map from the ratio of the minimum eigenvalue to the second largest eigenvalue of each updated parameterized map cube of the corresponding second high-precision map; and obtaining the second factor by subtracting the ratio of the second largest eigenvalue to the largest eigenvalue of each updated parameterized map cube of the first high-precision map from the ratio of the second largest eigenvalue to the largest eigenvalue of each updated parameterized map cube of the corresponding second high-precision map; Determining whether the absolute value of the first factor is less than or equal to a preset fourth threshold, and whether the absolute value of the second factor is less than or equal to the fourth threshold; When the preset conditions are met, the angle between the feature vector of the maximum eigenvalue mapping of the updated parameterized map cube of the second high-precision map and the feature vector of the maximum eigenvalue mapping of the corresponding updated parameterized map cube of the first high-precision map is calculated to obtain a fourth angle, and the angle between the feature vector of the second largest eigenvalue mapping of the updated parameterized map cube of the second high-precision map and the feature vector of the second largest eigenvalue mapping of the corresponding updated parameterized map cube of the first high-precision map is calculated to obtain a fifth angle, and the angle between the feature vector of the minimum eigenvalue mapping of the updated parameterized map cube of the second high-precision map and the feature vector of the minimum eigenvalue mapping of the corresponding updated parameterized map cube of the first high-precision map is calculated to obtain a sixth angle; Determining whether the fourth angle, the fifth angle, and the sixth angle are all greater than or equal to a preset fifth threshold; When the fourth angle, the fifth angle and the sixth angle are all greater than or equal to the fifth threshold, it is determined that the corresponding parameterized map cube of the first high-precision map needs to be updated; if the judgment result meets the preset conditions, the second high-precision map is used to replace the first high-precision map.

2. The method for updating a high-precision map according to claim 1, characterized in that: The step of determining whether the first high-precision map needs to be updated based on the eigenvalue or the eigenvector includes: Counting the number of the parameterized map cubes that need to be updated in the first high-precision map; Determining whether the total number of the parameterized map cubes that need to be updated for the first high-precision map exceeds a preset first threshold; If the total number of the parameterized map cubes that need to be updated for the first high-precision map exceeds a preset first threshold, the first high-precision map needs to be updated.

3. The method for updating a high-precision map according to claim 2, characterized in that: The step of determining whether the first high-precision map needs to be updated based on the eigenvalue or the eigenvector includes: When the autonomous driving vehicle is in an accelerating state, if the eigenvalue satisfies a condition, a first angle between a eigenvector of a maximum eigenvalue mapping of an updated parameterized map cube of the second high-precision map and a eigenvector of a maximum eigenvalue mapping of an updated parameterized map cube of the corresponding first high-precision map is calculated; If the first angle is greater than a preset second threshold, it is determined that the parameterized map cube of the corresponding first high-precision map needs to be updated.

4. The method for updating a high-precision map according to claim 1, characterized in that: The step of determining whether the first high-precision map needs to be updated based on the eigenvalue or the eigenvector includes: When the autonomous driving vehicle is in an accelerating state, if the eigenvalue satisfies a condition, a second angle between a eigenvector of the maximum eigenvalue mapped by the updated parameterized map cube of the second high-precision map and a eigenvector of the maximum eigenvalue mapped by the updated parameterized map cube of the corresponding first high-precision map, and a third angle between a eigenvector of the second largest eigenvalue mapped by the updated parameterized map cube of the second high-precision map and a eigenvector of the second largest eigenvalue mapped by the updated parameterized map cube of the corresponding first high-precision map are calculated; If the second angle is greater than or equal to a preset third threshold, and the third angle is greater than or equal to the third threshold, it is determined that the parameterized map cube of the corresponding first high-precision map needs to be updated.

5. The method for updating a high-precision map according to any one of claims 1 to 4, characterized in that: The calculation formulas for the center value of each map cube, the covariance matrix and the eigenvalues ​​of the eigenvector of the covariance matrix are respectively, In the formula, represents the center value, represents a point in each map cube, Represents the number of points in each map cube, represents the covariance matrix, represents the eigenvector of the covariance matrix, and λ represents the eigenvalue.

6. An updating device for an autonomous driving map, characterized in that: include, A map acquisition module, used to acquire a first high-precision map and a second high-precision map; A map parameterization module, used to parameterize the first high-precision map and the second high-precision map to obtain parameterized map cubes; A calculation module, used for calculating the center value and covariance matrix of each map cube to obtain eigenvalues ​​of the eigenvector of the covariance matrix, wherein the eigenvalues ​​are at least three; A sorting module, used for sorting the eigenvalues ​​in order of magnitude, and adjusting the order of the eigenvectors of the covariance matrix of each map cube according to the sorted order; A judgment module, for judging whether the first high-precision map needs to be updated based on the eigenvalue or the eigenvector, including obtaining a first factor and a second factor based on the eigenvalue, including calculating the ratio of the minimum eigenvalue to the second largest eigenvalue and the ratio of the second largest eigenvalue to the largest eigenvalue of each parameterized map cube updated in the first high-precision map, and calculating the ratio of the minimum eigenvalue to the second largest eigenvalue and the ratio of the second largest eigenvalue to the largest eigenvalue of each parameterized map cube updated in the corresponding second high-precision map; obtaining the first factor by subtracting the ratio of the minimum eigenvalue to the second largest eigenvalue of each parameterized map cube updated in the first high-precision map from the ratio of the minimum eigenvalue to the second largest eigenvalue of each parameterized map cube updated in the corresponding second high-precision map; and obtaining the second factor by subtracting the ratio of the second largest eigenvalue to the largest eigenvalue of each parameterized map cube updated in the first high-precision map from the ratio of the second largest eigenvalue to the largest eigenvalue of each parameterized map cube updated in the corresponding second high-precision map; Determining whether the absolute value of the first factor is less than or equal to a preset fourth threshold, and whether the absolute value of the second factor is less than or equal to the fourth threshold; When the preset conditions are met, the angle between the feature vector of the maximum eigenvalue mapping of the updated parameterized map cube of the second high-precision map and the feature vector of the maximum eigenvalue mapping of the corresponding updated parameterized map cube of the first high-precision map is calculated to obtain a fourth angle, and the angle between the feature vector of the second largest eigenvalue mapping of the updated parameterized map cube of the second high-precision map and the feature vector of the second largest eigenvalue mapping of the corresponding updated parameterized map cube of the first high-precision map is calculated to obtain a fifth angle, and the angle between the feature vector of the minimum eigenvalue mapping of the updated parameterized map cube of the second high-precision map and the feature vector of the minimum eigenvalue mapping of the corresponding updated parameterized map cube of the first high-precision map is calculated to obtain a sixth angle; Determining whether the fourth angle, the fifth angle, and the sixth angle are all greater than or equal to a preset fifth threshold; When the fourth angle, the fifth angle, and the sixth angle are all greater than or equal to the fifth threshold, it is determined that the corresponding parameterized map cube of the first high-precision map needs to be updated; The map update module is used to replace the first high-precision map with the second high-precision map when the judgment result meets the preset conditions.

7. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method according to 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, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images