A method, system, device, and medium for post-processing predicted trajectories

By obtaining map data and the prediction trajectory of the target vehicle, determining the lane center line and boundary line, forming a set of road boundary lines to be judged, and correcting the prediction trajectory under the coordinate system of the set, solving the problem of the prediction trajectory exceeding the boundary, ensuring driving safety and practicality of the trajectory.

CN116080687BActive Publication Date: 2025-07-25CHONGQING CHANGAN TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202310179054.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-02-27
Publication Date
2025-07-25
Estimated Expiration
2043-02-27

AI Technical Summary

Technical Problem

The prediction trajectory formed by existing trajectory prediction methods is easy to exceed road boundaries and is difficult to meet practical application needs.

Method used

By obtaining map data and the predicted trajectory of the target vehicle, determining the target lane center line, searching for the subsequent lane center line, and splicing the road boundary line to form a set of road boundary lines to be identified, and converting the predicted trajectory to the coordinate system of the set for correction, so that it is within the road boundary.

Benefits of technology

It effectively solves the problem that the predicted trajectory exceeds the boundary, ensures driving safety, and improves the practicality of the predicted trajectory.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116080687B_ABST
    Figure CN116080687B_ABST
Patent Text Reader

Abstract

The present application provides a method, a system, a device and a medium for post-processing a predicted trajectory. The method includes: obtaining map data and a predicted trajectory of a target vehicle, and determining a target lane center line of the current driving lane of the target vehicle according to the map data; searching for a corresponding subsequent lane center line in the map data according to the target lane center line, and splicing a road boundary line corresponding to the target lane center line and a road boundary line corresponding to the subsequent lane center line to obtain a set of road boundary lines to be discriminated; converting the predicted trajectory into a coordinate system with the road boundary lines in the set of road boundary lines to be discriminated as reference lines, and correcting the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to make the predicted trajectory within the corresponding road boundary lines. The present application can ensure that the predicted trajectory can meet the actual driving requirements and ensure driving safety.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of intelligent vehicle applications, and particularly to a method, system, device, and medium for post-processing predicted trajectories. Background Art

[0002] With the wide application of AI technologies represented by deep learning in the field of autonomous driving, the engineering adaptation of AI technologies has also received extensive attention. Environmental perception and prediction is an important part of autonomous driving. It predicts the trajectories of obstacles (vehicles, pedestrians, cyclists) that may interact with the host vehicle, and can pre-determine the future interaction between the autonomous driving vehicle and the environment, providing an input basis for path planning and decision-making control.

[0003] On the other hand, since the predicted trajectories output by AI algorithms are obtained by deep neural networks fitting real data, depending on the richness of training samples and the generalization ability of the network itself, the trajectories output by AI algorithms may not conform to kinematics or exceed the road boundaries. Summary of the Invention

[0004] In view of the above problems existing in the prior art, this application proposes a method, system, device, and medium for post-processing predicted trajectories, mainly solving the problem that the predicted trajectories formed by existing trajectory prediction methods are prone to exceeding the road boundaries and are difficult to meet the actual application requirements.

[0005] To achieve the above and other purposes, the technical solutions adopted in this application are as follows.

[0006] This application provides a method for post-processing predicted trajectories, including:

[0007] Obtain map data and the predicted trajectory of the target vehicle, and determine the target lane centerline of the current driving lane of the target vehicle according to the map data;

[0008] Search for the corresponding successor lane centerline in the map data according to the target lane centerline, and splice the road boundary line corresponding to the target lane centerline with the road boundary line corresponding to the successor lane centerline to obtain a set of road boundary lines to be discriminated;

[0009] Convert the predicted trajectory to a coordinate system with the road boundary lines in the set of road boundary lines to be discriminated as reference lines, and correct the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to make the predicted trajectory within the corresponding road boundary lines.

[0010] In an embodiment of the present application, the map data includes: all lane centerlines in the map, the numbers of all lane centerlines, the set of lane centerline points and the numbers of each point in the centerline point set, the numbers of the successor lane centerlines corresponding to each lane centerline, the road boundary lines corresponding to each lane centerline, and the numbers of the road boundary lines corresponding to each lane centerline.

[0011] In an embodiment of the present application, determining the target lane centerline of the current driving lane of the target vehicle according to the map data includes:

[0012] Obtain the global coordinates and heading angle of the target vehicle, and convert the road boundary lines in the map data to the vehicle coordinate system with the target vehicle as the origin according to the global coordinates and heading angle;

[0013] If the coordinates of two adjacent points on the same road boundary line have different signs along the extension direction of the boundary line, then use the corresponding road boundary line as the candidate road boundary line, and determine the current driving lane of the target vehicle according to two adjacent candidate road boundary lines.

[0014] In an embodiment of the present application, determining the current driving lane of the target vehicle according to two adjacent candidate road boundary lines includes:

[0015] Select points adjacent to the points with different signs on the two adjacent candidate road boundary lines respectively, and draw a straight line connecting the two candidate road boundary lines;

[0016] Determine whether the coordinate values on both sides of the relative position of the target vehicle are of different signs according to the straight line. If they are of different signs, then use the lane between the two adjacent candidate road boundary lines as the current driving lane of the target vehicle.

[0017] In an embodiment of the present application, searching for the corresponding successor lane centerline in the map data according to the target lane centerline includes:

[0018] Obtain the list of lane centerlines of the road where the current driving lane of the target vehicle is located in the map data;

[0019] Use DFS recursive search according to the list of lane centerlines to obtain at least two levels of successor lane centerline lists.

[0020] In an embodiment of the present application, splicing the road boundary line corresponding to the target lane centerline and the road boundary line corresponding to the successor lane centerline to obtain a set of road boundary lines to be discriminated, including:

[0021] After searching the map data to determine the corresponding list of lane centerlines, obtain the road boundary lines of the target lane in the list of lane centerlines and the road boundary lines of the subsequent lane corresponding to the target lane from the map data;

[0022] Splice the road boundary lines of the target lane and the road boundary lines of the subsequent lane corresponding to the target lane according to the numbers of the road boundary lines to obtain the set of road boundary lines to be discriminated.

[0023] In an embodiment of the present application, before correcting the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to be discriminated, it further includes:

[0024] Interpolate each road boundary line in the set of road boundary lines to be discriminated so that the point distribution density of the road boundary lines reaches a preset threshold.

[0025] In an embodiment of the present application, before correcting the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to be discriminated, it further includes:

[0026] In the coordinate system with the road boundary lines in the set of road boundary lines to be discriminated as the reference lines, multiply the coordinates of the predicted trajectory along the normal direction of any road boundary line in the set of road boundary lines to be discriminated by the coordinates of the corresponding road boundary line set. If the sign of the obtained result is negative, the predicted trajectory exceeds the road boundary line.

[0027] The present application also provides a predicted trajectory post-processing system, including:

[0028] A data acquisition module, configured to acquire map data and the predicted trajectory of the target vehicle, and determine the target lane centerline of the lane where the target vehicle is currently traveling according to the map data;

[0029] A boundary line determination module, configured to search for the corresponding subsequent lane centerlines in the map data according to the target lane centerline, and splice the road boundary lines corresponding to the target lane centerline and the road boundary lines corresponding to the subsequent lane centerlines to obtain a set of road boundary lines to be discriminated;

[0030] A correction module, configured to convert the predicted trajectory into a coordinate system with the road boundary lines in the set of road boundary lines to be discriminated as the reference lines, and correct the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to be discriminated so that the predicted trajectory is within the corresponding road boundary lines.

[0031] The present application also provides a computer device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, where when the processor executes the computer program, the steps of the above-mentioned predicted trajectory post-processing method are implemented.

[0032] The present application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned predicted trajectory post-processing method are implemented.

[0033] As described above, a predicted trajectory post-processing method, system, device, and medium of the present application have the following beneficial effects.

[0034] The present application obtains map data and the predicted trajectory of a target vehicle, determines the target lane center line of the current driving lane of the target vehicle according to the map data; searches for the corresponding successor lane center line in the map data according to the target lane center line, and splices the road boundary line corresponding to the target lane center line with the road boundary line corresponding to the successor lane center line to obtain a set of road boundary lines to be discriminated; converts the predicted trajectory into a coordinate system with the road boundary lines in the set of road boundary lines to be discriminated as reference lines, and corrects the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to make the predicted trajectory within the corresponding road boundary lines, which can effectively solve the problem that the predicted trajectory exceeds the boundary, ensure driving safety, and improve the practicability of the predicted trajectory. Description of the Drawings

[0035] Figure 1 It is a schematic diagram of the application scenario architecture of the predicted trajectory post-processing method in an embodiment of the present application.

[0036] Figure 2 It is a schematic flowchart of the predicted trajectory post-processing method in an embodiment of the present application.

[0037] Figure 3 It is a module diagram of the predicted trajectory post-processing system in an embodiment of the present application.

[0038] Figure 4 It is a schematic structural diagram of the device in an embodiment of the present application. Detailed Embodiments

[0039] The following uses specific specific examples to illustrate the implementation manners of the present application. Those skilled in the art can easily understand other advantages and effects of the present application from the content disclosed in this specification. The present application can also be implemented or applied through other different specific implementation manners, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that, without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0040] It should be noted that the diagrams provided in the following embodiments only illustrate the basic concept of the present application in a schematic manner. Therefore, only the components related to the present application are shown in the diagrams, rather than being drawn according to the number, shape, and size of the components in actual implementation. The types, quantities, and proportions of the components in actual implementation can be arbitrarily changed, and the component layout type may also be more complex.

[0041] Please refer to Figure 1 , Figure 1 , which is a schematic diagram of the application scenario architecture of the predicted trajectory post-processing method in an embodiment of the present application. First, obtain the map element point set, search for the road where the target vehicle is currently located based on the map element point set, call the topological relationship of the map elements based on the center lines of all lanes on the currently located road, and use DFS (Depth First Search algorithm) to search for and splice the successor road boundaries to obtain the road boundary constraints. Convert the predicted trajectory to the Frenet coordinate system, determine whether the predicted trajectory exceeds the boundary, correct the trajectory after exceeding the boundary, and convert the trajectory that satisfies the boundary constraints back to the original coordinate system to obtain the corrected trajectory information.

[0042] Please refer to Figure 2 , the present application provides a predicted trajectory post-processing method, which includes the following steps:

[0043] Step S200, obtain map data and the predicted trajectory of the target vehicle, and determine the target lane center line of the lane where the target vehicle is currently driving according to the map data;

[0044] Step S210, search for the corresponding successor lane center line in the map data according to the target lane center line, and splice the road boundary line corresponding to the target lane center line and the road boundary line corresponding to the successor lane center line to obtain a set of road boundary lines to be discriminated;

[0045] Step S220, convert the predicted trajectory to a coordinate system with the road boundary line in the set of road boundary lines to be discriminated as the reference line, and correct the predicted trajectory that exceeds the road boundary line in the set of road boundary lines to be discriminated so that the predicted trajectory is within the corresponding road boundary line.

[0046] In step S200, obtain map data and the predicted trajectory of the target vehicle, and determine the target lane center line of the lane where the target vehicle is currently driving according to the map data.

[0047] The predicted trajectory of the target vehicle can be obtained through trajectory planning based on the distance and speed between the obstacle and the vehicle. Any algorithm for trajectory planning of the vehicle can be applied to the target vehicle in the embodiments of the present application. The specific trajectory planning algorithm can be configured according to actual application requirements and is not limited here. After obtaining the predicted trajectory of the target vehicle through trajectory planning, a high-precision map can be further obtained. The map data included in the high-precision map can include: all lane centerlines in the map, the numbers of all lane centerlines, the set of lane centerline points and the numbers of each point in the centerline point set, the numbers of the successor lane centerlines corresponding to each lane centerline, the road boundary lines corresponding to each lane centerline, and the numbers of the road boundary lines corresponding to each lane centerline, etc. Specifically, in the high-precision map, each lane centerline will be numbered to obtain the centerline number, and at the same time, the number of each point in the lane centerline will be generated. The successor lane is the lane extending backward from the current lane, which can be the lane after lane merging or the lane entered after turning at an intersection. The specific successor lane can be determined according to the actual road conditions. The high-precision map will record the successor lane of each lane and the number of the successor lane centerline. The lane is marked by the centerline, and the road boundary lines on both sides of the lane and the adjacent lanes will also be recorded in the high-precision map. Only the corresponding data needs to be read in the high-precision map. The map data in the high-precision map is usually established under the global positioning coordinate system.

[0048] In one embodiment, determining the target lane centerline of the current driving lane of the target vehicle according to the map data includes the following steps:

[0049] Step S201, obtain the global coordinates and heading angle of the target vehicle, and convert the road boundary lines in the map data to the vehicle coordinate system with the target vehicle as the origin according to the global coordinates and heading angle.

[0050] In one embodiment, the global coordinates and heading angle of the target vehicle can be obtained. The heading angle can be used to determine the traveling direction of the target vehicle, and then the successor lane of the lane can be determined according to the traveling direction. The obtained map data is converted to the vehicle coordinate system with the target vehicle as the origin, and this vehicle coordinate system is a rectangular coordinate system. The specific coordinate conversion process is not described here again.

[0051] Step S202, if the coordinates of two adjacent points on the same road boundary line have different signs along the extension direction of the boundary line, then use the corresponding road boundary line as the candidate road boundary line, and determine the current driving lane of the target vehicle according to two adjacent candidate road boundary lines.

[0052] In one embodiment, longitudinal discrimination can be performed along the direction of the heading angle of the target vehicle following the lane line. Specifically, a point set of all road boundary lines in the area where the vehicle is located can be selected, and each road boundary line is calculated separately. It is calculated whether the coordinate values of two adjacent points on each road boundary line along the lane line direction in the vehicle coordinate system have different signs. For example, the sign of one point is positive and the sign of the other point is negative. If there are such two adjacent points, it is considered that the corresponding road boundary line longitudinally covers the target vehicle, that is, the positions of the target vehicle and these two adjacent points on the road boundary line are on the same horizontal plane. The corresponding road boundary line can be used as a candidate road boundary line. The target vehicle needs to drive between two road boundary lines when driving on the road. Through the foregoing calculation method, at least two candidate road boundary lines that meet the longitudinal constraints can be obtained. The current driving lane of the target vehicle can be determined according to two adjacent candidate road boundary lines.

[0053] In one embodiment, determining the current driving lane of the target vehicle according to two adjacent candidate road boundary lines includes the following steps:

[0054] Step S203, select adjacent points of points with different signs on the two adjacent candidate road boundary lines, and draw a straight line connecting the two candidate road boundary lines;

[0055] In one embodiment, after determining the candidate road boundary line through the foregoing steps, two adjacent road boundary lines are taken as a group. The adjacent points with different signs on the same road boundary line are obtained above, and the numbers of the points with different signs can be obtained, and the numbers of the points with different signs are associated with the corresponding candidate road boundary lines. Suppose there are adjacent candidate road boundary lines A and B, the points with different signs on candidate road boundary line A are a1 and a2 respectively, and the points with different signs on candidate road boundary line B are b1 and b2 respectively. A straight line equation can be established based on the coordinate values of a1, a2, b1, and b2 in the target vehicle coordinate system to obtain a straight line connecting candidate road boundary lines A and B.

[0056] Step S204, determine whether the coordinate values on both sides of the position relative to the target vehicle are of different signs according to the straight line. If they are of different signs, the lane between the two adjacent candidate road boundary lines is used as the current driving lane of the target vehicle.

[0057] In one embodiment, when the longitudinal coordinate in the target vehicle coordinate system is 0, if the coordinates of the intersection points of the straight line and the two candidate road boundary lines A and B are also of different signs, it means that the target vehicle is between candidate road boundary lines A and B. In this way, the road boundary line corresponding to the lane where the target vehicle is located can be determined from each candidate road boundary line.

[0058] In step S210, search for the corresponding successor lane centerline in the map data according to the target lane centerline, and splice the road boundary line corresponding to the target lane centerline and the road boundary line corresponding to the successor lane centerline to obtain a set of road boundary lines to be discriminated.

[0059] In one embodiment, searching for the corresponding successor lane centerline in the map data according to the target lane centerline includes the following steps:

[0060] Step S211, obtain the lane centerline list of the road where the target vehicle is currently traveling in the map data.

[0061] In one embodiment, all lane centerline numbers within the lane where the target vehicle is located can be obtained from the high-precision map, and the lane centerline numbers are recorded in the lane centerline list.

[0062] Step S212, use DFS recursive search according to the lane centerline list to obtain at least two levels of successor lane centerline lists.

[0063] In one embodiment, the lane centerline list contains the centerline of the lane where the target vehicle is currently traveling, that is, the target lane centerline. The adjacent lane centerlines of the target lane can be retrieved from the lane centerline list based on the target lane centerline to form an adjacent lane list containing adjacent lane centerline numbers.

[0064] In one embodiment, after searching for the successor lane centerlines of each adjacent lane in the adjacent lane list in the map data to obtain the first successor lane centerline list, it further includes:

[0065] Search for the successor lane centerlines of each lane centerline in the second successor lane centerline list in the map data to obtain the second successor lane centerline list, and use the second successor lane centerline list to update the lane centerline list.

[0066] In one embodiment, the successor lane centerline can be searched in the map data by DFS (Depth First Search). The depth-first search algorithm will search downward along each possible path until it can no longer go deeper, and each node can only be visited once. Specifically, the search depth can be set to 2. Starting from the target lane centerline as the starting node, search for the successor lane centerline numbers of the target lane and adjacent lanes. If the successor lane centerline number is empty, the recursion ends. If the successor lane centerline is not empty, a first successor lane centerline list can be constructed based on the searched successor lane centerline. The first successor lane centerline list can be incorporated into the lane centerline list of the target lane obtained at the beginning.

[0067] Further, during the second-level depth search, the subsequent lane centerline numbers can be continuously searched based on the first subsequent lane centerline list to obtain the subsequent lane centerlines of each subsequent lane centerline in the first subsequent lane centerline list, thereby forming a second subsequent lane centerline list, or the second subsequent lane centerline list can be added to the lane centerline list.

[0068] In one embodiment, splicing the road boundary line corresponding to the target lane centerline and the road boundary line corresponding to the subsequent lane centerline to obtain a set of road boundary lines to be discriminated, including the following steps:

[0069] Step S214, after searching the map data to determine the corresponding lane centerline list, obtaining the road boundary line of the target lane in the lane centerline list and the road boundary line of the subsequent lane corresponding to the target lane from the map data;

[0070] In one embodiment, the lane centerline list updated after DFS search includes the target lane centerline and the first-level subsequent lane centerline and the second-level subsequent lane centerline corresponding to the target lane centerline. The road boundary line of the target lane centerline and the road boundary line of the subsequent lane centerline can be called based on the target lane centerline and the corresponding subsequent lane centerlines.

[0071] Step S215, splicing the road boundary line of the target lane and the road boundary line of the subsequent lane corresponding to the target lane according to the numbers of the road boundary lines to obtain the set of road boundary lines to be discriminated.

[0072] In one embodiment, boundary line splicing can be performed based on the boundary lines of the target lane centerline and the corresponding subsequent lane centerlines to obtain one or more boundary lines and incorporate them into the set of road boundary lines to be discriminated. The set of road boundary lines to be discriminated can include the boundary lines of the target lane or the boundary lines of the adjacent lanes of the target lane.

[0073] In step S220, the predicted trajectory is converted into a coordinate system with the road boundary lines in the set of road boundary lines to be discriminated as the reference line, and the predicted trajectory exceeding the road boundary lines in the set of road boundary lines to be discriminated is corrected so that the predicted trajectory is within the corresponding road boundary lines.

[0074] In one embodiment, the road boundary lines in the set of road boundary lines to be discriminated can be used as the reference line to form a Frenet coordinate system. The Frenet coordinate system is a more intuitive way to represent the road position than the traditional x, y Cartesian coordinates. The Frenet coordinates use the variables s and d to describe the position of the vehicle on the road. The s coordinate represents the distance along the road (also known as the longitudinal displacement) and the d coordinate represents the left-right position on the road (also known as the lateral displacement).

[0075] In one embodiment, before correcting the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to be discriminated, it further includes:

[0076] Interpolate each road boundary line in the set of road boundary lines to be discriminated so that the point distribution density of the road boundary line reaches a preset threshold.

[0077] In one embodiment, the road boundary lines obtained by searching and splicing can be interpolated to increase the point distribution density, and then they are respectively used as Frenet reference lines, and the predicted trajectories are respectively transformed into these Frenet coordinate systems.

[0078] In one embodiment, before correcting the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to be discriminated, it further includes:

[0079] In the coordinate system with the road boundary lines in the set of road boundary lines to be discriminated as reference lines, multiply the coordinate along the normal direction of any road boundary line in the set of road boundary lines to be discriminated on the predicted trajectory by the coordinate of the corresponding road boundary line set. If the sign of the obtained result is negative, it means that the predicted trajectory exceeds the road boundary line.

[0080] In one embodiment, the zero elements in the d coordinate of the predicted trajectory can be removed, and the numbers of the non-zero elements before removing the zero elements are recorded.

[0081] Multiply the adjacent elements of the coordinates obtained in the previous step. If the sign of the product is -1 (from positive to negative or from negative to positive), it means that the d coordinate has a positive-negative jump, that is, the predicted trajectory intersects the reference line. Record the element number at this time.

[0082] Map the number obtained in the previous step back to the original coordinates before removing the zero elements to obtain the trajectory point number at the occurrence of the boundary crossing, and record the road boundary line where the crossing occurs.

[0083] Using the trajectory point number at the occurrence of the crossing, set the d coordinates of the trajectory coordinates from this number to the end to 0, and keep the s coordinates unchanged. Convert the Frenet coordinates of the adjusted predicted trajectory back to the Cartesian coordinate system. By keeping the S coordinate in the Frenet coordinate system unchanged, the predicted trajectory approximately maintains the original speed.

[0084] Please refer to Figure 3 , Figure 3The figure is a module diagram of a predicted trajectory post - processing system in an embodiment of the present application. The system includes: a data acquisition module 10, configured to acquire map data and the predicted trajectory of a target vehicle, and determine the target lane center line of the lane in which the target vehicle is currently traveling according to the map data; a boundary line determination module 11, configured to search for the corresponding successor lane center line in the map data according to the target lane center line, splice the road boundary line corresponding to the target lane center line and the road boundary line corresponding to the successor lane center line to obtain a set of road boundary lines to be discriminated; a correction module 12, configured to convert the predicted trajectory into a coordinate system with the road boundary lines in the set of road boundary lines to be discriminated as reference lines, and correct the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to make the predicted trajectory within the corresponding road boundary lines.

[0085] In an embodiment, the map data includes: all lane center lines in the map, the numbers of all lane center lines, the set of lane center line points and the numbers of each point in the set of center line points, the numbers of the corresponding successor lane center lines of each lane center line, the road boundary lines corresponding to each lane center line, and the numbers of the road boundary lines corresponding to each lane center line.

[0086] In an embodiment, the data acquisition module 10 is further configured to determine the target lane center line of the lane in which the target vehicle is currently traveling according to the map data, including: acquiring the global coordinates and the heading angle of the target vehicle, and converting the road boundary lines in the map data into a vehicle coordinate system with the target vehicle as the origin according to the global coordinates and the heading angle; if the coordinates of two adjacent points on the same road boundary line have different signs along the extension direction of the boundary line, then use the corresponding road boundary line as the candidate road boundary line, and determine the lane in which the target vehicle is currently traveling according to two adjacent candidate road boundary lines.

[0087] In an embodiment, the data acquisition module 10 is further configured to determine the lane in which the target vehicle is currently traveling according to two adjacent candidate road boundary lines, including: respectively selecting adjacent points of the points with different signs on the two adjacent candidate road boundary lines, and making a straight line connecting the two candidate road boundary lines; determining whether the coordinate values on both sides of the position relative to the target vehicle of the straight line have different signs, and if they have different signs, then use the lane between the two adjacent candidate road boundary lines as the lane in which the target vehicle is currently traveling.

[0088] In an embodiment, the boundary line determination module 11 is further configured to search for the corresponding successor lane center line in the map data according to the target lane center line, including: acquiring the list of lane center lines of the road in which the target vehicle is currently traveling in the map data; using DFS recursive search according to the list of lane center lines to obtain at least two - level successor lane center line lists.

[0089] In one embodiment, the boundary line determination module 11 is further configured to splice the road boundary line corresponding to the target lane center line and the road boundary line corresponding to the subsequent lane center line to obtain a set of road boundary lines to be discriminated, including: after searching the map data to determine the corresponding lane center line list, obtaining the road boundary line of the target lane and the road boundary line of the subsequent lane corresponding to the target lane from the map data; splicing the road boundary line of the target lane and the road boundary line of the subsequent lane corresponding to the target lane according to the number of the road boundary line to obtain the set of road boundary lines to be discriminated.

[0090] In one embodiment, before the correction module 12 is further configured to correct the prediction trajectory that exceeds the road boundary line in the set of road boundary lines to be discriminated, it further includes: interpolating each road boundary line in the set of road boundary lines to be discriminated so that the point distribution density of the road boundary line reaches a preset threshold.

[0091] In one embodiment, before the correction module 12 is further configured to correct the prediction trajectory that exceeds the road boundary line in the set of road boundary lines to be discriminated, it further includes: in a coordinate system with the road boundary line in the set of road boundary lines to be discriminated as the reference line, multiplying the coordinate along the normal direction of any road boundary line in the set of road boundary lines to be discriminated on the prediction trajectory by the coordinate of the corresponding road boundary line set. If the sign of the obtained result is negative, the prediction trajectory exceeds the road boundary line.

[0092] Each module in the above prediction trajectory post-processing system can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in or independent of the memory of the terminal in the form of hardware, or stored in the memory of the terminal in the form of software, so as to facilitate the processor to call and execute the operations corresponding to the above modules. The processor can be a central processing unit (CPU), a microprocessor, a single-chip microcomputer, etc.

[0093] Such as Figure 4As shown in the figure, it is a schematic diagram of the internal structure of a computer device in an embodiment. A computer device is provided, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are implemented: obtaining map data and the predicted trajectory of a target vehicle, and determining the target lane centerline of the current driving lane of the target vehicle according to the map data; searching for the corresponding successor lane centerline in the map data according to the target lane centerline, and splicing the road boundary line corresponding to the target lane centerline and the road boundary line corresponding to the successor lane centerline to obtain a set of road boundary lines to be discriminated; converting the predicted trajectory to a coordinate system with the road boundary lines in the set of road boundary lines to be discriminated as reference lines, and correcting the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to make the predicted trajectory within the corresponding road boundary lines.

[0094] In one embodiment, when the above-mentioned processor executes, the map data includes: all lane centerlines in the map, the numbers of all lane centerlines, the set of lane centerline points and the numbers of each point in the set of centerline points, the numbers of the corresponding successor lane centerlines of each lane centerline, the road boundary lines corresponding to each lane centerline, and the numbers of the road boundary lines corresponding to each lane line centerline.

[0095] In one embodiment, when the above-mentioned processor executes, the implementation of determining the target lane centerline of the current driving lane of the target vehicle according to the map data includes: obtaining the global coordinates and heading angle of the target vehicle, and converting the road boundary lines in the map data to a vehicle coordinate system with the target vehicle as the origin according to the global coordinates and heading angle; if the coordinates of two adjacent points on the same road boundary line have different signs along the extension direction of the boundary line, then the corresponding road boundary line is used as the candidate road boundary line, and the current driving lane of the target vehicle is determined according to two adjacent candidate road boundary lines.

[0096] In one embodiment, when the above-mentioned processor executes, the implementation of determining the current driving lane of the target vehicle according to two adjacent candidate road boundary lines includes: respectively selecting adjacent points of points with different signs on the two adjacent candidate road boundary lines, and making a straight line connecting the two candidate road boundary lines; determining whether the coordinate values on both sides of the position relative to the target vehicle are of different signs according to the straight line, and if they are of different signs, then the lane between the two adjacent candidate road boundary lines is used as the current driving lane of the target vehicle.

[0097] In one embodiment, when the above-mentioned processor executes, searching for the corresponding successor lane centerline in the map data according to the target lane centerline includes: obtaining a list of lane centerlines of the road where the target vehicle is currently driving in the map data; using DFS recursive search according to the list of lane centerlines to obtain at least two levels of successor lane centerline lists.

[0098] In one embodiment, when the above-mentioned processor executes, splicing the road boundary line corresponding to the target lane centerline and the road boundary line corresponding to the successor lane centerline to obtain a set of road boundary lines to be discriminated includes: after searching the map data to determine the corresponding list of lane centerlines, obtaining the road boundary line of the target lane and the road boundary line of the successor lane corresponding to the target lane from the map data; splicing the road boundary line of the target lane and the road boundary line of the successor lane corresponding to the target lane according to the number of the road boundary line to obtain the set of road boundary lines to be discriminated.

[0099] In one embodiment, before correcting the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to be discriminated when the above-mentioned processor executes, it further includes: interpolating each road boundary line in the set of road boundary lines to be discriminated so that the point distribution density of the road boundary line reaches a preset threshold.

[0100] In one embodiment, before correcting the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to be discriminated when the above-mentioned processor executes, it further includes: in the coordinate system with the road boundary lines in the set of road boundary lines to be discriminated as the reference line, multiplying the coordinate along the normal direction of any road boundary line in the set of road boundary lines to be discriminated on the predicted trajectory by the coordinate of the corresponding road boundary line set. If the sign of the obtained result is negative, the predicted trajectory exceeds the road boundary line.

[0101] In one embodiment, the above-mentioned computer device can be used as a server, including but not limited to an independent physical server, or a server cluster composed of multiple physical servers. The computer device can also be used as a terminal, including but not limited to mobile phones, tablets, personal digital assistants or smart devices, etc. As Figure 4 shown, the computer device includes a processor, a non-volatile storage medium, an internal memory, a display screen and a network interface connected through a system bus.

[0102] Among them, the processor of the computer device is used to provide computing and control capabilities to support the operation of the entire computer device. The non-volatile storage medium of the computer device stores an operating system and a computer program. The computer program can be executed by the processor to implement the prediction trajectory post-processing method provided by each of the above embodiments. The internal memory in the computer device provides a cache operating environment for the operating system and the computer program in the non-volatile storage medium. The display interface can display data through a display screen. The display screen can be a touch screen, such as a capacitive screen or an electronic screen, and can generate corresponding instructions by receiving click operations on the controls displayed on the touch screen.

[0103] Those skilled in the art can understand that Figure 4 the structure of the computer device shown in is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0104] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented: obtaining map data and a predicted trajectory of a target vehicle, and determining a target lane center line of the current driving lane of the target vehicle according to the map data; searching for a corresponding successor lane center line in the map data according to the target lane center line, and splicing the road boundary line corresponding to the target lane center line with the road boundary line corresponding to the successor lane center line to obtain a set of road boundary lines to be discriminated; converting the predicted trajectory into a coordinate system with the road boundary lines in the set of road boundary lines to be discriminated as reference lines, and correcting the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to make the predicted trajectory within the corresponding road boundary lines.

[0105] In one embodiment, when the computer program is executed by a processor, the map data includes: all lane center lines in the map, the numbers of all lane center lines, the set of lane center line points and the numbers of each point in the set of center line points, the numbers of the successor lane center lines corresponding to each lane center line, the road boundary lines corresponding to each lane center line, and the numbers of the road boundary lines corresponding to each lane line center line.

[0106] In one embodiment, when the computer program is executed by a processor, the target lane centerline for determining the current driving lane of the target vehicle according to the map data includes: obtaining the global coordinates and heading angle of the target vehicle, and converting the road boundary lines in the map data to the vehicle coordinate system with the target vehicle as the origin according to the global coordinates and heading angle; if the coordinates of two adjacent points on the same road boundary line have different signs along the extension direction of the boundary line, then the corresponding road boundary line is used as the candidate road boundary line, and the current driving lane of the target vehicle is determined according to two adjacent candidate road boundary lines.

[0107] In one embodiment, when the computer program is executed by a processor, the determination of the current driving lane of the target vehicle according to two adjacent candidate road boundary lines includes: respectively selecting points adjacent to the points with different signs on the two adjacent candidate road boundary lines, and drawing a straight line connecting the two candidate road boundary lines; determining whether the coordinate values on both sides of the position relative to the target vehicle are different in sign according to the straight line, and if they are different in sign, then the lane between the two adjacent candidate road boundary lines is used as the current driving lane of the target vehicle.

[0108] In one embodiment, when the computer program is executed by a processor, the search for the corresponding successor lane centerline in the map data according to the target lane centerline includes: obtaining the list of lane centerlines of the road where the target vehicle is currently driving in the map data; using DFS recursive search according to the list of lane centerlines to obtain at least two levels of successor lane centerline lists.

[0109] In one embodiment, when the instruction is executed by a processor, the splicing of the road boundary line corresponding to the target lane centerline and the road boundary line corresponding to the successor lane centerline to obtain a set of road boundary lines to be discriminated includes: after searching the map data to determine the corresponding list of lane centerlines, obtaining the road boundary line of the target lane and the road boundary line of the successor lane corresponding to the target lane from the map data; splicing the road boundary line of the target lane and the road boundary line of the successor lane corresponding to the target lane according to the number of the road boundary line to obtain the set of road boundary lines to be discriminated.

[0110] In one embodiment, before correcting the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to be discriminated when the instruction is executed by a processor, it further includes: interpolating each road boundary line in the set of road boundary lines to make the point distribution density of the road boundary line reach a preset threshold.

[0111] In one embodiment, before the predicted trajectory beyond the road boundary lines in the set of road boundary lines to be discriminated is corrected when the instruction is executed by the processor, it further includes: in a coordinate system with the road boundary lines in the set of road boundary lines to be discriminated as reference lines, multiplying the coordinate along the normal direction of any road boundary line in the set of road boundary lines to be discriminated on the predicted trajectory by the coordinate of the corresponding road boundary line set. If the sign of the obtained result is negative, then the predicted trajectory exceeds the road boundary line.

[0112] Those of ordinary skill in the art can understand that all or part of the processes in the above-described embodiment methods can be completed by instructing relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the above-described method embodiments. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), etc.

[0113] The above embodiments are only illustrative of the principles and effects of the present application, and are not used to limit the present application. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of the present application. Therefore, all equivalent modifications or changes completed by those with ordinary knowledge in the technical field without departing from the spirit and technical ideas disclosed in the present application should still be covered by the claims of the present application.

Claims

1. A method for post-processing predicted trajectories, characterized in that, Including: Obtain map data and the predicted trajectory of the target vehicle, and determine the target lane center line of the current driving lane of the target vehicle according to the map data; Determining the target lane center line of the current driving lane of the target vehicle according to the map data includes: obtaining the global coordinates and heading angle of the target vehicle, and converting the road boundary lines in the map data to the vehicle coordinate system with the target vehicle as the origin according to the global coordinates and heading angle; if the coordinates of two adjacent points on the same road boundary line have different signs along the extension direction of the boundary line, then use the corresponding road boundary line as a candidate road boundary line, and determine the current driving lane of the target vehicle according to two adjacent candidate road boundary lines; Search for the corresponding successor lane center line in the map data according to the target lane center line, and splice the road boundary line corresponding to the target lane center line and the road boundary line corresponding to the successor lane center line to obtain a set of road boundary lines to be discriminated; Convert the predicted trajectory to the coordinate system with the road boundary lines in the set of road boundary lines to be discriminated as the reference line, and correct the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to make the predicted trajectory within the corresponding road boundary line.

2. The post-processing method for predicted trajectory according to claim 1, wherein The map data includes: all lane center lines in the map, the numbers of all lane center lines, the set of lane center line points and the numbers of each point in the center line point set, the numbers of the successor lane center lines corresponding to each lane center line, the road boundary lines corresponding to each lane center line, and the numbers of the road boundary lines corresponding to each lane center line.

3. The post-processing method for predicted trajectory according to claim 1, wherein Determining the current driving lane of the target vehicle according to two adjacent candidate road boundary lines includes: Select adjacent points near the points with different coordinates on the two adjacent candidate road boundary lines, and draw a straight line connecting the two candidate road boundary lines; Determine whether the coordinate values on both sides of the relative position of the target vehicle are of different signs according to the straight line, and if they are of different signs, then use the lane between the two adjacent candidate road boundary lines as the current driving lane of the target vehicle.

4. The post-processing method for predicted trajectory according to claim 1, wherein Searching for the corresponding successor lane center line in the map data according to the target lane center line includes: Obtain the list of lane center lines of the road where the target vehicle is currently driving in the map data; Use DFS recursive search according to the list of lane center lines to obtain at least two levels of successor lane center line lists.

5. The post-processing method for predicted trajectory according to claim 2 or 4, characterized in that, Splicing the road boundary line corresponding to the target lane center line and the road boundary line corresponding to the successor lane center line to obtain a set of road boundary lines to be discriminated includes: After searching the map data to determine the corresponding list of lane center lines, obtain the road boundary line of the target lane and the road boundary line of the successor lane corresponding to the target lane from the map data; Splice the road boundary line of the target lane and the road boundary line of the successor lane corresponding to the target lane according to the number of the road boundary line to obtain the set of road boundary lines to be discriminated.

6. The post-processing method for predicted trajectory according to claim 1, wherein, Before correcting the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to be discriminated, it further includes: Interpolate each road boundary line in the set of road boundary lines to be discriminated so that the point distribution density of the road boundary line reaches a preset threshold.

7. The post-processing method for predicted trajectory according to claim 1, characterized in that Before correcting the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to be discriminated, it further includes: In the coordinate system with the road boundary lines in the set of road boundary lines to be discriminated as reference lines, multiply the coordinate of the predicted trajectory along the normal direction of any road boundary line in the set of road boundary lines to be discriminated by the coordinate of the corresponding road boundary line set. If the sign of the obtained result is negative, then the predicted trajectory exceeds the road boundary line.

8. A predicted trajectory post-processing system, characterized in that, It includes: A data acquisition module, configured to acquire map data and the predicted trajectory of the target vehicle, and determine the target lane center line of the current driving lane of the target vehicle according to the map data; Determining the target lane center line of the current driving lane of the target vehicle according to the map data includes: acquiring the global coordinates and heading angle of the target vehicle, and converting the road boundary lines in the map data to the vehicle coordinate system with the target vehicle as the origin according to the global coordinates and heading angle; if the coordinates of two adjacent points on the same road boundary line have different signs along the boundary line extension direction, then use the corresponding road boundary line as a candidate road boundary line, and determine the current driving lane of the target vehicle according to two adjacent candidate road boundary lines; A boundary line determination module, configured to search for the corresponding subsequent lane center line in the map data according to the target lane center line, and splice the road boundary line corresponding to the target lane center line and the road boundary line corresponding to the subsequent lane center line to obtain a set of road boundary lines to be discriminated; A correction module, configured to convert the predicted trajectory to the coordinate system with the road boundary lines in the set of road boundary lines to be discriminated as reference lines, and correct the predicted trajectory that exceeds the road boundary lines in the set of road boundary lines to be discriminated so that the predicted trajectory is within the corresponding road boundary line.

9. A computer device, comprising: A memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the computer program, it implements the steps of the predicted trajectory post-processing method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the predicted trajectory post-processing method according to any one of claims 1 to 7.

Citation Information

Patent Citations

  • Auxiliary forward track device for vehicle

    CN102745198A

  • ArcGIS-based map building and intelligent vehicle autonomous navigation method and system

    CN106840178A