A data processing method, device and electronic equipment

Abstract

The application discloses a data processing method and device and electronic equipment, and relates to the technical field of data processing. The method comprises the following steps: calculating the direction angle of each target object in a preset coordinate system and the distance between each target object and the origin of the preset coordinate system, and generating a mathematical model corresponding to each target object according to the direction angle and the distance. Further, the comprehensive matching error corresponding to each target object is calculated according to the mathematical model, and the direction angle of each target object in the preset coordinate system and the distance between each target object and the origin of the preset coordinate system are adjusted according to the comprehensive matching error. In this way, the coincidence degree between target objects and the direction consistency between target objects can be considered in the matching process of the target objects, and the matching precision is improved.

Description

Technical Field

[0001] This application relates to the field of image processing, and in particular to a data processing method, apparatus, and electronic device. Background Technology

[0002] High-precision maps are electronic maps with higher accuracy and more data dimensions. Higher accuracy is reflected in their centimeter-level precision, while more data dimensions are reflected in their inclusion of traffic-related static information beyond road details. As a key component in achieving autonomous driving, high-precision maps can effectively supplement existing sensors, providing vehicles with more reliable perception capabilities. Therefore, accurately constructing high-precision maps is essential.

[0003] Current methods for constructing high-precision maps generally involve collecting environmental information about the vehicle's surroundings, such as road and building information, from which the necessary map information for building the high-precision map is extracted. This map information extraction process requires matching multiple map vector fragments collected by the vehicle.

[0004] In the process of map vector segment matching, if target 'a' in map vector segment 1 and target 'b' in map vector segment 2 are the same target object in the actual scene, the matching of target 'a' and target 'b' will not be completely consistent due to the different acquisition devices used for them, meaning there is a certain representation error between them. Therefore, a data processing method is urgently needed to optimize the matching results between map vector segments. Summary of the Invention

[0005] This application discloses a data processing method, apparatus, and electronic device that considers not only the overlap between target objects but also the directional consistency between target objects during the matching process, thereby improving matching accuracy.

[0006] In a first aspect, this application provides a data processing method, the method comprising:

[0007] Obtain target objects from multiple map vector segments, and calculate the orientation angle of each target object in a preset coordinate system, as well as the distance between each target object and the origin of the preset coordinate system;

[0008] Based on each directional angle and each distance, generate a mathematical model corresponding to each target object;

[0009] Based on each mathematical model, calculate the comprehensive matching error common to each target object;

[0010] When the overall matching error is greater than the preset error, adjust the orientation angle of each target object in the preset coordinate system and the distance between each target object and the origin of the preset coordinate system until the overall matching error is less than or equal to the preset error.

[0011] By employing the above method, the mathematical model corresponding to the target object considers both the orientation angle and distance, thus improving matching accuracy by taking into account not only the overlap between target objects but also their directional consistency. Furthermore, the mathematical model unifies the dimensions of the orientation angle and distance, reducing the computational load during the matching process.

[0012] In one possible design, generating mathematical models for each target object based on various orientation angles and distances includes:

[0013] Perform Euler rotation calculations on each direction angle to obtain the rotation matrix corresponding to each direction angle;

[0014] Based on each rotation matrix and each distance, a mathematical model corresponding to each target object is generated, wherein the mathematical model is the SE(3) model.

[0015] The above method enables mathematical modeling of the target object from two dimensions: direction angle and distance. In addition, the dimensions of the direction angle and distance corresponding to the target object are standardized.

[0016] In one possible design, calculating the comprehensive matching error common to each target object based on various mathematical models includes:

[0017] Based on the mathematical model corresponding to each target object, the target objects are matched to obtain the first matching result for each target object;

[0018] Calculate the matching error for each first matching result;

[0019] Based on each matching error, the comprehensive matching error corresponding to each target object is calculated.

[0020] The comprehensive matching error is calculated using the above method. This comprehensive matching error can be used to evaluate the matching accuracy corresponding to the current matching result.

[0021] In one possible design, when the overall matching error is greater than a preset error, adjusting the orientation angle of each target object in the preset coordinate system and the distance between each target object and the origin of the preset coordinate system until the overall matching error is less than or equal to the preset error includes:

[0022] Determine whether the overall matching error is less than or equal to the preset error;

[0023] If so, then the comprehensive matching error is determined to meet the matching requirements between the target objects;

[0024] If not, adjust the orientation angle of each target object in the preset coordinate system and the distance between each target object and the origin of the preset coordinate system until the overall matching error is less than or equal to the preset error.

[0025] By using the above method, the orientation angle of each target object in the preset coordinate system and the distance between each target object and the origin of the preset coordinate system are adjusted, thereby improving the matching accuracy between target objects.

[0026] In one possible design, calculating the comprehensive matching error common to each target object based on various mathematical models includes:

[0027] Detect whether each target object is a vehicle pose vertex and / or lane line;

[0028] If so, obtain the constraint rules corresponding to each vehicle pose vertex and / or lane line, and calculate the comprehensive matching error corresponding to each vehicle pose vertex and / or lane line based on each constraint rule and each mathematical model.

[0029] The constraint rules corresponding to the vehicle pose vertices and / or lane lines include at least: constraint rules between vehicle pose vertices and / or lane lines at different times under the same trajectory; and / or constraint rules between vehicle pose vertices and / or lane lines with matching relationships under different trajectories; and / or constraint rules between vehicle pose vertices and lane lines; and / or constraint rules between lane lines under different vector segments.

[0030] Using the above method, when the target object is a vehicle pose vertex and / or lane line, the matching process corresponding to the vehicle pose vertex and / or lane line is constrained by the constraint rules corresponding to each vehicle pose vertex and / or lane line, thereby constraining the comprehensive matching error corresponding to the vehicle pose vertex and / or lane line.

[0031] Secondly, this application provides a data processing apparatus, the apparatus comprising:

[0032] The first calculation module is used to calculate the direction angle of each target object in the acquired map vector fragment, and the distance between each target object and the origin of the preset coordinate system.

[0033] The generation module is used to generate mathematical models for each target object based on various directional angles and distances.

[0034] The second calculation module is used to calculate the comprehensive matching error common to each target object based on each mathematical model.

[0035] The adjustment module is used to adjust the orientation angle of each target object in the preset coordinate system and the distance between each target object and the origin of the preset coordinate system when the overall matching error is greater than the preset error, until the overall matching error is less than or equal to the preset error.

[0036] In one possible design, the generation module is specifically used for:

[0037] Perform Euler rotation calculations on each direction angle to obtain the rotation matrix corresponding to each direction angle;

[0038] Based on each rotation matrix and each distance, a mathematical model corresponding to each target object is generated, wherein the mathematical model is the SE(3) model.

[0039] In one possible design, the second computing module is specifically used for:

[0040] Based on the mathematical model corresponding to each target object, the target objects are matched to obtain the first matching result for each target object;

[0041] Calculate the matching error for each first matching result;

[0042] Based on each matching error, the comprehensive matching error corresponding to each target object is calculated.

[0043] In one possible design, the adjustment module is specifically used for:

[0044] Determine whether the overall matching error is less than or equal to the preset error;

[0045] If so, then the comprehensive matching error is determined to meet the matching requirements between the target objects;

[0046] If not, adjust the orientation angle of each target object in the preset coordinate system and the distance between each target object and the origin of the preset coordinate system until the overall matching error is less than or equal to the preset error.

[0047] In one possible design, the second computing module is specifically used for:

[0048] Detect whether each target object is a vehicle pose vertex and / or lane line;

[0049] If so, obtain the constraint rules corresponding to each vehicle pose vertex and / or lane line, and calculate the comprehensive matching error corresponding to each vehicle pose vertex and / or lane line based on each constraint rule and each mathematical model.

[0050] The constraint rules corresponding to the vehicle pose vertices and / or lane lines include at least: constraint rules between vehicle pose vertices and / or lane lines at different times under the same trajectory; and / or constraint rules between vehicle pose vertices and / or lane lines with matching relationships under different trajectories; and / or constraint rules between vehicle pose vertices and lane lines; and / or constraint rules between lane lines under different vector segments.

[0051] Thirdly, this application provides an electronic device, comprising:

[0052] Memory, used to store computer programs;

[0053] When a processor executes a computer program stored in the memory, it implements the above-described data processing method steps.

[0054] Fourthly, this application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described data processing method steps.

[0055] Based on the data processing method described above, the mathematical model corresponding to the target object considers both the orientation angle and distance. This ensures that the matching process considers not only the overlap between target objects but also their directional consistency, thus improving the matching accuracy. Furthermore, the mathematical model unifies the dimensions of the orientation angle and distance, reducing the computational load during the matching process.

[0056] The technical effects of each of the second to fourth aspects mentioned above, as well as the technical effects that each aspect may achieve, are described above with reference to the technical effects that can be achieved for the first aspect or the various possible solutions in the first aspect, and will not be repeated here. Attached Figure Description

[0057] Figure 1 A flowchart of a data processing method provided in this application;

[0058] Figure 2 A schematic diagram of a map vector segment matching result provided in this application;

[0059] Figure 3 A schematic diagram of the structure of a data processing device provided in this application;

[0060] Figure 4This is a schematic diagram of an electronic device structure provided in this application. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The specific operational methods in the method embodiments can also be applied to the device embodiments or system embodiments. It should be noted that in the description of this application, "multiple" is understood as "at least two". "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. A connected to B can represent: A and B directly connected, and A and B connected through C. Furthermore, in the description of this application, terms such as "first" and "second" are used only for distinguishing the purpose of description and should not be construed as indicating or implying relative importance or order.

[0062] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0063] During map vector segment matching, if target 'a' in map vector segment 1 and target 'b' in map vector segment 2 are the same target object in the actual scene, the matching between them will not be completely identical due to the different acquisition devices used for them. In other words, there is a certain matching error between them. For example, if target 'a' and target 'b' are line segments corresponding to the boundary lines of the same lane, they will intersect in the matching result, rather than overlapping. In this case, a matching error can be considered to exist between target 'a' and target 'b'.

[0064] To reduce matching errors, it is necessary to optimize them, which involves the way matching errors are expressed. Currently, the way matching errors are expressed generally only considers distance errors without considering direction errors, or it considers both direction and distance errors. However, since direction and distance are expressed using different units, it increases the difficulty of optimizing matching errors, thus affecting the optimization effect.

[0065] To address the aforementioned problems, this application provides a data processing method that considers both orientation angle and distance when creating a mathematical model for target object pairs. This ensures that the matching process considers not only the overlap between target objects but also their directional consistency, thereby improving matching accuracy. Furthermore, the mathematical model unifies the dimensions of orientation angle and distance, reducing the computational load during the matching process. The methods and apparatus described in this application are based on the same technical concept. Since the principles by which the methods and apparatus solve the problems are similar, embodiments of the apparatus and methods can be referred to interchangeably, and repeated details will not be repeated.

[0066] like Figure 1 The diagram shown is a flowchart of a data processing method provided in this application, which specifically includes the following steps:

[0067] S11, acquire target objects in multiple map vector segments, and calculate the orientation angle of each target object in the preset coordinate system, as well as the distance between each target object and the origin of the preset coordinate system;

[0068] In this embodiment, during vehicle operation, pre-set onboard sensors such as GPS, cameras, and LiDAR (Light Detection and Ranging) collect environmental information around the vehicle, including road and building information. The collected environmental information is then locally fused to generate map vector fragments, which are transmitted to a cloud server. The cloud server then retrieves map vector fragments corresponding to different vehicles on the same road segment and identifies target objects within each fragment. These target objects include at least lane lines and vehicle pose vertices.

[0069] After acquiring each target object, the orientation angle of the target object in a preset coordinate system is further calculated. This preset coordinate system can be the world coordinate system, and the orientation angle is the angle between the target object and the preset coordinate system. When the target object is a line target, the orientation angle is the angle between the line target's direction vector and the preset coordinate system. If the preset coordinate system is a three-dimensional coordinate system including the X-axis, Y-axis, and Z-axis, then the orientation angle consists of angles in three dimensions: the angle between the line target's direction vector and the X-axis, the angle between the line target's direction vector and the Y-axis, and the angle between the line target and the Z-axis. Similarly, when the target object is a vehicle pose vertex, the orientation angle represents the vehicle's pose direction in the preset coordinate system, and the dimension of the orientation angle is the same as the dimension of the preset coordinate system.

[0070] After acquiring each target object, this embodiment of the application not only calculates the orientation angle of the target object in the preset coordinate system according to the above method, but also calculates the distance between the target object and the origin of the preset coordinate system. Specifically, when the target object is a line target, a preset point is first selected on the line target, and then the distance between the preset point and the origin of the preset coordinate system is used as the distance between the line target and the origin of the preset coordinate system. The preset point can be a single point, such as the midpoint of a line segment of the line target, or it can be multiple points, such as the endpoints of the line target.

[0071] Using the above method, the orientation angle of the line target object in the preset coordinate system, as well as the distance between the line target and the origin of the preset coordinate system, can be calculated.

[0072] S12, Generate mathematical models for each target object based on each direction angle and each distance;

[0073] After calculating the direction angle and distance corresponding to the target object, a mathematical model corresponding to the target object is further generated. In this embodiment, the specific method for generating the mathematical model may be: performing Euler rotation calculation on the direction angle to obtain the rotation matrix corresponding to the direction angle; and generating a mathematical model corresponding to each target object based on the rotation matrix and the distance. The mathematical model may be a Lie group SE(3) model, with the specific mathematical expression as follows:

[0074]

[0075] In formula (1), T is the target object, R is the rotation matrix corresponding to the direction angle, and t is the distance matrix between the target object and the origin of the preset coordinate system.

[0076] The mathematical model described above allows for the mathematical representation of target objects, considering not only their distances but also their corresponding orientation angles. Furthermore, it standardizes the units representing the orientation angles and distances, reducing the workload of subsequent matching calculations between target objects.

[0077] S13, Calculate the comprehensive matching error corresponding to each target object based on each mathematical model;

[0078] After calculating the mathematical model corresponding to each target object, the target objects are further matched according to the mathematical models to obtain the first matching result for each target object. The target matching algorithm can be the Fast Library for Approximate Nearest Neighbors (FLANN) algorithm, or the Random Sample Consensus (RANSAC) algorithm, etc. The specific target matching algorithm is not limited here. During the matching process, since the mathematical model corresponding to each target object considers both orientation and distance, when matching target objects, not only is the overlap between target objects considered through distance, but also the directional consistency between target objects is considered through orientation angle.

[0079] After calculating the first matching result for each target object, the matching error for each first matching result is calculated. The matching error can be calculated based on the similarity value of the first matching result. If the similarity value of the first matching result is 85%, then the matching error for the first result can be calculated as 100% - 85% = 15%. If the matching error for the first matching result is 0, it indicates that the two target objects corresponding to the first matching result completely overlap and are in the same direction.

[0080] Furthermore, based on each matching error, the comprehensive matching error corresponding to each target object is calculated. When calculating the comprehensive matching error, one can directly sum the individual matching error values ​​and normalize the result. Alternatively, one can first filter the matching results, deleting those with similarity values ​​below a preset threshold, then averaging the remaining matching errors and using this average as the comprehensive matching error.

[0081] For example, there are currently 5 matching results with similarity values ​​of 90%, 95%, 85%, 90%, and 0.1%. If the preset threshold is set to 5%, then the matching result with a similarity value of 0.1% can be considered an incorrect match. Although the target objects corresponding to this matching result are paired, it is an incorrect pairing. Therefore, the matching result with a similarity value of 0.1% is not considered. Then, the average of the matching errors corresponding to the remaining 4 matching results is calculated, and the average value is used as the comprehensive matching error.

[0082] Because the calculation process of the above-mentioned comprehensive matching error is based on the mathematical model corresponding to each target object, and these mathematical models take into account the orientation angle of each target object and the distance between each target object and the origin of the preset coordinate system, the calculation of the comprehensive matching error not only considers the overlap between target objects, but also the orientation consistency between target objects.

[0083] In this embodiment, the calculation of the comprehensive matching error corresponding to the target object is performed based on the mathematical model corresponding to the specific target object. Specifically, if each target object is detected as a vehicle pose vertex and / or lane line, the constraint rules corresponding to each vehicle pose vertex and / or lane line are obtained, and the comprehensive matching error corresponding to each vehicle pose vertex and / or lane line is calculated based on each constraint rule and each mathematical model. The constraint rules, used to constrain the calculation process of the comprehensive matching error, include at least: constraint rules between vehicle pose vertices and / or lane lines at different times under the same trajectory; and / or constraint rules between vehicle pose vertices and / or lane lines with matching relationships under different trajectories; and / or constraint rules between vehicle pose vertices and lane lines; and / or constraint rules between lane lines under different vector segments.

[0084] For example, such as Figure 2 The image shows the matching results for each target object. This represents the j-th pose vertex of the i-th trajectory. This represents the j-th trajectory segment of the i-th trajectory, where the two objects connected by the dashed line belong to the same matching result. For example, and This represents the matching results for a pair of lane lines under different trajectories; the rest are similar. and This represents a matching result between vehicle pose vertices under different trajectories; the rest are similar.

[0085] Depend on Figure 2 It can be seen that the matching process between vehicle pose vertices includes, but is not limited to, the following four constraint rules: (1) Constraint rules between front and rear pose vertices under the same trajectory. These constraint rules can be represented by Lie algebra distance, such as and (2) Constraint rules between vehicle pose vertices matched under different trajectories. These constraint rules can be represented by Lie algebraic distance, such as... and (3) Constraint rules between vehicle pose vertices and line targets. These constraints can be represented by Lie algebraic distance, such as... and (4) Constraints between lane lines under different vector segments, which can be represented by Lie algebraic distance, such as and The rules governing the constraints between them.

[0086] By applying the aforementioned constraint rules, the calculation process of the matching error between vehicle pose vertices and / or lane lines is constrained, ensuring that the matching accuracy between vehicle pose vertices and / or lane lines meets the matching requirements. Specifically, the constraint rules constrain the matching process between vehicle pose vertices and / or lane lines, thereby affecting the similarity value between them and achieving the effect of constraining the generation of matching errors. After obtaining the individual matching errors corresponding to each vehicle pose vertex and / or lane line, a comprehensive matching error corresponding to each vehicle pose vertex and / or lane line is further generated.

[0087] S14. When the overall matching error is greater than the preset error, adjust the orientation angle of each target object in the preset coordinate system and the distance between each target object and the origin of the preset coordinate system until the overall matching error is less than or equal to the preset error.

[0088] After calculating the comprehensive matching error for each target object, it is determined whether the comprehensive matching error is less than or equal to the preset error. The preset error is set according to the actual situation. If high accuracy is required for map vector segment matching, the preset error can be set to a relatively small value; otherwise, the value of the preset error can be increased. When the comprehensive matching error is less than or equal to the preset error, it is determined that the comprehensive matching error meets the matching requirements between target objects. That is, the overlap and directional consistency between the current target objects meet the matching requirements.

[0089] When the overall matching error exceeds the preset error, it indicates that the matching accuracy between the current map vector segments does not meet the requirements, and further optimization of the matching results is needed. Specifically, first, adjust the orientation angle of each target object in the preset coordinate system, as well as the distance between each target object and the origin of the preset coordinate system. For example, if there are three matching results: Matching result 1: (A, B), Matching result 2: (C, D), and Matching result 3: (E, F), and the overall matching error for these three matching results is 1%, while the preset error is 0.5%, then the current matching accuracy is considered unsatisfactory. Therefore, it is necessary to adjust the orientation angle and distance for each target object A, B, C, D, and E.

[0090] After adjusting the various orientation angles and distances, a new mathematical model is generated for each target object. Based on the regenerated mathematical model, the target objects are matched to obtain second matching results. The overall matching degree of each target object is then judged based on the comprehensive matching error of each second matching result. If yes, it indicates that the matching accuracy of the current second matching results meets the requirements. Otherwise, the orientation angles of each target object in the preset coordinate system and the distances between each target object and the origin of the preset coordinate system are adjusted until the comprehensive matching error is less than or equal to the preset error.

[0091] Based on the data processing method described above, the mathematical model corresponding to the target object considers both the orientation angle and distance. This ensures that the matching process considers not only the overlap between target objects but also their directional consistency, thus improving the matching accuracy. Furthermore, the mathematical model unifies the dimensions of the orientation angle and distance, reducing the computational load during the matching process.

[0092] Based on the same inventive concept, this application also provides a data processing device, such as... Figure 3 The diagram shown is a structural schematic of a data processing device, including:

[0093] The first calculation module 31 is used to calculate the direction angle of each target object in the acquired map vector fragment, and the distance between each target object and the origin of the preset coordinate system.

[0094] The generation module 32 is used to generate mathematical models corresponding to each target object based on each direction angle and each distance;

[0095] The second calculation module 33 is used to calculate the comprehensive matching error corresponding to each target object according to each mathematical model.

[0096] The adjustment module 34 is used to adjust the orientation angle of each target object in the preset coordinate system and the distance between each target object and the origin of the preset coordinate system when the overall matching error is greater than the preset error, until the overall matching error is less than or equal to the preset error.

[0097] In one possible design, the generation module is specifically used for:

[0098] Perform Euler rotation calculations on each direction angle to obtain the rotation matrix corresponding to each direction angle;

[0099] Based on each rotation matrix and each distance, a mathematical model corresponding to each target object is generated, wherein the mathematical model is the SE(3) model.

[0100] In one possible design, the second computing module is specifically used for:

[0101] Based on the mathematical model corresponding to each target object, the target objects are matched to obtain the first matching result for each target object;

[0102] Calculate the matching error for each first matching result;

[0103] Based on each matching error, the comprehensive matching error corresponding to each target object is calculated.

[0104] In one possible design, the adjustment module is specifically used for:

[0105] Determine whether the overall matching error is less than or equal to the preset error;

[0106] If so, then the comprehensive matching error is determined to meet the matching requirements between the target objects;

[0107] If not, adjust the orientation angle of each target object in the preset coordinate system and the distance between each target object and the origin of the preset coordinate system until the overall matching error is less than or equal to the preset error.

[0108] In one possible design, the second computing module is specifically used for:

[0109] Detect whether each target object is a vehicle pose vertex and / or lane line;

[0110] If so, obtain the constraint rules corresponding to each vehicle pose vertex and / or lane line, and calculate the comprehensive matching error corresponding to each vehicle pose vertex and / or lane line based on each constraint rule and each mathematical model.

[0111] The constraint rules corresponding to the vehicle pose vertices and / or lane lines include at least: constraint rules between vehicle pose vertices and / or lane lines at different times under the same trajectory; and / or constraint rules between vehicle pose vertices and / or lane lines with matching relationships under different trajectories; and / or constraint rules between vehicle pose vertices and lane lines; and / or constraint rules between lane lines under different vector segments.

[0112] Based on the aforementioned data processing device, since the mathematical model corresponding to the target object considers both the orientation angle and distance, the matching process not only takes into account the overlap between target objects but also their directional consistency, thus improving the matching accuracy. Furthermore, the mathematical model unifies the dimensions of the orientation angle and distance, reducing the computational load during the matching process.

[0113] Based on the same inventive concept, this application also provides an electronic device that can realize the functions of the aforementioned data processing method apparatus. (Refer to...) Figure 4 The electronic device includes:

[0114] At least one processor 41 and a memory 42 connected to the at least one processor 41. In this embodiment, the specific connection medium between the processor 41 and the memory 42 is not limited. Figure 4 The example shown is the connection between processor 41 and memory 42 via bus 40. Bus 40 is... Figure 4 The connections between other components are shown in thick lines only and are not intended to be limiting. Bus 40 can be divided into address bus, data bus, control bus, etc., for ease of representation. Figure 4 The term 41 is represented by a single thick line, but this does not imply that there is only one bus or one type of bus. Alternatively, processor 41 can also be called controller; there is no restriction on the name.

[0115] In this embodiment, memory 42 stores instructions executable by at least one processor 41. By executing the instructions stored in memory 42, at least one processor 41 can perform the data processing method described above. Processor 41 can implement... Figure 3 The functions of each module in the device shown.

[0116] The processor 41 is the control center of the device. It can connect to various parts of the control device through various interfaces and lines. By running or executing instructions stored in memory 42 and calling data stored in memory 42, the processor can perform various functions and process data, thereby monitoring the device as a whole.

[0117] In one possible design, processor 41 may include one or more processing units. Processor 41 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may also not be integrated into processor 41. In some embodiments, processor 41 and memory 42 may be implemented on the same chip; in some embodiments, they may also be implemented on separate chips.

[0118] Processor 41 can be a general-purpose processor, such as a central processing unit (CPU), digital signal processor, application-specific integrated circuit, field-programmable gate array or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the data processing method disclosed in the embodiments of this application can be directly manifested as being executed by a hardware processor, or executed by a combination of hardware and software modules within the processor.

[0119] Memory 42, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. Memory 42 may include at least one type of storage medium, such as flash memory, hard disk, multimedia card, card-type memory, random access memory (RAM), static random access memory (SRAM), programmable read-only memory (PROM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), magnetic memory, magnetic disk, optical disk, etc. Memory 42 can be any other medium capable of carrying or storing desired program code in the form of instructions or data structures that can be accessed by a computer, but is not limited thereto. In the embodiments of this application, memory 42 may also be a circuit or any other device capable of implementing storage functions for storing program instructions and / or data.

[0120] By designing and programming the processor 41, the code corresponding to the data processing method described in the foregoing embodiments can be embedded into the chip, enabling the chip to execute the code during operation. Figure 1 The steps of the data processing method in the illustrated embodiment are as follows. How to design and program the processor 41 is a technique well-known to those skilled in the art and will not be described further here.

[0121] Based on the same inventive concept, embodiments of this application also provide a storage medium storing computer instructions that, when executed on a computer, cause the computer to perform the data processing method described above.

[0122] In some possible implementations, various aspects of the data processing method provided in this application may also be implemented as a program product comprising program code that, when the program product is run on a device, causes the control device to perform the steps of the data processing method according to the various exemplary embodiments of this application described above.

[0123] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0124] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0125] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0126] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0127] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A data processing method, characterized in that, The method includes: Obtain target objects from multiple map vector segments, and calculate the orientation angle of each target object in a preset coordinate system, as well as the distance between each target object and the origin of the preset coordinate system; Based on each directional angle and each distance, generate a mathematical model corresponding to each target object; Based on each mathematical model, calculate the comprehensive matching error common to each target object; When the overall matching error is greater than the preset error, adjust the orientation angle of each target object in the preset coordinate system and the distance between each target object and the origin of the preset coordinate system until the overall matching error is less than or equal to the preset error. The step of calculating the comprehensive matching error common to each target object based on each mathematical model includes: Based on the mathematical models corresponding to each target object, the target objects are matched to obtain the first matching result for each target object; the matching error corresponding to each first matching result is calculated; based on the matching errors, the comprehensive matching error corresponding to all target objects is calculated; or Detect whether each target object is a vehicle pose vertex and / or lane line; if so, obtain the constraint rules corresponding to each vehicle pose vertex and / or lane line respectively, and calculate the comprehensive matching error corresponding to each vehicle pose vertex and / or lane line based on each constraint rule and each mathematical model; wherein, the constraint rules corresponding to the vehicle pose vertex and / or lane line include at least the constraint rules between vehicle pose vertices and / or lane lines at different times under the same trajectory; and / or the constraint rules between vehicle pose vertices and / or lane lines with matching relationship under different trajectories; and / or the constraint rules between vehicle pose vertices and lane lines; and / or the constraint rules between lane lines under different vector segments.

2. The method as described in claim 1, characterized in that, The process of generating mathematical models for each target object based on various directional angles and distances includes: Perform Euler rotation calculations on each direction angle to obtain the rotation matrix corresponding to each direction angle; Based on each rotation matrix and each distance, a mathematical model corresponding to each target object is generated, wherein the mathematical model is the SE(3) model.

3. The method as described in claim 1, characterized in that, When the overall matching error is greater than a preset error, adjusting the orientation angle of each target object in the preset coordinate system and the distance between each target object and the origin of the preset coordinate system until the overall matching error is less than or equal to the preset error includes: Determine whether the overall matching error is less than or equal to the preset error; If so, then the comprehensive matching error is determined to meet the matching requirements between the target objects; If not, adjust the orientation angle of each target object in the preset coordinate system and the distance between each target object and the origin of the preset coordinate system until the overall matching error is less than or equal to the preset error.

4. A data processing apparatus, characterized in that, The device includes: The first calculation module is used to acquire target objects in multiple map vector segments, and calculate the orientation angle of each target object in a preset coordinate system, as well as the distance between each target object and the origin of the preset coordinate system. The generation module is used to generate mathematical models for each target object based on various directional angles and distances. The second calculation module is used to calculate the comprehensive matching error common to each target object based on each mathematical model. The adjustment module is used to adjust the orientation angle of each target object in the preset coordinate system and the distance between each target object and the origin of the preset coordinate system when the overall matching error is greater than the preset error, until the overall matching error is less than or equal to the preset error; Specifically, the second calculation module is used for: Based on the mathematical models corresponding to each target object, the target objects are matched to obtain the first matching result for each target object; the matching error corresponding to each first matching result is calculated; based on the matching errors, the comprehensive matching error corresponding to all target objects is calculated; or Detect whether each target object is a vehicle pose vertex and / or lane line; if so, obtain the constraint rules corresponding to each vehicle pose vertex and / or lane line respectively, and calculate the comprehensive matching error corresponding to each vehicle pose vertex and / or lane line based on each constraint rule and each mathematical model; wherein, the constraint rules corresponding to the vehicle pose vertex and / or lane line include at least the constraint rules between vehicle pose vertices and / or lane lines at different times under the same trajectory; and / or the constraint rules between vehicle pose vertices and / or lane lines with matching relationship under different trajectories; and / or the constraint rules between vehicle pose vertices and lane lines; and / or the constraint rules between lane lines under different vector segments.

5. The apparatus as described in claim 4, characterized in that, The generation module is specifically used for: Perform Euler rotation calculations on each direction angle to obtain the rotation matrix corresponding to each direction angle; Based on each rotation matrix and each distance, a mathematical model corresponding to each target object is generated, wherein the mathematical model is the SE(3) model.

6. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, when executing a computer program stored in the memory, implements the method steps of any one of claims 1-3.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the method steps of any one of claims 1-3.

Images