Lane line adjustment method and device, vehicle and storage medium

Abstract

The application discloses a lane line adjustment method and device, a vehicle and a storage medium. The method comprises the following steps: acquiring a first self-vehicle lane boundary line and a second self-vehicle lane boundary line perceived by a self-vehicle; determining an initial error score corresponding to the first self-vehicle lane boundary line according to the first self-vehicle lane boundary line and the second self-vehicle lane boundary line; performing position adjustment on the first self-vehicle lane boundary line and the second self-vehicle lane boundary line for multiple times to obtain a candidate error score corresponding to the first self-vehicle lane boundary line after each position adjustment; determining a heading angle offset value of the self-vehicle lane boundary at a first time point based on the multiple candidate error scores and the initial error score; and adjusting the first self-vehicle lane boundary line according to the heading angle offset value to obtain a target self-vehicle lane boundary line for the self-vehicle lane boundary at the first time point. According to the method, the accuracy of the target self-vehicle lane boundary line is higher.

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and more specifically, to a lane adjustment method, device, vehicle, and storage medium. Background Art

[0002] In the field of autonomous driving, a vehicle can determine the lane boundary of the lane in which the vehicle is traveling based on the perceived surrounding environment information and obstacle information, and then plan a path based on the lane boundary, and automatically drive according to the planned path to achieve autonomous driving.

[0003] However, the lane boundary lines determined by the ego vehicle based on the perceived environmental information and obstacle information are inaccurate, resulting in inaccurate planned paths and low stability and safety of the vehicle's autonomous driving. Summary of the Invention

[0004] This application proposes a lane line adjustment method, device, vehicle and storage medium to improve the accuracy of the lane boundary line of the vehicle.

[0005] In a first aspect, an embodiment of the present application provides a lane adjustment method, the method comprising:

[0006] Obtaining a first lane boundary line perceived by the vehicle at a first time point and a second lane boundary line perceived by the vehicle at a second time point, wherein the first time point is later than the second time point;

[0007] Determining an initial error score corresponding to the first ego vehicle lane boundary line based on the first ego vehicle lane boundary line and the second ego vehicle lane boundary line;

[0008] Performing multiple position adjustments on the first ego vehicle lane boundary line and the second ego vehicle lane boundary line to obtain a candidate error score corresponding to the first ego vehicle lane boundary line after each position adjustment;

[0009] Determining a heading angle offset value of the ego vehicle lane boundary at a first time point based on a plurality of candidate error scores corresponding to the first ego vehicle lane boundary line after the plurality of position adjustments and the initial error score;

[0010] The first ego vehicle lane boundary line is adjusted according to the heading angle offset value to obtain a target ego vehicle lane boundary line for the ego vehicle lane boundary at a first time point.

[0011] In a second aspect, an embodiment of the present application further provides a lane adjustment device, comprising:

[0012] An acquisition module, configured to acquire a first lane boundary line perceived by the vehicle at a first time point and a second lane boundary line perceived by the vehicle at a second time point; the first time point is later than the second time point;

[0013] A first determining module is configured to determine an initial error score corresponding to the first ego vehicle lane boundary line based on the first ego vehicle lane boundary line and the second ego vehicle lane boundary line;

[0014] a first adjustment module, configured to perform multiple position adjustments on the first ego vehicle lane boundary line and the second ego vehicle lane boundary line, and obtain a candidate error score corresponding to the first ego vehicle lane boundary line after each position adjustment;

[0015] a second determining module, configured to determine a heading angle offset value of the ego vehicle lane boundary at a first time point based on a plurality of candidate error scores corresponding to the first ego vehicle lane boundary line after the plurality of position adjustments and the initial error score;

[0016] a second adjustment module, configured to adjust the first ego vehicle lane boundary line according to the heading angle offset value to obtain a target ego vehicle lane boundary line at a first time point; and a target ego vehicle lane boundary line acquisition module, configured to obtain a first ego vehicle lane boundary line sensed by the ego vehicle at the first time point and a second ego vehicle lane boundary line sensed at a second time point, wherein the first time point is later than the second time point.

[0017] A first determining module is configured to determine an initial error score corresponding to the first ego vehicle lane boundary line based on the first ego vehicle lane boundary line and the second ego vehicle lane boundary line;

[0018] a first adjustment module, configured to perform multiple position adjustments on the first ego vehicle lane boundary line and the second ego vehicle lane boundary line, and obtain a candidate error score corresponding to the first ego vehicle lane boundary line after each position adjustment;

[0019] a second determination module, configured to determine a heading angle offset value of the ego vehicle lane boundary at a first time point based on a plurality of candidate error scores corresponding to the ego vehicle lane boundary after the plurality of position adjustments and the initial error score;

[0020] The second adjustment module is configured to adjust the first ego vehicle lane boundary line according to the heading angle offset value to obtain a target ego vehicle lane boundary line corresponding to the ego vehicle lane boundary at a first time point.

[0021] In a third aspect, an embodiment of the present application further provides a vehicle, comprising: one or more processors; a memory; and one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to execute the method of the first aspect above.

[0022] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores a program code executable by a processor, and when the program code is executed by the processor, the processor executes the above method.

[0023] The present application provides a lane adjustment method, apparatus, vehicle, and storage medium. In the present application, an initial error score corresponding to a first ego-vehicle lane boundary and multiple candidate error scores corresponding to the first ego-vehicle lane boundary after multiple position adjustments are first determined. Then, based on the multiple candidate error scores and the initial error scores, a heading angle offset value of the ego-vehicle lane boundary at a first time point is determined. The first ego-vehicle lane boundary is adjusted according to the heading angle offset value to obtain a target ego-vehicle lane boundary for the ego-vehicle lane boundary at the first time point. In this way, multiple matching of the first ego-vehicle lane boundary and the second ego-vehicle lane boundary is achieved to determine the heading angle offset value based on the multiple candidate error scores and the initial error scores obtained from the multiple matching. Then, the first ego-vehicle lane boundary is corrected based on the determined heading angle offset value to obtain a corrected target ego-vehicle lane boundary. The target ego-vehicle lane boundary has a high accuracy, so that when path planning is performed according to the target ego-vehicle lane boundary, the planned trajectory has a high accuracy, thereby improving the safety and accuracy of the ego-vehicle in autonomous driving according to the planned trajectory.

[0024] Other features and advantages of the embodiments of the present application will be described in the following description and, in part, will become apparent from the description or be understood by practicing the embodiments of the present application. The objectives and other advantages of the embodiments of the present application can be achieved and obtained through the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 A schematic diagram of a vehicle hardware environment suitable for an embodiment of the present application is shown.

[0027] Figure 2 A flow chart of a lane adjustment method proposed according to an embodiment of the present application is shown.

[0028] Figure 3 A schematic diagram showing a lane line adjustment process in an embodiment of the present application is shown.

[0029] Figure 4A structural block diagram of a lane line adjustment device is shown. DETAILED DESCRIPTION

[0030] In order to enable persons skilled in the art to better understand the schemes of the present application, the technical schemes in the embodiments of the present application will be clearly and completely described below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by persons skilled in the art without creative work are within the scope of protection of the present application.

[0031] It should be noted that: similar reference numerals and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0032] Reference Figure 1 , Figure 1 A schematic diagram of a vehicle hardware environment suitable for the embodiments of the present application is shown, the vehicle 100 includes a driving system 110, the driving system 110 can be built-in with various automatic driving functions, the driving system 110 can store an electronic map, the driving system 110 can plan a driving path according to the electronic map stored by itself, and can also control the vehicle to automatically drive according to the planned driving path.

[0033] The driving system 110 can include a data acquisition device 111, one or more (only one is shown in the figure) processors 112, and a memory 113.

[0034] The data acquisition device 111 is used to detect various signals or data of the vehicle, for example, the data acquisition device 111 can acquire environmental information and obstacle information around the vehicle, the environmental information can include weather information, lane information and traffic light information, the obstacle information can include the position, speed and acceleration of the obstacle, etc., the obstacle can be a pedestrian, a vehicle and a vehicle signboard.

[0035] The processor 112 can be a micro control unit (MCU), the micro control unit is built-in with the memory 113, the memory 113 stores programs that can execute the contents in the following embodiments, and the processor 112 can execute the programs stored in the memory 113.

[0036] The processor 112 may include one or more processors. The processor 112 utilizes various interfaces and circuits to connect various components within the vehicle 100 and execute various functions and process data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 113 and accessing data stored in the memory 113.

[0037] The memory 113 may include a random access memory (RAM) or a read-only memory (ROM). The memory 15 may be used to store instructions, programs, codes, code sets, or instruction sets. The memory 15 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the following various method embodiments, etc.

[0038] See also Figure 2 , Figure 2 A flow chart of a lane adjustment method proposed according to one embodiment of the present application is shown, which is used for a vehicle. The method includes:

[0039] S110 , obtaining a first lane boundary line perceived by the vehicle at a first time point and a second lane boundary line perceived by the vehicle at a second time point.

[0040] The first time point is later than the second time point. That is, the first lane boundary line is the lane boundary line sensed for the lane in which the vehicle is traveling at the first time point, and the second lane boundary line is the lane boundary line sensed for the lane in which the vehicle is traveling at the second time point.

[0041] The vehicle in this embodiment can be an electric vehicle or a fuel vehicle, or a car, an SUV, a bus, a truck, etc., and the self-vehicle can refer to the vehicle itself.

[0042] The first time point may be any time point during the driving process of the vehicle, and may be each current time point during the driving process of the vehicle; or it may be a time point after each current time point during the driving process of the vehicle is delayed by a certain time. For example, the first time point may be a time point after each current time point during the driving process of the vehicle is delayed by 2 seconds, thereby achieving early correction of the first lane boundary line of the vehicle perceived at the first time point.

[0043] The second time point is any time point before the first time point. For example, the second time point may be a time point before the first time point. For another example, the second time point may be a time point before the first time point and 2 seconds away from the first time point.

[0044] In some embodiments, during the driving process of the vehicle, the boundary line of the vehicle lane is sensed to obtain a first lane boundary line and a second lane boundary line.

[0045] In another embodiment, the first lane boundary line and the second lane boundary line perceived by the vehicle may be obtained while the vehicle is traveling in a target road area on a highway; the target road area is the road area on the highway excluding the lane boundary line intersection area and the lane boundary line convergence area.

[0046] Among them, the lane boundary line intersection area refers to the area where the lane boundary lines intersect. For example, the range 100 meters before and after the lane boundary line intersection point can be used as the lane boundary line intersection area; the lane boundary line convergence area refers to the area where the lane boundary lines converge. For example, the range 100 meters before and after the lane boundary line convergence point can be used as the lane boundary line convergence area.

[0047] That is, the first lane boundary line and the second lane boundary line of the vehicle are perceived only when the vehicle is traveling in the target road area of ​​the highway, so as to correct the first lane boundary line in the subsequent process.

[0048] If the vehicle is traveling on a non-highway road, the speed of the vehicle is low, so the perceived lane boundary line of the vehicle is more accurate and does not need to be corrected. Therefore, the first lane boundary line correction is not performed.

[0049] If the vehicle is traveling in a non-target road area of ​​the highway (the non-target area refers to other areas of the highway excluding the target area), it means that the shape of the lane boundary line of the vehicle may be changeable and the shape of the lane boundary line of the vehicle is difficult to determine. At this time, there is no need to correct the lane boundary line of the vehicle, and thus the first lane boundary line correction is not performed.

[0050] In still another embodiment, before S110, the method further comprises: cutting a first ego vehicle lane boundary line from a first initial perceived lane boundary line perceived by the ego vehicle, and cutting a second ego vehicle lane boundary line from a second initial perceived lane boundary line perceived by the ego vehicle; wherein the first initial perceived lane boundary line is a lane boundary line perceived at a first time point for the ego vehicle lane boundary; the second initial perceived lane boundary line is a lane boundary line perceived at a second time point for the ego vehicle lane boundary; correspondingly, S110 comprises: if the first ego vehicle lane boundary line and the second ego vehicle lane boundary line satisfy the specified lane condition, obtaining the first ego vehicle lane boundary line and the second ego vehicle lane boundary line.

[0051] In the embodiment, the specified lane condition comprises at least one of the following: the perceived lane width of the ego vehicle for the ego vehicle lane at the first time point and the second time point is in a preset lane width interval; the first boundary distance and the second boundary distance are both greater than a specified distance; the first boundary distance is a distance between the center of the ego vehicle and the first ego vehicle lane boundary line at the first time point; the second boundary distance is a distance between the center of the ego vehicle and the second ego vehicle lane boundary line at the second time point. The preset lane width interval can be [3.5m, 5m], and the specified distance can be 1m.

[0052] During the driving of the ego vehicle, the ego vehicle lane boundary line perceived for the ego vehicle lane boundary is relatively long, and therefore, a part of the perceived ego vehicle lane boundary line can be cut off as the ego vehicle lane boundary line that needs to be corrected, that is, a part of the second initial perceived lane boundary line with a relatively long length perceived at the second time point is cut off as the second ego vehicle lane boundary line, and a part of the first initial perceived lane boundary line with a relatively long length perceived at the first time point is cut off as the first ego vehicle lane boundary line.

[0053] For example, the first 50 meters and the last 100 meters of the first initial perceived lane boundary can be cut off from the center of the position of the ego vehicle at the first time point as the first ego vehicle lane boundary line, and the first 50 meters and the last 100 meters of the second initial perceived lane boundary can be cut off from the center of the position of the ego vehicle at the second time point as the second ego vehicle lane boundary line.

[0054] When the perceived lane width of the self-vehicle lane perceived by the self-vehicle vehicle at the first time point and the second time point is within the preset lane width range, it means that the self-vehicle lane is neither too wide nor too narrow at the first time point and the second time point, and the perceived first self-vehicle lane boundary line and the second self-vehicle lane boundary line are both relatively accurate and can be used to correct the first self-vehicle lane boundary line; when the perceived lane width of the self-vehicle lane perceived by the self-vehicle vehicle at the first time point or the second time point is not within the preset lane width range, it means that the self-vehicle lane is too wide or too narrow at the first time point or the second time point, and the first self-vehicle lane boundary line or the second self-vehicle lane boundary line perceived by the self-vehicle vehicle may be accurate. If the first self-vehicle lane boundary line and the second self-vehicle lane boundary line are used to correct the first self-vehicle lane boundary line, the correction effect may be poor. At this time, the first self-vehicle lane boundary line and the second self-vehicle lane boundary line are not obtained, so that the first self-vehicle lane boundary line is no longer corrected.

[0055] When the first boundary distance and the second boundary distance are both greater than the specified distance, it means that the vehicle is not riding on the boundary line of the self-vehicle lane at the first time point and the second time point. The vehicle is traveling along the self-vehicle lane, and the first self-vehicle lane boundary line can be corrected. Therefore, the first self-vehicle lane boundary line and the second self-vehicle lane boundary line can be obtained. Conversely, when the first boundary distance or the second boundary distance is not greater than the specified distance, it means that the vehicle is riding on the boundary line of the self-vehicle lane at the first time point or the second time point. The vehicle may not be traveling along the self-vehicle lane, and the first self-vehicle lane boundary line does not need to be corrected. Therefore, the first self-vehicle lane boundary line and the second self-vehicle lane boundary line can no longer be obtained.

[0056] Of course, the aforementioned designated lane conditions may include the aforementioned two constraints, so as to obtain the first self-vehicle lane boundary line and the second self-vehicle lane boundary line when the self-vehicle lane is neither too wide nor too narrow and the self-vehicle does not appear to be riding on the self-vehicle lane boundary line, so as to realize the correction of the first self-vehicle lane boundary line.

[0057] S120 : Determine an initial error score corresponding to the first ego vehicle lane boundary line based on the first ego vehicle lane boundary line and the second ego vehicle lane boundary line.

[0058] In this application, an initial error score corresponding to the first lane boundary line can be determined based on the position difference between the first lane boundary line and the second lane boundary line. For example, the relative distance (either the lateral relative distance or the longitudinal relative distance) between the first lane boundary line and the second lane boundary line can be determined as the initial error score corresponding to the first lane boundary line. The lateral distance can refer to the lateral direction of the vehicle body, and the longitudinal distance refers to the longitudinal direction of the vehicle body.

[0059] In some embodiments, the first self-vehicle lane boundary line includes multiple first boundary line points; the second self-vehicle lane boundary line includes a second boundary line point corresponding to each first boundary line point; S120 may include: determining the position error corresponding to each first boundary line point based on the position difference between each first boundary line point and the corresponding second boundary line point; and determining the initial error score corresponding to the first self-vehicle lane boundary line based on the position errors corresponding to the multiple first boundary line points.

[0060] Exemplarily, the first boundary line points can be collected from the first lane boundary line of the vehicle in the longitudinal direction of the vehicle body with the position of the vehicle at the first time point as the center. For example, the first boundary line points are respectively selected in the longitudinal direction of the vehicle body [-10m, 0, 10m, 20m, 30m, 40m, 50m, 60m]. At this time, the number of first boundary line points is 16 - 8 on each side. Accordingly, the number of first boundary line points is 16 - 8 on each side.

[0061] In this embodiment, the first boundary line point and the second boundary line point are on the same side of the vehicle, and the first longitudinal distance and the second longitudinal distance are the same; the first longitudinal distance is the longitudinal relative distance between the position of the vehicle and the first boundary line point at the first time point; the second longitudinal distance is the longitudinal relative distance between the position of the vehicle and the second boundary line point at the first time point.

[0062] That is, if a first boundary line point and a second boundary line point are on the same side of the ego vehicle and have the same longitudinal distance from the ego vehicle, they are determined to correspond. For example, if the ego vehicle is at position s1 at the first time point, a first boundary line point d1 on the left boundary line of the first ego vehicle's lane boundary, which is at a longitudinal distance l1 from s1, and a second boundary line point d2 on the left boundary line of the second ego vehicle's lane boundary, which is at a longitudinal distance l1 from s1, are considered to correspond. However, a first boundary line point d1 on the left boundary line of the first ego vehicle's lane boundary, which is at a longitudinal distance l1 from s1, and a second boundary line point d3 on the right boundary line of the second ego vehicle's lane boundary, which is at a longitudinal distance l1 from s1, are not considered to correspond.

[0063] In one embodiment, the second self-vehicle lane boundary line can be projected onto the coordinate system where the first self-vehicle lane boundary line is located to obtain a boundary line projection result. After the first boundary line point is determined on the first self-vehicle lane boundary line, for each first boundary line point d1i, a projection point ti is sampled from the boundary line projection result (the projection point is a point where the boundary line projection result and the first boundary line point are on the same side of the self-vehicle and at the same longitudinal position as the self-vehicle at the first time point), and the projection result is that the second boundary line point d2i of the projection point ti corresponds to the first boundary line point d1i.

[0064] After determining the first boundary points and their corresponding second boundary points, the positional difference between each first boundary point and its corresponding second boundary point can be determined as the positional error corresponding to each first boundary point. The average of the positional errors corresponding to the multiple first boundary points is then calculated as the initial error score corresponding to the first lane boundary. The positional difference can be the Euclidean distance, Manhattan distance, or other distance between the first boundary point and its corresponding second boundary point.

[0065] In some embodiments, S120 may further include: determining a lateral error of a boundary line point corresponding to each first boundary line point based on a lateral distance difference between each first boundary line point and the corresponding second boundary line point; and determining an initial error score corresponding to the first vehicle lane boundary line based on an average value of the lateral errors of the boundary line points corresponding to multiple first boundary line points.

[0066] For example, the average of the lateral errors of the boundary points corresponding to the multiple first boundary points can be used as the initial error score corresponding to the first lane boundary. Alternatively, the average of the lateral errors of the boundary points corresponding to the multiple first boundary points can be multiplied by a preset coefficient to obtain the initial error score corresponding to the first lane boundary. The preset coefficient can be set based on demand and can be, for example, 1 or 0.8.

[0067] S130 , performing multiple position adjustments on the first ego vehicle lane boundary line and the second ego vehicle lane boundary line, and obtaining a candidate error score corresponding to the first ego vehicle lane boundary line after each position adjustment.

[0068] The first lane boundary line and the second lane boundary line may be repositioned multiple times. After each repositioning, an error score is determined based on the repositioned first lane boundary line and the repositioned second lane boundary line as a candidate repositioned error score. The process for determining the candidate error score based on the repositioned first lane boundary line and the repositioned second lane boundary line is similar to the process for determining the initial error score based on the first lane boundary line and the repositioned second lane boundary line, and is not further described here.

[0069] In this embodiment, multiple adjusted heading angle offset values ​​can be determined based on the initial error score. The first lane boundary line and the second lane boundary line are then adjusted based on the multiple adjusted heading angle offset values, resulting in the first lane boundary line and the second lane boundary line after each adjustment. For example, the first lane boundary line and the second lane boundary line can be rotated based on the adjusted heading angle offset values ​​to adjust the positions of the first lane boundary line and the second lane boundary line, resulting in the adjusted first lane boundary line and the adjusted second lane boundary line.

[0070] For example, multiple reference coefficients can be set, and the initial error fraction can be multiplied by the corresponding reference coefficients to obtain multiple adjusted heading angle offset values. The reference coefficients can be set based on demand, for example, the reference coefficients can be 1 / 25 or 1 / 30.

[0071] In some embodiments, S130 may further include: determining an initial heading angle offset value based on the initial error score; rotating the first self-vehicle lane boundary line and the second self-vehicle lane boundary line according to the initial heading angle offset value, respectively, to adjust the positions of the first self-vehicle lane boundary line and the second self-vehicle lane boundary line, and obtaining the first self-vehicle lane boundary line after position adjustment and the second self-vehicle lane boundary line after position adjustment; determining a candidate error score corresponding to the first self-vehicle lane boundary line after position adjustment according to the first self-vehicle lane boundary line after position adjustment and the second self-vehicle lane boundary line after position adjustment; determining a candidate error score corresponding to the first self-vehicle lane boundary line after position adjustment based on the candidate error score corresponding to the first self-vehicle lane boundary line after the most recent position adjustment. Determine a new initial heading angle offset value; obtain the first self-vehicle lane boundary line after the most recent position adjustment as the new first self-vehicle lane boundary line and the second self-vehicle lane boundary line after the most recent position adjustment as the new second self-vehicle lane boundary line; return to execute the steps of rotating the first self-vehicle lane boundary line and the second self-vehicle lane boundary line according to the initial heading angle offset value, so as to adjust the positions of the first self-vehicle lane boundary line and the second self-vehicle lane boundary line, and obtain the position-adjusted first self-vehicle lane boundary line and the position-adjusted second self-vehicle lane boundary line, until the number of position adjustments reaches the specified number, and obtain the candidate error score corresponding to the first self-vehicle lane boundary line after multiple position adjustments.

[0072] In this application, the initial error score can be multiplied by a specified coefficient to obtain an initial heading angle offset value. Similarly, the candidate error score corresponding to the lane boundary of the ego vehicle after the most recent position adjustment can be multiplied by a specified coefficient to obtain a new initial heading angle offset value. The specified coefficient can be, for example, 1 / 25 or 1 / 30. For example, when the specified coefficient is 1 / 25, the heading angle offset value increases by 0.02° for every 0.5m increase in position error.

[0073] After determining the initial heading angle offset value, the first self-vehicle lane boundary line and the second self-vehicle lane boundary line can be rotated according to the initial heading angle offset value to obtain the first self-vehicle lane boundary line with the rotated initial heading angle offset value as the first self-vehicle lane boundary line after position adjustment. Similarly, the second self-vehicle lane boundary line with the rotated initial heading angle offset value is obtained as the second self-vehicle lane boundary line after position adjustment. In this way, a position adjustment of the first self-vehicle lane boundary line and the second self-vehicle lane boundary line is achieved.

[0074] After the first lane boundary line and the second lane boundary line are repositioned, a candidate error score is determined according to the aforementioned process for determining the initial error score. A new initial heading angle offset value is determined based on the most recently determined candidate error score. The first lane boundary line after the most recently repositioned position adjustment is obtained as the new first lane boundary line, and the second lane boundary line after the most recently repositioned position adjustment is obtained as the new second lane boundary line. The new first lane boundary line and the new second lane boundary line are repositioned continuously based on the new initial heading angle offset value. This cycle is repeated until the number of position adjustments reaches a specified number, at which point the cycle is stopped and the specified number of candidate error scores are obtained. In this application, the specified number of times can be, for example, 50 times.

[0075] S140 : Determine a heading angle offset value of the ego vehicle lane boundary at a first time point based on multiple candidate error scores corresponding to the first ego vehicle lane boundary line after multiple position adjustments and the initial error score.

[0076] After obtaining multiple candidate error scores and initial error scores corresponding to the first ego vehicle lane boundary line after multiple position adjustments, a target error score can be determined based on the multiple candidate error scores and initial error scores after multiple position adjustments, and then the heading angle offset value of the ego vehicle lane boundary at the first time point can be determined based on the target error score.

[0077] For example, the target error score can be obtained by averaging, maximizing, or weighted summing multiple candidate error scores and the initial error score, and then the target error score is multiplied by a specified coefficient to obtain the heading angle offset value of the vehicle lane boundary at the first time point.

[0078] In some embodiments, S140 may include: obtaining the smallest value from multiple candidate error scores and the initial error score as the error score to be processed; determining the heading angle offset value to be processed based on the error score to be processed; and filtering the heading angle offset value to be processed to obtain the heading angle offset value of the vehicle lane boundary at the first time point.

[0079] The error fraction to be processed can be multiplied by a specified coefficient to obtain the heading angle offset value to be processed. The heading angle offset value to be processed is then filtered to obtain the heading angle offset value of the lane boundary at the first time point. The filtering process here can be smoothing filtering, mean filtering, Gaussian filtering, guided filtering, etc.

[0080] Exemplarily, the aforementioned filtering processing of the heading angle offset value to be processed to obtain the heading angle offset value of the vehicle lane boundary at the first time point may include: performing weighted summation on the heading angle offset value to be processed and the reference heading angle offset value to obtain the heading angle offset value of the vehicle lane boundary at the first time point; the reference heading angle offset value is the heading angle offset value of the vehicle lane boundary at the second time point; the weight of the heading angle offset value to be processed is less than the weight of the reference heading angle offset value.

[0081] It can be understood that the reference heading angle offset value can also be the heading angle offset value of the vehicle lane boundary at the second time point obtained according to the method of the present application, that is, the second time is taken as the first time point, and the time point before the second time point is obtained as the third time point. The heading angle offset value of the vehicle lane boundary at the second time point is determined according to the above-mentioned S110-S140 method as the reference heading angle offset value.

[0082] The aforementioned filtering process can be briefly described as Formula 1, which is as follows:

[0083]

[0084] in, is the heading angle offset value of the lane boundary line of the vehicle at the first time point, is the reference heading angle offset value, Δcur_yaw is the heading angle offset value to be processed, a and b are weights, a>b, for example, a is 0.95, b is 0.5.

[0085] It can be understood that for each time point, an initial error score and multiple (that is, a specified number) candidate error scores can be determined, and then based on the initial error score and the multiple candidate error scores, the heading angle offset value to be processed at the time point is determined, and then the above-mentioned filtering process is performed on the heading angle offset value to be processed at the time point to obtain the heading angle offset value at the time point.

[0086] S150: Adjust the first own-vehicle lane boundary line according to the heading angle offset value to obtain a target own-vehicle lane boundary line for the own-vehicle lane boundary at the first time point.

[0087] The first ego vehicle lane boundary line can be rotated, and the rotation angle is the heading angle offset value of the lane boundary line at the first time point, and the rotated first ego vehicle lane boundary line is obtained as the target ego vehicle lane boundary line for the ego vehicle lane boundary at the first time point.

[0088] For example, the lane line adjustment is as follows Figure 3 As shown, scene detection is first performed to determine whether the ego vehicle is traveling in the target road area of ​​the highway. If so, lane boundary lines are screened: a first initial perception lane boundary line and a second initial perception lane boundary line are obtained, and then the first ego vehicle lane boundary line is intercepted from the first initial perception lane boundary line, and the second ego vehicle lane boundary line is intercepted from the second initial perception lane boundary line, and the intercepted first ego vehicle lane boundary line and second ego vehicle lane boundary line meet the specified lane conditions.

[0089] Afterwards, the heading angle offset value is determined based on the first ego vehicle lane boundary line and the second ego vehicle lane boundary line to obtain the heading angle offset value to be processed, and then filtering is performed based on the heading angle offset value to obtain the heading angle offset value. Finally, the first ego vehicle lane boundary line is adjusted according to the heading angle offset value to obtain the target ego vehicle lane boundary line.

[0090] In this embodiment, an initial error score corresponding to the first lane boundary line and multiple candidate error scores corresponding to the first lane boundary line after multiple position adjustments are first determined. Then, based on the multiple candidate error scores and the initial error scores, a heading angle offset value of the lane boundary at the first time point is determined, and the first lane boundary line is adjusted according to the heading angle offset value to obtain a target lane boundary line for the lane boundary at the first time point. In this way, multiple matching of the first lane boundary line and the second lane boundary line is achieved to determine the heading angle offset value based on the multiple candidate error scores and the initial error scores obtained from the multiple matching. Then, the first lane boundary line is corrected based on the determined heading angle offset value to obtain a corrected target lane boundary line. The target lane boundary line has a high accuracy, so that when path planning is performed according to the target lane boundary line, the planned trajectory has a high accuracy, thereby improving the safety and accuracy of the autonomous driving of the vehicle according to the planned trajectory.

[0091] Secondly, based on the minimum value of the error scores (including the initial error score and the candidate error scores) - the error score to be processed, the heading angle offset value to be processed is determined, so that the accuracy of the heading angle offset value to be processed is higher; then the heading angle offset value to be processed is filtered to obtain the heading angle offset value, so as to avoid the heading angle offset value from jumping significantly, resulting in an unstable heading angle offset value and an inaccurate heading angle offset value, thereby further improving the accuracy of the determined heading angle offset value.

[0092] Furthermore, in this application, the above-mentioned correction of the lane boundary line of the self-vehicle is also performed in a specific scenario (that is, in the target area of ​​the highway), so as to avoid the situation where the lane boundary line of the self-vehicle is more accurate in other scenarios, or the shape of the lane boundary line of the self-vehicle is uncertain in other scenarios, and the above-mentioned correction of the lane boundary line of the self-vehicle is also performed, resulting in the correction of the lane boundary line of the self-vehicle being more inaccurate, thereby ensuring that the perceived lane boundary line of the self-vehicle is still accurate.

[0093] In addition, in the present application, when the first lane boundary line and the second lane boundary line meet the specified lane conditions, the lane boundary line of the self-vehicle is corrected to avoid the situation where the lane boundary line of the self-vehicle is corrected even when the first lane boundary line and the second lane boundary line do not meet the specified lane conditions and there is no need to correct the lane line of the self-vehicle, causing the self-vehicle to consume resources in the process of correcting the lane boundary line of the self-vehicle, resulting in serious waste of resources, thereby reducing the waste of resources of the self-vehicle.

[0094] See attached Figure 4 , Figure 4 The following is a block diagram of a lane adjustment device according to an embodiment of the present application. The device 800 is used in a vehicle and includes:

[0095] An acquisition module 810 is configured to acquire a first lane boundary line perceived by the vehicle at a first time point and a second lane boundary line perceived by the vehicle at a second time point; the first time point is later than the second time point;

[0096] A first determining module 820 is configured to determine an initial error score corresponding to the first ego vehicle lane boundary line based on the first ego vehicle lane boundary line and the second ego vehicle lane boundary line;

[0097] A first adjustment module 830 is configured to perform multiple position adjustments on the first ego-vehicle lane boundary line and the second ego-vehicle lane boundary line, and obtain a candidate error score corresponding to the first ego-vehicle lane boundary line after each position adjustment;

[0098] A second determining module 840 is configured to determine a heading angle offset value of the ego vehicle lane boundary at a first time point based on the multiple candidate error scores corresponding to the first ego vehicle lane boundary after the multiple position adjustments and the initial error score;

[0099] The second adjustment module 850 is configured to adjust the first ego vehicle lane boundary line according to the heading angle offset value to obtain a target ego vehicle lane boundary line corresponding to the ego vehicle lane boundary at a first time point.

[0100] Optionally, the first adjustment module 830 is further configured to determine an initial heading angle offset value based on the initial error score; rotate the first self-vehicle lane boundary line and the second self-vehicle lane boundary line according to the initial heading angle offset value to adjust the positions of the first self-vehicle lane boundary line and the second self-vehicle lane boundary line, thereby obtaining the first self-vehicle lane boundary line after position adjustment and the second self-vehicle lane boundary line after position adjustment; determine a candidate error score corresponding to the first self-vehicle lane boundary line after position adjustment based on the first self-vehicle lane boundary line after position adjustment and the second self-vehicle lane boundary line after position adjustment; determine a candidate error score corresponding to the first self-vehicle lane boundary line after position adjustment based on the candidate error score corresponding to the first self-vehicle lane boundary line after the most recent position adjustment. A new initial heading angle offset value; obtaining the first ego lane boundary line after the most recent position adjustment as the new first ego lane boundary line and the second ego lane boundary line after the most recent position adjustment as the new second ego lane boundary line; returning to execute the steps of rotating the first ego lane boundary line and the second ego lane boundary line according to the initial heading angle offset value respectively to adjust the positions of the first ego lane boundary line and the second ego lane boundary line, obtaining the position-adjusted first ego lane boundary line and the position-adjusted second ego lane boundary line, until the number of position adjustments reaches the specified number, and obtaining the candidate error scores corresponding to the first ego lane boundary line after multiple position adjustments.

[0101] Optionally, the second determination module 840 is further used to obtain the smallest value from multiple candidate error scores and the initial error score as the error score to be processed; determine the heading angle offset value to be processed based on the error score to be processed; and filter the heading angle offset value to be processed to obtain the heading angle offset value of the vehicle lane boundary at the first time point.

[0102] Optionally, the second determination module 840 is further used to perform a weighted summation of the heading angle offset value to be processed and the reference heading angle offset value to obtain the heading angle offset value of the vehicle lane boundary at the first time point; the reference heading angle offset value is the heading angle offset value of the vehicle lane boundary at the second time point; the weight of the heading angle offset value to be processed is less than the weight of the reference heading angle offset value.

[0103] Optionally, the first self-vehicle lane boundary line includes multiple first boundary line points; the second self-vehicle lane boundary line includes a second boundary line point corresponding to each first boundary line point; the first determination module 820 is also used to determine the boundary line point lateral error corresponding to each first boundary line point based on the lateral distance difference between each first boundary line point and the corresponding second boundary line point; based on the average value of the boundary line point lateral errors corresponding to the multiple first boundary line points, determine the initial error score corresponding to the first self-vehicle lane boundary line.

[0104] Optionally, the acquisition module 810 is also used to obtain the first lane boundary line and the second lane boundary line perceived by the vehicle when the vehicle is traveling in a target road area on the highway; the target road area is the road area on the highway excluding the lane boundary line intersection area and the lane boundary line convergence area.

[0105] Optionally, the acquisition module 810 is also used to intercept the first self-vehicle lane boundary line from the first initial perception lane boundary line perceived by the self-vehicle, and intercept the second self-vehicle lane boundary line from the second initial perception lane boundary line perceived by the self-vehicle; wherein the first initial perception lane boundary line is the lane boundary line perceived for the self-vehicle lane boundary at the first time point; the second initial perception lane boundary line is the lane boundary line perceived for the self-vehicle lane boundary at the second time point; if the first self-vehicle lane boundary line and the second self-vehicle lane boundary line meet the specified lane conditions, the first self-vehicle lane boundary line and the second self-vehicle lane boundary line are obtained; the specified lane conditions include at least one of the following: the perceived lane widths perceived by the self-vehicle at the first time point and the second time point for the self-vehicle lane are both within a preset lane width range; the first boundary distance and the second boundary distance are both greater than the specified distance; the first boundary distance is the distance between the center of the self-vehicle and the first self-vehicle lane boundary line at the first time point; the second boundary distance is the distance between the center of the self-vehicle and the second self-vehicle lane boundary line at the second time point.

[0106] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and modules can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0107] In addition, the functions in the various embodiments of the present application may be integrated into a single processing module, or each module may exist physically separately, or two or more modules may be integrated into a single module. The aforementioned integrated modules may be implemented in the form of hardware or software functional modules.

[0108] On the other hand, the present application also provides a computer-readable storage medium, which stores program code. The program code can be called by a processor to execute the method described in the above method embodiment.

[0109] The computer-readable storage medium can be an electronic, magnetic, optical, or other physical storage device that contains or stores a programmable code that can be read by a computer. The computer-readable storage medium can be a non-transitory computer-readable storage medium. The computer-readable storage medium can have a storage space that stores program codes for performing any of the method steps described above. The program codes can be read from or written to one or more computer program products. The program codes can be compressed in an appropriate form, for example.

[0110] Finally, it should be noted that the above examples are only used to illustrate the technical solutions of the present application, and are not intended to limit the same. Although the present application has been described in detail with reference to the foregoing examples, those skilled in the art will understand that the technical solutions described in the foregoing examples can still be modified, or some of the technical features can be replaced by equivalent features. Such modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A lane adjustment method, characterized in that: The method comprises: Obtaining a first lane boundary line perceived by the vehicle at a first time point and a second lane boundary line perceived by the vehicle at a second time point, wherein the first time point is later than the second time point; determining an initial error score corresponding to the first ego-vehicle lane boundary line based on the first ego-vehicle lane boundary line and the second ego-vehicle lane boundary line; Performing multiple position adjustments on the first ego-vehicle lane boundary line and the second ego-vehicle lane boundary line to obtain a candidate error score corresponding to the first ego-vehicle lane boundary line after each position adjustment; Determining a heading angle offset value of the ego vehicle lane boundary at the first time point based on a plurality of candidate error scores corresponding to the first ego vehicle lane boundary line after the multiple position adjustments and the initial error score; The first own-vehicle lane boundary line is adjusted according to the heading angle offset value to obtain a target own-vehicle lane boundary line for the own-vehicle lane boundary at the first time point.

2. The method according to claim 1, characterized in that The performing multiple position adjustments on the first ego vehicle lane boundary line and the second ego vehicle lane boundary line to obtain a candidate error score corresponding to the first ego vehicle lane boundary line after each position adjustment includes: determining an initial heading angle offset value based on the initial error score; Rotating the first lane boundary line and the second lane boundary line according to the initial heading angle offset value to adjust the positions of the first lane boundary line and the second lane boundary line, thereby obtaining the first lane boundary line and the second lane boundary line after the positions are adjusted; Determining a candidate error score corresponding to the first ego vehicle lane boundary line after the position adjustment and the second ego vehicle lane boundary line after the position adjustment; Determining a new initial heading angle offset value based on a candidate error score corresponding to the first ego-vehicle lane boundary line after the most recent position adjustment; Obtaining the first ego vehicle lane boundary line after the most recent position adjustment as a new first ego vehicle lane boundary line and the second ego vehicle lane boundary line after the most recent position adjustment as a new second ego vehicle lane boundary line; Return to executing the step of rotating the first ego vehicle lane boundary line and the second ego vehicle lane boundary line according to the initial heading angle offset value to adjust the positions of the first ego vehicle lane boundary line and the second ego vehicle lane boundary line, and obtaining the position-adjusted first ego vehicle lane boundary line and the position-adjusted second ego vehicle lane boundary line, until the number of position adjustments reaches a specified number, and obtaining the candidate error scores corresponding to the first ego vehicle lane boundary line after the multiple position adjustments.

3. The method according to claim 1, characterized in that The determining, based on the multiple candidate error scores corresponding to the first ego vehicle lane boundary line after the multiple position adjustments and the initial error score, a heading angle offset value of the ego vehicle lane boundary at the first time point includes: Obtaining the smallest one from the plurality of candidate error scores and the initial error score as the error score to be processed; Determining a heading angle offset value to be processed based on the error score to be processed; The to-be-processed heading angle offset value is filtered to obtain a heading angle offset value of the lane boundary of the vehicle at the first time point.

4. The method according to claim 3, characterized in that The filtering process on the to-be-processed heading angle offset value to obtain the heading angle offset value of the lane boundary of the vehicle at the first time point includes: A weighted sum is performed on the to-be-processed heading angle offset value and the reference heading angle offset value to obtain the heading angle offset value of the lane boundary of the own vehicle at the first time point; the reference heading angle offset value is the heading angle offset value of the lane boundary of the own vehicle at the second time point; the weight of the to-be-processed heading angle offset value is less than the weight of the reference heading angle offset value.

5. The method according to claim 1, wherein The first lane boundary line includes a plurality of first boundary line points; the second lane boundary line includes a second boundary line point corresponding to each of the first boundary line points; The determining, based on the first lane boundary line and the second lane boundary line, an initial error score corresponding to the first lane boundary line includes: determining a boundary line point lateral error corresponding to each first boundary line point based on a lateral distance difference between each first boundary line point and the corresponding second boundary line point; An initial error score corresponding to the first own-vehicle lane boundary line is determined based on an average of boundary line point lateral errors corresponding to the plurality of first boundary line points.

6. The method according to claim 5, characterized in that The correspondence between the first boundary line point and the second boundary line point includes: The first boundary line point and the second boundary line point are on the same side of the vehicle, and the first longitudinal distance and the second longitudinal distance are the same; the first longitudinal distance is the longitudinal relative distance between the position of the vehicle and the first boundary line point at the first time point; the second longitudinal distance is the longitudinal relative distance between the position of the vehicle and the second boundary line point at the first time point.

7. The method according to claim 1, characterized in that The obtaining of a first lane boundary line perceived by the vehicle at a first time point and a second lane boundary line perceived by the vehicle at a second time point includes: When the vehicle is traveling in a target road area on the highway, a first lane boundary line and a second lane boundary line perceived by the vehicle are obtained; the target road area is a road area on the highway excluding a lane boundary line intersection area and a lane boundary line convergence area.

8. The method according to claim 1, characterized in that Before acquiring a first lane boundary line perceived by the vehicle at a first time point and a second lane boundary line perceived by the vehicle at a second time point, the method further includes: intercepting the first lane boundary line of the self-vehicle from the first initial perceived lane boundary line perceived by the self-vehicle, and intercepting the second lane boundary line of the self-vehicle from the second initial perceived lane boundary line perceived by the self-vehicle; wherein the first initial perceived lane boundary line is the lane boundary line perceived for the self-vehicle lane boundary at the first time point; and the second initial perceived lane boundary line is the lane boundary line perceived for the self-vehicle lane boundary at the second time point; The obtaining of a first lane boundary line of the vehicle sensed by the vehicle at a first time point and a second lane boundary line of the vehicle sensed at a second time point includes: If the first lane boundary line and the second lane boundary line satisfy the designated lane condition, obtaining the first lane boundary line and the second lane boundary line; The designated lane condition includes at least one of the following: The lane width sensed by the vehicle at the first time point and the second time point for the lane of the vehicle is both within a preset lane width range; Both the first boundary distance and the second boundary distance are greater than the specified distance; the first boundary distance is the distance between the center of the vehicle and the boundary line of the first lane at the first time point; the second boundary distance is the distance between the center of the vehicle and the boundary line of the second lane at the second time point.

9. A lane adjustment device, characterized in that: The device comprises: an acquisition module, configured to acquire a first lane boundary line perceived by the vehicle at a first time point and a second lane boundary line perceived by the vehicle at a second time point; the first time point being later than the second time point; a first determining module, configured to determine an initial error score corresponding to the first ego vehicle lane boundary line based on the first ego vehicle lane boundary line and the second ego vehicle lane boundary line; a first adjustment module, configured to perform multiple position adjustments on the first ego-vehicle lane boundary line and the second ego-vehicle lane boundary line, and obtain a candidate error score corresponding to the first ego-vehicle lane boundary line after each position adjustment; a second determining module, configured to determine a heading angle offset value of the ego vehicle lane boundary at the first time point based on a plurality of candidate error scores corresponding to the first ego vehicle lane boundary line after the multiple position adjustments and the initial error score; A second adjustment module is configured to adjust the first ego vehicle lane boundary line according to the heading angle offset value to obtain a target ego vehicle lane boundary line for the ego vehicle lane boundary at the first time point.

10. A vehicle, characterized in that: include: one or more processors; Memory; One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more applications are configured to execute the method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that The computer-readable storage medium stores program codes executable by a processor, and when the program codes are executed by the processor, the processor is caused to perform the method according to any one of claims 1 to 8.

Images