Lane line vector smoothing method, high-precision map drawing method, device and equipment

Abstract

This disclosure relates to a method for smoothing lane line vectors, a method for drawing high-precision maps, an apparatus, and a device. This disclosure obtains the speed limit information of the lane containing the target lane line vector pair to be smoothed. Based on the speed limit information and the relative distance between lane line vectors in the target lane line vector pair, it adaptively determines the smoothing start point on one lane line vector and the smoothing end point on the other lane line vector of the target lane line vector pair under different road scenarios. This eliminates the need for manual intervention and improves the smoothing efficiency of lane line vectors. Furthermore, based on the smoothing start and end points, multiple smoothing control points are determined. These control points control the curve curvature, generating a smooth curve between the smoothing start and end points to smoothly connect the two lane line vectors of the target lane line vector pair. This ensures the accuracy of the smoothed portion of the original lane line vectors while also smoothing the unsmoothed portion, improving the stability and comfort of vehicle control based on high-precision maps.

Description

Technical Field

[0001] This disclosure relates to the field of map technology, specifically to a method for smoothing lane line vectors, a method for drawing high-precision maps, an apparatus, and a device. Background Technology

[0002] With technological advancements, electronic maps are evolving from standard maps to high-precision maps. High-precision maps offer greater accuracy and finer details, incorporating not only road network data from standard maps but also lane network data, lane lines, and traffic signs. The primary applications of high-precision maps are in autonomous driving and high-precision positioning. Particularly in autonomous driving scenarios, high-precision maps serve as the foundational data supporting autonomous driving functions. To enhance vehicle control stability and comfort, lane line vectors on the map require smoothing.

[0003] However, the relevant technologies usually involve manually setting multiple smoothing control points on the lane line vector, and then generating a smooth curve between these multiple smoothing control points to achieve smoothing of the lane line vector. It can be seen that the smoothing control points need to be set manually, the accuracy needs to be improved, and the smoothing efficiency is low. Summary of the Invention

[0004] At least one embodiment of this disclosure provides a method, apparatus, electronic device, and storage medium for smoothing lane line vectors.

[0005] In a first aspect, embodiments of this disclosure propose a method for smoothing lane line vectors, the method comprising:

[0006] Obtain the target lane line vector pair to be smoothed and the speed limit information of the lane where the target lane line vector pair is located;

[0007] Based on speed limit information and the relative distance between lane line vectors in the target lane line vector pair, the smoothing start point on the first lane line vector of the target lane line vector pair and the smoothing end point on the second lane line vector of the target lane line vector pair are determined, wherein the first lane line vector and the second lane line vector are sequentially arranged along the lane travel direction.

[0008] Based on the smoothing start point and smoothing end point, multiple smoothing control points are determined;

[0009] Based on multiple smoothing control points, a smoothing curve is determined between the smoothing start point and the smoothing end point, and the smoothing curve smoothly connects the target lane line vector pairs.

[0010] Secondly, this disclosure also proposes a high-precision map drawing method, which includes:

[0011] Based on the lane line vector smoothing method described in the first aspect, the target lane line vector pair to be smoothed is smoothed to obtain the smoothed lane line vector.

[0012] High-precision maps are drawn based on smoothed lane line vectors.

[0013] Thirdly, embodiments of this disclosure also provide a lane line vector smoothing device, the device comprising:

[0014] The acquisition unit is used to acquire the target lane line vector pair to be smoothed and the speed limit information of the lane where the target lane line vector pair is located;

[0015] The first determining unit is used to determine the smoothing start point on the first lane line vector of the target lane line vector pair and the smoothing end point on the second lane line vector of the target lane line vector pair based on speed limit information and the relative distance between lane line vectors in the target lane line vector pair, wherein the first lane line vector and the second lane line vector are sequentially arranged along the lane travel direction.

[0016] The second determining unit is used to determine multiple smoothing control points based on the smoothing start point and the smoothing end point;

[0017] A smoothing unit is used to determine a smoothing curve between a smoothing start point and a smoothing end point based on multiple smoothing control points. The smoothing curve smoothly connects the target lane line vector pairs.

[0018] Fourthly, embodiments of this disclosure also propose an electronic device, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the lane line vector smoothing method as described in the first aspect or the steps of the high-precision map drawing method as described in the second aspect.

[0019] Fifthly, embodiments of this disclosure also provide a computer-readable storage medium, wherein the computer-readable storage medium stores a program or instructions that cause a computer to perform steps of the lane line vector smoothing method as described in the first aspect or steps of the high-precision map drawing method as described in the second aspect.

[0020] In a sixth aspect, embodiments of this disclosure also provide a computer program product, wherein the computer program product includes a computer program stored in a computer-readable storage medium, and at least one processor of a computer reads from and executes the computer program from the computer-readable storage medium, causing the computer to perform steps of the lane line vector smoothing method as described in the first aspect or steps of the high-precision map drawing method as described in the second aspect.

[0021] As can be seen, in at least one embodiment of this disclosure, for different road scenarios, by acquiring the target lane vector pair to be smoothed and the speed limit information of the lane where the target lane vector pair is located, the smoothing start point on one lane vector and the smoothing end point on the other lane vector of the target lane vector pair can be adaptively determined based on the speed limit information and the relative distance between the lane vectors in the target lane vector pair, without manual intervention, thus improving the smoothing efficiency of the lane vectors. Furthermore, based on the smoothing start point and the smoothing end point, multiple smoothing control points can be determined, and the curvature of the curve can be controlled using multiple smoothing control points to generate a smooth curve between the smoothing start point and the smoothing end point, smoothly connecting the two lane vectors of the target lane vector pair. It can be seen that the smoothing curve smooths the local jumps between the two lane vectors, while the smoothed parts on the two lane vectors that have not jumped are still retained. This not only ensures the accuracy of the smoothed parts in the original lane vectors, but also smooths the unsmoothed parts to meet the continuous curvature change rate, which can meet the smoothing requirements of autonomous driving for lane vectors in high-precision maps, thereby improving the stability and comfort of autonomous driving vehicle control based on high-precision maps. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings.

[0023] Figure 1 A flowchart illustrating a method for smoothing lane line vectors provided in an embodiment of this disclosure;

[0024] Figure 2 A schematic flowchart illustrating a process for obtaining target lane line vector pairs to be smoothed, provided in an embodiment of this disclosure;

[0025] Figure 3 A flowchart illustrating a process for determining whether the relative distance between lane line vectors in a candidate lane line vector pair satisfies a preset smoothing distance condition, provided in an embodiment of this disclosure.

[0026] Figure 4 A flowchart illustrating the process of determining a smoothing start point and a smoothing end point is provided for an embodiment of this disclosure.

[0027] Figure 5 A schematic diagram of a scenario for determining the smooth length provided in an embodiment of this disclosure;

[0028] Figure 6 A schematic diagram of another process for determining the smoothing start point and smoothing end point provided in an embodiment of this disclosure;

[0029] Figure 7 A schematic diagram of a scenario for determining a smoothing start point, a smoothing end point, a first smoothing control point, and a second smoothing control point, provided as an embodiment of this disclosure;

[0030] Figure 8 A schematic diagram of a lane line vector smoothing device provided in an embodiment of this disclosure;

[0031] Figure 9 This is an exemplary block diagram of an electronic device provided in an embodiment of the present disclosure. Detailed Implementation

[0032] To better understand the above-described objectives, features, and advantages of this disclosure, the present disclosure will be further described in detail below with reference to the accompanying drawings and embodiments. It is to be understood that the described embodiments are only some, not all, of the embodiments of this disclosure. The specific embodiments described herein are merely for explaining this disclosure and are not intended to limit it. All other embodiments obtained by those skilled in the art based on the described embodiments of this disclosure are within the scope of protection of this disclosure.

[0033] It should be noted that in this article, relational terms such as “first” and “second” are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0034] In the map making process, lane line vectors are used to construct lane lines in the map. One lane line vector represents one lane line in the map. In some scenarios, such as turning, intersections, and lane changes, lane line vectors are not straight line vectors but curved vectors. Curved vectors may have unevenness issues, and even straight line vectors may have unevenness issues due to different accuracy requirements.

[0035] In different levels of intelligent driving scenarios, such as driverless driving, assisted driving, driver assistance, highly automated driving, and fully automated driving, lane line vectors in the map are required. If the lane line vectors are not smooth, it may lead to abnormal vehicle conditions such as swaying left and right, unstable vehicle control, and poor driving comfort. Therefore, during the production of lane line vectors, it is necessary to smooth them to ensure that the problem of uneven lane line vectors has been resolved when creating the map.

[0036] Some related technologies smooth lane line vectors by manually setting control points on the lane line vectors and then generating a Bézier curve that passes through all control points, thus ensuring the smoothness of the curve. However, since the control points are set manually, both accuracy and efficiency are subject to human intervention.

[0037] Therefore, embodiments of this disclosure provide a method, apparatus, electronic device, or storage medium for smoothing lane line vectors. For different road scenarios, by acquiring the target lane line vector pair to be smoothed and the speed limit information of the lane containing the target lane line vector pair, the smoothing start point on one lane line vector and the smoothing end point on the other lane line vector of the target lane line vector pair can be adaptively determined based on the speed limit information and the relative distance between the lane line vectors in the target lane line vector pair, without manual intervention, thus improving the smoothing efficiency of lane line vectors. Furthermore, multiple smoothing control points can be determined based on the smoothing start point and the smoothing end point. These multiple smoothing control points control the curve curvature, generating a smooth curve between the smoothing start point and the smoothing end point, smoothly connecting the two lane line vectors of the target lane line vector pair. It can be seen that the smoothing curve smooths the local jumps between the two lane line vectors, while retaining the smoothed portions without jumps on the two lane line vectors. This ensures the accuracy of the smoothed portions in the original lane line vectors and also smooths the unsmoothed portions to satisfy the continuous rate of curvature change, meeting the smoothing requirements of high-precision map lane line vectors for autonomous driving, thereby improving the stability and comfort of autonomous driving vehicle control based on high-precision maps.

[0038] Figure 1 This is a flowchart illustrating a lane line vector smoothing method provided in an embodiment of the present disclosure. The execution subject of the lane line vector smoothing method is an electronic device, including but not limited to in-vehicle devices, smartphones, PDAs, tablets, wearable devices with displays, desktop computers, laptops, all-in-one computers, smart home devices, servers, etc. The server can be a standalone server or a cluster of multiple servers, and can include servers built locally and servers set up in the cloud.

[0039] like Figure 1 As shown, the smoothing method for the lane line vector may include, but is not limited to, steps 101 to 104:

[0040] In step 101, the target lane line vector pair to be smoothed and the speed limit information of the lane where the target lane line vector pair is located are obtained.

[0041] In this embodiment, the target lane line vector pair to be smoothed consists of two lane line vectors, denoted as the first lane line vector and the second lane line vector. The first lane line vector and the second lane line vector are adjacent vectors, meaning there are no other lane line vectors between them. Furthermore, the first lane line vector and the second lane line vector are arranged sequentially along the lane travel direction. That is, if a vehicle is traveling in the lane containing the first lane line vector and the second lane line vector, it will pass the first lane line vector first and then the second lane line vector. Therefore, the sequential relationship between the first lane line vector and the second lane line vector can be determined based on the lane line topology in the map.

[0042] In this embodiment, after obtaining the target lane line vector pair to be smoothed, the lane where the target lane line vector pair is located can be determined, and then the speed limit information of the lane can be obtained. The speed limit information can be obtained from the ground speed limit number on the lane or the roadside speed limit sign of the lane.

[0043] In this embodiment, after obtaining the target lane line vector pair to be smoothed, the relative distance between the lane line vectors in the target lane line vector pair can be determined. The relative distance between the lane line vectors in the target lane line vector pair can be understood as the distance between the end point of the first lane line vector in the target lane line vector pair and the starting point of the second lane line vector in the target lane line vector pair.

[0044] In step 102, based on the speed limit information and the relative distance between lane line vectors in the target lane line vector pair, the smoothing start point on the first lane line vector of the target lane line vector pair and the smoothing end point on the second lane line vector of the target lane line vector pair are determined, wherein the first lane line vector and the second lane line vector are sequentially arranged along the lane travel direction.

[0045] In this embodiment, the smoothing starting point is located on the first lane line vector. Therefore, the first line segment between the smoothing starting point and the end point of the first lane line vector will be smoothed into a curve during subsequent smoothing processing.

[0046] In this embodiment, the smoothing endpoint is located on the second lane line vector. Therefore, the second line segment between the starting point and the smoothing endpoint of the second lane line vector will be smoothed into a curve during subsequent smoothing processing.

[0047] In this process, the first and second line segments can be smoothed to form a smooth curve that replaces the first and second line segments, thus smoothly connecting the target lane line vectors to obtain the smoothed lane line vectors. The specific smoothing process is described below.

[0048] In step 103, multiple smoothing control points are determined based on the smoothing start point and the smoothing end point.

[0049] The continuity of a curve is an important property, including G0 continuity, G1 continuity, G2 continuity, G3 continuity, ..., Gn continuity. G0 continuity, also known as point continuity, refers to the continuity of the curve equation, meaning there are no discontinuities. G1 continuity, also known as tangent continuity, refers to the continuity of the first derivative of the curve equation, meaning it satisfies G0 continuity and is smooth without sharp angles. G2 continuity, also known as curvature continuity, refers to the continuity of the second derivative of the curve equation, meaning it satisfies G1 continuity and the radius of curvature changes continuously without abrupt changes. G3 continuity, also known as curvature tangent continuity, refers to the continuity of the third derivative of the curve equation, meaning it satisfies G2 continuity and the radius of curvature changes smoothly without sharp angles. Gn continuity refers to the continuity of the Nth derivative of the curve equation.

[0050] Lane vectors used for high-precision maps need to satisfy at least G2 continuity. To draw a G2 continuous curve, in addition to specifying the start and end points of the curve, one control point is also required. The control point is used to control the curvature of the curve. Lane vectors used for autonomous driving need to satisfy at least G3 continuity. To draw a G3 continuous curve, in addition to specifying the start and end points of the curve, two control points are also required. And so on, for a curve that satisfies Gn continuity, in addition to specifying the start and end points of the curve, n-1 control points are also required.

[0051] In this embodiment, in order for the smoothed lane line vector to be used for autonomous driving, the smoothed lane line vector needs to satisfy at least G3 continuity. That is, in addition to specifying the start and end points of the smooth curve, at least two (i.e. more) smooth control points are also required to control the curvature of the smooth curve.

[0052] In step 104, a smoothing curve is determined between the smoothing start point and the smoothing end point based on multiple smoothing control points, and the smoothing curve smoothly connects the target lane line vector pairs.

[0053] In this embodiment, multiple smoothing control points are used together to control the curvature of the smoothing curve so that the smoothing curve can smoothly connect the first lane line vector and the second lane line vector in the target lane line vector pair to be smoothed, so that the first lane line vector, the smoothing curve and the second lane line vector are merged into a continuous lane line vector, which is called the smoothed lane line vector. The smoothed lane line vector is at least G3 continuous lane line vector, which can meet the smoothing requirements of autonomous driving for lane line vectors in high-precision maps, thereby improving the stability and comfort of autonomous driving relying on high-precision maps for vehicle control.

[0054] In some embodiments, the generated smooth curve is a Bézier curve. After determining the start point, end point and multiple control points of the Bézier curve, generating the Bézier curve is a mature technology in the field of computer science, and will not be elaborated further.

[0055] As can be seen, in the above embodiments, for different road scenarios, by acquiring the target lane vector pair to be smoothed and the speed limit information of the lane where the target lane vector pair is located, the smoothing start point on one lane vector and the smoothing end point on the other lane vector of the target lane vector pair can be adaptively determined based on the speed limit information and the relative distance between the lane vectors in the target lane vector pair, without manual intervention, thus improving the smoothing efficiency of the lane vectors. Furthermore, based on the smoothing start point and the smoothing end point, multiple smoothing control points can be determined, and the curvature of the curve can be controlled using multiple smoothing control points to generate a smooth curve between the smoothing start point and the smoothing end point, smoothly connecting the two lane vectors of the target lane vector pair. It can be seen that the smoothing curve smooths the local jumps between the two lane vectors, while the smoothed parts on the two lane vectors that have not jumped are still retained. This not only ensures the accuracy of the smoothed parts in the original lane vectors, but also smooths the unsmoothed parts to meet the continuous curvature change rate, which can meet the smoothing requirements of autonomous driving for lane vectors in high-precision maps, thereby improving the stability and comfort of autonomous driving vehicle control based on high-precision maps.

[0056] Based on the above embodiments, Figure 1 Step 101, "obtaining the target lane line vector pair to be smoothed", includes, for example... Figure 2 Steps 201 and 202 are shown below:

[0057] In step 201, obtain any two adjacent but unconnected lane line vectors.

[0058] In this embodiment, the first lane line vector and the second lane line vector in the target lane line vector pair to be smoothed are two adjacent lane line vectors. That is, there are no other lane line vectors between the first lane line vector and the second lane line vector. If the two lane line vectors are not adjacent, then these two lane line vectors do not need to be smoothed. Otherwise, it would turn two separate lane lines in the real world into one lane line, which is inconsistent with the real world.

[0059] Therefore, in this embodiment, any two adjacent and unconnected lane line vectors are obtained as candidate lane line vector pairs. There can be multiple candidate lane line vector pairs. For each candidate lane line vector pair, step 202 is executed to determine whether the candidate lane line vector pair is the target lane line vector pair to be smoothed.

[0060] In step 202, if the relative distance between two lane line vectors meets the preset smoothing distance condition, then the two lane line vectors are determined as the target lane line vector pair to be smoothed.

[0061] In this embodiment, the preset smoothing distance condition indicates that the two lane line vectors should be connected vectors. For any candidate lane line vector pair, the relative distance between the two lane line vectors in the candidate lane line vector pair is determined. Then, it can be determined whether the relative distance satisfies the preset smoothing distance condition. If the preset smoothing distance condition is satisfied, it means that the two lane line vectors in the candidate lane line vector pair should be connected vectors, and these two lane line vectors need to be smoothly connected. Therefore, the candidate lane line vector pair is determined as the target lane line vector pair to be smoothed.

[0062] Figure 3 A flowchart illustrating how to determine the relative distance between lane line vectors in a candidate lane line vector alignment meets a preset smoothing distance condition, as provided in this embodiment of the disclosure, includes steps 301 and 303:

[0063] In step 301, the relative distance between the target adjacent shape points on the two lane line vectors is determined, wherein the target adjacent shape points are the first and last adjacent shape points on the two lane line vectors.

[0064] In this embodiment, a lane line vector is typically composed of multiple line segments (each segment having the same length), and the two endpoints of each line segment are called the shape points of the lane line vector. The relative distances between the target adjacent shape points of the first lane line vector and the target adjacent shape points of the second lane line vector of the candidate lane line vector pair are determined. The first lane line vector and the second lane line vector are ordered sequentially along the lane travel direction. The target adjacent shape point of the first lane line vector is its endpoint, and the target adjacent shape point of the second lane line vector is its starting point.

[0065] In step 302, the longitudinal distance and lateral distance corresponding to the relative distance are determined, wherein the longitudinal distance is the projected length of the relative distance along the direction of the target adjacent shape point on the first lane line vector, and the lateral distance is the projected length of the relative distance in the vertical direction along the direction of the target adjacent shape point on the first lane line vector.

[0066] In this embodiment, the relative distance between candidate lane line vectors and center lane line vectors is decomposed into longitudinal distance and lateral distance. The direction of the target adjacent shape point on the first lane line vector is the end point direction of the first lane line vector, and the end point direction is tangent to the first lane line vector.

[0067] In step 303, if the longitudinal distance is less than or equal to the longitudinal distance threshold and the lateral distance is greater than or equal to the lateral distance threshold, then the relative distance is determined to meet the preset smooth distance condition, which includes the longitudinal distance threshold and the lateral distance threshold.

[0068] In this embodiment, the smoothing distance condition includes two: a longitudinal distance threshold and a lateral distance threshold. The longitudinal distance threshold is the maximum longitudinal distance between two connected lane line vectors. If the longitudinal distance is greater than the longitudinal distance threshold, it means that the two lane line vectors are definitely not connected. The lateral distance threshold is the minimum lateral distance that causes vehicle control instability. If the lateral distance is less than the lateral distance threshold, it means that the lateral distance will not cause vehicle control instability, that is, it will not cause the vehicle to sway left and right, or although it will cause the vehicle to sway left and right, the swaying amplitude is small and imperceptible to the user.

[0069] It is evident that if the longitudinal distance obtained from the relative distance decomposition of the lane line vectors in the candidate lane line vector pair is less than or equal to the longitudinal distance threshold, it indicates that the two lane line vectors in the candidate lane line vector pair should be connected. At the same time, if the lateral distance obtained from the relative distance decomposition of the candidate lane line vector pair is greater than or equal to the lateral distance threshold, it indicates that the two lane line vectors in the candidate lane line vector pair cause vehicle control instability. Therefore, it is determined that the relative distance of the candidate lane line vector pair meets the preset smoothing distance condition, that is, the candidate lane line vector pair needs to be smoothed. Thus, the candidate lane line vector pair is determined as the target lane line vector pair to be smoothed.

[0070] Based on the above embodiments, Figure 1 Step 102, "based on speed limit information and the relative distance between lane vectors in the target lane vector pair, determines the smoothing start point on the first lane vector of the target lane vector pair and the smoothing end point on the second lane vector of the target lane vector pair," includes... Figure 4 Steps 401 to 404 are shown below:

[0071] In step 401, the target turning radius corresponding to the speed limit information is determined based on the speed limit information and the preset correspondence between the road speed limit and the range of turning radius values.

[0072] In this embodiment, the preset correspondence between road speed limits and the range of turning radii is shown in Table 1 below.

[0073] Table 1. Preset Correspondence Between Road Speed ​​Limits and Turning Radius Ranges

[0074]

[0075] For example, based on Table 1, if the speed limit information is 80km / h, the turning radius ranges from 250m to 400m. The target turning radius corresponding to the speed limit information is determined to be 400m. Alternatively, any value within the range of turning radius values ​​can be selected as the target turning radius based on the actual scenario to ensure that the vehicle can smoothly pass through the smooth curve after subsequent smoothing processing to generate a smooth curve.

[0076] In step 402, the projected length of the relative distance in the direction perpendicular to the end point of the first lane line vector is determined as the target lateral distance.

[0077] In this embodiment, the endpoint direction of the first lane line vector is tangent to the first lane line vector. Therefore, the relative distance between the first lane line vector and the second lane line vector in the target lane line vector pair is the projected length of the first lane line vector's endpoint direction as the target lateral distance.

[0078] In step 403, the smoothing length is determined based on the target turning radius and the target lateral distance.

[0079] For example, Figure 5 This is a schematic diagram of a scenario for determining a smooth length according to an embodiment of the present disclosure. Figure 5 In the diagram, the first lane line vector and the second lane line vector are two lane line vectors in the target lane line vector pair to be smoothed. The first lane line vector and the second lane line vector are adjacent vectors, that is, there are no other lane line vectors between the first lane line vector and the second lane line vector. Furthermore, the first lane line vector and the second lane line vector are sequentially arranged along the direction of lane travel. In other words, if a vehicle is traveling in the lane containing the first lane line vector and the second lane line vector, it will pass the first lane line vector first and then the second lane line vector.

[0080] Figure 5 In the diagram, point B is the endpoint of the first lane line vector, and the direction of point B is the arrow direction of the first lane line vector. The projection of the relative distance between the first lane line vector and the second lane line vector onto the direction of point B is the longitudinal distance. Figure 5 The longitudinal distance is zero. The projection length of the relative distance between the first lane line vector and the second lane line vector in the direction perpendicular to point B is the target lateral distance. Figure 5 In this context, H represents the target lateral distance, R represents the target turning radius corresponding to the speed limit information of the lane where the target lane line vector is located, and L represents the smoothing length. The smoothing length L satisfies the following:

[0081] sinθ=0.5L / R; cosθ=(R-0.5H) / R.

[0082] Therefore, the formula for calculating the smoothing length L can be obtained as follows:

[0083] In step 404, based on the smoothing length, the smoothing start point on the first lane line vector of the target lane line vector pair and the smoothing end point on the second lane line vector of the target lane line vector pair are determined.

[0084] For example, Figure 5In the diagram, the smoothing starting point determined on the first lane line vector is point A, and the smoothing ending point determined on the second lane line vector is point D. The sum of the first length between point A and the ending point B of the first lane line vector, and the second length between the starting point C of the second lane line vector and point D, is the smoothing length L. Figure 5 In this embodiment, the first length and the second length are the same, both being 0.5L. In some embodiments, the first length and the second length may be different, but the sum of the first length and the second length must be the smooth length L.

[0085] As can be seen, in this embodiment, if the target turning radius R is different, the smoothing starting point determined on the first lane line vector will be different, and the smoothing ending point determined on the second lane line vector will be different. The target turning radius R is determined by the speed limit information of the lane to which the target lane line vector is located. Different speed limit information corresponds to different target turning radii R. Different speed limit information can be understood as different road scenarios, such as urban expressways, highways and other road scenarios. That is, for different road scenarios, this embodiment can adaptively determine the smoothing length, and then adaptively determine the smoothing starting point and smoothing ending point without manual intervention, thereby improving the smoothing efficiency of lane line vectors and thus improving the production efficiency and accuracy of high-precision map lane line vectors.

[0086] Based on the above embodiments, Figure 4 Step 404, "Based on the smoothing length, determine the smoothing start point on the first lane vector of the target lane vector pair, and determine the smoothing end point on the second lane vector of the target lane vector pair," includes, for example: Figure 6 Steps 601 and 602 are shown below:

[0087] In step 601, taking the end point of the first lane line vector as the starting position, a smoothing starting point is determined on the first lane line vector in the opposite direction.

[0088] For example, Figure 7 This is a schematic diagram of a scenario for determining a smoothing start point, a smoothing end point, a first smoothing control point, and a second smoothing control point, provided by an embodiment of this disclosure. Figure 7 In the diagram, the first lane line vector is OB. Taking the end point B of the first lane line vector OB as the starting position, a smooth starting point A is determined on the first lane line vector OB in the opposite direction.

[0089] In step 602, starting from the beginning of the second lane line vector, a smooth endpoint is determined along the direction of the second lane line vector.

[0090] For example, Figure 7In the diagram, the second lane line vector is CE. Starting from the starting point C of the second lane line vector CE, a smooth endpoint D is determined along the direction of the second lane line vector CE.

[0091] The sum of the length between the smoothing starting point A and the ending point B of the first lane line vector, and the length between the starting point C and the smoothing ending point D of the second lane line vector, is the smoothing length. In some embodiments, the lengths of line segments AB and CD are both half of the smoothing length.

[0092] Based on the above embodiments, Figure 1 Step 103, "determine multiple smoothing control points based on the smoothing start point and smoothing end point," specifically refers to:

[0093] Starting from the smoothing start point, extend the smoothing length by half along the direction of the smoothing start point to obtain the first smoothing control point; starting from the smoothing end point, extend the smoothing length by half in the opposite direction of the smoothing end point to obtain the second smoothing control point.

[0094] For example, Figure 7 In the process, the direction of the smoothing starting point A is tangent to the first lane line vector OB. Extending half the smoothing length along the direction of the smoothing starting point A yields the first smoothing control point B'. The direction of the smoothing ending point D is tangent to the second lane line vector CE. Extending half the smoothing length in the opposite direction of the smoothing ending point D yields the second smoothing control point C'.

[0095] For example, in Figure 5 In the process, since the first lane line vector is a straight line, the first smoothing control point coincides with the end point B of the first lane line vector. Similarly, since the second lane line vector is a straight line, the second smoothing control point coincides with the starting point C of the second lane line vector.

[0096] In some embodiments, Figure 1 One implementation of step 104, "determining a smooth curve between the smoothing start point and the smoothing end point based on multiple smoothing control points," is as follows: Figure 7 In this method, based on the first smoothing control point B' and the second smoothing control point C', and taking the smoothing start point A as the start point of the Bézier curve and the smoothing end point D as the end point of the Bézier curve, a third-order Bézier curve (such as...) can be generated. Figure 7 As shown in the dashed curve, the dashed curve smoothly connects the first lane line vector OB and the second lane line vector CE. The dashed curve replaces line segment AB in the first lane line vector OB and line segment CD in the second lane line vector CE, so that OA, the dashed curve and DE are merged into a continuous lane line vector OADE, which is called the merged lane line vector.

[0097] It is evident that lane line vectors used for autonomous driving must at least satisfy G3 continuity (i.e., curvature tangency continuity). The third derivative of the curve equation of a third-order Bézier curve is continuous, i.e., curvature tangency continuity, satisfying G3 continuity. This ensures that the fused lane line vector is a lane line vector that satisfies G3 continuity. Specifically, the fused lane line vector smooths the local transitions (i.e., AB and CD) between the two lane line vectors, while retaining the smoothed portions (i.e., OA and DE) on the two lane line vectors that do not transition. This not only ensures the accuracy of the smoothed portions in the original lane line vectors but also smooths the unsmoothed portions to satisfy the curvature change rate continuity. This meets the smoothing requirements (G3 continuity) of lane line vectors in high-precision maps for autonomous driving, thereby improving the stability and comfort of autonomous driving vehicle control based on high-precision maps.

[0098] Based on the above embodiments, Figure 1 After "determine the smooth curve" in step 104, the smoothing method for lane line vectors also includes the following steps:

[0099] Based on the preset point spacing or number of points, the smooth curve is sampled to obtain multiple sampling points as points on the smooth curve.

[0100] Lane vectors are typically composed of multiple line segments (each of which is of equal length). The two endpoints of each line segment are called the shape points of the lane vector. Considering that a smooth curve is a newly generated curve, and that multiple line segments are needed to represent the curve for map creation, this embodiment samples the smooth curve, dividing it into multiple line segments. The sampling points can be understood as the endpoints of these line segments (also called the shape points of the lane vector). The sampling basis can be a preset shape point spacing, which can be the length between any two adjacent shape points in the lane vector. The shape point spacing can also be a configurable item, with its value manually configured. The sampling basis can also be the number of shape points, which can also be a configurable item, with its value manually configured. The number of shape points can also be calculated using the shape point spacing and the length of the equilibrium curve: Number of shape points = (Length of equilibrium curve / Shape point spacing) - 1.

[0101] In some embodiments, this disclosure also proposes a high-precision map drawing method, including: smoothing the target lane line vector pair to be smoothed based on the lane line vector smoothing method disclosed in the above embodiments to obtain smoothed lane line vectors; and then drawing a high-precision map based on the smoothed lane line vectors. The drawing method is a mature technology in the field and will not be described in detail here.

[0102] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art will understand that the embodiments of this disclosure are not limited to the described order of actions, because according to the embodiments of this disclosure, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art will understand that the embodiments described in the specification are all optional embodiments.

[0103] Figure 8 This diagram illustrates a lane line vector smoothing device provided in an embodiment of this disclosure. This smoothing device can be applied to electronic devices, including but not limited to in-vehicle devices, smartphones, PDAs, tablets, wearable devices with displays, desktop computers, laptops, all-in-one computers, smart home devices, and servers. The server can be a standalone server or a cluster of multiple servers, and can include locally located servers and cloud-based servers. The lane line vector smoothing device provided in this disclosure can execute the processing flow provided in various embodiments of the lane line vector smoothing method, such as... Figure 8 As shown, the lane line vector smoothing device includes, but is not limited to: acquisition unit 801, first determination unit 802, second determination unit 803, and smoothing unit 804. The functions of each unit are described below:

[0104] The acquisition unit 801 is used to acquire the target lane line vector pair to be smoothed and the speed limit information of the lane where the target lane line vector pair is located;

[0105] The first determining unit 802 is used to determine the smoothing start point on the first lane line vector of the target lane line vector pair and the smoothing end point on the second lane line vector of the target lane line vector pair based on speed limit information and the relative distance between lane line vectors in the target lane line vector pair, wherein the first lane line vector and the second lane line vector are sequentially arranged along the lane forward direction.

[0106] The second determining unit 803 is used to determine multiple smoothing control points based on the smoothing start point and the smoothing end point;

[0107] Smoothing unit 804 is used to determine a smoothing curve between a smoothing start point and a smoothing end point based on multiple smoothing control points, the smoothing curve smoothly connecting the target lane line vector pairs.

[0108] In some embodiments, the acquisition unit 801 acquires the target lane line vector pair to be smoothed, including:

[0109] Get any two adjacent but unconnected lane line vectors;

[0110] If the relative distance between two lane line vectors meets the preset smoothing distance condition, then the two lane line vectors are determined as the target lane line vector pair to be smoothed.

[0111] In some embodiments, the first determining unit 802 determines that the relative distance between two lane line vectors satisfies a preset smooth distance condition, including:

[0112] Determine the relative distance between adjacent shape points on two lane line vectors, where adjacent shape points are the first and last adjacent shape points on the two lane line vectors;

[0113] Determine the longitudinal distance and lateral distance corresponding to the relative distance, where the longitudinal distance is the projected length of the relative distance along the direction of the target adjacent shape point on the first lane line vector, and the lateral distance is the projected length of the relative distance in the direction perpendicular to the direction of the target adjacent shape point on the first lane line vector.

[0114] If the vertical distance is less than or equal to the vertical distance threshold and the horizontal distance is greater than or equal to the horizontal distance threshold, then the relative distance is determined to meet the preset smooth distance condition, which includes the vertical distance threshold and the horizontal distance threshold.

[0115] In some embodiments, the first determining unit 802 is configured to:

[0116] Based on speed limit information and the preset correspondence between road speed limits and turning radius ranges, the target turning radius corresponding to the speed limit information is determined.

[0117] The projected length of the relative distance in the direction perpendicular to the end point of the first lane line vector is determined as the target lateral distance;

[0118] The smoothing length is determined based on the target turning radius and the target lateral distance;

[0119] Based on the smoothing length, the smoothing start point on the first lane line vector of the target lane line vector pair is determined, and the smoothing end point on the second lane line vector of the target lane line vector pair is determined.

[0120] In some embodiments, the first determining unit 802 determines, based on the smoothing length, a smoothing start point on the first lane vector of the target lane vector pair and a smoothing end point on the second lane vector of the target lane vector pair, including:

[0121] Starting from the end point of the first lane line vector, determine the smoothing starting point on the first lane line vector in the opposite direction.

[0122] Starting from the beginning of the second lane line vector, determine the smooth endpoint along the direction of the second lane line vector;

[0123] The smoothing length is the sum of the length between the smoothing start point and the end point of the first lane line vector and the length between the start point and the smoothing end point of the second lane line vector.

[0124] In some embodiments, the second determining unit 803 is configured to:

[0125] Starting from the smoothing start point, extend the smoothing length by half along the direction of the smoothing start point to obtain the first smoothing control point;

[0126] Starting from the smoothing endpoint, extend the smoothing length by half in the opposite direction of the smoothing endpoint to obtain the second smoothing control point.

[0127] In some embodiments, the smoothing unit 804 is used for:

[0128] Based on the first and second smoothing control points, the smoothing start point is taken as the start point of the Bézier curve and the smoothing end point is taken as the end point of the Bézier curve to generate a third-order Bézier curve.

[0129] In some embodiments, the lane line vector smoothing device further includes:

[0130] The sampling unit is used to sample the smooth curve based on a preset point spacing or number of points, and obtain multiple sampling points as points on the smooth curve.

[0131] As can be seen, in at least one embodiment of the lane line vector smoothing device disclosed herein, for different road scenarios, by acquiring the target lane line vector pair to be smoothed and the speed limit information of the lane where the target lane line vector pair is located, the smoothing start point on one lane line vector and the smoothing end point on the other lane line vector of the target lane line vector pair can be adaptively determined based on the speed limit information and the relative distance between the lane line vectors in the target lane line vector pair under different road scenarios, without manual intervention, thus improving the smoothing efficiency of lane line vectors. Furthermore, based on the smoothing start point and the smoothing end point, multiple smoothing control points can be determined, and the curvature of the curve can be controlled using multiple smoothing control points to generate a smooth curve between the smoothing start point and the smoothing end point, smoothly connecting the two lane line vectors of the target lane line vector pair. It can be seen that the smoothing curve smooths the local jumps between the two lane line vectors, while the smoothed parts on the two lane line vectors that have not jumped are still retained. This not only ensures the accuracy of the smoothed parts in the original lane line vectors, but also smooths the unsmoothed parts to meet the continuous curvature change rate, which can meet the smoothing requirements of autonomous driving for lane line vectors in high-precision maps, thereby improving the stability and comfort of autonomous driving vehicle control based on high-precision maps.

[0132] Figure 9This is an exemplary block diagram of an electronic device provided in an embodiment of this disclosure. Figure 9 As shown, the electronic device includes a memory 901, a processor 902, and a computer program stored on the memory 901. It is understood that the memory 901 in this embodiment may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory.

[0133] In some implementations, memory 901 stores elements such as executable modules or data structures, or subsets thereof, or extended sets thereof: operating systems and applications.

[0134] The operating system includes various system programs, such as the framework layer, core library layer, and driver layer, used to implement various basic tasks and handle hardware-based tasks. The application programs include various applications, such as media players and browsers, used to implement various application tasks. The program implementing the lane line vector smoothing method or high-precision map drawing method provided in the embodiments of this disclosure can be included in the application programs.

[0135] In this embodiment of the disclosure, at least one processor 902 executes the steps of the lane line vector smoothing method or high-precision map drawing method embodiments provided in this disclosure by calling a program or instruction stored in at least one memory 901, specifically, a program or instruction stored in an application.

[0136] The lane line vector smoothing method or high-precision map drawing method provided in this disclosure can be applied to, or implemented by, the processor 902. The processor 902 can be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by integrated logic circuits in the hardware of the processor 902 or by instructions in software form. The processor 902 can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor.

[0137] The steps of the lane line vector smoothing method or high-precision map drawing method provided in this disclosure can be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory 901, and processor 902 reads the information in memory 901 and combines it with hardware to complete the steps of the method.

[0138] This disclosure also proposes a computer-readable storage medium that stores a program or instructions that cause a computer to perform steps, such as those in embodiments of a lane line vector smoothing method or a high-precision map drawing method. To avoid repetition, these steps will not be repeated here. The computer-readable storage medium can be a non-transitory computer-readable storage medium.

[0139] This disclosure also proposes a computer program product comprising a computer program stored in a computer-readable storage medium, which may be a non-transitory computer-readable storage medium. At least one processor of a computer reads and executes the computer program from the computer-readable storage medium, causing the computer to perform steps, such as those in embodiments of a lane line vector smoothing method or a high-precision map drawing method, which will not be repeated here to avoid repetition.

[0140] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0141] Those skilled in the art will understand that although some embodiments described herein include certain features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of this disclosure and form different embodiments.

[0142] Those skilled in the art will understand that the descriptions of the various embodiments have different focuses, and for parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0143] Although embodiments of the present disclosure have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present disclosure, and all such modifications and variations fall within the scope defined by the appended claims.

Claims

1. A method for smoothing lane line vectors, the method comprising: Obtain the target lane line vector pair to be smoothed and the speed limit information of the lane where the target lane line vector pair is located; Based on the speed limit information and the preset correspondence between the road speed limit and the turning radius range, the target turning radius corresponding to the speed limit information is determined; the projection length of the relative distance between the first lane line vector and the second lane line vector in the target lane line vector pair in the direction perpendicular to the end point of the first lane line vector is determined as the target lateral distance; based on the target turning radius and the target lateral distance, the smoothing length is determined; taking the end point of the first lane line vector as the starting position, along the opposite direction of the first lane line vector, a smoothing starting point is determined on the first lane line vector; taking the starting point of the second lane line vector as the starting position, along the direction of the second lane line vector, a smoothing ending point is determined on the second lane line vector, wherein the sum of the length between the smoothing starting point and the end point of the first lane line vector and the length between the starting point of the second lane line vector and the smoothing ending point is the smoothing length, and the first lane line vector and the second lane line vector are sequentially arranged along the lane forward direction; Based on the smoothing start point and the smoothing end point, multiple smoothing control points are determined; Based on the plurality of smoothing control points, a smoothing curve is determined between the smoothing start point and the smoothing end point, the smoothing curve smoothly connecting the target lane line vector pairs.

2. The method according to claim 1, wherein, The process of obtaining the target lane line vector pair to be smoothed includes: Get any two adjacent but unconnected lane line vectors; If the relative distance between the two lane line vectors meets the preset smoothing distance condition, then the two lane line vectors are determined to be the target lane line vector pair to be smoothed.

3. The method according to claim 2, wherein, The relative distance between the two lane line vectors satisfies a preset smooth distance condition, including: Determine the relative distance between adjacent target shape points on the two lane line vectors, wherein the adjacent target shape points are the first and last adjacent shape points on the two lane line vectors; Determine the longitudinal distance and lateral distance corresponding to the relative distance, wherein the longitudinal distance is the projected length of the relative distance along the direction of the target adjacent shape point on the first lane line vector, and the lateral distance is the projected length of the relative distance along the direction perpendicular to the direction of the target adjacent shape point on the first lane line vector; If the longitudinal distance is less than or equal to the longitudinal distance threshold and the lateral distance is greater than or equal to the lateral distance threshold, then the relative distance is determined to meet a preset smooth distance condition, which includes the longitudinal distance threshold and the lateral distance threshold.

4. The method according to claim 1, wherein, The determination of multiple smoothing control points based on the smoothing start point and the smoothing end point includes: Starting from the smoothing starting point, extend half of the smoothing length along the direction of the smoothing starting point to obtain the first smoothing control point; Starting from the smoothing endpoint, extend the smoothing length by half in the opposite direction of the smoothing endpoint to obtain the second smoothing control point.

5. The method according to claim 4, wherein, The step of determining a smooth curve between the smoothing start point and the smoothing end point based on the plurality of smoothing control points includes: Based on the first and second smoothing control points, a third-order Bézier curve is generated by taking the smoothing start point as the start point of the Bézier curve and the smoothing end point as the end point of the Bézier curve.

6. The method according to claim 1, wherein, After determining the smooth curve, the method further includes: Based on a preset point spacing or number of points, the smooth curve is sampled to obtain multiple sampling points as points on the smooth curve.

7. A method for high-precision map creation, the method comprising: Based on the lane line vector smoothing method according to any one of claims 1 to 6, the target lane line vector pair to be smoothed is smoothed to obtain the smoothed lane line vector. A high-precision map is drawn based on the smoothed lane line vectors.

8. A lane line vector smoothing device, the device comprising: The acquisition unit is used to acquire the target lane line vector pair to be smoothed and the speed limit information of the lane where the target lane line vector pair is located; The first determining unit is configured to: determine the target turning radius corresponding to the speed limit information based on the speed limit information and a preset correspondence between the road speed limit and the turning radius range; determine the target lateral distance as the projected length of the relative distance between the first lane line vector and the second lane line vector in the target lane line vector pair in the direction perpendicular to the end point of the first lane line vector; determine a smoothing length based on the target turning radius and the target lateral distance; determine a smoothing starting point on the first lane line vector with the end point of the first lane line vector as the starting position and along the opposite direction of the first lane line vector, determine a smoothing ending point on the second lane line vector with the starting point of the second lane line vector as the starting position and along the direction of the second lane line vector, wherein the smoothing length is the sum of the length between the smoothing starting point and the end point of the first lane line vector and the length between the starting point of the second lane line vector and the smoothing ending point, and the first lane line vector and the second lane line vector are sequentially arranged along the lane forward direction; The second determining unit is used to determine multiple smoothing control points based on the smoothing start point and the smoothing end point; A smoothing unit is used to determine a smoothing curve between the smoothing start point and the smoothing end point based on the plurality of smoothing control points, the smoothing curve smoothly connecting the target lane line vector pairs.

9. An electronic device, wherein, The system includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the lane line vector smoothing method as claimed in any one of claims 1 to 6 or the steps of the high-precision map drawing method as claimed in claim 7.

10. A computer-readable storage medium, wherein, The computer-readable storage medium stores a program or instructions that cause a computer to perform the steps of the lane line vector smoothing method as claimed in any one of claims 1 to 6 or the steps of the high-precision map drawing method as claimed in claim 7.

Images