Positioning method and device fusing dynamic and static point cloud maps, equipment and storage medium

Abstract

The invention discloses a positioning method and device fusing a dynamic and static point cloud map, equipment and a storage medium. The method comprises the following steps: acquiring a corresponding static point cloud map and a dynamic point cloud map for point cloud of a current frame, performing three-dimensional grid preprocessing, performing statistics on a mean value and a three-dimensional variance of distribution of all point cloud in each grid, taking the grid closest to a predicted position after projection as a static target grid and a dynamic target grid, and performing three-dimensional grid preprocessing on the static point cloud map and the dynamic point cloud map; different constraints of the point cloud of the current frame based on the static point cloud map and the dynamic point cloud map are determined respectively; calculating a first combined constraint value and a second combined constraint value, and comparing the ratio of the first combined constraint value to the second combined constraint value with a set threshold value; and if the ratio is not greater than the set threshold value, taking the ratio as an increment to compensate the initial prediction position so as to position the point cloud of the current frame. Therefore, the positioning result is determined in a manner of tightly coupling the static point cloud map and the dynamic point cloud map, and the positioning accuracy and stability are improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, specifically to technical fields such as intelligent transportation and autonomous driving, and in particular to a positioning method, device, equipment and storage medium that integrates dynamic and static point cloud maps. Background Art

[0002] In existing unmanned driving or robotic systems, LiDAR is often used as the main sensor to obtain accurate real-time positioning. This type of positioning solution relies on offline high-precision point cloud maps and uses the matching of the previous frame of the current point cloud and the point cloud map to obtain the positioning result.

[0003] And point cloud Figure 1 Generally, it is made in advance based on historical data. In the actual positioning process, there may be changes in the real scene, or the scene itself is a highly dynamic scene such as a dock or warehouse. In these cases, the historical point cloud map cannot reflect the real scene map data, but carries wrong information; the positioning information obtained based on wrong map information is bound to be inaccurate. In addition, sometimes intelligent systems such as vehicles or robots need to temporarily go to unmapped areas to complete tasks, and there is no historical point cloud map to rely on.

[0004] In view of the above situation, the existing technology can improve it by collecting data in a small area. Generally, it is necessary to collect data for map update after the scene changes or requirements are put forward. However, in the face of positioning deviation caused by unknown scene changes, the security risk is relatively large before the map is updated; moreover, for small-scale scene changes or highly dynamic scenes, frequent data collection and continuous maintenance of high-precision maps have high operating costs.

[0005] Therefore, it is urgent to find a low-risk and low-cost positioning solution to solve the above-mentioned problem of inaccurate positioning caused by low credibility of point cloud maps due to scene changes or high dynamic scenes. Summary of the invention

[0006] The present application provides a positioning method, device, equipment and storage medium for integrating dynamic and static point cloud maps to solve the problem of inaccurate positioning caused by low credibility of point cloud maps due to scene changes or high dynamic scenes.

[0007] The technical solution is as follows:

[0008] In a first aspect, a positioning method for fusing dynamic and static point cloud maps is provided, comprising:

[0009] Obtaining a corresponding static point cloud map and a dynamic point cloud map for the current frame point cloud, and performing three-dimensional grid preprocessing on the static point cloud map and the dynamic point cloud map, respectively, and counting the mean and three-dimensional variance of all point cloud distributions in each grid;

[0010] Project the initial predicted position of each point in the current frame point cloud, and search the grid closest to the projected predicted position in the static point cloud map and the dynamic point cloud map respectively as the static target grid and the dynamic target grid;

[0011] Determine a first static constraint and a second static constraint of the current frame point cloud based on the static point cloud map according to the mean and three-dimensional variance of the static target grid; and determine a first dynamic constraint and a second dynamic constraint of the current frame point cloud based on the dynamic point cloud map according to the mean and three-dimensional variance of the dynamic target grid;

[0012] Calculating a first joint constraint value based on the first static constraint and the first dynamic constraint and the set static constraint weight and dynamic constraint weight; and calculating a second joint constraint value based on the second static constraint and the second dynamic constraint and the static constraint weight and the dynamic constraint weight;

[0013] The ratio of the first joint constraint value to the second joint constraint value is compared with a set threshold; if the ratio is not greater than the set threshold, the ratio is used as an incremental compensation for the initial predicted position to locate the current frame point cloud.

[0014] In a possible implementation, the method further includes:

[0015] If the ratio is greater than the set threshold, the ratio is used as an incremental compensation for the initial predicted position, and the compensated predicted position is used as a new initial predicted position for projection;

[0016] Recalculating the first joint constraint value and the second joint constraint value until the ratio of the recalculated first joint constraint value to the recalculated second joint constraint value is no greater than a set threshold;

[0017] The ratio is used as an incremental compensation for the initial predicted position to locate the point cloud of the current frame.

[0018] In a possible implementation, the initial predicted position of each point in the current frame point cloud is projected, and the grid closest to the projected predicted position is searched in the static point cloud map and the dynamic point cloud map respectively as the static target grid and the dynamic target grid, specifically including:

[0019] For the initial predicted position of each point in the point cloud of the current frame, projection is performed based on the rotation parameters and translation parameters of the sensor in the map to obtain the projected predicted position;

[0020] Based on the projected predicted position, a grid closest to the projected predicted position is searched in the static point cloud map as a static target grid; and a grid closest to the projected predicted position is searched in the dynamic point cloud map as a dynamic target grid.

[0021] In a possible implementation, determining the first static constraint and the second static constraint of the current frame point cloud based on the static point cloud map according to the mean and the three-dimensional variance of the static target grid; and determining the first dynamic constraint and the second dynamic constraint of the current frame point cloud based on the dynamic point cloud map according to the mean and the three-dimensional variance of the dynamic target grid, specifically includes:

[0022] According to the mean and three-dimensional variance of the static target grid, the first static constraint and the second static constraint of each point in the current frame point cloud based on the static point cloud map are calculated; the first static constraint of all points in the current frame point cloud based on the static point cloud map is accumulated, and the second static constraint of all points in the current frame point cloud based on the static point cloud map is accumulated to obtain the first static constraint and the second static constraint of the current frame point cloud based on the static point cloud map;

[0023] According to the mean and three-dimensional variance of the dynamic target grid, the first dynamic constraint and the second dynamic constraint of each point in the current frame point cloud based on the dynamic point cloud map are calculated; the first dynamic constraint of all points in the current frame point cloud based on the dynamic point cloud map is accumulated, and the second dynamic constraint of all points in the current frame point cloud based on the dynamic point cloud map is accumulated to obtain the first dynamic constraint and the second dynamic constraint of the current frame point cloud based on the dynamic point cloud map.

[0024] In a possible implementation, according to the mean and three-dimensional variance of the static target grid, calculating the first static constraint and the second static constraint of each point in the current frame point cloud based on the static point cloud map, specifically including:

[0025] Based on the mean and 3D variance of the static target grid, the first static constraint is calculated using the following formula:

[0026]

[0027] And the second static constraint:

[0028]

[0029] in, The R preis the rotation parameter of the radar in the map coordinate system in the previous frame of point cloud, and the t pre is the translation parameter of the radar in the map in the previous frame of point cloud. is the coordinate vector of the i-th point in the radar coordinate system; μ in the formula (2) is the mean of the static target grid, and ∑ in the formulas (1) and (2) is the three-dimensional variance of the static target grid;

[0030] According to the mean and three-dimensional variance of the dynamic target grid, a first dynamic constraint and a second dynamic constraint based on the dynamic point cloud map are calculated for each point in the current frame point cloud, specifically including:

[0031] According to the mean and three-dimensional variance of the dynamic target grid, the first dynamic constraint is calculated by the following formula:

[0032]

[0033] And the second dynamic constraint:

[0034]

[0035] in, The R pre is the rotation parameter of the radar in the map coordinate system in the previous frame of point cloud, and the t pre is the translation parameter of the radar in the map in the previous frame of point cloud. is the coordinate vector of the i-th point in the radar coordinate system; μ in formula (4) is the mean of the dynamic target grid, and ∑ in formulas (3) and (4) is the three-dimensional variance of the dynamic target grid.

[0036] In a possible implementation manner, after using the ratio as an incremental compensation for the initial predicted position to locate the current frame point cloud, the method further includes:

[0037] Based on the predicted position after positioning update, the dynamic point cloud map is updated;

[0038] Among them, the effective grids in the dynamic point cloud map are dynamically adjusted and the number remains unchanged.

[0039] In a second aspect, a positioning device for fusing dynamic and static point cloud maps is provided, comprising:

[0040] A statistical module is used to obtain a corresponding static point cloud map and a dynamic point cloud map for the current frame point cloud, and perform three-dimensional grid preprocessing on the static point cloud map and the dynamic point cloud map, respectively, and count the mean and three-dimensional variance of all point cloud distributions in each grid;

[0041] A search module is used to project the initial predicted position of each point in the current frame point cloud, and search the grid closest to the projected predicted position in the static point cloud map and the dynamic point cloud map respectively as the static target grid and the dynamic target grid;

[0042] A determination module, configured to determine a first static constraint and a second static constraint of a current frame point cloud based on the static point cloud map according to the mean and three-dimensional variance of the static target grid; and to determine a first dynamic constraint and a second dynamic constraint of the current frame point cloud based on the dynamic point cloud map according to the mean and three-dimensional variance of the dynamic target grid;

[0043] a calculation module, configured to calculate a first joint constraint value based on the first static constraint and the first dynamic constraint and the set static constraint weight and the dynamic constraint weight; and to calculate a second joint constraint value based on the second static constraint and the second dynamic constraint and the static constraint weight and the dynamic constraint weight;

[0044] A positioning module is used to compare the ratio of the first joint constraint value to the second joint constraint value with a set threshold; if the ratio is not greater than the set threshold, the ratio is used as an incremental compensation for the initial predicted position to locate the current frame point cloud.

[0045] In a third aspect, a computer-readable storage medium is provided, wherein at least one instruction is stored in the storage medium, and the at least one instruction is loaded and executed by a processor to implement the method of the above-mentioned aspect and any possible implementation manner.

[0046] In a fourth aspect, an electronic device is provided, including:

[0047] at least one processor; and

[0048] a memory communicatively connected to the at least one processor; wherein,

[0049] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any possible implementation manner and the aspects described above.

[0050] In a fifth aspect, an autonomous driving vehicle is provided, comprising the electronic device as described above.

[0051] The beneficial effects of the technical solution provided by this application include at least:

[0052] It can be seen from the above technical solution that the present application can obtain the corresponding static point cloud map and dynamic point cloud map for the current frame point cloud, and perform three-dimensional grid preprocessing respectively, and count the mean and three-dimensional variance of all point cloud distributions in each grid. After that, the grid closest to the predicted position after projection is used as the static target grid and the dynamic target grid, and respectively determine the first static constraint and the second static constraint of the current frame point cloud based on the static point cloud map; and the first dynamic constraint and the second dynamic constraint of the current frame point cloud based on the dynamic point cloud map; then calculate the first joint constraint value and the second joint constraint value, and compare the ratio of the first joint constraint value to the second joint constraint value with the set threshold; if the ratio is not greater than the set threshold, the ratio is used as an incremental compensation for the initial predicted position to locate the current frame point cloud. Thus, by using the tight coupling method of the static point cloud map and the dynamic point cloud map, the constraint relationship of the static point cloud map and the dynamic point cloud map is obtained respectively, and the positioning result is finally determined by establishing an association relationship through the constraint relationship. In the case of low credibility of the historical point cloud map, the positioning accuracy and stability are improved.

[0053] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0055] Figure 1 It is a schematic diagram of the steps of a positioning method for fusing dynamic and static point cloud maps provided by an embodiment of the present application.

[0056] Figure 2 It is a schematic diagram of the positioning process of integrating dynamic and static point cloud maps provided by another embodiment of the present application.

[0057] Figure 3 A structural block diagram of a positioning device for fusing dynamic and static point cloud maps provided in yet another embodiment of the present application.

[0058] Figure 4 It is a block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0059] The following is a description of exemplary embodiments of the present application in conjunction with the accompanying drawings, including various details of the embodiments of the present application to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0060] Obviously, the described embodiments are only part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0061] It should be noted that the terminal devices involved in the embodiments of the present application may include but are not limited to mobile phones, personal digital assistants (PDAs), wireless handheld devices, tablet computers and other smart devices; display devices may include but are not limited to personal computers, televisions and other devices with display functions.

[0062] In addition, the term "and / or" in this article is only a description of the association relationship between the 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 generally indicates that the associated objects before and after are in an "or" relationship.

[0063] Considering the positioning deviation caused by unknown scene changes, the security risk is relatively high before the map is updated; in addition, in the face of small-scale scene changes or highly dynamic scenes, frequent data collection and continuous maintenance of high-precision maps have relatively high operating costs. As a result, the credibility of point cloud maps is low and positioning is inaccurate. In view of this, the embodiment of the present application proposes a positioning scheme that integrates dynamic and static point cloud maps. The inventive concept is: obtain the corresponding static point cloud map and dynamic point cloud map for the current frame point cloud, and perform three-dimensional grid preprocessing respectively, and count the mean and three-dimensional variance of all point cloud distributions in each grid. After that, the grid closest to the predicted position after projection is used as the static target grid and the dynamic target grid, and respectively determine the first static constraint and the second static constraint of the current frame point cloud based on the static point cloud map; and the first dynamic constraint and the second dynamic constraint of the current frame point cloud based on the dynamic point cloud map; then calculate the first joint constraint value and the second joint constraint value, and compare the ratio of the first joint constraint value to the second joint constraint value with the set threshold; if the ratio is not greater than the set threshold, the ratio is used as an incremental compensation for the initial predicted position to locate the current frame point cloud. Thus, by using the tight coupling method of the static point cloud map and the dynamic point cloud map, the constraint relationship of the static point cloud map and the dynamic point cloud map is obtained respectively, and the positioning result is finally determined by establishing an association relationship through the constraint relationship. When the credibility of historical point cloud maps is low, improve positioning accuracy and stability.

[0064] Reference Figure 1 The figure is a schematic diagram of the steps of the positioning method for integrating dynamic and static point cloud maps provided in an embodiment of the present application. The execution subject of the method is a positioning device for integrating dynamic and static point cloud maps, which can be a hardware device or software module with storage and data processing capabilities, such as a terminal device such as a computer, a tablet computer, a smart phone, a smart wearable device, or, for example, a software module or component that can be integrated or installed in these terminal devices.

[0065] like Figure 1 As shown, the positioning method for integrating dynamic and static point cloud maps may specifically include the following steps:

[0066] Step 102: Obtain a corresponding static point cloud map and a dynamic point cloud map for the current frame point cloud, and perform three-dimensional grid preprocessing on the static point cloud map and the dynamic point cloud map respectively, and count the mean and three-dimensional variance of all point cloud distributions in each grid.

[0067] First, based on the environment where the current frame point cloud is located, a static point cloud map determined based on the historical point cloud data of the environment can be obtained, and at the same time, a dynamic point cloud map maintained based on the real-time point cloud data of the environment can also be obtained. Among them, the dynamic point cloud map can be established by, for example, CoppeliaSim software. It should be understood that the establishment and acquisition of the static point cloud map and the dynamic point cloud map in this step can be implemented in any existing feasible way, and this application is not limited to this.

[0068] Then, for the obtained static point cloud map and dynamic point cloud map, three-dimensional grid preprocessing is performed respectively. Specifically, the static point cloud map and the dynamic point cloud map can be divided into three-dimensional grids according to appropriate parameter distribution. Then, the mean μ and three-dimensional variance ∑ of all point cloud distributions falling in the same grid in the static point cloud map and the dynamic point cloud map are respectively counted. The formula is as follows:

[0069]

[0070]

[0071] Among them, P i and P k A three-dimensional coordinate vector representing a single point, where m is the number of points in the same grid.

[0072] Through the above formulas, the mean and three-dimensional variance of all point cloud distributions in each grid in the static point cloud map, as well as the mean and three-dimensional variance of all point cloud distributions in each grid in the dynamic point cloud map can be obtained respectively.

[0073] Step 104: Project the initial predicted position of each point in the current frame point cloud, and search in the static point cloud map and the dynamic point cloud map respectively, and find the grid closest to the projected predicted position as the static target grid and the dynamic target grid.

[0074] Optionally, in the scheme of the present application, the initial predicted position of each point in the current frame point cloud can be projected based on the rotation parameters and translation parameters of the sensor in the map to obtain the projected predicted position; based on the projected predicted position, the grid closest to the projected predicted position is searched in the static point cloud map as the static target grid; and, the grid closest to the projected predicted position is searched in the dynamic point cloud map as the dynamic target grid.

[0075] Specifically, the initial predicted position of each point in the current frame point cloud can be realized by referring to the following formula, and the projection is performed based on the rotation parameters and translation parameters of the sensor in the map to obtain the projected predicted position:

[0076]

[0077] in, is the coordinate vector of the i-th point in the map coordinate system, is the coordinate vector of the i-th point in the radar coordinate system, R pre is the rotation parameter of the radar in the map coordinate system in the previous frame of point cloud, and the t pre It is the translation parameter of the radar in the map in the previous frame of point cloud.

[0078] Similarly, in step 104, projection processing and search operations are performed in the static point cloud map and the dynamic point cloud map respectively, and then the static target grid is determined from the static point cloud map, and the dynamic point cloud grid is determined from the dynamic point cloud map.

[0079] Step 106: Determine the first static constraint and the second static constraint of the current frame point cloud based on the static point cloud map according to the mean and the three-dimensional variance of the static target grid; and determine the first dynamic constraint and the second dynamic constraint of the current frame point cloud based on the dynamic point cloud map according to the mean and the three-dimensional variance of the dynamic target grid.

[0080] In the solution of the present application, when determining the first static constraint and the second static constraint of the current frame point cloud based on the static point cloud map according to the mean and the three-dimensional variance of the static target grid, the following steps may be included:

[0081] Step 1: Calculate the first static constraint and the second static constraint of each point in the current frame point cloud based on the static point cloud map according to the mean and three-dimensional variance of the static target grid.

[0082] In specific implementation, the first static constraint can be calculated according to the mean and three-dimensional variance of the static target grid using the following formula:

[0083]

[0084] And the second static constraint:

[0085]

[0086] in, The R pre is the rotation parameter of the radar in the map coordinate system in the previous frame of point cloud, and the t pre is the translation parameter of the radar in the map in the previous frame of point cloud. is the coordinate vector of the i-th point in the radar coordinate system; μ in formula (2) is the mean of the static target grid, and ∑ in formulas (1) and (2) is the three-dimensional variance of the static target grid.

[0087] Step 2: Accumulate the first static constraints of all points in the current frame point cloud based on the static point cloud map, and accumulate the second static constraints of all points in the current frame point cloud based on the static point cloud map to obtain the first static constraints and the second static constraints of the current frame point cloud based on the static point cloud map.

[0088] In specific implementation, the first static constraint of all points in the current frame point cloud based on the static point cloud map can be accumulated to obtain the first static constraint H of the current frame point cloud based on the static point cloud map. s , as shown in the following formula:

[0089]

[0090] And, the first static constraints of all points in the current frame point cloud based on the static point cloud map are accumulated to obtain the second static constraint g of the current frame point cloud based on the static point cloud map. s , as shown in the following formula:

[0091]

[0092] At the same time, in the solution of the present application, when determining the first dynamic constraint and the second dynamic constraint of the current frame point cloud based on the dynamic point cloud map according to the mean and the three-dimensional variance of the dynamic target grid, the following steps may be included:

[0093] In the first step, according to the mean and three-dimensional variance of the dynamic target grid, the first dynamic constraint and the second dynamic constraint of each point in the current frame point cloud based on the dynamic point cloud map are calculated.

[0094] In specific implementation, the first dynamic constraint of each point based on the dynamic point cloud map can be calculated according to the mean and three-dimensional variance of the dynamic target grid by the following formula:

[0095]

[0096] And the second dynamic constraint:

[0097]

[0098] in, The R pre is the rotation parameter of the radar in the map coordinate system in the previous frame of point cloud, and the t pre is the translation parameter of the radar in the map in the previous frame of point cloud. is the coordinate vector of the i-th point in the radar coordinate system; in formula (4), μ is the mean of the dynamic target grid, in formulas (3) and (4), ∑ is the three-dimensional variance of the dynamic target grid, and I represents a unit positive.

[0099] In the second step, the first dynamic constraints of all points in the current frame point cloud based on the dynamic point cloud map are accumulated, and the second dynamic constraints of all points in the current frame point cloud based on the dynamic point cloud map are accumulated to obtain the first dynamic constraints and the second dynamic constraints of the current frame point cloud based on the dynamic point cloud map.

[0100] In a specific implementation, the first dynamic constraint of all points in the current frame point cloud based on the dynamic point cloud map can be accumulated to obtain the first dynamic constraint H of the current frame point cloud based on the dynamic point cloud map. d , as shown in the following formula:

[0101]

[0102] And, the second dynamic constraint of all points in the current frame point cloud based on the dynamic point cloud map is accumulated to obtain the second dynamic constraint g of the current frame point cloud based on the dynamic point cloud map. d , as shown in the following formula:

[0103]

[0104] In this way, based on the above formula group, the first static constraint and the second static constraint of the current frame point cloud based on the static point cloud map, and the first dynamic constraint and the second dynamic constraint based on the dynamic point cloud map can be calculated respectively.

[0105] It should be understood that in the present application, R pre is a 3*3 matrix, which physically means a matrix describing rotation. I is a 3*3 identity matrix. refers to the coordinate vector of the i-th point in the lidar coordinate system, ^ is the antisymmetric matrix, here is based on The calculated antisymmetric matrix is ​​also a 3*3 matrix. Therefore, the entire J is a 3*6 matrix.

[0106] Step 108: Calculate a first joint constraint value based on the first static constraint and the first dynamic constraint and the set static constraint weight and dynamic constraint weight; and calculate a second joint constraint value based on the second static constraint and the second dynamic constraint and the static constraint weight and the dynamic constraint weight.

[0107] In the present application, the first joint constraint H can be calculated by the following formula:

[0108] H=ω s H s +ω d H d

[0109] Among them, ω sis the weight of the first static constraint of the current frame point cloud based on the static point cloud map in the first joint constraint, ω d The weight of the first dynamic constraint of the current frame point cloud based on the dynamic point cloud map in the first joint constraint;

[0110] And, the second joint constraint g is calculated by the following formula,

[0111] g=ω s g s +ω d g d

[0112] Among them, ω s is the weight of the second static constraint of the current frame point cloud based on the static point cloud map in the second joint constraint, ω d The weight of the second dynamic constraint in the second joint constraint for the current frame point cloud based on the dynamic point cloud map.

[0113] By accumulating the constraints of different point cloud maps for each point in the current frame point cloud, we get H s , g s and H d , g d , and the joint constraints H and g are accumulated based on their respective defined weights, thereby preliminarily establishing the coupling relationship between the dynamic point cloud map and the dynamic point cloud map.

[0114] In this application, the weight ω s and ω d The definition of can be flexibly set according to business needs, or determined according to empirical values ​​obtained by repeated testing of historical positioning data. In short, this application does not limit this.

[0115] Step 110: Compare the ratio of the first joint constraint value to the second joint constraint value with a set threshold; if the ratio is not greater than the set threshold, use the ratio as an incremental compensation for the initial predicted position to locate the current frame point cloud.

[0116] After obtaining the first joint constraint H and the second joint constraint g, the following association formula can be established:

[0117] Hδx=g

[0118] Among them, δx is the ratio of the first joint constraint H and the second joint constraint g, and its actual physical meaning is the predicted position x pre The final updated position x new The difference or increment between.

[0119] If the ratio result δx is greater than the set threshold, it means that the predicted position is very different from the final more determined position, and it is necessary to re-determine the predicted position to establish a coupling relationship, such as returning to step 104 to re-select the initial predicted position, and then executing subsequent steps 104-110 until the ratio result is no greater than the set threshold. In specific implementation, the ratio can be used as an incremental compensation for the initial predicted position, and the compensated predicted position is used as a new initial predicted position for projection; the first joint constraint value and the second joint constraint value are recalculated until the ratio of the recalculated first joint constraint value to the recalculated second joint constraint value is no greater than the set threshold; the ratio is used as an incremental compensation for the initial predicted position to locate the current frame point cloud.

[0120] If the ratio result δx is not greater than the set threshold, the ratio can be used as an incremental compensation for the initial predicted position to obtain the updated position x new , so far, one positioning is completed.

[0121]

[0122] After using the ratio as an incremental compensation for the initial predicted position to locate the current frame point cloud, the dynamic point cloud map can also be supplemented and updated based on the predicted position after positioning update; wherein the effective grids in the dynamic point cloud map are dynamically adjusted and the number remains unchanged.

[0123] In the above technical solution, by using the tight coupling method of static point cloud map and dynamic point cloud map, the constraint relationship of static point cloud map and dynamic point cloud map is obtained respectively, and the association relationship is established through the constraint relationship to finally determine the positioning result. In the case of low credibility of historical point cloud map, the positioning accuracy and stability are improved.

[0124] Reference Figure 2 The figure shows a schematic diagram of the positioning process of integrating dynamic and static point cloud maps provided in an embodiment of the present application.

[0125] Step 202: Perform three-dimensional grid preprocessing on the static point cloud map and the dynamic point cloud map respectively, and calculate the mean and three-dimensional variance of all point cloud distributions in each grid.

[0126]

[0127]

[0128] Among them, P i and P k A three-dimensional coordinate vector representing a single point, where m is the number of points in the same grid.

[0129] Through the above formulas, the mean and three-dimensional variance of all point cloud distributions in each grid in the static point cloud map, as well as the mean and three-dimensional variance of all point cloud distributions in each grid in the dynamic point cloud map can be obtained respectively.

[0130] Step 204: For the initial predicted position of each point in the point cloud of the current frame, projection is performed based on the rotation parameters and translation parameters of the sensor in the map to obtain a projected predicted position.

[0131] After obtaining the initialization of positioning on the static point cloud map, start calculating the constraints of the current frame point cloud for the static point cloud map and the dynamic point cloud map respectively. The method of calculating the constraints from the point cloud to the map will not be affected by the map type.

[0132] Specifically, the projection prediction position can be determined by the following formula:

[0133]

[0134] in, is the coordinate vector of the i-th point in the map coordinate system, is the coordinate vector of the i-th point in the radar coordinate system, R pre is the rotation parameter of the radar in the map coordinate system in the previous frame of point cloud, and the t pre It is the translation parameter of the radar in the map in the previous frame of point cloud.

[0135] Step 206: Based on the projected predicted position, search the grid closest to the projected predicted position in the static point cloud map as the static target grid; and search the grid closest to the projected predicted position in the dynamic point cloud map as the dynamic target grid.

[0136] You can search for specific The nearest grid in the static point cloud map and the dynamic point cloud map marks the nearest grid in each map as the target grid.

[0137] Step 208: Calculate the first static constraint and the second static constraint of each point in the current frame point cloud based on the static point cloud map, and the first dynamic constraint and the second dynamic constraint of each point in the current frame point cloud based on the dynamic point cloud map.

[0138] The specific formula in this step can be found in Figure 1 The formulas for the corresponding steps in the scheme are shown in Figure 2.

[0139] Step 210: Accumulate the constraints of each point based on different point cloud maps respectively, and obtain the first joint constraint and the second joint constraint based on the customized static weight and dynamic weight.

[0140] The specific formula in this step can be found in Figure 1 The formulas for the corresponding steps in the scheme are shown in Figure 2.

[0141] Step 212: Establish an association relationship between the first joint constraint and the second joint constraint, and solve to obtain the position increment.

[0142] Step 214: Determine whether the position increment is greater than a set threshold.

[0143] If the position increment is greater than the set threshold, step 216 is executed; otherwise, the process jumps to step 204 .

[0144] Step 216: Update the predicted position based on the position increment to perform accurate positioning.

[0145] Step 218: Update the dynamic point cloud map based on the updated position.

[0146] In specific implementation, the current frame point cloud can be based on x new Perform projection transformation, calculate the mean and three-dimensional variance in the same way as in step 202, compress the point cloud of the current frame after projection transformation, and define the calculated information in a single grid as a new grid V i , V i The specific contents include: adding grid markers, such as the mth i grids, the mean μ of the newly added grids i , the three-dimensional variance of the new grid ∑ i ;

[0147] For each newly added grid V i First, through a strict query strategy, determine whether there is a historical grid in the corresponding grid in the dynamic point cloud map If there is no historical grid, just use Otherwise, it will be based on V i and Update the content to

[0148] It should be understood that the update of the dynamic point cloud map can be implemented according to the existing way of establishing the dynamic point cloud map. The above is only used as an example and is not limited to this.

[0149] In the present application, as the positioning system continues to run, the number of valid grids, i.e., voxels, in the dynamic point cloud map will also increase. The present application can use a recent cache mechanism to maintain the valid voxels of the entire dynamic point cloud map, keeping a certain number of recently added or updated voxels, while the old voxels are eliminated.

[0150] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0151] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0152] Figure 3 A structural block diagram of a positioning device for integrating dynamic and static point cloud maps provided by an embodiment of the present application is shown. Figure 3 As shown. The positioning device 300 for integrating dynamic and static point cloud maps of this embodiment may include a statistical module 301, a search module 302, a determination module 303, a calculation module 304 and a positioning module 305. Among them, the statistical module 301 is used to obtain the corresponding static point cloud map and dynamic point cloud map for the current frame point cloud, and perform three-dimensional grid preprocessing on the static point cloud map and the dynamic point cloud map respectively, and count the mean and three-dimensional variance of all point cloud distributions in each grid. The search module 302 is used to project the initial predicted position of each point in the current frame point cloud, and search in the static point cloud map and the dynamic point cloud map respectively, and use the grid closest to the predicted position after projection as the static target grid and the dynamic target grid. The determination module 303 is used to determine the first static constraint and the second static constraint of the current frame point cloud based on the static point cloud map according to the mean and three-dimensional variance of the static target grid; and determine the first dynamic constraint and the second dynamic constraint of the current frame point cloud based on the dynamic point cloud map according to the mean and three-dimensional variance of the dynamic target grid. The calculation module 304 is used to calculate the first joint constraint value based on the first static constraint and the first dynamic constraint and the set static constraint weight and dynamic constraint weight; and the second joint constraint value is calculated based on the second static constraint and the second dynamic constraint and the static constraint weight and the dynamic constraint weight. The positioning module 305 is used to compare the ratio of the first joint constraint value to the second joint constraint value with a set threshold value; if the ratio is not greater than the set threshold value, the ratio is used as an incremental compensation for the initial predicted position to locate the current frame point cloud.

[0153] It should be noted that part or all of the positioning device for integrating dynamic and static point cloud maps in this embodiment may be an application located in the local terminal, or may also be a functional unit such as a plug-in or software development kit (SDK) set in the application located in the local terminal, or may also be a processing engine located in a network-side server, or may also be a distributed system located on the network side, for example, a processing engine or distributed system in an autonomous driving platform on the network side, etc. This embodiment does not specifically limit this.

[0154] It is understandable that the application may be a local program (nativeApp) installed on the local terminal, or may be a webpage program (webApp) of a browser on the local terminal, which is not limited in this embodiment.

[0155] Optionally, in a possible implementation of this embodiment, the positioning device for integrating dynamic and static point cloud maps further includes: a compensation module; wherein,

[0156] The compensation module is configured to, if the ratio is greater than the set threshold, use the ratio as an incremental compensation for the initial prediction position, and then use the compensated prediction position as a new initial prediction position for projection;

[0157] The calculation module 304 is used to recalculate the first combined constraint value and the second combined constraint value until the ratio of the recalculated first combined constraint value to the recalculated second combined constraint value is not greater than a set threshold;

[0158] The positioning module 305 is used to use the ratio as an incremental compensation for the initial predicted position to position the current frame point cloud.

[0159] Optionally, in a possible implementation of the present embodiment, the search module 302 can be specifically used to project the initial predicted position of each point in the current frame point cloud based on the rotation parameters and translation parameters of the sensor in the map to obtain the projected predicted position; based on the projected predicted position, search the grid closest to the projected predicted position in the static point cloud map as the static target grid; and search the grid closest to the projected predicted position in the dynamic point cloud map as the dynamic target grid.

[0160] Optionally, in a possible implementation of the present embodiment, the determination module 303 may be specifically used to calculate the first static constraint and the second static constraint of each point in the current frame point cloud based on the static point cloud map according to the mean and the three-dimensional variance of the static target grid; accumulate the first static constraints of all points in the current frame point cloud based on the static point cloud map, and accumulate the second static constraints of all points in the current frame point cloud based on the static point cloud map to obtain the first static constraint and the second static constraint of the current frame point cloud based on the static point cloud map; calculate the first dynamic constraint and the second dynamic constraint of each point in the current frame point cloud based on the dynamic point cloud map according to the mean and the three-dimensional variance of the dynamic target grid; accumulate the first dynamic constraint of all points in the current frame point cloud based on the dynamic point cloud map, and accumulate the second dynamic constraint of all points in the current frame point cloud based on the dynamic point cloud map to obtain the first dynamic constraint and the second dynamic constraint of the current frame point cloud based on the dynamic point cloud map.

[0161] Optionally, in a possible implementation of this embodiment, when the determination module 303 calculates the first static constraint and the second static constraint of each point in the current frame point cloud based on the static point cloud map according to the mean and the three-dimensional variance of the static target grid, it can be specifically used to calculate the first static constraint according to the mean and the three-dimensional variance of the static target grid by the following formula:

[0162]

[0163] And the second static constraint:

[0164]

[0165] in, , the R pre is the rotation parameter of the radar in the map coordinate system in the previous frame of point cloud, and the t pre is the translation parameter of the radar in the map in the previous frame of point cloud. is the coordinate vector of the i-th point in the radar coordinate system; μ in the formula (2) is the mean of the static target grid, and ∑ in the formulas (1) and (2) is the three-dimensional variance of the static target grid;

[0166] When the determination module 303 calculates the first dynamic constraint and the second dynamic constraint of each point in the current frame point cloud based on the dynamic point cloud map according to the mean value and the three-dimensional variance of the dynamic target grid, it can be specifically used to calculate the first dynamic constraint according to the mean value and the three-dimensional variance of the dynamic target grid by the following formula:

[0167]

[0168] And the second dynamic constraint:

[0169]

[0170] in, The R pre is the rotation parameter of the radar in the map coordinate system in the previous frame of point cloud, and the t pre is the translation parameter of the radar in the map in the previous frame of point cloud. is the coordinate vector of the i-th point in the radar coordinate system; μ in formula (4) is the mean of the dynamic target grid, and ∑ in formulas (3) and (4) is the three-dimensional variance of the dynamic target grid.

[0171] Optionally, in a possible implementation of this embodiment, the positioning device for fusing dynamic and static point cloud maps further includes: an updating module;

[0172] The updating module is used to supplement and update the dynamic point cloud map based on the predicted position after the positioning module uses the ratio as an incremental compensation for the initial predicted position to locate the current frame point cloud; wherein the effective grids in the dynamic point cloud map are dynamically adjusted and the number remains unchanged.

[0173] In this embodiment, the corresponding static point cloud map and dynamic point cloud map can be obtained for the current frame point cloud, and three-dimensional grid preprocessing is performed respectively, and the mean and three-dimensional variance of all point cloud distributions in each grid are counted. After that, the grid closest to the predicted position after projection is used as the static target grid and the dynamic target grid, and the first static constraint and the second static constraint of the current frame point cloud based on the static point cloud map are determined respectively; and the first dynamic constraint and the second dynamic constraint of the current frame point cloud based on the dynamic point cloud map are determined respectively; then the first joint constraint value and the second joint constraint value are calculated, and the ratio of the first joint constraint value to the second joint constraint value is compared with the set threshold value; if the ratio is not greater than the set threshold value, the ratio is used as an incremental compensation for the initial predicted position to locate the current frame point cloud. Thus, by using the tight coupling method of the static point cloud map and the dynamic point cloud map, the constraint relationship of the static point cloud map and the dynamic point cloud map is obtained respectively, and the positioning result is finally determined by establishing an association relationship through the constraint relationship. In the case of low credibility of the historical point cloud map, the positioning accuracy and stability are improved.

[0174] An embodiment of the present application provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is loaded and executed by a processor to implement the positioning method for fusing dynamic and static point cloud maps as described above.

[0175] An embodiment of the present application provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one instruction, and the instruction is loaded and executed by the processor to implement the positioning method for fusing dynamic and static point cloud maps as described above.

[0176] An embodiment of the present application provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable the computer to execute the positioning method for fusing dynamic and static point cloud maps as described above.

[0177] An embodiment of the present application provides an autonomous driving vehicle, comprising the electronic device as described above. Specifically, the autonomous driving vehicle may be a vehicle of level L2 or above.

[0178] In the technical solution of this application, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the relevant laws and regulations and do not violate public order and good morals.

[0179] Figure 4 A schematic block diagram of an example electronic device 400 that can be used to implement an embodiment of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or required herein.

[0180] like Figure 4 As shown, the electronic device 400 includes a computing unit 401, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 402 or a computer program loaded from a storage unit 408 into a random access memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the electronic device 400 can also be stored. The computing unit 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0181] Multiple components in the electronic device 400 are connected to the I / O interface 405, including: an input unit 406, such as a keyboard, a mouse, etc.; an output unit 407, such as various types of displays, speakers, etc.; a storage unit 408, such as a disk, an optical disk, etc.; and a communication unit 409, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 409 allows the electronic device 400 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0182] The computing unit 401 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 401 performs the various methods and processes described above, such as the positioning method of the fused dynamic and static point cloud map as described above. For example, in some embodiments, the positioning method of the fused dynamic and static point cloud map as described above may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 408. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 400 via the ROM 402 and / or the communication unit 409. When the computer program is loaded into the RAM 403 and executed by the computing unit 401, one or more steps of the positioning method of the fused dynamic and static point cloud map as described above may be performed. Alternatively, in other embodiments, the computing unit 401 may be configured in any other appropriate manner (for example, by means of firmware) to execute the positioning method for fusing dynamic and static point cloud maps as described above.

[0183] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0184] The program code for implementing the method of the present application can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, implements the functions / operations specified in the flow chart and / or block diagram. The program code can be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0185] In the context of the present application, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0186] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0187] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0188] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0189] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps disclosed in this application can be performed in parallel, sequentially or in different orders, as long as the desired results of the technical solution disclosed in this application can be achieved, and this document does not limit this.

[0190] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A positioning method for integrating dynamic and static point cloud maps, characterized in that: include: Obtaining a corresponding static point cloud map and a dynamic point cloud map for the current frame point cloud, and performing three-dimensional grid preprocessing on the static point cloud map and the dynamic point cloud map, respectively, and counting the mean and three-dimensional variance of all point cloud distributions in each grid; Project the initial predicted position of each point in the current frame point cloud, and search the grid closest to the projected predicted position in the static point cloud map and the dynamic point cloud map respectively as the static target grid and the dynamic target grid; Determine a first static constraint and a second static constraint of the current frame point cloud based on the static point cloud map according to the mean and three-dimensional variance of the static target grid; and determine a first dynamic constraint and a second dynamic constraint of the current frame point cloud based on the dynamic point cloud map according to the mean and three-dimensional variance of the dynamic target grid; Calculating a first joint constraint value based on the first static constraint and the first dynamic constraint and the set static constraint weight and dynamic constraint weight; and calculating a second joint constraint value based on the second static constraint and the second dynamic constraint and the static constraint weight and the dynamic constraint weight; The ratio of the first joint constraint value to the second joint constraint value is compared with a set threshold; if the ratio is not greater than the set threshold, the ratio is used as an incremental compensation for the initial predicted position to locate the current frame point cloud.

2. The method according to claim 1, characterized in that The method further comprises: If the ratio is greater than the set threshold, the ratio is used as an incremental compensation for the initial predicted position, and the compensated predicted position is used as a new initial predicted position for projection; Recalculating the first joint constraint value and the second joint constraint value until the ratio of the recalculated first joint constraint value to the recalculated second joint constraint value is no greater than a set threshold; The ratio is used as an incremental compensation for the initial predicted position to locate the point cloud of the current frame.

3. The method according to claim 1 or 2, characterized in that Project the initial predicted position of each point in the current frame point cloud, and search the grid closest to the projected predicted position in the static point cloud map and the dynamic point cloud map respectively as the static target grid and the dynamic target grid, including: For the initial predicted position of each point in the point cloud of the current frame, projection is performed based on the rotation parameters and translation parameters of the sensor in the map to obtain the projected predicted position; Based on the projected predicted position, a grid closest to the projected predicted position is searched in the static point cloud map as a static target grid; and a grid closest to the projected predicted position is searched in the dynamic point cloud map as a dynamic target grid.

4. The method according to claim 1 or 2, characterized in that: Determining the first static constraint and the second static constraint of the current frame point cloud based on the static point cloud map according to the mean and the three-dimensional variance of the static target grid; and determining the first dynamic constraint and the second dynamic constraint of the current frame point cloud based on the dynamic point cloud map according to the mean and the three-dimensional variance of the dynamic target grid, specifically including: According to the mean and three-dimensional variance of the static target grid, the first static constraint and the second static constraint of each point in the current frame point cloud based on the static point cloud map are calculated; the first static constraint of all points in the current frame point cloud based on the static point cloud map is accumulated, and the second static constraint of all points in the current frame point cloud based on the static point cloud map is accumulated to obtain the first static constraint and the second static constraint of the current frame point cloud based on the static point cloud map; According to the mean and three-dimensional variance of the dynamic target grid, the first dynamic constraint and the second dynamic constraint of each point in the current frame point cloud based on the dynamic point cloud map are calculated; the first dynamic constraint of all points in the current frame point cloud based on the dynamic point cloud map is accumulated, and the second dynamic constraint of all points in the current frame point cloud based on the dynamic point cloud map is accumulated to obtain the first dynamic constraint and the second dynamic constraint of the current frame point cloud based on the dynamic point cloud map.

5. The method according to claim 4, characterized in that According to the mean and three-dimensional variance of the static target grid, a first static constraint and a second static constraint of each point in the current frame point cloud based on the static point cloud map are calculated, specifically including: Based on the mean and 3D variance of the static target grid, the first static constraint is calculated using the following formula: And the second static constraint: in, The R pre is the rotation parameter of the radar in the map coordinate system in the previous frame of point cloud, and the t pre is the translation parameter of the radar in the map in the previous frame of point cloud. is the coordinate vector of the i-th point in the radar coordinate system; μ in the formula (2) is the mean of the static target grid, and ∑ in the formulas (1) and (2) is the three-dimensional variance of the static target grid; According to the mean and three-dimensional variance of the dynamic target grid, a first dynamic constraint and a second dynamic constraint based on the dynamic point cloud map are calculated for each point in the current frame point cloud, specifically including: According to the mean and three-dimensional variance of the dynamic target grid, the first dynamic constraint is calculated by the following formula: And the second dynamic constraint: in, The R pre is the rotation parameter of the radar in the map coordinate system in the previous frame of point cloud, and the t pre is the translation parameter of the radar in the map in the previous frame of point cloud. is the coordinate vector of the i-th point in the radar coordinate system; μ in formula (4) is the mean of the dynamic target grid, and ∑ in formulas (3) and (4) is the three-dimensional variance of the dynamic target grid.

6. The method according to claim 1 or 2, characterized in that: After using the ratio as an incremental compensation for the initial predicted position to locate the current frame point cloud, the method further includes: Based on the predicted position after positioning update, the dynamic point cloud map is updated; Among them, the effective grids in the dynamic point cloud map are dynamically adjusted and the number remains unchanged.

7. A positioning device integrating dynamic and static point cloud maps, characterized in that: include: A statistical module is used to obtain a corresponding static point cloud map and a dynamic point cloud map for the current frame point cloud, and perform three-dimensional grid preprocessing on the static point cloud map and the dynamic point cloud map, respectively, and count the mean and three-dimensional variance of all point cloud distributions in each grid; A search module is used to project the initial predicted position of each point in the current frame point cloud, and search the grid closest to the projected predicted position in the static point cloud map and the dynamic point cloud map respectively as the static target grid and the dynamic target grid; A determination module, configured to determine a first static constraint and a second static constraint of a current frame point cloud based on the static point cloud map according to the mean and three-dimensional variance of the static target grid; and to determine a first dynamic constraint and a second dynamic constraint of the current frame point cloud based on the dynamic point cloud map according to the mean and three-dimensional variance of the dynamic target grid; a calculation module, configured to calculate a first joint constraint value based on the first static constraint and the first dynamic constraint and the set static constraint weight and the dynamic constraint weight; and to calculate a second joint constraint value based on the second static constraint and the second dynamic constraint and the static constraint weight and the dynamic constraint weight; A positioning module is used to compare the ratio of the first joint constraint value to the second joint constraint value with a set threshold; if the ratio is not greater than the set threshold, the ratio is used as an incremental compensation for the initial predicted position to locate the current frame point cloud.

8. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-6.

10. An autonomous driving vehicle comprising the electronic device as claimed in claim 8.

Images