Method and device for evaluating transverse and longitudinal errors of map lane lines, electronic equipment and medium

Abstract

The invention relates to the technical field of intelligent driving, and discloses a map lane line transverse and longitudinal error evaluation method, which comprises the following steps: S10, obtaining lane line data of a true-value lane line L1 and a to-be-measured lane line L2, the lane line data comprising geographic positions; s20, carrying out equidistant data point acquisition on the L2; s30, the true value lane line L1KD-tree is subjected to structuring, and the true value lane line L1KD-tree is obtained; s40, calculating the point-to-broken line distance from the midpoint of L2 to L1, and normalizing the distance according to the number of m points to obtain a final result d (L1, L2), the formula being # imgabs0 #, m being the sampling number of L1, p being a two-dimensional space point expressed by horizontal and vertical coordinates, and d (L1, L2) being the distance between L1 and L2; the final result d (L1, L2) is a transverse and longitudinal absolute error of L2 relative to L1.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving technology, and in particular, to a method, device, electronic device and medium for evaluating the horizontal and vertical errors of map lane lines. Background Art

[0002] In the field of intelligent driving, mapping technology has always played a role in escorting intelligent driving. Although the industry increasingly emphasizes perception capabilities, end-to-end model technology capabilities, and mapless intelligent driving capabilities, and most have achieved mapless autonomous driving solutions on highways and urban expressways. However, in the complex urban areas of our country, mapless intelligent driving technology still faces great challenges. For example, at urban intersections, without the help of a base map, it is very difficult for the vehicle's path planning and control module to make reliable decisions for areas outside the perception range. At this time, it is necessary to rely on maps and positioning to generate pre-anchor points or reference trajectories for driving; another example is scenarios such as highway toll stations, above-ground and ground parking lots, and urban dense intersections, where maps play a guarantee role in real-time perception and intelligent vehicles. Therefore, in recent years, the industry has re-recognized the importance of maps for intelligent driving and has begun to vigorously develop crowd-sourcing technology. Among them, crowd-sourcing technology is a low-cost lightweight concept born compared to the traditional high-precision map production process. The difference is that it uses a large number of sub-high-precision sensors, adopts a strategy of multiple collections in multiple trips for crowd-sourcing mapping, and uses cloud algorithms to realize the production of quasi-high-precision maps, thereby assisting the use of downstream intelligent driving. Since the production of crowd-sourcing maps is a process of technological iteration, how to effectively measure the error between the constructed map and the ground truth map during the production and algorithm development process is crucial, otherwise the algorithm will lack the target of iteration.

[0003] In the field of traditional map accuracy evaluation, it is all based on the matching method of feature points. However, in actual maps, most of the expressed lane lines are parallel lane lines, and longitudinal errors only occur at intersections and when encountering stop lines. And the lane line perception output by general perception models is independent for the categories of solid and dashed lines and stop lines. This will make the horizontal and vertical errors in the map to be measured relatively independent, and the effect of the feature point-based method will be poor when measuring the horizontal error. For this situation, there is a method of calculating the distance between lines called the one-way distance calculation method, which can effectively consider the calculation of horizontal error, but the object it faces is not the absolute accuracy of the error of high-precision maps but the similarity.

[0004] Therefore, there is an urgent need for a method that can efficiently and quickly calculate the absolute error on the basis of considering the horizontal and vertical errors of two lane lines, providing calculation support for subsequent calculation of the error between two maps. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to solve the problem that the horizontal and vertical errors of the existing map lane lines are not calculated accurately enough.

[0006] To solve the above technical problem, the present invention provides a method for evaluating the horizontal and vertical errors of map lane lines, and the method includes:

[0007] S10. Obtain the lane line data of the true value lane line L1 and the lane line L2 to be measured, and the lane line data includes geographical locations;

[0008] S20. Collect data points at equal distances for L2;

[0009] S30. Structure the true value lane line L1 into a KD-tree;

[0010] S40. Calculate the distance from the midpoint of L2 to L1 as the distance from a point to a polyline, and normalize the distance according to the number of m points to obtain the final result d(L1, L2). The formula is:

[0011]

[0012] where m is the number of samples of L1, and p represents a two-dimensional space point expressed by horizontal and vertical coordinates. d(L1, L2) is the distance between L1 and L2;

[0013] The final result d(L1, L2) is the horizontal and vertical absolute error of L2 compared to L1.

[0014] Furthermore, the lane line data of the true value lane line L1 and the lane line L2 to be measured is a two-dimensional polyline in the form of an ordered point series, denoted as:

[0015]

[0016]

[0017] Furthermore, the method includes

[0018] Define the distance from a point p to a line segment S = {p_s, p_e} as d(p, S), and the line segment S is composed of a starting point p_s and an ending point p_e;

[0019] When the perpendicular from the point p to the line segment S exists and the intersection point is on the line segment S, d(p, S) is the length of the perpendicular from the point p to the line segment, which is dp

[0020] When the perpendicular from the point p to the line segment S exists but the intersection point is not on the line segment S, d(p, S) is the minimum value ds of the distances from the point p to the starting point ps and pe of the line segment S;

[0021] According to the distance from the point to the line segment, when the point p is far away from the line segment, a larger value will be calculated, and the larger value is the longitudinal constraint.

[0022] Furthermore, the method includes:

[0023] Define the distance d(p, L) from a point p to a curve L, and the calculation method is to take the minimum value of the distances from d to the line segments formed by two points in sequence on L1. The formula is:

[0024]

[0025] where S i ={p i , p i+1}.

[0026] Furthermore, the specific steps of structuring the true value lane line L1 into a KD-tree are

[0027] S31, obtain the true value lane line L1;

[0028] S32, use the KD-Tree algorithm to construct the points in L1 into a KD-Tree structure S33, obtain a search point p;

[0029] S33, use the nearest neighbor search algorithm of the KD-Tree to search for 6 nearest points in the true value KD-Tree with p as the center;

[0030] S34, connect the 6 nearest points found in sequence to form a new curve L_{KD}, and the L_{KD} is an approximate representation of L1 near the point p and is used to calculate the distance from a point to a curve;

[0031] S35, record that d(p, L1) is approximately equal to d(p, L_{KD}).

[0032] According to another aspect of the present invention, there is provided an evaluation device for the transverse and longitudinal errors of a map lane line. When using the evaluation device for evaluation, it includes the above-mentioned evaluation method for the transverse and longitudinal errors of the map lane line.

[0033] Furthermore, the device includes:

[0034] An information acquisition module, which is used to acquire the lane line data of the true value lane line L1 and the lane line L2 to be measured, and the lane line data includes geographical locations;

[0035] A data point acquisition module, which is used to equally distance collect data points for L2;

[0036] An information processing module, which is used to structure the true value lane line L1 into a KD-tree;

[0037] A calculation module, which traverses the distance d(p, L1) from the midpoint p of L2 to L1, where p represents a two-dimensional spatial point expressed by horizontal and vertical coordinates;

[0038] An error evaluation module, which is used to calculate the point-to-polyline distance from the midpoint of L2 to L1, and normalize the distance according to the number of points m to obtain the final result d(L1, L2). The formula is:

[0039]

[0040] where m is the number of samples of L1, and p represents a two-dimensional spatial point expressed by horizontal and vertical coordinates;

[0041] The final result d(L1, L2) is the horizontal and vertical absolute error of L2 compared to L1.

[0042] According to another aspect of the present invention, there is provided an electronic device, including: at least one processor, and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute any method for evaluating the horizontal and vertical errors of map lane lines in the embodiments of the present invention.

[0043] Compared with the prior art, the method for evaluating the horizontal and vertical errors of map lane lines in the embodiments of the present invention has the following beneficial effects:

[0044] In the embodiments of the present invention, by structuring the true value lane line L1 into a KD-tree, the true value lane line point closest to each point on the lane line L2 to be measured is searched, so as to more accurately calculate the point-to-polyline distance. The method of collecting data points at equal distances ensures that the points on L2 are evenly distributed, avoiding error accumulation caused by uneven data point distribution. Through normalization processing, the present invention combines the point-to-polyline distances from all points on L2 to L1 into a final result d(L1, L2), which reflects the overall horizontal and vertical error of L2 compared to L1 and enhances the accuracy of the evaluation result. Description of the Drawings

[0045] Figure 1 is a flowchart of the method for evaluating the horizontal and vertical errors of map lane lines provided by the embodiments of the present invention;

[0046] Figure 2 is a schematic diagram of the relationship between point p and line segment s provided by the embodiments of the present invention;

[0047] Figure 3It is a schematic diagram for calculating the distance between the true value lane and the lane line to be measured provided by an embodiment of the present invention;

[0048] Figure 4 It is a schematic diagram of an evaluation device for the horizontal and vertical errors of the map lane line provided by an embodiment of the present invention;

[0049] Figure 5 It is a block diagram of an electronic device for implementing an embodiment of the present invention.

[0050] In the figure, 10 is an information acquisition module; 20 is a data point acquisition module; 30 is an information processing module; 40 is an error evaluation module; 600 is an electronic device; 601 is a calculation unit; 602 is a ROM; 603 is a RAM; 604 is a bus; 605 is an I / O interface; 606 is an input unit; 607 is an output unit; 608 is a storage unit; 609 is a communication unit. Detailed implementation manners

[0051] The following describes exemplary embodiments of the present invention with reference to the accompanying drawings. Various details of the embodiments of the present invention are included to facilitate understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope of the present invention. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following.

[0052] As Figures 1 - 3 shown, in an alternative embodiment of the present invention, the method for evaluating the horizontal and vertical errors of the map lane line includes:

[0053] S10. Obtain the lane line data of the true value lane line L1 and the lane line L2 to be measured, where the lane line data includes geographical locations;

[0054] S20. Collect data points at equal distances for L2;

[0055] S30. Structure the true value lane line L1 into a KD-tree;

[0056] S40. Traverse the distance d(p, L1) from the point p on L2 to L1, where p represents a two-dimensional space point expressed by horizontal and vertical coordinates;

[0057] S50. Calculate the distance from the point on L2 to the polyline of L1, and normalize the distance according to the number of points m to obtain the final result d(L1, L2). The formula is:

[0058]

[0059] where m is the number of samples of L1, and p represents a two-dimensional space point expressed by horizontal and vertical coordinates;

[0060] The final result d(L1, L2) is the absolute horizontal and vertical error of L2 compared to L1.

[0061] Specifically, combined with Figures 1 - 3 Further explanation:

[0062] The true value map lane line L1 refers to a determined lane line reference standard. In applications such as autonomous driving and map matching, L1 is used to compare with the lane line L2 to be measured to evaluate the error of L2. L1 is represented in the form of a two-dimensional polyline, connected by a series of ordered point strings, and these points precisely describe the shape and position of the lane line. All lane line inputs are two-dimensional polylines and are represented in the form of ordered point strings. Define the true value map lane line L1 and the lane line L2 to be measured. p represents a two-dimensional space point expressed by horizontal and vertical coordinates, where:

[0063]

[0064] First, define the distance from a point p to a line segment S = {p_s, p_e} as d(p, S), and the specific calculation method is combined with Figure 2 Explanation:

[0065] When the perpendicular from point p to line segment S exists and the intersection point is on line segment S, d(p, S) is the length of the perpendicular from point p to line segment S, which is dp;

[0066] When the perpendicular from point p to line segment S exists but the intersection point is not on line segment S, d(p, S) is the minimum of the distances from point p to the starting point p_s and the ending point p_e of line segment S, which is ds in the schematic diagram

[0067] Through this definition of the distance from a point to a line segment, it can be observed that it will calculate a larger value when point p is far from the line segment, thus providing a longitudinal constraint.

[0068] Define the distance from a point p to a curve L as d(p, L), and the calculation method is to take the minimum value of the distances from d to the line segments formed by two adjacent points of L1, which is represented by the following formula:

[0069]

[0070] Among them, S i ={p i , p i+1}.

[0071] In the above formula \(d(p, L1)=\min_{d(p, S_i)}\), it can be seen that when calculating the distance from a point to a curve, it is necessary to traverse the distances from the point to all sub-line segments and take the minimum value, which is a linear search. For a curve with a large number of points or a long curve, the efficiency of using the traversal method is not high. This method introduces the use of a KD-Tree structure to accelerate the calculation of the minimum value. The KD-Tree itself is an extension of the binary tree structure in the K-dimensional space, which can reduce the linear search degree from \(O(N)\) to \(O(\log N)\). The specific process is as follows: 1. First, obtain the true value lane line \(L1\); 2. Structure the true value lane line into a KD-Tree; 3. Obtain a search point \(p\); 4. Search for the 6 nearest points of the true value KD-Tree centered on \(p\); 5. Form a new curve \(L_{KD}\) with the 6 nearest points; 6. Denote \(d(p, L1)\approx d(p, L_{KD})\).

[0072] Finally, define the improved one-way distance method for calculating the lane line distance \(d(L1, L2)\) in this patent, where \(L1\) is the true value map lane line and \(L2\) is the lane line to be measured.

[0073]

[0074] First, obtain the true value lane line \(L1\) and the lane line \(L2\) to be measured that need to be calculated; perform equidistant resampling on \(L2\) so that \(L2\) is composed of \(m\) points connected in series; construct a KdTree structure for \(L1\); for the \(L1\) KdTree structure, quickly calculate the distance from the points on \(L2\) to the broken line of \(L1\) in sequence, and normalize according to the number of \(m\) points; output the final result, which is recorded as the absolute error of \(L2\) compared to \(L1\).

[0075] In the embodiment of the present invention, by structuring the true value lane line \(L1\) into a KD-tree, the true value lane line points closest to each point on the lane line \(L2\) to be measured are searched, so as to more accurately calculate the distance from a point to a broken line. The method of collecting data points at equal distances ensures that the points on \(L2\) are evenly distributed, avoiding error accumulation caused by uneven data point distribution. Through normalization processing in the present invention, the distances from all points on \(L2\) to the broken line of \(L1\) are integrated into a final result \(d(L1, L2)\), which reflects the overall horizontal and vertical errors of \(L2\) compared to \(L1\) and enhances the accuracy of the evaluation result.

[0076] In an optional embodiment of the present invention, the lane line data of the true value lane line \(L1\) and the lane line \(L2\) to be measured are two-dimensional broken lines in the form of an ordered point series, denoted as:

[0077]

[0078] Specifically, the lane line data is represented as a two-dimensional polyline formed by a series of ordered points connected in series, which is convenient for computer processing and comparison. This simplifies the representation of lane line data and improves the calculation efficiency and accuracy. At the same time, this representation method is also convenient for calculating the distance from a point to a polyline, providing a basis for error evaluation.

[0079] In an alternative embodiment of the present invention, it is characterized in that the method includes

[0080] Define the distance from a point p to a line segment S = {p_s, p_e} as d(p, S), where the line segment S is composed of a starting point p_s and an ending point p_e;

[0081] When the perpendicular from point p to line segment S exists and the intersection point is on line segment S, d(p, S) is the length of the perpendicular from point p to line segment S, which is dp

[0082] When the perpendicular from point p to line segment S exists but the intersection point is not on line segment S, d(p, S) is the minimum value ds of the distances from point p to the starting point ps and pe of line segment S;

[0083] According to the distance from the point to the line segment, a larger value will be calculated when point p is far from the line segment, and this larger value is the longitudinal constraint.

[0084] Among them, line segment S is the straight-line part determined by two end points p_s (starting point) and p_e (ending point). Line segment S = {p_s, p_e}, where p_s and p_e are the starting point and ending point of the line segment respectively. Line segment S is used as the reference object for distance calculation to determine the shortest distance from point p to it.

[0085] In an alternative embodiment of the present invention, the method includes:

[0086] Define the distance d(p, L) from a point p to a curve L. The calculation method is to take the minimum value of the distances from d to the line segments formed by two points in sequence on L1. The formula is:

[0087]

[0088] Among them, S i = {p i , p i+1}.

[0089] In an alternative embodiment of the present invention, the specific steps for structuring the true value lane line L1 into a KD-tree are as follows:

[0090] S31. Obtain the true value lane line L1;

[0091] S32. Use the KD-Tree algorithm to construct the points in L1 into a KD-Tree structure S33. Obtain a search point p;

[0092] S33. Use the nearest neighbor search algorithm of the KD-Tree to search for the 6 nearest points in the true value KD-Tree centered on p;

[0093] S34. Connect the 6 nearest points found in sequence to form a new curve L_{KD}. The L_{KD} is an approximate representation of L1 near point p and is used to calculate the distance from a point to the curve;

[0094] S35. Denote that d(p, L1) is approximately equal to d(p, L_{KD}).

[0095] Among them, d(p, L1) represents the distance from point p to L1.

[0096] According to another aspect of the present invention, there is provided an evaluation device for the horizontal and vertical errors of a map lane line. When using the evaluation device for the horizontal and vertical errors of the map lane line for evaluation, it includes the above-mentioned evaluation method for the horizontal and vertical errors of the map lane line.

[0097] As Figure 4 shown, the evaluation device for the horizontal and vertical errors of the map lane line includes:

[0098] An information acquisition module 10, which is used to acquire the lane line data of the true value lane line L1 and the lane line L2 to be measured. The lane line data includes geographical locations;

[0099] A data point acquisition module 20, which is used to equally distance collect data points for L2;

[0100] An information processing module 30, which is used to structure the true value lane line L1 into a KD-tree;

[0101] An error evaluation module 40, which is used to calculate the distance from a point on L2 to the polyline of L1 and normalize the distance according to the number of m points to obtain the final result d(L1, L2). The formula is:

[0102]

[0103] Among them, m is the number of samples of L1, p represents a two-dimensional space point expressed by horizontal and vertical coordinates, and d(L1, L2) is the distance between L1 and L2;

[0104] The final result d(L1, L2) is the horizontal and vertical absolute error of L2 compared to L1.

[0105] In the embodiments of the present invention, by structuring the true lane line L1 with a KD-tree, the true lane line points closest to each point on the lane line L2 to be measured are searched, so as to calculate the distance from a point to a broken line more accurately. The method of collecting data points at equal distances ensures uniform distribution of points on L2 and avoids error accumulation caused by uneven distribution of data points. Through normalization processing, the present invention synthesizes the distances from all points in L2 to the points on L1 into a final result d(L1, L2), which reflects the overall horizontal and vertical errors of L2 compared to L1 and enhances the accuracy of the evaluation result.

[0106] According to an embodiment of the present invention, the present invention also provides an electronic device, a readable storage medium, and a computer program product.

[0107] Figure 5 A schematic block diagram of an exemplary electronic device 600 that can be used to implement the embodiments of the present invention is shown. The electronic device 600 is intended to represent various forms of digital computers, such as, for example, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, for example, a personal digital assistant, a cellular phone, a smart phone, a wearable device, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely exemplary and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0108] As Figure 5 shown, the electronic device 600 includes a computing unit 601, which can execute various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 602 or a computer program loaded from a storage unit 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the device 600 can also be stored. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0109] Each component in the electronic device 600 is connected to the I / O interface 605, including: an input unit 606, such as a keyboard, a mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a magnetic disk, an optical disc, etc.; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 609 allows the electronic device 600 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0110] The computing unit 601 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 executes the various methods and processes described above, such as a method for evaluating the horizontal and vertical errors of map lane lines. For example, in some embodiments, a method for evaluating the horizontal and vertical errors of map lane lines can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as the storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 600 via the ROM 602 and / or the communication unit 609. When the computer program is loaded into the RAM 603 and executed by the computing unit 601, one or more steps of the method for evaluating the horizontal and vertical errors of map lane lines described above can be executed. Alternatively, in other embodiments, the computing unit 601 can be configured to execute a method for evaluating the horizontal and vertical errors of map lane lines in any other suitable manner (e.g., by means of firmware).

[0111] Various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip systems (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a dedicated or general-purpose programmable processor, receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0112] The program code for implementing the methods of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to the processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing devices, such that when the program codes are executed by the processor or controller, the functions / operations specified in the flowchart and / or block diagram are implemented. The program codes can be executed entirely on the machine, partially on the machine, as an independent software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0113] In the context of the present invention, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0114] To provide for interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can also be used to provide for interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0115] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0116] A computer system may include a client and a server. The client and the server are generally far away from each other and usually interact via a communication network. The relationship between the client and the server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.

[0117] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in the present invention can be achieved, and no limitations are imposed herein.

[0118] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for evaluating the lateral and longitudinal errors of map lane lines, characterized in that: The method comprises: S10, obtaining lane line data of a true value lane line L1 and a lane line L2 to be measured, wherein the lane line data includes a geographic location; S20, collecting data points at equal distances from L2; S30, structuring the true value lane line L1 KD-tree; S40, calculating the point-to-polyline distance from the midpoint of L2 to L1, and normalizing the distance according to the number of m points to obtain the final result d(L1, L2), the formula is: Where m is the number of L1 samples, p is a two-dimensional space point expressed by horizontal and vertical coordinates, and d(L1,L2) is the distance between L1 and L2; The final result d(L1, L2) is the absolute horizontal and vertical errors of L2 compared to L1.

2. The method for evaluating the lateral and longitudinal errors of map lane lines according to claim 1, characterized in that: The lane line data of the true value lane line L1 and the lane line to be measured L2 are two-dimensional polylines in the form of ordered point series, which are recorded as:

3. The method for evaluating the lateral and longitudinal errors of map lane lines according to claim 2, characterized in that: The method comprises: Define a point p to a line segment S = {p_s, p_e}, the distance is d(p, S), the line segment S consists of a starting point p_s and an end point p_e; When a perpendicular line from point p to line segment S exists and the intersection point is on line segment S, d(p,S) is the length of the perpendicular line from point p to the line segment, which is dp; When a perpendicular line from point p to line segment S exists but the intersection point is not on line segment S, d(p,S) is the minimum value ds of the distances from point p to the starting points ps and pe of line segment S; According to the distance from the point to the line segment, a larger value will be calculated when point p is far away from the line segment, and the larger value is the longitudinal constraint.

4. The method for evaluating the lateral and longitudinal errors of map lane lines according to claim 3, characterized in that: The method comprises: Define the distance d(p,L) from a point p to a curve L. The calculation method is to take the minimum value of the distance from d to the line segment composed of two points in sequence on L1. The formula is: Among them, S i ={p i ,p i+1 }.

5. The method for evaluating the lateral and longitudinal errors of map lane lines according to claim 4, characterized in that: The specific steps of structuring the true value lane line L1 KD-tree are: S31, obtaining the true value lane line L1; S32, construct the points in L1 into a KD-Tree structure using the KD-Tree algorithm S33, obtain a search point p; S33, using the nearest neighbor search algorithm of KD-Tree, search for the 6 nearest points in the true value KD-Tree with p as the center; S34, connecting the six nearest points found in sequence to form a new curve L_{KD}, where L_{KD} is an approximate representation of L1 near point p and is used to calculate the distance from the point to the curve; S35, let d(p,L1) be approximately equal to d(p,L_{KD}).

6. A device for evaluating the lateral and longitudinal errors of map lane lines, characterized in that: The device for evaluating the lateral and longitudinal errors of map lane lines includes the method for evaluating the lateral and longitudinal errors of map lane lines as described in any one of claims 1-5 when performing evaluation.

7. The device for evaluating the lateral and longitudinal errors of map lane lines according to claim 6, characterized in that: The device comprises: An information acquisition module, the information acquisition module is used to obtain lane line data of the true value lane line L1 and the lane line L2 to be measured, the lane line data including the geographical location; A data point acquisition module, the data point acquisition module is used to collect data points at equal distances from L2; An information processing module, the information processing module is used to structure the true value lane line L1 KD-tree; Error evaluation module, which is used to calculate the point-to-polyline distance from the midpoint of L2 to L1, and normalize the distance according to the number of m points to obtain the final result d(L1, L2), the formula is: Where m is the number of L1 samples, p represents a two-dimensional space point expressed by horizontal and vertical coordinates, and d(L1,L2) is the distance between L1 and L2; The final result d(L1, L2) is the absolute horizontal and vertical errors of L2 compared to L1.

8. An electronic device, characterized in that: include: at least one processor, and a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method described in any one of claims 1 to 5.

9. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-5.

Images