A path guide line generation method, device, vehicle, medium and equipment
By acquiring and recording the feature data of road trajectory and centerline, binding and training a deep learning model, the problem of low efficiency in generating path guidance lines is solved, and efficient generation of path guidance lines that conform to human driving experience is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- MOMENTA (SUZHOU) TECHNOLOGY CO LTD
- Filing Date
- 2022-03-23
- Publication Date
- 2026-07-31
AI Technical Summary
Existing technologies suffer from low efficiency in generating path guide lines and inefficient acquisition of related data, as well as cumbersome and time-consuming neural network training processes.
By acquiring the trajectory and centerline points of the road, using the trajectory database to record relevant feature data, binding the trajectory points with the center point, calculating the relative positional relationship, training a deep learning model, and generating path guidance lines.
It improves data extraction efficiency, enabling deep learning models to better learn from human driving experience and generate path guidance lines that conform to human driving experience, thus improving generation efficiency.
Smart Images
Figure CN116860810B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of data processing, and in particular to a method, apparatus, vehicle, medium and equipment for generating path guidance lines. Background Technology
[0002] In general map navigation, guide lines are generated to guide drivers. These guide lines are generated using neural networks. However, these networks require a large amount of data to generate the guide lines. Obtaining this large amount of data depends on a large number of trajectory points and road center points, and the calculation of related data for each data point is quite tedious. When inputting the relevant data into the neural network for training, each data point needs to be recorded and input one by one, which is very labor-intensive and time-consuming. Summary of the Invention
[0003] To address the issues of low efficiency in generating path guide lines and inefficient acquisition of related data, this application mainly provides a method, apparatus, vehicle, medium, and equipment for generating path guide lines.
[0004] In a first aspect, embodiments of this application provide a method for generating path guide lines, which includes:
[0005] Obtain the first number of trajectories in a road segment and the first road centerline in the base map for the road segment. Obtain the second preset number of trajectory points on each trajectory and the third preset number of first road center points on the first road centerline. Record the directional position information of the trajectory points and the directional position information of the first road center points in the trajectory database.
[0006] Bind the trajectory points in each trajectory to the center point of the first road to obtain the nearest neighbor trajectory points associated with each center point of the first road. Mark the corresponding trajectory point ID of each nearest neighbor trajectory point and record the nearest neighbor trajectory point and its trajectory point ID in the trajectory database.
[0007] Using the directional position information of the nearest neighbor trajectory points and the directional position information of the center point of the first road, the relative positional relationship of the nearest neighbor trajectory points with respect to the center line of the first road is obtained and recorded in the trajectory database;
[0008] By utilizing the relative positional relationships of nearest neighbor trajectory points in the trajectory database with respect to the center line of the first road, a deep learning model is trained to obtain a guide line fitting model.
[0009] The location information of the center line of the second road in the map is input into the guide line fitting model. The predicted associated trajectory points corresponding to the center line of the second road are used as path guide points, and the path guide points are fitted to generate the path guide line output.
[0010] Secondly, embodiments of this application provide a path guide line generation device, which includes:
[0011] The acquisition module is used to acquire a first number of trajectories in a road segment and a first road centerline in the base map for the road segment, respectively obtain a second preset number of trajectory points on each trajectory and a third preset number of first road center points on the first road centerline, and record the directional position information of the trajectory points and the directional position information of the first road center points into the trajectory database.
[0012] The binding module is used to bind the trajectory points in each trajectory to the center point of the first road, obtain the nearest neighbor trajectory points associated with each center point of the first road, mark the corresponding trajectory point ID of each nearest neighbor trajectory point, and record the nearest neighbor trajectory points and their trajectory point IDs in the trajectory database.
[0013] The relevant feature calculation module is used to obtain the relative positional relationship of the nearest neighbor trajectory point with respect to the center line of the first road by using the directional positional information of the nearest neighbor trajectory point and the directional positional information of the center point of the first road, and record it in the trajectory database.
[0014] The training module is used to train the deep learning model by utilizing the relative positional relationship of the nearest neighbor trajectory points in the trajectory database with respect to the center line of the first road, so as to obtain the guide line fitting model.
[0015] The guide line generation module is used to input the location information of the center line of the second road in the map into the guide line fitting model, use the predicted associated trajectory points corresponding to the center line of the second road as path guide points, and fit the path guide points to generate path guide lines as output.
[0016] Thirdly, embodiments of this application provide a vehicle, wherein the vehicle includes the path guidance line generation device of Scheme 2.
[0017] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer instructions, which, when executed, cause the computer to perform the path guide line generation method in Scheme 1.
[0018] Fifthly, embodiments of this application provide a computer device including a processor and a memory, the memory storing computer instructions, which, when executed by the processor, implement the path guide line generation method in Scheme 1.
[0019] The technical solution of this application embodiment calculates and obtains a large number of actual trajectories and the relevant features between the center point of the first road on the center line of the first road in the map, and records them in the trajectory database. This makes it easier to directly extract a large amount of relevant feature data from the trajectory database, saving data extraction efficiency. By training the deep learning model with a large amount of relevant feature data in the trajectory database, the deep learning model can learn the experience trajectory of human driving well and fit a route that is more in line with human driving experience. Attached Figure Description
[0020] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description exemplarily illustrate some embodiments of this application.
[0021] Figure 1 This is a schematic diagram illustrating a specific implementation of a path guide line generation method according to this application;
[0022] Figure 2 This is a schematic diagram of a specific embodiment of a path guide line generation device according to this application.
[0023] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation
[0024] The preferred embodiments of this application will now be described in detail with reference to the accompanying drawings, so that the advantages and features of this application can be more easily understood by those skilled in the art, thereby providing a clearer and more definite definition of the scope of protection of this application.
[0025] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.
[0026] The proposed method for generating path guidance lines by extracting relevant feature data from a trajectory database can be applied to scenarios such as highways and urban roads.
[0027] In this application, the base map refers to the layer placed at the bottom of multiple layers during the map drawing process, which constitutes the basic framework of the base map. The base map contains many geographical features that identify geographical locations, such as transportation, roads, landforms, etc.
[0028] In this application, the first road center point refers to the one existing in the base map, and the second road center point refers to the one existing in the map. Both are used to mark a series of location points of the road. The road center point is the location point on the center line of each road.
[0029] This application obtains trajectory points in each trajectory of a road segment and the first road center point on the first road centerline of that road segment in the base map. The number of trajectory points or first road center points is determined according to the sampling frequency and the length of the trajectory points or the first road centerline. The information corresponding to each trajectory point and the first road center point is recorded in the trajectory database. The process of binding trajectory points in each trajectory with the first road center point includes using the location information of the trajectory points and the location information of the first road center point to bind the coordinates of the trajectory points with the coordinates of the first road center point, obtaining the nearest neighbor trajectory point that is most relevant and closest to each first road center point, marking the corresponding trajectory point ID of each nearest neighbor trajectory point so that the nearest neighbor trajectory point related to the first road centerline can be queried and recorded in the trajectory database. Based on the corresponding information of the nearest neighbor trajectory points and the first road center point, the process is further refined. The process of obtaining the relative positional relationship of the nearest trajectory points with respect to the centerline of the first road using the directional positional information of the nearest trajectory points and the centerline of the first road includes: calculating the azimuth deflection angle and the nearest distance between the centerline of the first road and its corresponding nearest trajectory points using the directional positional information of the nearest trajectory points and the centerline of the first road, and recording them in the trajectory database. The azimuth deflection angle is obtained based on the azimuth information, and the nearest distance is obtained based on the positional information. The projected distance and the first lateral offset of the nearest trajectory point on the centerline of the first road are calculated using the azimuth deflection angle and the nearest distance, and recorded as relevant feature data in the trajectory database. The relevant feature data, namely the projected distance and the first lateral offset, are directly extracted from the trajectory database to train a deep learning model, resulting in a guide line fitting model. Directly extracting a large amount of relevant feature data from the trajectory database can improve the efficiency of data extraction. The guide line fitting model predicts the second lateral offset from the centerline of the second road based on the centerline of the second road in the map. Based on the second lateral offset, the corresponding predicted associated trajectory points are obtained as path guide points, and the path guide points are fitted to generate a path guide line output. By training a deep learning model with a large amount of relevant feature data in the trajectory database, the deep learning model can learn human driving experience trajectories well and fit routes that are more in line with human driving experience.
[0030] The technical solutions of this application and how they solve the aforementioned technical problems will be described in detail below with specific embodiments. The specific embodiments described below can be combined with each other to form new embodiments. The same or similar ideas or processes described in one embodiment may not be repeated in other embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.
[0031] Figure 1 This illustration shows a specific implementation of a path guide line generation method according to this application.
[0032] exist Figure 1 In the specific implementation shown, a method for generating path guide lines mainly includes:
[0033] Step S101: Obtain a first number of trajectories in a road segment and a first road centerline in the base map for the road segment. Obtain a second preset number of trajectory points on each trajectory and a third preset number of first road centerlines on the first road centerline. Record the directional position information of the trajectory points and the directional position information of the first road centerlines in the trajectory database.
[0034] In this embodiment, each frame of the image pre-records the directional and positional information corresponding to each trajectory point. The positional and orientational information of each trajectory point can be directly obtained. The first quantity is determined based on the specific actual situation; the actual trajectory in a road segment is statistically determined. The base image also records the relevant information value of the first road center point on the road centerline to facilitate subsequent calculation of feature data.
[0035] In one specific embodiment of this application, the corresponding second preset quantity and third preset quantity are determined according to the length of the trajectory and the length of the center line of the first road, respectively.
[0036] In this embodiment, a road segment will have multiple trajectories, each composed of several trajectory points. The second number is determined based on the length of the trajectory and the sampling frequency of the trajectory points. The centerline of a road segment also has several first road center points, the third number determined based on the length of the road and the sampling frequency of the center points.
[0037] exist Figure 1 In the specific embodiment shown, a path guide line generation method further includes:
[0038] Step S102: Bind the trajectory points in each trajectory to the center point of the first road to obtain the nearest neighbor trajectory points associated with each center point of the first road. Mark the corresponding trajectory point ID of each nearest neighbor trajectory point and record the nearest neighbor trajectory point and its trajectory point ID in the trajectory database.
[0039] In this embodiment, when there are many trajectory passes, there will be many trajectory points associated with each first road center point. Based on the multiple trajectory point IDs recorded in the trajectory database, the nearest neighbor trajectory point associated with each first road center point can be queried to determine the relationship between the trajectory point and the first road center point.
[0040] In one specific embodiment of this application, the trajectory points in each trajectory are bound to the center point of the first road to obtain the nearest neighbor trajectory point associated with each first road center point. The method further includes: using the directional position information of the trajectory points and the directional position information of the first road center point, binding the coordinates of the trajectory points to the coordinates of the first road center point respectively to obtain the nearest neighbor trajectory point that is most relevant to and closest to each first road center point.
[0041] In this embodiment, the process of binding trajectory points and road center points is a process of point coordinate binding, that is, finding the closest trajectory point associated with each first road center point, so as to provide a basis for calculating the relevant feature data between trajectory points and first road center points.
[0042] exist Figure 1 In the specific embodiment shown, a path guide line generation method further includes:
[0043] Step S103: Using the directional position information of the nearest neighbor trajectory point and the directional position information of the center point of the first road, the relative positional relationship of the nearest neighbor trajectory point with respect to the center line of the first road is obtained and recorded in the trajectory database.
[0044] In this embodiment, the relative positional relationship between the nearest neighbor trajectory point and the center point of the first road is recorded in the trajectory database, which facilitates data extraction and calculation in the future, saves time, and improves efficiency.
[0045] In one specific embodiment of this application, the relative positional relationship of the nearest neighbor trajectory point with respect to the centerline of the first road is obtained using the directional positional information of the nearest neighbor trajectory point and the directional positional information of the first road center point. This includes: calculating the angle between the first road center point and its corresponding nearest neighbor trajectory point using the azimuth information of the nearest neighbor trajectory point and the azimuth information of the first road center point to obtain the azimuth deflection angle in the relative positional relationship; and calculating the distance between the first road center point and its corresponding nearest neighbor trajectory point using the position information of the nearest neighbor trajectory point and the position information of the first road center point to obtain the shortest distance in the relative positional relationship.
[0046] In this embodiment, the azimuth information of the nearest neighbor trajectory point records its azimuth angle, and the azimuth information of the first road center point also records its azimuth angle. Therefore, the angle between the line connecting the nearest neighbor trajectory point and the first road center point and the centerline of the first road can be obtained, i.e., the azimuth deflection angle. The position information of the nearest neighbor trajectory point records its position coordinates, and the position information of the first road center point also records its position coordinates. Therefore, the distance between the nearest neighbor trajectory point and the first road center point can be calculated, i.e., the closest distance. The azimuth deflection angle and the closest distance in the relative positional relationship provide the basis for calculating the projection distance and the first lateral offset.
[0047] In one specific embodiment of this application, the relative positional relationship of the nearest neighbor trajectory point with respect to the center line of the first road is obtained by using the directional positional information of the nearest neighbor trajectory point and the directional positional information of the center point of the first road. This includes: using the azimuth deflection angle and according to the cosine theorem of the nearest distance, obtaining the projected distance in the relative positional relationship of the nearest neighbor trajectory point with respect to the center line of the first road; and using the azimuth deflection angle and according to the sine theorem of the nearest distance, obtaining the first lateral offset in the relative positional relationship of the nearest neighbor trajectory point with respect to the center line of the first road.
[0048] In this embodiment, the projection distance of the nearest neighbor trajectory point onto the centerline of the first road and the first lateral offset between the first road center point and its nearest neighbor trajectory point are obtained using the azimuth deflection angle and the nearest distance. These values are recorded in the trajectory database, facilitating subsequent feature data extraction, saving time, and improving efficiency. The product of the nearest distance and the cosine of the azimuth deflection angle is the projection distance, and the product of the nearest distance and the sine of the azimuth deflection angle is the first lateral offset. The calculation is simple, convenient, and efficient.
[0049] exist Figure 1 In the specific embodiment shown, a path guide line generation method further includes:
[0050] Step S104: Using the relative positional relationship between the nearest neighbor trajectory points in the trajectory database and the center line of the first road, the deep learning model is trained to obtain the guide line fitting model.
[0051] In this embodiment, a large amount of relevant feature data, namely a large amount of projection distance and first lateral offset, is directly extracted from the trajectory database to train the deep learning model, so that the deep learning model can learn the experience trajectory of human driving well and fit a route that is more in line with human driving experience.
[0052] exist Figure 1In the specific embodiment shown, a path guide line generation method further includes:
[0053] Step S105: Input the location information of the center line of the second road in the map into the guide line fitting model, take the predicted associated trajectory points corresponding to the center line of the second road as path guide points, and fit the path guide points to generate path guide lines.
[0054] In this embodiment, the deep learning model trained with relevant feature data from the trajectory database can better fit routes that conform to human driving experience and improve the efficiency of guide line fitting.
[0055] In one specific embodiment of this application, the location information of the center line of the second road in the map is input into the guide line fitting model, and the predicted associated trajectory point corresponding to the center line of the second road is used as the path guide point. The process includes: the guide line fitting model predicts a second lateral offset that deviates from the center point of the second road in the center line of the second road based on the location information of the center line of the second road in the map; the predicted associated trajectory point is obtained based on the second lateral offset, and it is used as the path guide point.
[0056] In this embodiment, the guide line fitting model predicts a second lateral offset from the center point of the second road centerline based on the centerline of the second road in the map. The corresponding predicted trajectory points are then obtained as path guidance points, and these points are fitted to generate the path guidance line output. This process fits a route that better reflects human driving experience and improves the efficiency of guide line fitting.
[0057] This application calculates and obtains a large number of actual trajectories and the relevant features between the center point of the first road on the center line of the first road in the map, and records them in the trajectory database. This makes it easier to directly extract a large amount of relevant feature data from the trajectory database, saving data extraction time and improving the efficiency of guide line generation. By training a deep learning model with a large amount of relevant feature data in the trajectory database, the deep learning model can learn human driving experience trajectories well and fit a route that is more in line with human driving experience.
[0058] Figure 2 This paper illustrates a specific embodiment of a path guide line generation device according to the present application.
[0059] exist Figure 2 In the specific embodiment shown, a path guide line generation device mainly includes:
[0060] The acquisition module 201 is used to acquire a first number of trajectories in a road segment and a first road centerline in the base map about the road segment, respectively obtain a second preset number of trajectory points on each trajectory and a third preset number of first road center points on the first road centerline, and record the directional position information of the trajectory points and the directional position information of the first road center points into the trajectory database.
[0061] The binding module 202 is used to bind the trajectory points in each trajectory to the center point of the first road, obtain the nearest neighbor trajectory points associated with each center point of the first road, mark the corresponding trajectory point ID of each nearest neighbor trajectory point, and record the nearest neighbor trajectory points and their trajectory point IDs in the trajectory database.
[0062] The relevant feature calculation module 203 is used to obtain the relative positional relationship of the nearest neighbor trajectory point with respect to the center line of the first road by using the directional positional information of the nearest neighbor trajectory point and the directional positional information of the center point of the first road, and record it in the trajectory database.
[0063] Training module 204 is used to train a deep learning model by utilizing the relative positional relationship of the nearest neighbor trajectory points in the trajectory database with respect to the center line of the first road, thereby obtaining a guide line fitting model.
[0064] The guide line generation module 205 is used to input the location information of the center line of the second road in the map into the guide line fitting model, use the predicted associated trajectory points corresponding to the center line of the second road as path guide points, and fit the path guide points to generate path guide lines.
[0065] In this embodiment, the acquisition module 201 can acquire the trajectory points in each trajectory of a road segment and the first road center point on the first road centerline of the road segment in the base map according to the sampling frequency, and record the direction and position information corresponding to the trajectory points and the first road center point into the trajectory database; the binding module 202 binds the coordinates of the trajectory points to the coordinates of the first road center point by using the positions of the trajectory points and the first road center point, finds the nearest neighbor trajectory point that is most relevant to each first road center point and is closest to it, and marks the corresponding trajectory point ID of each nearest neighbor trajectory point so that the nearest neighbor trajectory point related to the first road centerline can be queried and recorded in the trajectory database; the relevant feature calculation module 203 calculates the orientation and position in the direction and position information of the nearest neighbor trajectory point and the direction and position information of the first road center point. The system calculates the azimuth and position of the first road center point and its nearest neighbor trajectory point, recording these values in the trajectory database. It then calculates the projection distance and first lateral offset of the nearest neighbor trajectory point on the first road centerline using the azimuth and position, recording these as relevant feature data in the trajectory database. The training module 204 directly extracts these relevant feature data (projection distance and first lateral offset) from the trajectory database to train the deep learning model, resulting in a guide line fitting model. Directly extracting a large amount of relevant feature data from the trajectory database improves data extraction efficiency. Furthermore, training the deep learning model with this data allows it to better learn human driving experience trajectories and fit routes that better align with human driving experience. The guide line generation module 205 uses the guide line fitting model and, based on the second road centerline in the map, predicts the second lateral offset from the second road center point. Based on this second lateral offset, it obtains the corresponding guide point, fits the guide point to generate the path guide line, and outputs it, thus improving the efficiency of guide line fitting.
[0066] The path guide line generation device provided in this application can be used to execute the path guide line generation method described in any of the above embodiments. Its implementation principle and technical effect are similar, and will not be repeated here.
[0067] In one specific embodiment of this application, the functional modules in the path guidance line generation device of this application can be directly in hardware, in software modules executed by a processor, or in a combination of both.
[0068] Software modules may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium known in this art. An exemplary storage medium is coupled to the processor, enabling the processor to read information from and write information to the storage medium.
[0069] The processor 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, discrete hardware components, or any combination thereof. A general-purpose processor can be a microprocessor, but alternatively, it can be any conventional processor, controller, microcontroller, or state machine. The processor can also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors incorporating a DSP core, or any other such configuration. Alternatively, the storage medium can be integrated with the processor. The processor and storage medium can reside in an ASIC. The ASIC can reside in the user terminal. Alternatively, the processor and storage medium can reside as discrete components in the user terminal.
[0070] In another specific embodiment of this application, a vehicle is provided, wherein the vehicle includes the path guidance line generation device of any embodiment. Optionally, the vehicle includes a processor and a memory, the processor and the memory being coupled, the vehicle being used to implement the appendix to this application specification. Figure 1 The path guide line generation method in any of the embodiments shown.
[0071] In another specific embodiment of this application, a computer-readable storage medium is provided, which stores computer instructions that, when executed, cause a computer to perform the path guide line generation method in any embodiment.
[0072] In another specific embodiment of this application, a computer device includes a processor and a memory, the memory storing computer instructions that, when executed by the processor, implement the path guide line generation method in any embodiment.
[0073] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.
[0074] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0075] The above description is merely an embodiment of this application and does not limit the patent scope of this application. Any equivalent structural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.
Claims
1. A method for generating path guide lines, characterized in that, include: Obtain a first number of trajectories in a road segment and a first road centerline in the base map for the road segment. Obtain a second preset number of trajectory points on each trajectory and a third preset number of first road center points on the first road centerline. Record the directional position information of the trajectory points and the directional position information of the first road center points in the trajectory database. The trajectory points in each trajectory are bound to the center point of the first road to obtain the nearest neighbor trajectory points associated with each center point of the first road. Each nearest neighbor trajectory point is marked with its corresponding trajectory point ID, and the nearest neighbor trajectory point and its trajectory point ID are recorded in the trajectory database. Using the directional position information of the nearest neighbor trajectory point and the directional position information of the first road center point, the relative positional relationship of the nearest neighbor trajectory point with respect to the center line of the first road is obtained and recorded in the trajectory database; Using the relative positional relationship between the nearest neighbor trajectory points in the trajectory database and the center line of the first road, the deep learning model is trained to obtain the guide line fitting model; The location information of the centerline of the second road in the map is input into the guide line fitting model. The predicted associated trajectory points corresponding to the centerline of the second road are used as path guide points, and the path guide points are fitted to generate a path guide line output. The guide line fitting model predicts a second lateral offset from the center point of the second road in the centerline of the second road based on the centerline of the second road in the map. The predicted associated trajectory points corresponding to the second lateral offset are used as path guide points, and the path guide points are fitted to generate a path guide line output.
2. The path guide line generation method as described in claim 1, characterized in that, The step of obtaining the relative positional relationship of the nearest neighbor trajectory point with respect to the centerline of the first road using the directional positional information of the nearest neighbor trajectory point and the directional positional information of the center point of the first road includes: Using the azimuth information in the directional position information of the nearest neighbor trajectory point and the azimuth information in the directional position information of the first road center point, the angle between the first road center point and its corresponding nearest neighbor trajectory point is calculated to obtain the azimuth deflection angle in the relative position relationship; Using the positions in the directional position information of the nearest neighbor trajectory point and the positions in the directional position information of the first road center point, the distance between the first road center point and its corresponding nearest neighbor trajectory point is calculated to obtain the closest distance in the relative position relationship.
3. The path guide line generation method as described in claim 2, characterized in that, The step of obtaining the relative positional relationship of the nearest neighbor trajectory point with respect to the centerline of the first road using the directional positional information of the nearest neighbor trajectory point and the directional positional information of the center point of the first road includes: Using the azimuth deflection angle, and according to the cosine theorem of the nearest distance, the projected distance of the nearest neighbor trajectory point relative to the center line of the first road is obtained; Using the azimuth deflection angle and the sine theorem of the nearest distance, the first lateral offset in the relative positional relationship of the nearest neighbor trajectory point with respect to the centerline of the first road is obtained.
4. The path guide line generation method as described in claim 1, characterized in that, The step of binding trajectory points in each trajectory with the center point of the first road to obtain the nearest neighbor trajectory point associated with each center point of the first road also includes: By using the directional position information of the trajectory points and the directional position information of the first road center point, the coordinates of the trajectory points are respectively bound to the coordinates of the first road center point to obtain the nearest neighbor trajectory point that is most relevant to and closest to each first road center point.
5. The path guide line generation method as described in claim 1, characterized in that, The step of inputting the location information of the second road centerline in the map into the guide line fitting model, and using the predicted associated trajectory points corresponding to the second road centerline as path guide points, includes: The guide line fitting model predicts a second lateral offset that deviates from the center point of the second road in the second road center line based on the position information of the second road center line in the map. The predicted associated trajectory point is obtained based on the second lateral offset and used as the path guidance point.
6. The path guide line generation method as described in claim 1, characterized in that, The second preset quantity and the third preset quantity are determined according to the length of the trajectory and the length of the center line of the first road, respectively.
7. A path guide line generation device, characterized in that, include: The acquisition module is used to acquire a first number of trajectories in a road segment and a first road centerline in the base map related to the road segment, respectively obtain a second preset number of trajectory points on each trajectory and a third preset number of first road center points on the first road centerline, and record the directional position information of the trajectory points and the directional position information of the first road center points into the trajectory database. The binding module is used to bind the trajectory points in each trajectory to the center point of the first road, obtain the nearest neighbor trajectory points associated with each center point of the first road, mark the corresponding trajectory point ID of each nearest neighbor trajectory point, and record the nearest neighbor trajectory points and their trajectory point IDs in the trajectory database. The relevant feature calculation module is used to use the directional position information of the nearest neighbor trajectory point and the directional position information of the first road center point to obtain the relative positional relationship of the nearest neighbor trajectory point with respect to the center line of the first road, and record it in the trajectory database. The training module is used to train the deep learning model using the relative positional relationship of the nearest neighbor trajectory points in the trajectory database with respect to the center line of the first road, so as to obtain the guide line fitting model. The guide line generation module is used to input the location information of the second road centerline in the map into the guide line fitting model, use the predicted associated trajectory points corresponding to the second road centerline as path guide points, and fit the path guide points to generate a path guide line output; wherein, the guide line fitting model predicts a second lateral offset from the center point of the second road centerline based on the second road centerline in the map, obtains the corresponding predicted associated trajectory points based on the second lateral offset as path guide points, and fits the path guide points to generate a path guide line output.
8. A vehicle, characterized in that, The vehicle includes the path guidance line generation device as described in claim 7.
9. A computer-readable storage medium storing computer instructions, characterized in that, When the computer instructions are executed, the computer performs the path guide line generation method as described in any one of claims 1-6.
10. A computer device comprising a processor and a memory, the memory storing computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the path guide line generation method as described in any one of claims 1-6.