High-precision map lane line smoothing processing method and device and electronic equipment

Abstract

The application relates to a high-precision map lane line smoothing processing method and device and electronic equipment. The method comprises the following steps: acquiring line string data; performing merging preprocessing on the line string data to obtain merged and preprocessed line string data; constructing a target function, wherein the target function comprises a first target function determined according to a curvature change parameter of the line string data; and performing smoothing processing on the merged and preprocessed line string data based on the constructed target function and a preset constraint condition. According to the scheme provided in the application, the line string data is subjected to merging preprocessing and then smoothing processing, so that the output data amount during lane line production can be effectively reduced, the operation speed is improved, and the smoothness and production precision of the lane line are improved through constraint optimization setting.

Description

Technical Field

[0001] This application relates to the field of navigation technology, and in particular to a method, apparatus and electronic device for smoothing lane lines in high-precision maps. Background Technology

[0002] In high-precision maps, lane lines are typically formed by connecting multiple data points. Lane lines help vehicles anticipate complex road surface information, such as slope, curvature, and heading, and also serve as a reference for subsequent route navigation. The smoothness of the lane lines directly affects the curvature of the navigation trajectory points, thus impacting the vehicle's lateral and longitudinal performance.

[0003] In related technologies, fitting-based methods are often used to create lane lines in high-precision maps. Specifically, by first selecting the function type, such as a polynomial function or a spline function, and then solving for the unknowns, a curve that describes the changing trend of the data points as much as possible is obtained.

[0004] However, the above-mentioned lane line creation method generates a large amount of data, which is not conducive to storage and loading, and the calculation speed is slow. Furthermore, after adding thinning processing to the obtained data, there is a tendency for the curvature to not meet the constraint conditions, which affects the smoothness of the lane lines. Summary of the Invention

[0005] To address or partially address the problems existing in related technologies, this application provides a high-precision map lane line smoothing method, apparatus, and electronic device, which can effectively reduce the amount of output data during lane line creation, improve the calculation speed, and improve the smoothness and creation accuracy of lane lines through constraint optimization settings.

[0006] The first aspect of this application provides a method for smoothing lane lines in high-precision maps, including:

[0007] Get the string data;

[0008] The string data is preprocessed by merging to obtain preprocessed string data.

[0009] Construct an objective function, the objective function including a first objective function determined based on the curvature change parameter of the line string data;

[0010] The merged preprocessed string data is smoothed based on the constructed objective function and preset constraints.

[0011] In some embodiments, the step of merging and preprocessing the wire string data to obtain merged and preprocessed wire string data includes:

[0012] The string data is divided into multiple string sub-data to be merged according to a preset number of nodes;

[0013] Merge adjacent nodes in multiple sub-strings of data to be merged based on the distance between two nodes being less than the distance threshold, and obtain the merged preprocessed sub-string data.

[0014] In some embodiments, the two adjacent nodes to be merged are two consecutive nodes in the middle of at least four consecutive nodes in the string data.

[0015] In some embodiments, the smoothing process of the merged preprocessed string data based on the constructed objective function and preset constraints includes:

[0016] A derivative-free optimization algorithm is used to find the minimum value that satisfies the preset constraints based on the constructed objective function, so as to smooth the string data.

[0017] In some embodiments, the preset constraints include:

[0018] A first constraint condition corresponding to the curvature variation parameter of the line string data, wherein the first constraint condition may include:

[0019] The curvature change value of the circle formed by the preset number of nodes in the line string data is less than the preset curvature change threshold.

[0020] In some embodiments, the objective function further includes at least: a second objective function determined based on the relative offset parameter of the string data.

[0021] In some embodiments, the preset constraints include:

[0022] A second constraint condition corresponding to the relative offset parameter of the wire string data, wherein the second constraint condition may include:

[0023] The relative offset of the string data consisting of a preset number of nodes is less than a preset offset threshold.

[0024] A second aspect of this application provides a high-precision map lane line smoothing processing device, comprising:

[0025] The data module is used to acquire wire string data;

[0026] The merging module is used to perform merging preprocessing on the wire string data to obtain merged preprocessed wire string data.

[0027] A construction module is used to construct an objective function, the objective function including a first objective function determined based on the curvature change parameters of the line string data;

[0028] The smoothing module is used to smooth the merged preprocessed string data based on the constructed objective function and preset constraints.

[0029] A third aspect of this application provides an electronic device, comprising:

[0030] Processor; and

[0031] A memory that stores executable code, which, when executed by the processor, causes the processor to perform the method described above.

[0032] A fourth aspect of this application provides a computer-readable storage medium having executable code stored thereon, which, when executed by a processor of an electronic device, causes the processor to perform the method described above.

[0033] The technical solution provided in this application may include the following beneficial effects:

[0034] The technical solution of this application can effectively reduce the amount of output data when creating lane lines and improve the calculation speed by merging and preprocessing the line string data and then smoothing it. Furthermore, by setting constraint optimization, it can improve the smoothness and accuracy of lane lines.

[0035] The technical solution of this application also uses a derivative-free optimization algorithm to solve the objective function to achieve smooth optimization of line string data, eliminating the need for the use of fitting functions and multiple parameter tuning processes, and effectively improving the efficiency of smooth processing of line string data.

[0036] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0037] The above and other objects, features and advantages of this application will become more apparent from the more detailed description of exemplary embodiments thereof in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments thereof.

[0038] Figure 1 This is a schematic flowchart illustrating the high-precision map lane line smoothing method in an embodiment of this application;

[0039] Figure 2 This is another schematic flowchart illustrating the high-precision map lane line smoothing method shown in the embodiments of this application;

[0040] Figure 3 This is a schematic diagram of the structure of the high-precision map lane line smoothing device shown in the embodiments of this application;

[0041] Figure 4 This is another structural schematic diagram of the high-precision map lane line smoothing processing device shown in the embodiments of this application;

[0042] Figure 5This is a schematic diagram of the structure of an electronic device shown in an embodiment of this application. Detailed Implementation

[0043] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While embodiments of this application are shown in the drawings, it should be understood that this application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to make this application more thorough and complete, and to fully convey the scope of this application to those skilled in the art.

[0044] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0045] It should be understood that although the terms "first," "second," "third," etc., may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0046] In related technologies, fitting-based methods are often used to create lane lines in high-precision maps. However, this results in a large amount of data, which is not conducive to storage and loading. Furthermore, the processing speed is slow, and after thinning the obtained data, the curvature may not meet the constraints, affecting the smoothness of the lane lines.

[0047] To address the aforementioned issues, this application provides a high-precision map lane line smoothing method that effectively reduces the amount of output data during lane line creation, improves computation speed, and enhances lane line smoothness and creation accuracy through constraint optimization settings.

[0048] The technical solutions of the embodiments of this application are described in detail below with reference to the accompanying drawings.

[0049] Figure 1 This is a schematic flowchart illustrating the high-precision map lane line smoothing method shown in the embodiments of this application.

[0050] See Figure 1 The high-precision map lane line smoothing method of this application includes:

[0051] S110, retrieve string data.

[0052] Line string data can be the raw line string data of lane lines in a high-precision map. It should be understood that line string data is an ordered array generated from two or more points, used to describe the shape of map elements. In this application, the high-precision map lane lines can be lane boundaries or lane centerlines.

[0053] The string data can be obtained locally or downloaded from the server.

[0054] S120, perform preprocessing to merge the string data to obtain the merged preprocessed string data.

[0055] Before smoothing the acquired line string data, a merging preprocessing is performed. Specifically, points in the line string data that meet the merging conditions are merged to obtain the merged preprocessed line string data.

[0056] Merging the line string data before smoothing it reduces the amount of data compared to the traditional method of directly solving the curve corresponding to the line string data using fitting. This makes the line string data obtained after subsequent smoothing easier to store and load.

[0057] S130, Construct the objective function, which includes a first objective function determined based on the curvature variation parameter of the line string data.

[0058] The objective function, also known as the performance index, cost function, or objective criterion, is primarily used to define the optimization criteria for a process; in other words, it is the mathematical expression of the target variable that the process optimization aims to achieve.

[0059] The objective function in this application includes at least a first objective function determined based on the curvature variation parameter of the line string data. Curvature is the rate of rotation of the tangent direction angle with respect to the arc length at a point on the curve. Curvature can be defined by differentiation and indicates the degree to which the curve deviates from a straight line. A larger curvature indicates a greater degree of curvature in the curve.

[0060] In some embodiments, the objective function in the technical solution of this application further includes at least: a second objective function determined based on the relative offset parameter of the string data. The relative offset can be the offset of the optimized node relative to the original initial node.

[0061] S140, based on the constructed objective function and preset constraints, smooths the pre-processed string data after merging.

[0062] After constructing the objective function, the preprocessed string data after merging is smoothed based on the constructed objective function and preset constraints. The smoothing process can involve solving for the minimum value of the objective function that satisfies the preset constraints.

[0063] In some embodiments, after smoothing the line string data, the method further includes detecting whether the line string data meets preset output conditions. These preset output conditions may be whether the curvature change parameter and / or relative offset parameter of the line string data meet certain conditions. For example, the preset output condition may be whether the curvature change parameter of the line string data is less than a preset curvature change threshold. Another example is whether the relative offset parameter of the line string data is less than a preset relative offset threshold. Data output is only performed when the smoothed line string data meets the preset output conditions, further improving the reliability of the data smoothing process. Of course, if the smoothed line string data does not meet the preset output conditions, data output may be withheld or an alarm may be triggered.

[0064] In this embodiment, the technical solution of this application can effectively reduce the amount of output data when creating lane lines by merging and preprocessing the line string data before smoothing it, thereby improving the calculation speed. Furthermore, by setting constraint optimization, it can improve the smoothness and accuracy of lane lines.

[0065] Figure 2 This is another schematic diagram of the high-precision map lane line smoothing method shown in the embodiments of this application.

[0066] See Figure 2 The high-precision map lane line smoothing method of this application includes:

[0067] S210: Obtain the raw line string data of lane lines in the high-precision map.

[0068] The raw line string data of lane lines in high-precision maps can be obtained through data acquisition devices. For example, data can be collected on-site using a data acquisition vehicle equipped with such devices. The acquired data is then organized and categorized to obtain the corresponding line string data for the lane lines on the high-precision map. The raw line string data can be of the LineString data type.

[0069] S220 divides the string data into multiple sub-string data to be merged according to the preset number of nodes.

[0070] The line string data is divided according to a preset number of nodes. For example, if the preset number of nodes is 6, the line string data is divided into multiple line string sub-data to be merged, with each sub-data having 6 consecutive nodes. The line string data is divided into sub-data with the same number of nodes and then merged. This effectively improves the efficiency of line string data merging and processing, while also improving the uniformity and stability of the merging process.

[0071] In some embodiments, the end point of the previous sub-data string to be merged coincides with the beginning point of the next sub-data string to be merged. This effectively ensures the connectivity and smoothness of the sub-data strings obtained after merging. Only one node is divided into overlapping sub-data strings to be merged, thus effectively avoiding data redundancy.

[0072] S230: Merge two adjacent nodes in multiple sub-data strings to be merged based on the distance between two nodes being less than the distance threshold, and obtain the merged preprocessed sub-data strings.

[0073] The rule that adjacent nodes in multiple substrings to be merged should be less than a certain threshold is used to determine whether they need to be merged. Specifically, the threshold can be set according to the actual application; for example, it can be any value between 0.005cm and 0.02cm. When the distance between two adjacent nodes in the substrings to be merged is less than the threshold, the two nodes are merged. The distance between the two nodes can be obtained from their coordinate parameters.

[0074] The process of merging and preprocessing multiple sub-data strings to be merged can be carried out by traversal, which effectively ensures the sufficiency of the merging and preprocessing of the sub-data strings.

[0075] In some embodiments, merging two adjacent nodes can be achieved using the median method. It should be understood that the string data can be a vector or a non-vector. When merging two adjacent nodes, the median of the coordinate parameters of the two adjacent nodes can be taken as the merged point value; that is, the parameter value of the new node obtained after merging can be the median of the parameter values ​​of the two adjacent nodes before merging. In other embodiments, merging two adjacent nodes can also be achieved by removing either of the two adjacent nodes using a deletion method, thus realizing the merging of the two adjacent nodes. By merging the original string data, the amount of data that needs to be processed subsequently is reduced, thereby effectively improving the efficiency of smoothing the string data.

[0076] In some embodiments, during the merging process of line string data, the two adjacent nodes to be merged can be two consecutive nodes in the middle of at least four consecutive nodes in the line string data. It can be understood that the two adjacent nodes to be merged are at least two nodes in the middle of at least four consecutive nodes. In other words, during the merging process, the first and last nodes need to remain fixed. For example, when merging four consecutive nodes A, B, C, and D in the line string data, nodes A and D remain fixed. If the distance between nodes B and C is less than a distance threshold, nodes B and C are merged. This ensures that the line string data maintains a certain continuity after merging, preventing discontinuity caused by the merging process, thereby guaranteeing the connectivity of the line string after smoothing.

[0077] In some embodiments, the two adjacent nodes to be merged can be the new node obtained after the merge and the adjacent original node. That is, when there are at least five consecutive nodes, at least three nodes in the middle can be merged once or more. For example, merging four consecutive nodes A, B, C, D, and E in a line string data, where merging nodes B and C generates a new node B1, and the distance between nodes B1 and D is less than a distance threshold, then nodes B1 and D are merged.

[0078] S240, Construct the objective function, which includes a first objective function determined based on the curvature change parameter of the line string data and a second objective function determined based on the relative offset parameter of the line string data.

[0079] The first objective function, determined based on the curvature variation parameter of the line string data, can be determined by the square of the difference between the curvature of a curve formed by at least three consecutive points and the curvature of adjacent curves formed by at least three consecutive points.

[0080] In some embodiments, the first objective function can be determined using the following formula:

[0081]

[0082] Where the nodes of the line string are p0, p1, p2, ..., pn-1; curvchange represents the curvature change. The above formula is determined by the square of the difference between the curvature of the curve formed by the three nodes pi, pi+1, and pi+2 and the curvature of the curve formed by the three nodes pi+1, pi+2, and pi+3.

[0083] The second objective function, determined based on the relative offset parameter of the string data, can be determined based on the squared distance between the optimized node and the initial node.

[0084] In some embodiments, the second objective function can be determined using the following formula:

[0085]

[0086] Where pi represents the optimized node, and srcpi represents the initial node before optimization. The above formula is determined by the squared distance between the optimized pi and the initial srcpi. The objective function above keeps the first and last nodes unchanged.

[0087] S250 employs a derivative-free optimization algorithm to solve for the minimum value that satisfies preset constraints based on the constructed objective function, thereby smoothing the string data.

[0088] Using a derivative-free optimization algorithm for solving the problem eliminates the need for a fitting function and multiple parameter tuning processes. Specifically, the derivative-free optimization algorithm used in this application can be a Cobyla (linear approximation constrained optimization) algorithm.

[0089] It is understandable that existing technologies employ derivative optimization to optimize data points in string data. These derivative optimization methods commonly use first-order or second-order derivative algorithms, such as gradient descent (first-order) and Newton's method (second-order). When optimizing string data using derivative optimization, the constructed objective function must be first- or second-order differentiable for the data points. For example, this involves iteratively solving for the minimum value of the objective function that satisfies the constraints using a first-order Jacobian matrix or a second-order Hessian matrix. In other words, the optimization process requires first- or second-order derivative information of both the objective function and the constraints. Constructing this objective function requires sophisticated design, necessitates certain construction techniques, involves complex data processing, and requires multiple parameter tuning operations, resulting in high time consumption and low efficiency.

[0090] The Cobyla algorithm, which employs derivative-free optimization, does not require derivative information when smoothing lane lines in high-precision maps. It achieves fast and effective smoothing of line string data by solving for the minimum value that satisfies preset constraints based on the constructed objective function.

[0091] In some embodiments, the preset constraints include a first constraint corresponding to the curvature change parameter of the line string data, such that the preset constraints can correspond to a first objective function. The first constraint may include the curvature change value of the circle formed by a preset number of nodes in the line string data being less than a preset curvature change threshold. By comparing with the preset curvature change threshold, the curvature change in the smoothed line string data is controlled within a certain range, effectively ensuring the smoothing effect of the line string data.

[0092] For example, the first constraint condition can be:

[0093] curvchange(pi,pi+1,pi+2,pi+3) 2 <curvthresh 2

[0094] Here, curvthresh is a preset curvature change threshold. For example, curvthresh can be 0.07, then the optimized line string data corresponding to the first constraint condition satisfies the curvature change value being less than 0.07.

[0095] In some embodiments, the preset constraint may further include a second constraint corresponding to the relative offset parameter of the string data, such that the preset constraint corresponds to the second objective function. The second constraint may include the relative offset of the string data consisting of a preset number of nodes being less than a preset offset threshold. By comparing with the preset offset threshold, the relative offset in the smoothed string data is controlled within a certain range.

[0096] For example, the second constraint could be:

[0097] (pi-srcpi) 2 <jitterthresh 2

[0098] Among them, jitterthresh 2 The preset offset threshold is used. For example, if jitterthresh can be 0.05m, then the optimized string data corresponding to the second constraint will satisfy the relative offset being less than 0.05m.

[0099] The optimization process of the line string data is constrained by the first and second constraints, thereby effectively improving the smoothness and accuracy of the smoothed line string data. Specifically, the method of this application performs a merging preprocessing after acquiring the line string data. In particular, the merging preprocessing involves dividing the line string data into multiple sub-data sets to be merged according to a preset number of nodes. By performing separate merging preprocessing on each sub-data set, reasonable node data is ensured in each sub-data set, such as the absence of adjacent nodes with excessively small spacing. This allows for the early removal of some undesirable data nodes during the constraint optimization process under the first and second constraints, improving the efficiency and data stability of the subsequent smoothing process of the line string data.

[0100] In this application, the preset number of nodes can be at least three.

[0101] For example, the following uses a smoothed line string ABCDE consisting of five nodes as an example to illustrate the effect of the first and second constraints on the line string ABCDE:

[0102] (1) Curvature variation constraint: In the ABCDE line string, the absolute value of the difference between the curvature of the curve formed by the three nodes A, B, and C and the curvature of the curve formed by the three nodes B, C, and D does not exceed 0.07, and the absolute value of the difference between the curvature of the curve formed by the three nodes B, C, and D and the curvature of the curve formed by the three nodes C, D, and E does not exceed 0.07.

[0103] (2) Relative offset constraint: Node B is within a circle with a radius of 0.05m or a square with a side length of 0.1m relative to the original point B'; Node C is within a circle with a radius of 0.05m or a square with a side length of 0.1m relative to the original point C'; Node D is within a circle with a radius of 0.05m or a square with a side length of 0.1m relative to the original point D'; the first point A relative to the first point of the original line string cannot change; and the last point E relative to the last point of the original line string cannot change.

[0104] In this embodiment, the technical solution of this application uses a derivative-free optimization algorithm to solve the objective function to achieve smooth optimization of line string data, eliminating the need for fitting functions and multiple parameter tuning processes, effectively improving the efficiency of smoothing line string data, and adapting to different smoothing requirements by making only minor modifications to the objective function and constraints.

[0105] Corresponding to the aforementioned application function implementation method embodiments, this application also provides a high-precision map lane line smoothing processing device, electronic device, and corresponding embodiments.

[0106] Figure 3 This is a schematic diagram of the structure of the high-precision map lane line smoothing device shown in the embodiments of this application.

[0107] See Figure 3 The high-precision map lane line smoothing processing device of this application includes a data module 310, a merging module 320, a construction module 330, and a smoothing module 340.

[0108] Data module 310 is used to acquire wire string data.

[0109] In some embodiments, the string data acquired by the data module 310 can be obtained locally or downloaded from a server.

[0110] The merging module 320 is used to perform preprocessing on the line string data to obtain the preprocessed line string data.

[0111] In some embodiments, the merging module 320 can be used to divide the string data into multiple string sub-data to be merged according to a preset number of nodes; and merge two adjacent nodes in the multiple string sub-data to be merged according to the distance between two nodes being less than a distance threshold, to obtain the merged preprocessed string data.

[0112] In some embodiments, the merging module 320 may perform the merging process of two adjacent nodes by using the median method.

[0113] In some embodiments, during the merging process of line string data, the merging module 320 merges two adjacent nodes that are two consecutive nodes in the middle of at least four consecutive nodes in the line string data.

[0114] In some embodiments, the two adjacent nodes for which the merging module 320 performs the merging process can be the new node obtained after the merging process and the adjacent original node.

[0115] Module 330 is used to construct the objective function. The objective function may include a first objective function determined based on the curvature variation parameters of the line string data.

[0116] In some embodiments, the objective function of the construction module 330 may further include at least a second objective function determined based on the relative offset parameter of the string data.

[0117] The smoothing module 340 is used to smooth the pre-processed string data after merging based on the constructed objective function and preset constraints.

[0118] In some embodiments, the smoothing module 340 may employ a derivative-free optimization algorithm to solve for the minimum value that satisfies preset constraints based on the constructed objective function, thereby smoothing the string data. The derivative-free optimization algorithm employed by the smoothing module 340 may be the Cobyla (linear approximation constraint optimization) algorithm.

[0119] In some embodiments, the preset constraints of the smoothing module 340 include a first constraint corresponding to the curvature change parameter of the line string data. The first constraint may include the curvature change value of the circle formed by a preset number of nodes in the line string data being less than a preset curvature change threshold.

[0120] In some embodiments, the preset constraints of the smoothing module 340 may further include a second constraint corresponding to the relative offset parameter of the line string data. The second constraint may include the relative offset of the line string data consisting of a preset number of nodes being less than a preset offset threshold.

[0121] In this embodiment, the high-precision map lane line smoothing processing device of this application can effectively reduce the amount of output data during lane line creation and improve the calculation speed by merging and preprocessing the lane line data before smoothing. Furthermore, it can improve the smoothness and creation accuracy of lane lines through constraint optimization settings. In addition, it uses a derivativeless optimization algorithm to solve the objective function to achieve smooth optimization of the lane line data, eliminating the need for the use of fitting functions and multiple parameter tuning processes, thus effectively improving the efficiency of smoothing the lane line data.

[0122] Figure 4 This is another structural schematic diagram of the high-precision map lane line smoothing processing device shown in the embodiments of this application.

[0123] See Figure 4 The high-precision map lane line smoothing processing device of this application includes a data module 310, a merging module 320, a construction module 330, a smoothing module 340, and a condition detection module 350.

[0124] The functions of data module 310, merging module 320, building module 330, and smoothing module 340 can be found in [reference needed]. Figure 3 As shown, it will not be elaborated further here.

[0125] The condition detection module 350 can be used to detect whether the line string data meets preset output conditions after smoothing. The preset output conditions can be whether the curvature change parameter and / or relative offset parameter of the line string data meet certain conditions. For example, the preset output condition can be whether the curvature change parameter of the line string data is less than a preset curvature change threshold. Another example is whether the preset output condition can be whether the relative offset parameter of the line string data is less than a preset relative offset threshold. Data is output only after the smoothed line string data is detected to meet the preset output conditions.

[0126] In some embodiments, the condition detection module 350 can also be used to prevent data output or issue an alarm when it detects that the string data after smoothing does not meet the preset output conditions.

[0127] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated further here.

[0128] Figure 5 This is a schematic diagram of the structure of an electronic device shown in an embodiment of this application.

[0129] See Figure 5 The electronic device 1000 includes a memory 1010 and a processor 1020.

[0130] The processor 1020 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0131] Memory 1010 may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage devices. ROM may store static data or instructions required by processor 1020 or other modules of the computer. Permanent storage devices may be read-write storage devices. Permanent storage devices may be non-volatile storage devices that retain stored instructions and data even when the computer is powered off. In some embodiments, permanent storage devices use mass storage devices (e.g., magnetic or optical disks, flash memory) as permanent storage devices. In other embodiments, permanent storage devices may be removable storage devices (e.g., floppy disks, optical drives). System memory may be a read-write storage device or a volatile read-write storage device, such as dynamic random access memory. System memory may store some or all of the instructions and data required by the processor during operation. Furthermore, memory 1010 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (e.g., DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and disks and / or optical disks may also be used. In some embodiments, the memory 1010 may include a removable storage device that is readable and / or writable, such as a laser disc (CD), a read-only digital multifunction optical disc (e.g., DVD-ROM, dual-layer DVD-ROM), a read-only Blu-ray disc, a high-density optical disc, a flash memory card (e.g., SD card, mini SD card, Micro-SD card, etc.), a magnetic floppy disk, etc. Computer-readable storage media do not contain carrier waves or transient electronic signals transmitted wirelessly or via wired connections.

[0132] The memory 1010 stores executable code, which, when processed by the processor 1020, can cause the processor 1020 to execute part or all of the methods described above.

[0133] Furthermore, the method according to this application can also be implemented as a computer program or computer program product, which includes computer program code instructions for performing some or all of the steps in the method described above.

[0134] Alternatively, this application may be implemented as a computer-readable storage medium (or a non-transitory machine-readable storage medium or a machine-readable storage medium) storing executable code (or computer program or computer instruction code) thereon, which, when executed by a processor of an electronic device (or server, etc.), causes the processor to perform part or all of the steps of the methods described above according to this application.

[0135] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for smoothing lane lines in high-precision maps, characterized in that, include: Get the string data; The string data is preprocessed by merging to obtain preprocessed string data. The process includes: dividing the line string data into multiple line string sub-data to be merged according to a preset number of nodes; merging two adjacent nodes in the multiple line string sub-data to be merged based on the distance between two nodes being less than a distance threshold, to obtain merged preprocessed line string data; the two adjacent nodes to be merged are two consecutive nodes in the middle of at least four consecutive nodes in the line string sub-data to be merged that have not undergone merging processing or have undergone at least one merging processing; the two adjacent nodes to be merged are two consecutive nodes in the middle of at least four consecutive nodes in the line string data; when merging two adjacent nodes, the median value of the coordinate parameters of the two adjacent nodes is taken as the point value obtained after merging; Construct an objective function, which includes a first objective function determined based on the curvature change parameter of the line string data and a second objective function determined based on the relative offset parameter of the line string data; the first objective function is shown in equation (1): （1） In equation (1), the nodes of the string data are p0, p1, p2, ..., pn-1. For curvature variation; The second objective function is shown in equation (2): （2） In equation (2), pi is the optimized node, and srcpi is the initial node before optimization; The pre-processed line string data is smoothed based on the constructed objective function and preset constraints. This includes: employing a derivative-free optimization algorithm to find the minimum value satisfying the preset constraints based on the constructed objective function, thereby smoothing the line string data. The preset constraints include: a first constraint corresponding to the curvature change parameter of the line string data, wherein the first constraint includes: the curvature change value of the circle formed by a preset number of nodes in the line string data is less than a preset curvature change threshold.

2. The method of claim 1, wherein, The preset constraints include: A second constraint condition corresponding to the relative offset parameter of the wire string data, wherein the second constraint condition may include: The relative offset of the string data consisting of a preset number of nodes is less than a preset offset threshold.

3. A high-definition map lane line smoothing device, characterized by, include: The data module is used to acquire wire string data; The merging module is used to perform merging preprocessing on the wire string data to obtain merged preprocessed wire string data; The process includes: dividing the line string data into multiple line string sub-data to be merged according to a preset number of nodes; merging two adjacent nodes in the multiple line string sub-data to be merged based on the distance between two nodes being less than a distance threshold, to obtain merged preprocessed line string data; the two adjacent nodes to be merged are two consecutive nodes in the middle of at least four consecutive nodes in the line string sub-data to be merged that have not undergone merging processing or have undergone at least one merging processing; the two adjacent nodes to be merged are two consecutive nodes in the middle of at least four consecutive nodes in the line string data; when merging two adjacent nodes, the median value of the coordinate parameters of the two adjacent nodes is taken as the point value obtained after merging; A construction module is used to construct an objective function, which includes a first objective function determined based on the curvature change parameter of the line string data and a second objective function determined based on the relative offset parameter of the line string data; the first objective function is shown in equation (1): （1） In equation (1), the nodes of the string data are p0, p1, p2, ..., pn-1. For curvature variation; The second objective function is shown in equation (2): （2） In equation (2), pi is the optimized node, and srcpi is the initial node before optimization; A smoothing module is used to smooth the merged preprocessed line string data based on a constructed objective function and preset constraints. This includes: employing a derivative-free optimization algorithm to find the minimum value satisfying the preset constraints based on the constructed objective function, thereby smoothing the line string data. The preset constraints include: a first constraint corresponding to the curvature change parameter of the line string data, wherein the first constraint includes: the curvature change value of the circle formed by a preset number of nodes in the line string data is less than a preset curvature change threshold.

4. An electronic device, characterized in that, include: processor; as well as A memory having executable code stored thereon, which, when executed by the processor, causes the processor to perform the method as described in any one of claims 1-2.

5. A computer-readable storage medium having executable code stored thereon, characterized in that: When the executable code is executed by the processor of the electronic device, the processor performs the method as described in any one of claims 1-2.

Images