CN115009284A

CN115009284A - Lane reference line optimization method, device, equipment and storage medium

Info

Publication number: CN115009284A
Application number: CN202210742508.3A
Authority: CN
Inventors: 刘立; 曲和政; 金凌鸽
Original assignee: Suzhou Yihang Yuanzhi Intelligent Technology Co ltd
Current assignee: Suzhou Yihang Yuanzhi Intelligent Technology Co ltd
Priority date: 2022-06-27
Filing date: 2022-06-27
Publication date: 2022-09-06

Abstract

The disclosure provides a lane reference line optimization method, a lane reference line optimization device, lane reference line optimization equipment and a storage medium. The lane reference line optimization method comprises the following steps: acquiring an original lane reference line corresponding to the environment where the first vehicle is located from the high-definition map; fitting according to randomly selected discrete points on the original lane reference line to obtain a first lane reference line; carrying out curvature optimization on sampling points in the first lane reference line according to a maximum curvature threshold of the first vehicle to obtain a control point set, fitting according to the control point set to obtain a second lane reference line, and determining the maximum curvature threshold according to vehicle parameters of the first vehicle; performing collision detection on the second lane reference line to update the control point set, and fitting according to the updated control point set to obtain a third lane reference line; and obtaining an optimized lane reference line corresponding to the environment where the first vehicle is located according to the third lane reference line. The lane reference line acquisition method and the lane reference line acquisition device are suitable for various scenes and various vehicles and enable the vehicles to smoothly pass through a curve area.

Description

Lane reference line optimization method, device, equipment and storage medium

Technical Field

The present disclosure relates to the field of automatic driving technologies, and in particular, to a method, an apparatus, a device, and a storage medium for optimizing a lane reference line.

Background

The lane reference line, i.e. the lane center line, is one of the important data in the automatic driving application, and can provide decision basis for vehicle control and driving planning. For example, a lane reference line may provide a trajectory line that the vehicle travels when the vehicle is performing lane keeping; when the vehicle changes lanes, the vehicle should travel towards the lane reference line of the lane after the change.

At present, the traditional methods for optimizing the lane reference line include a least quadratic fitting method, a quadratic programming method, and the like. And fitting points on the lane reference line by a least squares fitting method according to the least quadratic fitting method, and then sampling the lane reference line according to the fitted parameters so as to obtain the optimized lane reference line. Although the least quadratic fit method is convenient and simple in principle, the method can only process simple scenes, and lane reference lines may be composed of different types of lines, such as straight line segments, relaxation curves and the like. And, the accuracy of the lane reference line obtained by the least squares fit method will also decrease as the length of the lane reference line increases. In addition, the situation that the curvature of the optimized lane reference line is too large in the curve area may occur by adopting the traditional methods, and when a vehicle runs to the area, the situation that the vehicle cannot pass through the area due to the too large curvature of the lane reference line may occur, so that potential safety hazards are brought.

Disclosure of Invention

In order to solve at least one of the above technical problems, the present disclosure provides a lane reference line optimization method, apparatus, device, and storage medium, which can obtain a lane reference line that is suitable for various scenes and various vehicles and enables the vehicles to smoothly pass through a curve area.

A first aspect of the present disclosure provides a lane reference line optimization method, including:

acquiring an original lane reference line corresponding to the environment where the first vehicle is located from the high-definition map;

fitting according to the randomly selected discrete points on the original lane reference line to obtain a first lane reference line;

carrying out curvature optimization on sampling points in the first lane reference line according to the maximum curvature threshold of the first vehicle to obtain a control point set, fitting according to the control point set to obtain a second lane reference line, wherein the maximum curvature threshold is determined according to vehicle parameters of the first vehicle;

performing collision detection on the second lane reference line to update the control point set, and fitting according to the updated control point set to obtain a third lane reference line;

and obtaining an optimized lane reference line corresponding to the environment where the first vehicle is located according to the third lane reference line.

In some possible implementations of the first aspect of the present disclosure, fitting to obtain the first lane reference line according to a randomly selected discrete point on the origin lane reference line includes: determining control points and coordinates thereof according to a curve equation of the reference line of the original lane obtained by fitting the discrete points so as to obtain a control point set; and according to the control point set, fitting through a rotating curve equation and a cubic B-spline curve equation to obtain the first lane reference line, wherein the rotating curve equation is constructed based on rigid motion characteristics.

In some possible implementations of the first aspect of the present disclosure, the vehicle parameter may include one or more of: the wheel base of the vehicle, the width of the front wheel of the vehicle, the distance from the center of the front axle of the vehicle to the center of the rear axle of the vehicle, the maximum rotating angle of a steering wheel of the vehicle, the wheel base of the front wheel of the vehicle and the center distance of a kingpin of the vehicle.

In some possible embodiments of the first aspect of the present disclosure, the first lane reference line is obtained by fitting the following formula:

wherein s (t) is a first lane reference line, r is a coefficient matrix related to discrete points, r is determined according to coordinates of the discrete points, B (t) is a cubic B spline curve set constructed according to a control point set, B (t) _i For cubic B-spline curve segment, sigma is summation operator, C _i Is the ith control point, F _i (t) is a basis function, t is a variable parameter, t is an element [ t ] _a ,t _b ],t _a Is the minimum value of the abscissa of the control point, t _b Exp is the operation of exponential mapping, which is the maximum value of the abscissa of the control point.

In some possible embodiments of the first aspect of the present disclosure, the maximum curvature threshold is obtained by:

wherein, K _max Is the maximum curvature threshold, L is the distance from the center of the front axle to the center of the rear axle of the vehicle,

the maximum turning angle of the steering wheel of the vehicle, b the front wheel track of the vehicle and M the kingpin center distance of the vehicle.

In some possible implementations of the first aspect of the present disclosure, the curvature optimizing the sampling points in the first lane reference line according to the maximum curvature threshold of the first vehicle to obtain the set of control points includes: taking sampling points obtained by uniformly sampling the first lane reference line as control points to form a control point set; calculating the curvature of each control point in the control point set; and performing curvature optimization on the control points with the curvatures larger than the maximum curvature threshold value in the control point set to update the control point set until the curvatures of all the control points in the control point set are smaller than or equal to the maximum curvature threshold value.

In some possible implementations of the first aspect of the present disclosure, performing curvature optimization on a control point of the set of control points whose curvature is greater than the maximum curvature threshold value to update the set of control points includes: and inserting a new control point into a lane line segment formed by the closest control point of the control point with the curvature larger than the maximum curvature threshold value and the adjacent control point of the closest control point according to the maximum turning angle of the steering wheel of the vehicle, and adding the new control point into the control point set.

In some possible implementations of the first aspect of the present disclosure, performing collision detection on the second lane reference line to update the set of control points includes: and performing collision detection on the second lane reference line, determining a serial number, a starting point and an end point of a curve segment where the second lane reference line collides with an obstacle when the second lane reference line collides with a static obstacle, inserting a new control point between the starting point and the end point of the curve segment corresponding to the serial number, and incorporating the new control point into the control point set.

In some possible embodiments of the first aspect of the present disclosure, the collision detection comprises: and judging whether the minimum directional bounding box of the second lane reference line is intersected with the bounding box of the obstacle, wherein the intersection indicates that the second lane reference line is collided with the obstacle.

In some possible implementations of the first aspect of the present disclosure, the obtaining an optimized lane reference line corresponding to an environment in which the first vehicle is located according to the third lane reference line includes: and carrying out speed optimization and acceleration optimization on the third lane reference line according to a preset maximum speed value and a preset maximum acceleration value to obtain a fourth lane reference line.

In some possible embodiments of the first aspect of the present disclosure, the speed optimization comprises: and obtaining the vehicle speed corresponding to each control point by carrying out first-order derivation on the third lane reference line, and adjusting the position of the control point with the vehicle speed being greater than the maximum speed value until the vehicle speeds corresponding to all the control points are not greater than the maximum speed value.

In some possible embodiments of the first aspect of the present disclosure, the acceleration optimization includes: and performing second-order derivation on the lane reference line obtained by optimizing the speed to obtain the vehicle acceleration corresponding to each control point, and adjusting the positions of the control points with the vehicle acceleration greater than the maximum acceleration until the vehicle accelerations corresponding to all the control points are not greater than the maximum acceleration.

In some possible implementations of the first aspect of the present disclosure, the obtaining an optimized lane reference line corresponding to an environment in which the first vehicle is located according to the third lane reference line further includes: selecting one of the plurality of feasible lane reference lines as the optimized lane reference line according to one or more of the following items: the length of the feasible lane reference line, the generation time of the lane reference line, the driving comfort level of the first vehicle on the feasible lane reference line, and the displacement between the feasible lane reference line and the original lane reference line, wherein the feasible lane reference line is the fourth lane reference line or the third lane reference line.

In some possible embodiments of the first aspect of the present disclosure, the selecting one of a plurality of feasible lane reference lines as the optimized lane reference line includes: determining a preference coefficient of each feasible lane reference line according to one or more of the following factors, and selecting a feasible lane reference line as the optimized lane reference line according to the preference coefficient: the length of the feasible lane reference line, the generation time of the feasible lane reference line, the driving comfort level of a first vehicle on the feasible lane reference line, and the displacement between the feasible lane reference line and an original lane reference line; wherein the feasible lane reference line is the fourth lane reference line or the third lane reference line.

A second aspect of the present disclosure provides a lane reference line optimizing apparatus, including:

the acquisition unit is used for acquiring an original lane reference line corresponding to the environment where the first vehicle is located from the high-definition map;

the fitting unit is used for fitting according to the randomly selected discrete points on the original lane reference line to obtain a first lane reference line; the second lane reference line is obtained according to the control point set obtained by the curvature optimization unit; the collision detection unit is used for updating the control point set of the control point set to obtain a control point set;

the curvature optimization unit is used for carrying out curvature optimization on the sampling points in the first lane reference line according to the maximum curvature threshold of the first vehicle so as to obtain a control point set;

a collision detection unit for performing collision detection on the second lane reference line to update the control point set;

and the third lane reference line optimizing unit is used for obtaining an optimized lane reference line corresponding to the environment where the first vehicle is located according to the third lane reference line.

In some possible embodiments of the second aspect of the present disclosure, the fitting unit is configured to fit the first lane reference line by:

determining control points and coordinates thereof according to a curve equation of the reference line of the original lane obtained by fitting the discrete points so as to obtain a control point set;

and according to the control point set, fitting through a rotating curve equation and a cubic B-spline curve equation to obtain the first lane reference line, wherein the rotating curve equation is constructed based on rigid motion characteristics.

In some possible embodiments of the second aspect of the present disclosure, the vehicle parameter may include one or more of: the wheel base of the vehicle, the width of the front wheel of the vehicle, the distance from the center of the front axle of the vehicle to the center of the rear axle of the vehicle, the maximum rotating angle of a steering wheel of the vehicle, the wheel base of the front wheel of the vehicle and the center distance of a kingpin of the vehicle.

In some possible embodiments of the second aspect of the present disclosure, the curvature optimizing unit is further configured to obtain the maximum curvature threshold by:

In some possible embodiments of the second aspect of the present disclosure, the curvature optimizing unit is specifically configured to: taking sampling points obtained by uniformly sampling the first lane reference line as control points to form a control point set; calculating the curvature of each control point in the control point set; and performing curvature optimization on the control points with the curvatures larger than the maximum curvature threshold value in the control point set to update the control point set until the curvatures of all the control points in the control point set are smaller than or equal to the maximum curvature threshold value.

In some possible embodiments of the second aspect of the present disclosure, the curvature optimizing unit is specifically configured to insert a new control point into the set of control points according to a maximum rotation angle of a steering wheel of a vehicle, in a lane segment formed by a control point with a curvature greater than the maximum curvature threshold value and a nearest control point to the nearest control point.

In some possible embodiments of the second aspect of the present disclosure, the collision detecting unit is specifically configured to: and performing collision detection on the second lane reference line, determining a serial number, a starting point and an end point of a curve segment where the second lane reference line collides with an obstacle when the second lane reference line collides with a static obstacle, inserting a new control point between the starting point and the end point of the curve segment corresponding to the serial number, and incorporating the new control point into the control point set.

In some possible embodiments of the second aspect of the present disclosure, the collision detecting unit is specifically configured to: and judging whether the minimum directional bounding box of the second lane reference line is intersected with the bounding box of the obstacle, wherein the intersection indicates that the second lane reference line is collided with the obstacle.

In some possible embodiments of the second aspect of the present disclosure, the third lane reference line optimizing unit includes: and the speed optimization unit is used for carrying out speed optimization and acceleration optimization on the third lane reference line according to preset maximum speed values and maximum acceleration values to obtain a fourth lane reference line.

In some possible embodiments of the second aspect of the present disclosure, the speed optimization unit is specifically configured to: and obtaining the vehicle speed corresponding to each control point by carrying out first-order derivation on the third lane reference line, and adjusting the position of the control point with the vehicle speed being greater than the maximum speed value until the vehicle speeds corresponding to all the control points are not greater than the maximum speed value.

In some possible embodiments of the second aspect of the present disclosure, the speed optimization unit is specifically configured to: and performing second-order derivation on the lane reference line obtained by optimizing the speed to obtain the vehicle acceleration corresponding to each control point, and adjusting the positions of the control points with the vehicle acceleration greater than the maximum acceleration until the vehicle accelerations corresponding to all the control points are not greater than the maximum acceleration.

In some possible embodiments of the second aspect of the present disclosure, the third lane reference line optimizing unit further includes: an evaluation unit, configured to select one of the plurality of feasible lane reference lines as the optimized lane reference line according to one or more of the following: the length of the feasible lane reference line, the generation time of the lane reference line, the driving comfort level of the first vehicle on the feasible lane reference line, and the displacement between the feasible lane reference line and the original lane reference line, wherein the feasible lane reference line is the fourth lane reference line or the third lane reference line.

In some possible embodiments of the second aspect of the present disclosure, the evaluation unit is specifically configured to: determining a preference coefficient of each feasible lane reference line according to one or more of the following factors, and selecting a feasible lane reference line as the optimized lane reference line according to the preference coefficient: the length of the feasible lane reference line, the generation time of the feasible lane reference line, the driving comfort level of the first vehicle on the feasible lane reference line, and the displacement between the feasible lane reference line and the original lane reference line.

A third aspect of the present disclosure provides an electronic device, comprising:

a memory storing execution instructions; and

a processor executing the execution instructions stored by the memory, such that the processor performs the lane reference line optimization method described above.

A fourth aspect of the present disclosure provides a readable storage medium having stored therein execution instructions, which when executed by a processor, are used to implement the lane reference line optimization method described above.

Curvature constraint and collision detection related to vehicle parameters are introduced into the optimization of the lane reference line, the method is applicable to various scenes and various vehicles, the vehicles can smoothly and safely pass through a curve area when running along the optimized lane reference line, and excessive obstacle avoidance control is avoided, so that the running safety of the vehicles is improved.

Drawings

The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the disclosure and together with the description serve to explain the principles of the disclosure.

FIG. 1 is a schematic flow diagram of a lane reference line optimization method of one embodiment of the present disclosure;

FIG. 2 is a schematic flow chart of fitting a first lane reference line according to one embodiment of the present disclosure;

FIG. 3 is a schematic diagram of a curvature optimization implementation of one embodiment of the present disclosure;

FIG. 4 is a schematic diagram of interpolation of new control points in a lane segment according to one embodiment of the present disclosure;

FIG. 5 is an exemplary plot of cubic B-spline curve versus obstacle position in one embodiment of the present disclosure;

FIG. 6 is a schematic flow chart illustrating the optimization of collision detection according to one embodiment of the present disclosure;

FIG. 7 is a schematic diagram of interpolation of new control points in a set of control points during collision detection optimization in one embodiment of the present disclosure;

fig. 8 is a block diagram schematically illustrating a structure of a lane reference line optimizing apparatus using a hardware implementation of a processing system according to an embodiment of the present disclosure.

Detailed Description

The present disclosure will be described in further detail with reference to the drawings and embodiments. It is to be understood that the specific embodiments described herein are for purposes of illustration only and are not to be construed as limitations of the present disclosure. It should be further noted that, for the convenience of description, only the portions relevant to the present disclosure are shown in the drawings.

It should be noted that the embodiments and features of the embodiments in the present disclosure may be combined with each other without conflict. Technical solutions of the present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

Unless otherwise indicated, the illustrated exemplary embodiments/examples are to be understood as providing exemplary features of various details of some ways in which the technical concepts of the present disclosure may be practiced. Accordingly, unless otherwise indicated, features of the various embodiments may be additionally combined, separated, interchanged, and/or rearranged without departing from the technical concept of the present disclosure.

The use of cross-hatching and/or shading in the drawings is generally used to clarify the boundaries between adjacent components. As such, unless otherwise noted, the presence or absence of cross-hatching or shading does not convey or indicate any preference or requirement for a particular material, material property, size, proportion, commonality between the illustrated components and/or any other characteristic, attribute, property, etc., of a component. Further, in the drawings, the size and relative sizes of components may be exaggerated for clarity and/or descriptive purposes. While example embodiments may be practiced differently, the specific process sequence may be performed in a different order than that described. For example, two processes described consecutively may be performed substantially simultaneously or in reverse order to that described. In addition, like reference numerals denote like parts.

When an element is referred to as being "on" or "on," "connected to" or "coupled to" another element, it can be directly on, connected or coupled to the other element or intervening elements may be present. However, when an element is referred to as being "directly on," "directly connected to" or "directly coupled to" another element, there are no intervening elements present. For purposes of this disclosure, the term "connected" may refer to physically, electrically, etc., and may or may not have intermediate components.

The terminology used herein is for the purpose of describing particular embodiments and is not intended to be limiting. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. Furthermore, when the terms "comprises" and/or "comprising" and variations thereof are used in this specification, the presence of stated features, integers, steps, operations, elements, components and/or groups thereof are stated but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and/or groups thereof. It is also noted that, as used herein, the terms "substantially," "about," and other similar terms are used as approximate terms and not as degree terms, and as such, are used to interpret inherent deviations in measured values, calculated values, and/or provided values that would be recognized by one of ordinary skill in the art.

Brief description of related art:

related art 1: chinese patent CN112562372 discloses a method for processing trajectory data and related apparatus. Firstly, sampling points on a lane reference line, and solving a third derivative of the sampling points to obtain the smoothness of the lane reference line; then calculating the distance between two adjacent sampling points, and superposing the obtained distances to obtain the length of the reference line of the whole lane; and finally, performing weighted optimization on the smoothness and the length of the lane reference line and the displacement of the track points on the lane reference line before and after optimization by a quadratic programming method, thereby obtaining the optimized lane reference line. However, the optimized lane reference line obtained according to the method is only as close to the original lane reference line as possible, and it cannot be guaranteed that the vehicle can travel along the optimized lane reference line. For example, the method does not add curvature limitation, the optimized lane reference line may have too large curvature in the curve area, and when the vehicle travels to the area, the vehicle may not pass through the area because of the too large curvature of the lane reference line. Secondly, collision detection is not added in the method, and the optimized lane reference line cannot be ensured to be in the lane boundary; and the method obtains a smooth lane reference line by restricting the smoothness of the lane reference line, but the smoothness is difficult to control, and the optimized lane reference line may be distorted.

Related art 2: chinese patent CN109521763 discloses a lane reference line optimization method based on constrained smooth spline for automatic driving vehicle. The method comprises the steps that firstly, a lane reference line is divided into a plurality of path sections, each path section is represented by a different polynomial function, the polynomial functions of two adjacent path sections need to be smooth in connection and continuous in first-order and second-order derivatives, and a target function is generated according to the polynomial functions; then the method generates a constraint set of polynomial functions according to lane boundaries, obstacles and/or curvatures; and finally, performing quadratic programming optimization on the objective function according to the added constraint set of the polynomial function, so that the output of the objective function is minimum on the premise of meeting the constraint set. Although the related art 2 adds the curvature to the polynomial function constraint set, it can be ensured that the vehicle runs smoothly along the generated lane reference line, and the emergency braking situation does not occur; however, the method requires a path generated by referring to the collected driving data, and the curvature of the generated optimized path is made to meet the requirement by restricting the difference between the generated lane reference line and the path generated by the collected driving data. In practical application, all scenes cannot have collected driving data, so that the method relying on the collected driving data is poor in practicability; the minimum turning radius of the vehicle is related to parameters of the vehicle itself, such as the wheel base, the front wheel width, and the like. Since each vehicle parameter is different, if the vehicle parameter of the currently driving vehicle and the vehicle parameter of the collected driving data are very different, the currently driving vehicle may still not pass through the curve area by using the collected driving data for curvature constraint. Furthermore, this method requires optimization (e.g., quadratic programming) of discrete points on all lane reference lines, which not only affects the speed and efficiency of optimization, but also causes additional consumption of computational resources.

Related art 3: chinese patent CN113327448 discloses a vehicle track optimization method under the phase special for automatic driving. Firstly, collecting physical parameters of a road intersection, basic information of vehicles and signal states of each exit lane and each entrance lane; secondly, dividing the road intersection into a track control area and a coordinated braking area, and establishing a vehicle running track model and a coordinated control model for conflict separation of the automatic driving vehicles; and finally, optimizing to obtain the position, the speed and the acceleration of the vehicle by taking the safe and rapid passing of the vehicle on all the entrance and exit lanes of the intersection as an optimization target. The method can only be applied to vehicles passing through the road intersection safely, has a single use scene, and is not suitable for global lane reference line optimization.

In view of the above, the present disclosure provides an improved method, an apparatus, a device, and a storage medium for optimizing a lane reference line, which introduce curvature constraints and collision detection related to vehicle parameters in optimization of the lane reference line, and are not only applicable to various scenes and various vehicles, but also enable the vehicle to smoothly and safely pass through a curve region when the vehicle travels along the optimized lane reference line, and avoid excessive obstacle avoidance control, thereby improving the safety of vehicle traveling.

Hereinafter, a detailed description will be given of a specific embodiment of the present disclosure with reference to fig. 1 to 8.

Fig. 1 illustrates a flow diagram of a lane reference line optimization method of some embodiments of the present disclosure. Referring to fig. 1, an exemplary implementation flow S10 of the lane reference line optimization method may include:

step S12, acquiring an original lane reference line corresponding to the environment where the first vehicle is located from the high-definition map;

in the present disclosure, the first vehicle refers to any vehicle, and may be, but is not limited to, various different types of vehicles such as an electric vehicle, a fuel vehicle, a hybrid vehicle, and the like, and may also be, but is not limited to, a passenger vehicle, a private vehicle, a bus, or other various vehicles for different purposes.

In the present disclosure, the high-definition map may be a high-definition map built in the vehicle, a high-definition map from a cloud server, or a high-definition map obtained in other various ways. Illustratively, the high-definition map may be a semantic map, a high-definition map, a high-precision map, or any other type of map, and the map may include a lane reference line.

And step S14, fitting the random discrete points on the original lane reference line to obtain a first lane reference line.

Specifically, discrete points are randomly selected from the original lane reference line, a first lane reference line is obtained by fitting the discrete points, the fitted first lane reference line may be composed of multiple segments of curves, and the connection of adjacent curves is continuous.

In some embodiments, the method of fitting in step S14 may be, but is not limited to, a B-spline curve, a bezier curve, a cubic spline curve, or other curve fitting method. Because the B-spline curve is flexible and easy to control, and the B-spline curve is formed by combining a plurality of sections of B-spline curves, the position of a certain control point is changed without influencing other sections of B-spline curves, and the local part of the B-spline curve is easy to adjust by adjusting the position of the control point. Therefore, preferably, step S12 uses B-spline curve fitting to obtain the first lane reference line.

Since the origin lane reference lines are provided in the form of discrete points, which are inconvenient to directly use, the origin lane reference lines may be fitted in advance before the origin lane reference lines are optimized. The fitting may be, but is not limited to, a B-spline curve, a bezier curve, a cubic spline curve, or other curve fitting method.

In some embodiments, the randomly selected discrete points uniformly cover the entire origin lane reference line. In order to ensure that the first lane reference line passes through the starting point and the end point of the original lane reference line, the selected discrete points comprise the starting point and the end point of the original lane reference line.

In some embodiments, the number of discrete points selected at a time should be no less than 20% of the total number of discrete points in the reference line of the origin lane. If 20% of the total number of the original lane line reference lines is less than 4, 4 discrete points are selected.

Taking a B-spline curve as an example, P +1 discrete points can be randomly selected, and since the highest order of each segment of the B-spline curve is 3, at least 4 points are needed to solve a 3-order curve, and therefore, P is a positive integer not less than 3. It should be noted that P +1 points are selected instead of P points because the matrix coefficients are described starting from 0, and P +1 is selected in the present embodiment for convenience of description of the matrix coefficients.

In some embodiments, the first lane reference line may be directly fitted with a randomly chosen discrete point set V as the control point set.

Since the first lane reference line obtained when the curve is directly fitted with the discrete points as the control points is not necessarily able to pass through the discrete points, in order to enable the first lane reference line to describe the original lane reference line more accurately, in some embodiments, referring to fig. 2, the curve fitting in step S14 may include:

step S22, determining control points and coordinates thereof according to a curve equation of the original lane reference line obtained by fitting discrete points to obtain a control point set;

and S24, fitting a rotating curve equation and a cubic B-spline curve equation to obtain a first lane reference line according to the control point set obtained in the step S22, wherein the rotating curve equation is constructed based on rigid motion characteristics.

In some embodiments, step S22 may determine the coordinates of the control point by the following equation (1):

in the formula (1), C ₀ ,C ₁ ,....C _P For a control point, r is a coefficient matrix associated with the discrete point, and a is a coefficient matrix of (P +1) × (P + 1).

In some embodiments, the coefficient matrix a and the coefficient matrix r may be obtained according to the following equation (2):

in the formula (2), B ₀ (t) a curve equation representing the reference line of the origin lane based on a random selection of discrete points, B ₀ (t _k ) Coordinates representing the k +1 th discrete point of the origin lane reference line, B ₀ (t ₀ ) Represents the coordinate of the 1 st sampling point of the reference line of the origin lane, B ₀ (t _P ) Represents the coordinates of the P +1 sampling point in the reference line of the original lane, u _i+1 Represents the time at which the vehicle passes through the origin lane reference line sampling point i +1, u _i Representing the time, Δ, at which the vehicle passes the origin lane reference line sample point i _i Representing the time required for the vehicle to pass through sample point i and sample point i +1 in sequence at the maximum allowable speed, since ideally the vehicle is stationary at the start and end points, the present disclosure adds to the coefficient matrix r the vehicle initial speed constraints at the start and end points, i.e., Δ _0- ＝Δ _P ＝0，Δ ₀ -represents the initial speed of the vehicle at the starting point, Δ _P Indicating the initial speed, v, of the vehicle at the end point _max The maximum allowable speed for the vehicle to travel on the lane may be set according to actual conditions.

Curve B of original lane reference line obtained based on random selected discrete point fitting ₀ And (t) sampling points are discrete points of the original lane reference line.

In some embodiments, in step S24, a cubic B-spline curve equation is constructed based on the control point set C, and a rotational curve equation of the vehicle may be constructed based on the cubic B-spline curve equation and the rigid body motion characteristics.

In some embodiments, in step S24, the first lane reference line may be obtained according to equations (3) to (5):

in the expressions (3) to (5), s (t) is a first lane reference line, r is a coefficient matrix related to discrete points, r can be determined according to a curve equation of an original lane reference line, for example, it can be obtained by the expression (2), a translation function t (t) can be represented by B (t), a rotation function r (t) can also be represented by B (t), B (t) is a cubic B spline curve set { B (t) constructed according to a control point set C ₀ ,B ₁ ,B ₂ ,...B _i ,…B _P }，B _i Is a cubic B-spline curve segment (i ═ 0,1,2, … …, P), Σ is the summation operator, C _i Is the ith control point, F _i (t) is a basis function, F _i (t) can be calculated by a code-de Boor recursive formula, the code-de Boor recursive formula can be obtained by inquiring books related to numerical analysis, t is a variable parameter, and t belongs to [ t [ [ t ] _a ,t _b ]，t _a Is the minimum value of the abscissa of the control point, t _b Equation (4) is a cubic B-spline curve equation, equation (5) is a rotational curve equation of the vehicle, exp is an operation of the exponential mapping, R is a maximum value of the abscissa of the control point _SO3 (t) is a spatial rotation curve of the vehicle.

Regarding the construction process of formula (5), the principle is as follows:

considering a vehicle as a rigid body point, the motion of the vehicle from the origin to the destination can be decomposed into translation t (t) and rotation r (t). Since the cubic B-spline curve is continuous in the real space of the curve, the translation t (t) of the vehicle from the start to the destination can be represented by B (t). Since the translation of the rigid body in the three-dimensional space can be obtained through a series of rotations, i.e., t (t) is also a function of r (t), a cubic spline curve is used to describe the whole movement process of the vehicle from the starting point to the destination, and r (t) can also be represented by b (t) for simplicity.

The rotational motion of the rigid body is not necessarily continuous in the real number space, but is continuous in the space formed by the special orthogonal group SO (3), i.e., the rotational motion R of the vehicle can be represented by a continuous and smooth curve in the special orthogonal group SO (3). Assuming Euler angles (alpha, beta, gamma) for rotation of the vehicle ^T Indicating that alpha indicates that the vehicle rotates anticlockwise around the X axis and forms an angle with the X axis; beta represents the angle formed by the vehicle rotating anticlockwise around the Y axis and the Y axis; γ represents the angle formed by the vehicle rotating counterclockwise about the Z axis with respect to the Z axis. Although the Euler angle (alpha, beta, gamma) ^T Indicating that rotation is easily understood, but (α, β, γ) ^T Derivation and the like cannot be calculated, and therefore, in some embodiments of the present disclosure, the euler angle is converted into a quaternion q, and the conversion formula is expressed by the following equation (6):

quaternions can be rewritten as a vector of rotation, assuming a vector

Is a rotation axis, and theta is an angle of counterclockwise rotation around the rotation axis, a rotation vector

Quaternions q may be used _v Expressed as shown in formula (7):

in the formula (7), the reaction mixture is,

is composed of

Theta is the angle of rotation.

Assuming that the coordinate vector of the vehicle in the real number space is v, since the coordinate vector v can be obtained by rotating the unit vector around the rotation vector, the coordinate vector can be equivalent to a certain rotation vector. When v is not 0, it can be attributed to the real number space according to the exponential mapping principle

Mapping to a special orthogonal group SO (3) space, wherein the mapping formula is as follows (8):

in equation (8), exp is an exponential mapping operation, and θ is an angle at which the vehicle rotates in a real number space.

By combining equation (7) with equation (8), it can be seen that the rotation of the vehicle in the real number space can be regarded as the rotation of the vehicle in the special orthogonal group SO (3) space through the exponential mapping. Since the special orthogonal group SO (3) space is closed to multiplication, and the rotation of the rigid body can be represented by multiplication of a series of rotation vectors, the vehicle rotates in the special orthogonal group SO (3) space

Is continuous.

I.e. the rotation curve sought. It should be noted that, according to the property of the derivative of the matrix exponential function, the derivative of the matrix exponential function does not change the property of the original function, R (t) and

the properties of (A) are the same,

likewise a cubic B-spline curve. Thus constructed

As shown in formula (5).

As can be seen from the above, in the embodiment of the present disclosure, by using the property that lie algebra is smooth and continuous in a number domain space, a curve is mapped to a lie group space through exponential mapping, so that any two points on the curve can be mutually converted through rigid transformation (i.e., rotation and translation), and the obtained lane reference line can be ensured to be smooth and continuous in any scene. If the vehicle is regarded as a rigid body, the transformation (rotation and translation) of the rigid body is continuous in the lie algebraic number domain space, so that the motion of the vehicle between two points of the lane reference line can be completed through the rigid body transformation, namely, the vehicle starts from one point on the lane reference line and can reach any other point on the lane reference line along the lane reference line, thereby ensuring that the obtained lane reference line can be suitable for different scenes.

Step S16, curvature optimization is carried out on sampling points in the first lane reference line according to the maximum curvature threshold of the first vehicle to obtain a control point set, and a second lane reference line is obtained according to fitting of the control point set;

if the vehicle is regarded as a rigid body point, the first lane reference line can meet the kinematic plan of the vehicle, and the vehicle can be ensured to run smoothly along the curve. However, in practice, the vehicle has a length and a width, and if the curvature of some points in the lane reference line is too large, the vehicle is likely to be unable to pass through these areas, particularly the curved portions in the lane. Therefore, the curvature control is performed on the curve through the step S16, so that the lane reference line can meet the requirements of different vehicles and is suitable for various vehicles.

In some embodiments, in step S16, curvature optimizing the sampling points in the first lane reference line according to the maximum curvature threshold of the first vehicle to obtain a control point set includes: and taking sampling points obtained by uniformly sampling the first lane reference line as control points to form a control point set, calculating the curvature of each control point in the control point set, and performing curvature optimization on the control points of which the curvatures are greater than the maximum curvature threshold value in the control point set to update the control point set until the curvatures of all the control points in the control point set are less than or equal to the maximum curvature threshold value.

The maximum curvature threshold represents the maximum curvature that is allowed for smooth vehicle passage, i.e., the maximum curvature required for the vehicle to smoothly complete a turn. In some embodiments of the present disclosure, the maximum curvature threshold of the first vehicle may be determined from a vehicle parameter of the first vehicle. Since the maximum curvature requirement of a vehicle for a point on a curve is related to the vehicle's own parameters, such as wheel base, width of the front wheel, etc., the use of a uniform threshold for curvature control cannot meet the requirements of different vehicles. In order to enable the lane reference lines to be suitable for different vehicles, vehicle parameters are introduced into curvature control, maximum curvature threshold values suitable for different vehicles are generated according to the vehicle parameters of the different vehicles, so that the lane reference lines suitable for the different vehicles are obtained through fitting, and the requirements of the different vehicles are met.

In some embodiments, the vehicle parameters may include, but are not limited to, vehicle wheelbase, front wheel width, distance from vehicle front axle center to rear axle center, maximum turn angle of the steering wheels of the vehicle, front wheel base of the vehicle, kingpin center distance of the vehicle, and the like.

In some embodiments, curvature optimization of the control points of the set of control points having a curvature greater than the maximum curvature threshold to update the set of control points includes: and inserting a new control point into a lane line segment formed by the closest control point of the control point with the curvature larger than the maximum curvature threshold value and the adjacent control point of the closest control point according to the maximum turning angle of the steering wheel of the vehicle, and adding the new control point into the control point set.

Fig. 3 illustrates a flow chart of a specific implementation of curvature optimization in some embodiments of the present disclosure. Referring to fig. 3, an exemplary flow of curvature optimization may include:

step S32, uniformly sampling the first lane reference line to obtain sampling points, forming a control point set by taking the sampling points as control points, and calculating the curvature K of each control point to determine the control points needing curvature optimization;

in some embodiments, the curvature K of the control point can be obtained by the following equation (9):

in the formula (9), K is the curvature of the control point, s is the curve equation of the first lane reference line, s' is the second derivative of the curve equation s, s ^’ Is the first derivative of the curve equation s, | | is the operation of solving for the absolute value.

Before step S32 or in step S32, a maximum curvature threshold of the vehicle may be determined according to vehicle parameters, and it may be detected whether the curvature of each control point satisfies a requirement (i.e., whether the curvature of the control point is not greater than the maximum curvature threshold) according to the maximum curvature threshold of the vehicle, and curvature optimization need not be performed on a sampling point if the sampling point satisfies the curvature requirement, and curvature optimization needs to be performed on the control point if the control point does not satisfy the curvature requirement.

In some embodiments, the maximum curvature threshold may be determined by the following equation (10):

in the formula (10), K _max Is the maximum curvature threshold, L is the distance from the center of the front axle to the center of the rear axle of the vehicle,

the maximum turning angle of the steering wheel of the vehicle, b the front wheel track of the vehicle and M the kingpin center distance of the vehicle. These parameters may be obtained from vehicle design parameters.

Step S34, for the cubic B-spline curve segment B where the control point with excessive curvature (i.e. curvature greater than the maximum curvature threshold) is located _i Under the condition of B _i Of the sampling points (2), the curvature is selected to be too large from the beginning to the end (i.e., the curvature is larger than K) _max ) And selecting a control point C nearest to the control point p _p And C _p Of adjacent control points, e.g. control point C _p Two nearest control points are respectively selected before and after the control points.

Because the cubic B-spline curve modification local point cannot influence the whole cubic B-spline curve, only the B-spline curve segment with the curvature larger than the maximum curvature threshold is processed in the embodiment of the disclosure, and if the point with the curvature larger than the maximum curvature does not exist in a certain cubic B-spline curve segment, the cubic B-spline curve segment is not processed, so that the processing efficiency is improved.

Step S36, using the 5 control points (i.e., control point C) selected in step S34 _p And C _p Two control points in front and at the back) to generate a lane segment;

step S38, interpolating a new control point in the lane line segment, and adding the new control point into the control point set;

fig. 4 shows a schematic diagram of interpolation of new control points in a lane segment. As shown in fig. 4, with this control point C _p Rotates delta by taking the ray of the origin and the adjacent preorder point as an axis _min Intercepting the length L on the rotated ray by taking the point as an end point _min Adding the newly generated control point to the cubic B-spline curve segment B _i Of the control point set.

Wherein, delta _min And L _min Can be determined by the following formula (11):

in the formula (11), w is the maximum steering angle of the steering wheel of the vehicle, and can be obtained by referring to vehicle parameters, δ _min To an angle of rotation, K _max Is the maximum curvature threshold.

Step S310 to step S314, similar to step S34 to step S38, selecting a new combination of control points satisfying the conditions (i.e., selecting a new combination of control points around the control point with the excessive curvature), generating a lane segment, interpolating the new control points, and repeating the steps to optimize the cubic B-spline curve segment B _i All curvatures being greater than K _max If it is the greatest from a control point of excessive curvatureIf the near control point is processed, the processing is not needed again;

step S316, obtaining a processed control point set, that is, an updated control point set;

and step S318, performing curve fitting by using the control point set obtained in the step S316 in the same manner as in the step S14 to generate a new cubic B-spline curve, wherein the new cubic B-spline curve is the reference line of the second lane.

Therefore, the curvature optimization of the lane reference line is realized by introducing the curvature constraint based on vehicle reference, and the lane reference line suitable for different vehicles can be obtained.

Because different vehicles have different requirements on the maximum curvature of the lane reference line in the curve area, the vehicle parameters are introduced in the curvature optimization, and different lane reference lines are generated according to the vehicle parameters of different vehicles, so that different vehicles can smoothly pass through the curve area when running along the optimized lane reference line. And the vehicle can smoothly travel on the second lane reference line after passing the curvature constraint without an obstacle.

And step S16, performing collision detection on the second lane reference line to update the control point set, and fitting the updated control point set to obtain a third lane reference line, wherein the third lane reference line is a lane reference line which cannot collide with an obstacle.

In some embodiments, performing collision detection on the second lane reference line to update the set of control points may include: and performing collision detection on the second lane reference line, determining a serial number, a starting point and an end point of a curve segment where the second lane reference line collides with the obstacle when the second lane reference line collides with the static obstacle, inserting a new control point between the starting point and the end point of the curve segment corresponding to the serial number, and incorporating the new control point into the control point set to obtain a new control point set so as to perform curve fitting according to the new control point set to obtain a third lane reference line.

In some embodiments, the collision detection may include: and judging whether the minimum directional bounding box of the second lane reference line is intersected with the bounding box of the obstacle, wherein the intersection indicates that the second lane reference line can collide with the obstacle, and the non-intersection indicates that the second lane reference line cannot collide with the obstacle.

In some embodiments, step S16 may further include: it is determined whether the resulting third lane reference line collides with the obstacle, and if not, the collision detection may be ended, and if still colliding, step S16 is repeated until the third lane reference line does not collide with the obstacle.

Although modifying the control points of the cubic B-spline curve portion does not affect the entire cubic B-spline curve, there may be a case where a certain cubic B-spline curve segment is contained inside an obstacle, which is difficult to optimize effectively. As shown in fig. 5, the cubic B-spline curve is formed by connecting 5 segments of B-spline curves first, and the red dotted line is a virtual segment line of two adjacent segments of B-spline curves. In fig. 5(a), only the third B-spline is collided with the obstacle, and only the third B-spline can be optimized for obstacle avoidance; however, the obstacle in fig. 5(B) includes the entire third segment of B-spline curve, and the obstacle avoidance processing cannot be performed by optimizing the third segment of B-spline curve. Therefore, in order to avoid a situation where the obstacle contains a certain cubic B-spline curve segment, in some embodiments of the present disclosure, it is preferable to use the entire cubic B-spline curve for collision detection.

In some embodiments, the optimization procedure of collision detection in step S16 may include the following steps:

step S62, finding out a B-spline curve segment intersecting with the barrier;

generating a minimum directional Bounding Box (OBB) of the second lane reference line, judging whether the OBB of the second lane reference line is intersected with the Bounding Box of the obstacle, and if not, indicating that the second lane reference line is not intersected with the obstacle without processing; if the OBB intersects with the bounding box of the obstacle, the second lane reference line is required to be optimized, which shows that the second lane reference line intersects with the obstacle.

When the second lane reference line needs to be optimized, a first section intersected with the obstacle polygon is found according to the position relation between the cubic B-spline curve section and the obstacle polygon (namely the bounding box of the obstacle)B spline curve segment B _q And a last cubic B-spline curve segment B _q+n 。

Step S64, a control point set is obtained: finding a control cubic B-spline curve segment B _q To cubic B-spline curve segment B _q+n A control point set C' of all cubic B-spline curve segments in between;

step S66, insert new control point: inserting a new control point into the control point set C';

as shown in FIG. 7, first, a cubic B-spline curve segment B is found _q The control point C ' closest to the intersection point is found in the control point set C ' at the intersection point with the obstacle polygon ' _k And the intersection C' _k Front and rear adjacent control points C' _k-1 And control point C' _k+1 (ii) a Then, find out cubic B-spline curve segment B _q+n The control point C ' nearest to the intersection is found in the control point set C ' at the intersection with the obstacle polygon ' _g And the intersection C' _g Front and rear adjacent control points C' _g-1 ，C′ _g+1 (ii) a Finally, a new control point C' is interpolated into the set of control points C _k+1 ,C″ _g+1 ，

Wherein the new control point C ″) _k+1 ,C″ _g+1 Can be determined by equation (12):

in formula (12), C ″) _k-1 ,C″ _k+1 ,C″ _g-1 ,C″ _g+1 The coordinates of the control point newly added to the control point set C'.

Step S68, fitting a curve and determining whether to collide with an obstacle: performing curve fitting and curvature optimization according to the control point set obtained in the step S66 to obtain a new cubic B-spline curve segment, and judging whether the new cubic B-spline curve segment collides with the obstacle (namely judging whether the minimum oriented bounding box of the new cubic B-spline curve segment intersects with the bounding box of the obstacle), if so, continuing to the step S610-S614, if not, ending collision detection optimization, and marking the new cubic B-spline curve segment as a third lane reference line;

and S610-614, solving a cubic B spline curve intersected with the obstacle and a corresponding control point set, inserting a new control point, fitting the curve, and repeating the steps S62-S68 on the obtained cubic B spline curve segment until the minimum directional bounding box of the cubic B spline curve is not intersected with the bounding box of the obstacle, namely the cubic B spline curve segment is not collided with the obstacle, wherein the cubic B spline curve segment is a third lane reference line. .

Therefore, the lane reference line capable of avoiding the static obstacle can be obtained through collision detection, and the applicability of the lane reference line to various application scenes is further improved.

In step S18, an optimized lane reference line corresponding to the environment where the first vehicle is located is obtained according to the third lane reference line, and the optimized lane reference line is used for executing subsequent decisions such as route planning, vehicle control or other decisions related to automatic driving.

After curvature optimization and collision detection, the vehicle can smoothly run along the reference line of the third lane. However, speed and acceleration control is not added to the reference line of the third lane, so that phenomena such as jerking and rushing can occur when the vehicle runs, and bad riding experience is brought to passengers. Therefore, in some embodiments of the present disclosure, speed and acceleration of the third lane reference line may be optimized to avoid occurrence of jerking, jogging, and the like during vehicle driving, thereby improving the comfort of vehicle driving.

In some embodiments, step S18 may include: and carrying out speed optimization and acceleration optimization on the third lane reference line according to the preset maximum speed value and maximum acceleration value to obtain a fourth lane reference line. In some embodiments, the performing speed optimization and acceleration optimization on the third lane reference line according to preset maximum speed and maximum acceleration values may include: performing first-order derivation and second-order derivation on control points of a third lane reference line, checking first-order derivative values and second-order derivative values of the control points one by one, and adjusting the positions of the control points to update a control point set if the control points have overlarge first-order derivative values or overlarge second-order derivative values; and fitting a curve according to the updated control point set, and executing curvature optimization and collision detection on the fitted curve so as to obtain a fourth lane reference line, wherein the fourth lane reference line is a feasible lane reference line.

Since the speed is obtained by obtaining a first derivative of the lane reference line, the acceleration is obtained by obtaining a second derivative of the lane reference line, and the value ranges of the speed and the acceleration are different, in some embodiments of the present disclosure, the speed optimization and the acceleration optimization may be performed separately.

According to the definition of the derivative, when the vehicle travels along the third lane reference line and the speed assumes an extreme value, the value of the second derivative of the cubic B-spline curve at this point is 0. For simplicity, a uniformly sampled cubic B-spline curve is used, at [ t ] _i ,t _i+1 ]In the interval, i is a, a +1, … …, b-1, and the extreme value of the vehicle running speed can be obtained according to equation (13):

in the formula (13), the reaction mixture is,

for vehicles in the interval [ t _i ,t _i+1 ]Inner velocity limit, s' (t) _i ) Is a third lane reference line s (t) at t _i The first derivative of (d), s' (t) _i+1 ) Is a third lane reference line s (t) at t _i+1 The first derivative of (a) is,

is the third lane reference line s (t)

First derivative of (A), C _i ,C _i+1 ,C _i+2 ,C _i+3 Is a control point.

As can be seen from equation (13), the vehicle is at [ t ] _i ,t _i+1 ]Speed limit value in interval and vehicle at t _i 、

t _i+1 The velocity of (c) is related. Since the vehicle is at t _i ,t _i+1 ]The velocity at a location within the interval is the first derivative of s (t) at that location, so that it is only necessary to control the function of s (t) at t _i 、

t _i+1 The first derivative of (d) can control the vehicle at [ t ] _i ,t _i+1 ]The speed at any position in the interval meets the requirement. It should be noted that although s (t) is at [ t ] _a ,t _b ]Intervals are continuous and first order derivable, but s (t) is a piecewise function, which cannot be controlled at t by controlling the s (t) function alone _a 、

t _b The first derivative of (d) controls the vehicle at t _a ,t _b ]Speed at any position within the interval, but rather for each interval t _i ,t _i+1 ]And i is more than or equal to a and less than or equal to b-1 to carry out first derivative constraint.

In some embodiments, the speed optimization may include: and obtaining the vehicle speed corresponding to each control point by carrying out first-order derivation on the third lane reference line, and adjusting the position of the control point with the vehicle speed being greater than the maximum speed value until the vehicle speeds corresponding to all the control points are not greater than the maximum speed value. That is, the first derivative is carried out on the control points of the third lane reference line to obtain the first derivative value of each control point, the positions of the control points of which the first derivative values are greater than the preset speed maximum value are adjusted until the first derivative values of all the control points are less than or equal to the speed maximum value, so that a new control point set is obtained, and a new lane reference line is obtained through fitting according to the new control point set.

In some embodiments, the specific implementation flow of the speed optimization may include the following steps a1 to a 4:

step a1, for interval [ t _a ,t _a+1 ]Find the function of s (t) at t _a 、

The first derivative of (a);

note that the s (t) function is at t _a+1 The derivative of (d) in the next interval t _a+1 ,t _a+2 ]The calculation is convenient;

step a2, if

Or s' (t) _a ) Has a value of greater than

C is to be _i+2 Is updated to C _i+2 Is multiplied by λ if

And s' (t) _a ) Are not more than

Step a4 is entered;

namely, the coordinate values of the control points are updated according to equation (14):

in the formula (14), the compound represented by the formula (I),

for vehicles in the interval [ t _a ,t _a+1 ]The maximum speed of the vehicle can be driven at different speeds on different roads, different time periods and different scenes,

the value of (1) needs to be set according to the actual condition of the road, and default values cannot be uniformly set; 0<λ<1, the value of lambda is largeThe speed of the small influence iteration, for example, λ may take 0.7.

Step a3, repeating the steps a 1-a 2 until the step a3 is finished

And s' (t) _a ) All values of are less than

Stopping updating the control points, re-fitting the curve by using the updated control points, carrying out curvature optimization and collision detection on the new curve, and regenerating a new lane reference line;

if multiple iterations are performed

Or s' (t) _a ) Is still greater than

Or the curve obtained by fitting the updated control point coordinates does not meet the curvature constraint or the collision detection, and a third lane reference line can be used.

Step a4, selecting the next interval, repeating the steps a 1-a 3 until [ t ] _a ,t _b ]All the intervals are processed.

In some embodiments, the acceleration optimization may include: and performing second-order derivation on the lane reference lines obtained by optimizing the speed to obtain the vehicle acceleration corresponding to each control point, and adjusting the positions of the control points with the vehicle accelerations larger than the maximum acceleration value until the vehicle accelerations corresponding to all the control points are not larger than the maximum acceleration value. Namely, the second-order derivation is carried out on the control points of the new lane reference line obtained through speed optimization to obtain second-order derivative values of the control points, the positions of the control points with the second-order derivative values larger than the preset maximum acceleration value are adjusted until the second-order derivative values of all the control points are smaller than or equal to the maximum acceleration value to obtain a new control point set, and a fourth lane reference line is obtained through fitting according to the new control point set.

In some embodiments, the vehicle at-variable parameter t may be calculated by the following equation (15) _i Acceleration of (2):

in formula (15), s' (t) _i ) For vehicle at variable parameter t _i Acceleration of (C) _i ,C _i+1 ,C _i+2 As a control point, t _i+1 Is a variable parameter t _i The next variable parameter.

In some embodiments, an exemplary implementation flow of acceleration optimization may include the following steps b1 to b 5:

step b1, selecting a first variable parameter interval [ t _a ,t _a+1 ]Determining the value of the acceleration s' (t) according to the control point updated by the speed optimization by using the formula (15);

step b2, if the value of s' (t) is greater than the maximum value of acceleration

Then C needs to be adjusted _i+1 The coordinates of (a);

generally, if s' (t) _i ) Value greater than maximum value of acceleration

Only C needs to be adjusted _i+1 The position of (a). But adjust C _i+1 Will influence s "(t) _i-1 )，s″(t _i+1 ) Therefore, it needs to iterate for many times to ensure the adjustment C _i+1 After position(s) "(t) _i )、s″(t _i-1 ) The values of (A) and (B) all meet the requirements.

Due to s' (t) _i+1 ) The optimization can be performed in the next variable parameter interval, thus in the current variable parameter interval t _a ,t _a+1 ]Only s' (t) needs to be considered _i-1 ) If s' (t) is present _i-1 ) Then updated control point coordinates C _i+1 It is necessary to satisfy s' (t) at the same time _i-1 ) Is less than

And s' (t) is less than

If s' (t) is not present _i-1 ) Then it need only be satisfied that s' (t) is less than

Step b3, adopting greedy strategy, selecting the distance C which satisfies the formula (16) _i+1 The coordinate of the nearest point is taken as C _i+1 New coordinates of (2);

that is, the control point coordinates are updated according to equation (16):

in the formula (16), the compound represented by the formula,

for vehicles in the interval t _a ,t _b ]The maximum value of the acceleration of (a),

the value of (b) can be set according to the actual condition of the road.

B4, re-fitting a curve by using the updated control point coordinates, and carrying out curvature optimization and collision detection on the new curve to generate a new lane reference line; if the lane reference line obtained by using the updated control point coordinate fitting does not meet the curvature constraint or the collision detection, the cubic B spline curve segment after the speed optimization processing is still used as a new lane reference line;

b5, selecting the next time interval, and repeating the steps b 1-b 4 until t _a ,t _b ]All the intervals of (2) are processed to completion.

And taking the lane reference line obtained through speed optimization and acceleration optimization as a fourth lane reference line.

In some embodiments, step S18 may further include: when the iteration number reaches the upper limit during speed optimization, the first derivative value of the control point is still larger than the maximum speed value, and the third lane reference line is directly used as a new lane reference line obtained through speed optimization; and when the iteration number reaches the upper limit during acceleration optimization, the second derivative value of the control point is still larger than the maximum value of the acceleration, and a new lane reference line obtained by speed optimization is directly used as a fourth lane reference line.

In some embodiments, when the speed optimization and the acceleration optimization are successful, the fourth lane reference line may be obtained, and when the speed optimization or the acceleration optimization fails, the fourth lane reference line may not be obtained, and at this time, the third lane reference line obtained in step S16 may be directly used, that is, the following two cases exist: 1) when the acceleration and the speed are successfully optimized, the fourth lane reference line is a feasible lane reference line; 2) and when the acceleration and speed optimization fails, the third lane reference line is a feasible lane reference line.

Through the feasible lane reference line obtained in the steps S12 to S18, the vehicle can smoothly run along the feasible lane reference line, and the motion planning and the speed and acceleration constraints are satisfied.

From the nature of cubic B-splines, the generation of cubic B-splines is related to the selection of control points. Therefore, cubic B-spline curves generated by different control points are different, feasible lane reference lines obtained at a single time are not necessarily optimal, and the conditions of long length, long driving time, curve knotting and the like can occur, so that an optimal lane reference line can be selected from a plurality of feasible lane reference lines to serve as an optimized lane reference line. That is, in some embodiments, step S18 may further include: and selecting one of the feasible lane reference lines as an optimized lane reference line, wherein the feasible lane reference line is a fourth lane reference line or a third lane reference line.

In some embodiments, one of the plurality of feasible lane reference lines may be selected as the optimized lane reference line according to one or more of the following: the length of the feasible lane reference line, the generation time of the feasible lane reference line, the driving comfort of the first vehicle on the feasible lane reference line, and the displacement between the feasible lane reference line and the original lane reference line.

In some embodiments, the preference coefficient of each feasible lane reference line is determined according to one or more of the following factors, and one feasible lane reference line is selected as the optimized lane reference line according to the preference coefficient: the length of the feasible lane reference line, the generation time of the feasible lane reference line, the driving comfort of the first vehicle on the feasible lane reference line, and the displacement between the feasible lane reference line and the original lane reference line.

For example, the preference coefficients of the feasible lane reference lines can be determined according to the lengths of the feasible lane reference lines, the driving comfort of the vehicle on the feasible lane reference lines and/or the displacement between the feasible lane reference lines and the original lane reference lines, and the feasible lane reference line with the smallest preference coefficient is selected as the optimized lane reference line.

In some embodiments, an exemplary method of selecting one feasible lane reference line among a plurality of feasible lane reference lines as a preferred lane reference line may include the steps of:

and c1, after obtaining a plurality of feasible lane reference lines through the steps S14-S18, calculating the preference coefficient of each feasible lane reference line according to the length of the lane reference line, the generation time of the lane reference line, the driving comfort of the lane reference line, the displacement of the lane reference line and the original lane reference line and the like.

Discrete points of the original lane reference lines corresponding to different feasible lane reference lines cannot be completely consistent. In addition, the number of the feasible lane reference lines can be set according to the actual requirement, and for example, can be set to 200.

The length of the lane reference line indicates the distance that the vehicle travels along the lane reference line, and the longer the distance, the longer the vehicle travels, and thus a lane reference line of a long length needs to be penalized.

The time for generating the lane reference line represents the ease of generating the lane reference line, and longer time means that the steps of processing the lane reference line are more complicated, and thus the lane reference line having a long generation time needs to be punished.

In order to enhance the experience of the passengers, in the embodiment of the present disclosure, the driving comfort is taken as one of the criteria for evaluating the lane reference line. In a specific application, other indexes such as the driving safety of the vehicle can be replaced according to needs, and of course, driving comfort or other similar indexes can be not considered in the evaluation of the lane reference line according to needs.

In some embodiments, the driving comfort of the lane reference line can be represented by a third derivative of the lane reference line, and the larger the third derivative is, the worse the driving comfort is, therefore, the lane reference line with too large third derivative needs to be punished.

The displacement of the lane reference line from the origin lane reference line represents the deviation of the lane reference line from the origin lane reference line. If the optimized lane reference line is required to be closer to the origin lane reference line, the lane reference line which is displaced from the origin lane reference line can be punished.

In some embodiments, the preference factor for the feasible lane reference line may be obtained by the following equation (17):

in the formula (17), I _total A priority coefficient for a feasible lane reference line; w is a _l ，w _t ，w _j ，w _o For weighting, it is necessary to set according to specific requirements, for example, a short-length lane reference line is required, w can be set _l Set to a larger value; a lane reference line with small calculation amount is needed, and w can be calculated _t Set to a larger value; to emphasize the comfort level, can change w _j Set to a larger value; the lane reference line to be generated is close to the original lane reference line, and w can be set _o Set to a larger value; integral operation, and the length of the cubic B spline curve is calculated by the formula(s) (t); t is ₁ ，T ₀ Respectively calculating the ending time and the starting time of the feasible lane reference line; s '(t) represents the third derivative of the cubic B-spline curve, [ integral ] s', and' (t) denotes the integration of the third derivative of the cubic B-spline curve; sigma is a summation operation, sigma (s (t) -s _O ) ² Representing the sum of squares of the displacements, s, of the lane reference line and the origin lane reference line _O Coordinates representing discrete points on the origin lane reference line.

It should be noted that since the origin lane reference line is composed of discrete points, the displacement between the optimized lane reference line and the origin lane reference line can only be calculated by cumulative summation, not by integration.

Step c2, calculating the optimal coefficient I of each feasible lane reference line _total And then selecting a feasible lane reference line with the minimum optimization coefficient as an optimized lane reference line for the vehicle to use.

In the embodiment of the disclosure, by setting various evaluation parameters including the length of the lane reference line, the generation time of the lane reference line, the driving comfort level of the lane reference line, and the displacement of the lane reference line and the original lane reference line, the optimal lane reference line suitable for the current vehicle and the current scene is selected, so that the vehicle can be ensured to smoothly and safely pass through a curve area, and to safely drive and smoothly drive when driving along the lane reference line.

In summary, the lane reference line optimization method of the embodiment of the present disclosure can obtain the following beneficial effects:

1) by utilizing the smooth and continuous property of lie algebra in a number domain space and mapping a curve to a lie group space through exponential mapping, any two points on the curve can be mutually converted through rigid transformation (namely rotation and translation), and the obtained lane reference line can be ensured to be smooth and continuous in any scene. Namely, the vehicle starts from one point on the lane reference line and can reach any other point on the lane reference line along the lane reference line, so that the generated lane reference line can be suitable for different scenes;

2) by introducing curvature constraints and taking into account vehicle parameters of different vehicles in the curvature constraints; different lane reference lines are generated according to different vehicle parameters, and different vehicles can pass through the curve area according to the optimized lane reference lines.

3) By introducing collision detection and motion planning constraint, the vehicle can smoothly and comfortably run on the optimized lane reference line, and excessive obstacle avoidance control is avoided.

4) Introducing various external constraints, specifically comprising curvature constraints, collision detection, speed constraints and acceleration constraints, ensuring that an optimized lane reference line is smooth and continuous and meeting the requirements of driving safety and driving;

5) only the sampling points are needed to be calculated, so that a large amount of calculation is avoided, and the efficiency of each iteration is improved;

6) a plurality of parameters are set to evaluate the feasible lane reference lines, so that the lane reference lines with good comfort and safety can be obtained, and the riding experience of passengers and drivers is improved.

7) The method can be suitable for different scenes, curvature optimization is not required to be carried out by utilizing the collected path track, the characteristics of different vehicles are considered, and the condition that the optimized lane reference line can be suitable for different vehicles is ensured.

The apparatus may include corresponding means for performing each or several of the steps of the flowcharts described above. Thus, each step or several steps in the above-described flow charts may be performed by a respective module, and the apparatus may comprise one or more of these modules. The modules may be one or more hardware modules specifically configured to perform the respective steps, or implemented by a processor configured to perform the respective steps, or stored within a computer-readable medium for implementation by a processor, or by some combination.

The hardware architecture may be implemented with a bus architecture. The bus architecture may include any number of interconnecting buses and bridges depending on the specific application of the hardware and the overall design constraints. The bus 900 connects together various circuits including one or more processors 1000, memories 1100, and/or hardware modules. The bus 900 may also connect various other circuits 1200 such as peripherals, voltage regulators, power management circuits, external antennas, and the like.

The bus 900 may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one connection line is shown, but no single bus or type of bus is shown.

Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps in the process, and alternate implementations are included within the scope of the preferred embodiment of the present disclosure in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the implementation of the present disclosure. The processor performs the various methods and processes described above. For example, method embodiments in the present disclosure may be implemented as a software program tangibly embodied in a machine-readable medium, such as a memory. In some embodiments, some or all of the software program may be loaded and/or installed via memory and/or a communication interface. When the software program is loaded into memory and executed by a processor, one or more steps of the method described above may be performed. Alternatively, in other embodiments, the processor may be configured to perform one of the methods described above by any other suitable means (e.g., by means of firmware).

The logic and/or steps represented in the flowcharts or otherwise described herein may be embodied in any readable storage medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions.

For the purposes of this description, a "readable storage medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the readable storage medium include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable read-only memory (CDROM). In addition, the readable storage medium may even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in the memory.

It should be understood that portions of the present disclosure may be implemented in hardware, software, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.

It will be understood by those skilled in the art that all or part of the steps of the method implementing the above embodiments may be implemented by hardware that is instructed to be associated with a program, which may be stored in a readable storage medium, and which, when executed, includes one or a combination of the steps of the method embodiments.

In addition, each functional unit in the embodiments of the present disclosure may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a separate product, may also be stored in a readable storage medium. The storage medium may be a read-only memory, a magnetic or optical disk, or the like.

As shown in fig. 8, a lane reference line optimizing apparatus 800 according to some embodiments of the present disclosure may include:

the acquiring unit 802 is configured to acquire an original lane reference line corresponding to an environment where a first vehicle is located from a high-definition map;

the fitting unit 804 is configured to fit the randomly selected discrete points on the original lane reference line to obtain a first lane reference line; the second lane reference line is obtained according to the control point set obtained by the curvature optimization unit; the collision detection unit is used for updating the control point set of the control point set to obtain a control point set;

a curvature optimization unit 806, configured to perform curvature optimization on the sampling points in the first lane reference line according to a maximum curvature threshold of the first vehicle to obtain a set of control points;

a collision detection unit 808 configured to perform collision detection on the second lane reference line to update the set of control points;

and a third lane reference line optimizing unit 810, configured to obtain, according to the third lane reference line, an optimized lane reference line corresponding to an environment where the first vehicle is located.

In some embodiments, the fitting unit 804 is configured to fit the first lane reference line by: determining control points and coordinates thereof according to a curve equation of the reference line of the original lane obtained by fitting the discrete points so as to obtain a control point set; and according to the control point set, fitting through a rotating curve equation and a cubic B-spline curve equation to obtain the first lane reference line, wherein the rotating curve equation is constructed based on rigid motion characteristics.

In some embodiments, the vehicle parameters may include one or more of: the wheel base of the vehicle, the width of the front wheel of the vehicle, the distance from the center of the front axle of the vehicle to the center of the rear axle of the vehicle, the maximum rotating angle of a steering wheel of the vehicle, the wheel base of the front wheel of the vehicle and the center distance of a kingpin of the vehicle.

In some embodiments, the curvature optimization unit 806 is further configured to derive the maximum curvature threshold by equation (10).

In some embodiments, the curvature optimization unit 806 is specifically configured to: taking sampling points obtained by uniformly sampling the first lane reference line as control points to form a control point set; calculating the curvature of each control point in the control point set; and performing curvature optimization on the control points with the curvatures larger than the maximum curvature threshold value in the control point set to update the control point set until the curvatures of all the control points in the control point set are smaller than or equal to the maximum curvature threshold value.

In some embodiments, the curvature optimization unit 806 is specifically configured to: and inserting a new control point into a lane line segment formed by the closest control point of the control point with the curvature larger than the maximum curvature threshold value and the adjacent control point of the closest control point according to the maximum turning angle of the steering wheel of the vehicle, and adding the new control point into the control point set.

In some embodiments, the collision detection unit 808 is specifically configured to: and performing collision detection on the second lane reference line, determining a serial number, a starting point and an end point of a curve segment where the second lane reference line collides with an obstacle when the second lane reference line collides with a static obstacle, inserting a new control point between the starting point and the end point of the curve segment corresponding to the serial number, and incorporating the new control point into the control point set.

In some embodiments, the collision detection unit 808 is specifically configured to: and judging whether the minimum directional bounding box of the second lane reference line is intersected with the bounding box of the obstacle, wherein the intersection indicates that the second lane reference line is collided with the obstacle.

In some embodiments, the third lane reference line optimizing unit 810 includes: and a speed optimization unit 812, configured to perform speed optimization and acceleration optimization on the third lane reference line according to preset maximum speed values and maximum acceleration values to obtain a fourth lane reference line.

In some embodiments, the speed optimization unit 812 is specifically configured to: and obtaining the vehicle speed corresponding to each control point by carrying out first-order derivation on the third lane reference line, and adjusting the position of the control point with the vehicle speed being greater than the maximum speed value until the vehicle speeds corresponding to all the control points are not greater than the maximum speed value.

In some embodiments, the speed optimization unit 812 is specifically configured to: and performing second-order derivation on the lane reference line obtained by optimizing the speed to obtain the vehicle acceleration corresponding to each control point, and adjusting the positions of the control points with the vehicle acceleration greater than the maximum acceleration until the vehicle accelerations corresponding to all the control points are not greater than the maximum acceleration.

In some embodiments, the third lane reference line optimizing unit 810 further includes: an evaluation unit 814, configured to select one of the plurality of feasible lane reference lines as the optimized lane reference line according to one or more of the following: the method comprises the following steps of obtaining a feasible lane reference line, wherein the feasible lane reference line is a fourth lane reference line or a third lane reference line, and comprises the length of the feasible lane reference line, the generation time of the lane reference line, the driving comfort level of a first vehicle on the feasible lane reference line, and the displacement between the feasible lane reference line and an original lane reference line.

In some embodiments, the evaluation unit 814 is specifically configured to: determining a preference coefficient of each feasible lane reference line according to one or more of the following factors, and selecting a feasible lane reference line as the optimized lane reference line according to the preference coefficient: the length of the feasible lane reference line, the generation time of the feasible lane reference line, the driving comfort of the first vehicle on the feasible lane reference line, and the displacement between the feasible lane reference line and the original lane reference line.

The present disclosure also provides an electronic device, including: a memory storing execution instructions; and a processor or other hardware module that executes the execution instructions stored by the memory, such that the processor or other hardware module performs the lane reference line optimization method described above.

By way of example, the electronic device may be implemented as, but is not limited to: an onboard device, an onboard module, an onboard controller, or other similar device. In a particular application, the electronic device may be deployed in a vehicle. The present disclosure is not limited to specific implementation forms and deployment modes of the electronic device.

The present disclosure also provides a readable storage medium having stored therein execution instructions, which when executed by a processor, are used to implement the lane reference line optimization method described above.

In the description herein, reference to the description of the terms "one embodiment/implementation," "some embodiments/implementations," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment/implementation or example is included in at least one embodiment/implementation or example of the present application. In this specification, the schematic representations of the terms described above are not necessarily the same embodiment/mode or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments/modes or examples. Furthermore, the various embodiments/aspects or examples and features of the various embodiments/aspects or examples described in this specification can be combined and combined by one skilled in the art without conflicting therewith.

Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.

It will be understood by those skilled in the art that the foregoing embodiments are merely for clarity of illustration of the disclosure and are not intended to limit the scope of the disclosure. Other variations or modifications may occur to those skilled in the art, based on the foregoing disclosure, and are still within the scope of the present disclosure.

Claims

1. A lane reference line optimization method, comprising:

2. The method of claim 1, wherein the fitting a first lane reference line according to randomly selected discrete points on the origin lane reference line comprises:

3. The lane reference line optimization method of claim 1, wherein the vehicle parameters may include one or more of: the distance between the vehicle wheelbase and the front wheel, the distance from the center of the front axle of the vehicle to the center of the rear axle, the maximum rotating angle of the steering wheel of the vehicle, the front wheelbase of the vehicle and the center distance of a kingpin of the vehicle;

preferably, the maximum curvature threshold is obtained by:

K _max ＝1/(L/sin(θ _max )+(b-M)/2)

wherein, K _max Is a maximum curvature threshold, L is the distance from the center of the front axle to the center of the rear axle of the vehicle, theta _max The maximum turning angle of the steering wheel of the vehicle, b the front wheel track of the vehicle and M the kingpin center distance of the vehicle.

4. The lane reference line optimization method of claim 1, wherein said curvature optimizing sample points in the first lane reference line according to the maximum curvature threshold of the first vehicle to obtain a set of control points comprises:

taking sampling points obtained by uniformly sampling the first lane reference line as control points to form a control point set;

calculating the curvature of each control point in the control point set;

performing curvature optimization on the control points with the curvatures larger than the maximum curvature threshold value in the control point set to update the control point set until the curvatures of all the control points in the control point set are smaller than or equal to the maximum curvature threshold value;

preferably, performing curvature optimization on the control points in the set of control points whose curvature is greater than the maximum curvature threshold value to update the set of control points includes: and inserting a new control point into a lane line segment formed by the closest control point of the control point with the curvature larger than the maximum curvature threshold value and the adjacent control point of the closest control point according to the maximum turning angle of the steering wheel of the vehicle, and adding the new control point into the control point set.

5. The lane reference line optimization method of claim 1, wherein collision detection of the second lane reference line to update the set of control points comprises: performing collision detection on the second lane reference line, determining a serial number, a starting point and an end point of a curve segment where the second lane reference line collides with an obstacle when the second lane reference line collides with a static obstacle, inserting a new control point between the starting point and the end point of the curve segment corresponding to the serial number, and incorporating the new control point into the control point set;

preferably, the collision detection includes: and judging whether the minimum directional bounding box of the second lane reference line is intersected with the bounding box of the obstacle, wherein the intersection indicates that the second lane reference line is collided with the obstacle.

6. The lane reference line optimization method of claim 1, wherein the obtaining an optimized lane reference line corresponding to an environment in which the first vehicle is located according to the third lane reference line comprises: carrying out speed optimization and acceleration optimization on the third lane reference line according to preset maximum speed values and maximum acceleration values to obtain a fourth lane reference line;

preferably, the speed optimization comprises: obtaining the vehicle speed corresponding to each control point by carrying out first-order derivation on the third lane reference line, and adjusting the position of the control point with the vehicle speed being greater than the maximum speed value until the vehicle speeds corresponding to all the control points are not greater than the maximum speed value;

preferably, the acceleration optimization comprises: obtaining the vehicle acceleration corresponding to each control point by carrying out second-order derivation on the lane reference line obtained by optimizing the speed, and adjusting the positions of the control points with the vehicle acceleration greater than the maximum acceleration value until the vehicle accelerations corresponding to all the control points are not greater than the maximum acceleration value;

preferably, the obtaining an optimized lane reference line corresponding to an environment where the first vehicle is located according to the third lane reference line further includes:

selecting one of the plurality of feasible lane reference lines as the optimized lane reference line according to one or more of the following items: the length of the feasible lane reference line, the generation time of the lane reference line, the driving comfort level of a first vehicle on the feasible lane reference line, and the displacement between the feasible lane reference line and an original lane reference line, wherein the feasible lane reference line is the fourth lane reference line or the third lane reference line;

preferably, the selecting one of the feasible lane reference lines as the optimized lane reference line comprises:

determining a preference coefficient of each feasible lane reference line according to one or more of the following factors, and selecting a feasible lane reference line as the optimized lane reference line according to the preference coefficient:

the length of the feasible lane reference line, the generation time of the feasible lane reference line, the driving comfort of a first vehicle on the feasible lane reference line, and the displacement between the feasible lane reference line and an original lane reference line;

wherein the feasible lane reference line is the fourth lane reference line or the third lane reference line.

7. A lane reference line optimizing apparatus, comprising:

8. The lane reference line optimization device of claim 7, wherein the fitting unit is configured to fit the first lane reference line by:

according to the control point set, fitting through a rotating curve equation and a cubic B-spline curve equation to obtain the first lane reference line, wherein the rotating curve equation is constructed based on rigid motion characteristics;

preferably, the vehicle parameters may include one or more of: the distance between the vehicle wheelbase and the front wheel, the distance from the center of the front axle of the vehicle to the center of the rear axle, the maximum rotating angle of the steering wheel of the vehicle, the front wheelbase of the vehicle and the center distance of a kingpin of the vehicle;

preferably, the curvature optimizing unit is further configured to obtain the maximum curvature threshold value by:

K _max ＝1/(L/sin(θ _max )+(b-M)/2)

wherein, K _max Is a maximum curvature threshold, L is the distance from the center of the front axle to the center of the rear axle of the vehicle, theta _max The maximum turning angle of a steering wheel of the vehicle, b is the front wheel track of the vehicle, and M is the center distance of a kingpin of the vehicle;

preferably, the curvature optimization unit is specifically configured to:

calculating the curvature of each control point in the control point set;

preferably, the curvature optimizing unit is specifically configured to insert a new control point into a lane line segment formed by a control point with a curvature greater than the maximum curvature threshold value and a control point adjacent to the control point with the closest curvature threshold value according to the maximum steering angle of the steering wheel of the vehicle, and add the new control point into the control point set;

preferably, the collision detection unit is specifically configured to: performing collision detection on the second lane reference line, determining a serial number, a starting point and an end point of a curve segment where the second lane reference line collides with an obstacle when the second lane reference line collides with a static obstacle, inserting a new control point between the starting point and the end point of the curve segment corresponding to the serial number, and incorporating the new control point into the control point set;

preferably, the collision detection unit is specifically configured to: judging whether the minimum directional bounding box of the second lane reference line is intersected with the bounding box of the obstacle, if so, indicating that the second lane reference line is collided with the obstacle;

preferably, the third lane reference line optimizing unit includes: the speed optimization unit is used for carrying out speed optimization and acceleration optimization on the third lane reference line according to preset maximum speed values and maximum acceleration values to obtain a fourth lane reference line;

preferably, the speed optimization unit is specifically configured to: obtaining the vehicle speed corresponding to each control point by carrying out first-order derivation on the third lane reference line, and adjusting the position of the control point with the vehicle speed being greater than the maximum speed value until the vehicle speeds corresponding to all the control points are not greater than the maximum speed value;

preferably, the speed optimization unit is specifically configured to: obtaining the vehicle acceleration corresponding to each control point by carrying out second-order derivation on the lane reference line obtained by optimizing the speed, and adjusting the positions of the control points with the vehicle acceleration greater than the maximum acceleration value until the vehicle accelerations corresponding to all the control points are not greater than the maximum acceleration value;

preferably, the third lane reference line optimizing unit further includes: an evaluation unit, configured to select one of the plurality of feasible lane reference lines as the optimized lane reference line according to one or more of the following: the length of the feasible lane reference line, the generation time of the lane reference line, the driving comfort level of a first vehicle on the feasible lane reference line, and the displacement between the feasible lane reference line and an original lane reference line, wherein the feasible lane reference line is the fourth lane reference line or the third lane reference line;

preferably, the evaluation unit is specifically configured to: determining a preference coefficient of each feasible lane reference line according to one or more of the following factors, and selecting a feasible lane reference line as the optimized lane reference line according to the preference coefficient: the length of the feasible lane reference line, the generation time of the feasible lane reference line, the driving comfort level of the first vehicle on the feasible lane reference line, and the displacement between the feasible lane reference line and the original lane reference line.

9. An electronic device, comprising:

a memory storing execution instructions; and

a processor executing execution instructions stored by the memory such that the processor performs the lane reference line optimization method of any of claims 1 to 6.

10. A readable storage medium, characterized in that the readable storage medium has stored therein execution instructions for implementing the lane reference line optimization method of any one of claims 1 to 6 when executed by a processor.