Method and device for constructing distance truth value library, electronic device and storage medium

Abstract

The disclosed embodiments relate to a method and apparatus for constructing a distance truth library, a method and apparatus for detecting high-precision maps, an electronic device, and a storage medium, wherein the method for constructing the distance truth library comprises: obtaining map data, the map data being collected along the entire road by a collection device that meets set accuracy conditions; performing semantic segmentation based on the map data to obtain at least two first map elements; obtaining the distance truth value between two first map elements that are within an accuracy benchmark range among the at least two first map elements; and constructing a distance truth library based on the distance truth value between the two first map elements that are within the accuracy benchmark range. The disclosed embodiments can cause the distance truth library to include the distance truth value between any two first map elements that are within the accuracy benchmark range on the entire road section, thereby improving the accuracy of the map accuracy assessment when the map accuracy of the high-precision map is subsequently detected based on the distance truth library.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a method and device for constructing a distance truth library, a method and device for detecting high-precision maps, an electronic device, and a storage medium. Background Art

[0002] Compared to standard maps, HD maps offer higher precision and richer map information, making them widely used in autonomous driving. For autonomous driving, HD maps are generally considered to have an accuracy of 5 cm / 100 m to 20 cm / 100 m; otherwise, they pose significant safety risks.

[0003] In reality, during the HD map production process, various factors, such as vectorization errors, can lead to accuracy issues in the final HD map. Therefore, after HD map production is completed, surveying equipment such as RTK surveyors and total stations are typically used to perform random distance measurements between map features to verify the map's accuracy. However, due to objective factors such as the inability to stop for distance measurement on main highways, sampling can be uneven, resulting in significant deviations between the estimated HD map accuracy and actual accuracy. Summary of the Invention

[0004] In order to solve the above technical problems or at least partially solve the above technical problems, the present disclosure provides a method and device for constructing a distance truth value library, a method and device for detecting high-precision maps, an electronic device, a storage medium and a program product.

[0005] In a first aspect, an embodiment of the present disclosure provides a method for constructing a distance truth library, the method comprising:

[0006] Acquiring map data, wherein the map data is collected along the entire route by a collection device that meets set accuracy conditions;

[0007] Performing semantic segmentation based on the map data to obtain at least two first map elements;

[0008] Obtaining a true distance value between two first map elements among the at least two first map elements that are within the accuracy reference range;

[0009] A distance truth value library is constructed based on the distance truth value between the two first map elements within the accuracy reference range.

[0010] In a second aspect, an embodiment of the present disclosure provides a method for detecting a high-precision map, the method comprising:

[0011] Acquire two second map elements within an accuracy reference range in the high-precision map, wherein the two second map elements constitute a second map element pair;

[0012] Matching the two second map elements with first map elements in a distance truth library to obtain two first map elements corresponding to the two second map elements, wherein the corresponding two first map elements constitute a first map element pair, wherein the distance truth library stores the distance truth values ​​of the two first map elements within an accuracy reference range;

[0013] respectively obtaining the distance mark value of the second map element pair from the high-precision map, and obtaining the distance true value corresponding to the first map element pair from the distance true value library;

[0014] Based on the distance mark value and the distance true value, determine whether the map accuracy of the high-precision map meets the requirements.

[0015] In a third aspect, an embodiment of the present disclosure further provides a device for constructing a distance truth value library, the device comprising:

[0016] A first acquisition module is used to acquire map data, wherein the map data is acquired by an acquisition device that meets a set accuracy condition along the entire road;

[0017] a segmentation module, configured to perform semantic segmentation based on the map data to obtain at least two first map elements;

[0018] A second acquisition module is configured to acquire a true distance value between two first map elements located within the accuracy reference range among the at least two first map elements;

[0019] A construction module is used to construct a distance truth value library based on the distance truth value between the two first map elements within the accuracy reference range.

[0020] In a fourth aspect, an embodiment of the present disclosure further provides a high-precision map detection device, the device comprising:

[0021] A third acquisition module is configured to acquire two second map elements within an accuracy reference range in the high-precision map, wherein the two second map elements constitute a second map element pair;

[0022] a fourth acquisition module, configured to match the two second map elements with first map elements in a distance truth library to obtain two first map elements corresponding to the two second map elements, wherein the two corresponding first map elements constitute a first map element pair, wherein the distance truth library stores the distance truth values ​​of the two first map elements within a reference accuracy range;

[0023] a fifth acquisition module, configured to respectively acquire the distance mark value of the second map element pair from the high-precision map, and acquire the distance true value corresponding to the first map element pair from the distance true value library;

[0024] A determination module is used to determine whether the map accuracy of the high-precision map meets the requirements based on the distance mark value and the distance true value.

[0025] In a fifth aspect, an embodiment of the present disclosure further provides an electronic device, comprising: a memory and a processor; the memory is used to store executable instructions of the processor; the processor is used to read the executable instructions from the memory and execute the executable instructions to implement any of the distance truth library construction methods or high-precision map detection methods provided in the embodiments of the present disclosure.

[0026] In a sixth aspect, an embodiment of the present disclosure further provides a computer-readable storage medium, wherein the storage medium stores a computer program, and the computer program is used to execute any of the distance truth library construction methods or high-precision map detection methods provided in the embodiments of the present disclosure.

[0027] In a seventh aspect, an embodiment of the present disclosure further provides a computer program product, which is used to execute any of the distance truth library construction methods or high-precision map detection methods provided in the embodiments of the present disclosure.

[0028] The technical solution provided by the embodiments of the present disclosure has at least the following advantages compared with the prior art:

[0029] In the embodiment of the present disclosure, map data collected by a collection device that meets the set accuracy conditions for the entire road can be obtained, and semantic segmentation can be performed based on the map data to obtain at least two first map elements; the true distance value between two first map elements located within the accuracy reference range among the at least two first map elements is obtained, and a distance truth value library is constructed based on the true distance value between the two first map elements located within the accuracy reference range, so that the distance truth value library can include the true distance value between any two first map elements located within the accuracy reference range on the entire road section, so that when the map accuracy of the high-precision map is subsequently tested based on the distance truth value library, two second map elements located within the accuracy reference range on the entire road section in the high-precision map can be covered. Compared with the sampling detection method in the prior art, the number of covered second map elements is greater, which is conducive to improving the problem in the prior art that the evaluation result of map accuracy is greatly different from the actual result due to uneven sampling, and improving the accuracy of the evaluation of map accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The above and other features, advantages, and aspects of the various embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that the originals and elements are not necessarily drawn to scale.

[0031] Figure 1 A schematic diagram of a process for constructing a distance truth value library provided in an embodiment of the present disclosure;

[0032] Figure 2 A schematic diagram of the structure of a distance truth value network provided in an embodiment of the present disclosure;

[0033] Figure 3 A schematic diagram of a specific process of S120 provided in an embodiment of the present disclosure;

[0034] Figure 4 A schematic diagram of a specific process of S130 provided in an embodiment of the present disclosure;

[0035] Figure 5 A schematic diagram of a process for constructing another distance truth value library provided in an embodiment of the present disclosure;

[0036] Figure 6 A schematic diagram of a flow chart of a high-precision map detection method provided in an embodiment of the present disclosure;

[0037] Figure 7 A schematic diagram of a specific process of S610 provided in an embodiment of the present disclosure;

[0038] Figure 8 A schematic diagram of a specific process of S620 provided in an embodiment of the present disclosure;

[0039] Figure 9 A flowchart of another high-precision map detection method provided in an embodiment of the present disclosure;

[0040] Figure 10 A schematic diagram of a specific process of S940 provided in an embodiment of the present disclosure;

[0041] Figure 11 A schematic diagram of the structure of a device for constructing a distance truth value library provided in an embodiment of the present disclosure;

[0042] Figure 12 A schematic diagram of the structure of a high-precision map detection device provided in an embodiment of the present disclosure;

[0043] Figure 13 A schematic structural diagram of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0044] The following describes embodiments of the present disclosure in more detail with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of protection of the present disclosure.

[0045] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0046] As used herein, the term "including" and its variations are open-ended, i.e., "including but not limited to." The term "based on" means "based, at least in part, on." The term "one embodiment" means "at least one embodiment," the term "another embodiment" means "at least one additional embodiment," and the term "some embodiments" means "at least some embodiments." Other terms are defined in the following description.

[0047] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0048] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, they should be understood as "one or more".

[0049] Figure 1 This is a flowchart of a method for constructing a distance truth library, provided in an embodiment of the present disclosure. This method can be applied to constructing a distance truth library in the context of detecting high-precision map accuracy. This construction method can be executed by a construction device, which can be implemented using software and / or hardware and integrated into any electronic device with computing capabilities.

[0050] like Figure 1 As shown, the method for constructing a distance truth library provided by the embodiment of the present disclosure may include:

[0051] S110: Acquire map data.

[0052] Among them, the map data is collected along the entire road by collection equipment that meets the set accuracy conditions.

[0053] Specifically, the map data may include image data and / or point cloud data, but is not limited thereto.

[0054] Here, image data refers to data captured by an image acquisition device (such as a camera) installed in a mobile device (such as a vehicle) and recorded in the form of pixels of the road environment. Each pixel contains a grayscale value or red, green, and blue (RGB) color information, but is not limited to this.

[0055] Among them, point cloud data refers to data that is scanned by a scanning device (such as a 3D laser scanner) installed in a mobile device to scan the road environment and record it in the form of 3D points. Each 3D point contains the 3D coordinates of the 3D point and attribute information of the corresponding 3D point, such as red, green, and blue (RGB) color information, or reflection intensity information (Intensity), etc., but is not limited to these.

[0056] Specifically, the acquisition device meets the set accuracy conditions. The set accuracy conditions may include accuracy conditions corresponding to relative accuracy. Relative accuracy is used to characterize the difference between the measured distance between two map elements determined based on map data and the true value of the distance between the two map elements in the real world (i.e., the actual distance). For example, when the relative accuracy is 3 cm / 100 m, it means that when the true value of the distance between the two map elements is 100 meters, the error of the measured distance between the two map elements determined based on the map data is within 3 cm compared with 100 meters.

[0057] The specific content of the accuracy condition corresponding to the relative accuracy can be set by those skilled in the art according to actual conditions and is not limited here. For example, the accuracy condition corresponding to the relative accuracy is that the relative accuracy is less than or equal to 3 cm / 100 m, but is not limited to this.

[0058] Of course, the set accuracy conditions may also include accuracy conditions corresponding to absolute accuracy or other accuracy types known to those skilled in the art, which are not limited here.

[0059] Specifically, those skilled in the art may select an acquisition device according to actual conditions, as long as the selected acquisition device meets the set accuracy conditions.

[0060] For example, if the accuracy condition includes a relative accuracy of 3 cm / 100 m, then a collection device that meets this accuracy condition has the following characteristics: when the true distance between two map features is 100 m, the measured distance between the two map features determined based on the map data is within 3 cm of 100 m. In this case, the collection device can be a professional surveying and mapping vehicle equipped with a surveying-grade single-line lidar and combined inertial navigation, such as a RIEGL collection vehicle. RIEGL collection vehicles typically have a relative accuracy of less than 3 cm / 100 m and an absolute accuracy of less than 50 cm. Other collection devices known to those skilled in the art are not limited here.

[0061] It should be understood that when the acquisition equipment meets the set accuracy conditions, the error between the measured distance between two map elements determined based on the map data and the true value of the distance between the two map elements is extremely small. At this time, the measured distance can be regarded as the true value of the distance between the two map elements.

[0062] Specifically, full road collection means that the collection device collects at least the first map elements in the road environment while moving or while moving with the mobile device, for example, collects all the map elements in the road environment.

[0063] For example, when a RIEGL data collection vehicle is driving on a certain road section, it collects the road environment of the road section in real time to obtain various map elements in the road environment.

[0064] It should be noted that the collection device can collect the entire route in a single trip or in multiple trips, which is not limited here. In addition, the round trip of the collection device (or the accompanying mobile device) along the same track can be regarded as two round trips of full route collection.

[0065] In some embodiments, the distance truth library construction device may construct the distance truth library in real time based on map data collected by a collection device. In this case, S110 may include: receiving map data collected by the collection device in real time. In this way, the distance truth library can be constructed as quickly as possible, so that it can be promptly applied to map accuracy testing.

[0066] In other implementations, the map data collected by the collection device may be stored in a storage device. When the distance truth value library construction device constructs the distance truth value library, S110 may include: reading the map data from a preset storage device. In this way, the time when the distance truth value library construction device constructs the distance truth value library is relatively flexible.

[0067] S120: Perform semantic segmentation based on the map data to obtain at least two first map elements.

[0068] Specifically, semantic segmentation is the process of classifying input into semantically interpretable categories. Semantic segmentation based on map data involves segmenting the map data into multiple independent subsets, each of which is expected to correspond to a physically meaningful map feature. Map features are elements contained in the map, such as, but not limited to, lane markings, traffic signs, billboards, roadside trees, streetlights, and houses.

[0069] In some embodiments, the collection device collects all map elements in the road environment to obtain map data, and the first map element is a part of all map elements.

[0070] Optionally, the first map element includes at least one of a lane line, a pole, a traffic sign, a guardrail, a curb, and a gantry.

[0071] Specifically, the lane line is a lane marking used to guide the direction, which can be a solid line or a dashed line, but is not limited thereto.

[0072] Specifically, the rod-shaped object may include a lamp pole for setting a street lamp, a flag pole for hanging a flag, a heat rod, etc., but is not limited thereto.

[0073] Specifically, traffic signs use graphic symbols and text to convey specific information. They are used to manage traffic, indicate driving directions, and ensure smooth roads and safe driving. They are mainly applicable to highways, urban roads, and all dedicated highways. They have the nature of laws and must be obeyed by vehicles and pedestrians.

[0074] At this time, see Figure 3 , S120 may include:

[0075] S121a, performing semantic segmentation based on the map data to obtain at least two map elements;

[0076] S122a, selecting at least two first map elements from the at least two map elements;

[0077] Alternatively, S120 may include:

[0078] S121b, filtering out map data corresponding to the first map element from the map data;

[0079] S122b: Perform semantic segmentation based on the filtered map data to obtain at least two first map elements.

[0080] Specifically, a semantic segmentation network model may be used to perform semantic segmentation on map data, but is not limited thereto.

[0081] Among them, the semantic segmentation network model can include networks based on fully convolutional networks (FCNs), SegNet, E-Net, Link-Net, Mask R-CNN, Cylinder3D, or other networks that can realize the semantic segmentation function of map data, without limitation.

[0082] In another example, the acquisition device only acquires the first map element in the road environment to obtain map data. In this case, the at least two map elements obtained by semantic segmentation based on the map data are the first map elements.

[0083] S130: Obtain a true distance value between two first map elements that are within an accuracy reference range among at least two first map elements.

[0084] Specifically, the specific range of the accuracy reference range can be set by those skilled in the art according to actual conditions and is not limited here.

[0085] Optionally, the accuracy reference range is 80 meters to 120 meters.

[0086] It is understandable that the applicant considers that when evaluating the relative accuracy of high-precision maps used in autonomous driving, an accuracy reference range that is too short or too long relative to the perception range will lead to inaccurate relative accuracy evaluation. The perception range of autonomous driving is generally 100 meters. Therefore, the accuracy reference range can be set to around 100 meters, for example, 80 meters to 120 meters in the embodiment of this disclosure. This can bring the accuracy reference range closest to the perception range of autonomous driving, facilitating subsequent accurate evaluation of the relative accuracy of high-precision maps.

[0087] In some embodiments, see Figure 4 , S130 may include:

[0088] S131a, obtaining the true distance value between each two first map elements, and obtaining at least one true distance value;

[0089] S132a. Filter out a distance true value within a reference accuracy range from the at least one distance true value.

[0090] Specifically, the true value of the distance between the two first map elements may be determined based on the map data corresponding to the two first map elements.

[0091] For example, the distance between the reference points on two first map elements is used as the true value of the distance between the two first map elements. Those skilled in the art can select the reference points on each first map element according to actual conditions and are not limited here.

[0092] For another example, among the two first map elements, one first map element includes N1 three-dimensional points (pixel points), and the other first map element includes N2 three-dimensional points (pixel points). For each three-dimensional point (pixel point) in the N1 three-dimensional points (pixel points), the true distance value between it and each three-dimensional point (pixel point) in the N2 three-dimensional points (pixel points) is determined to obtain N1*N2 distance true values. The maximum value, minimum value, or average value of the N1*N2 distance true values ​​is used as the true distance value between the two first map elements, but is not limited to this.

[0093] In some other embodiments, S130 may include:

[0094] S131b, obtaining a rough distance between every two first map elements to obtain at least one rough distance value;

[0095] S132b, obtaining a true distance value between two first map elements whose roughly measured distance values ​​are within a rough measurement accuracy reference range, to obtain at least one true distance value;

[0096] S133b. Filter out the distance true value that is within the accuracy reference range from the at least one distance true value.

[0097] Specifically, the distance mark value of the two first map elements in a common navigation map, that is, a map with relatively low accuracy (for example, 30 cm / 100 m) can be used as the rough distance, but the present invention is not limited thereto.

[0098] It can be understood that the maximum value of the coarse measurement accuracy reference range can be greater than the maximum value of the accuracy reference range, and the minimum value of the coarse measurement accuracy reference range can be greater than the minimum value of the accuracy reference range. For example, when the accuracy reference range is 80 meters to 120 meters, the coarse measurement accuracy reference range is 75 meters to 125 meters, so as to avoid missing the first map element whose true distance value is within the accuracy reference range in S132b.

[0099] It should be noted that, when the collection device performs multiple full-road collections, S120 - S130 may be executed separately for the map data obtained from each full-road collection.

[0100] S140: Construct a distance truth value library based on the distance truth values ​​between two first map elements within the accuracy reference range.

[0101] Specifically, each true distance value within the accuracy reference range is located in the true distance value library.

[0102] In the embodiment of the present disclosure, there is no limit on the number of times the collection device collects data on the entire road section, or the number of repeated collection times. If multiple collection trips are required for the entire road section, a distance truth value sub-library can be constructed based on the map data obtained from each collection trip, and each distance truth value sub-library constitutes a distance truth value library.

[0103] In one example, a collection device performs a single pass to collect data for the entire road section. Based on the map data from this single pass, a distance truth sub-library, known as the distance truth library, can be constructed. In this case, two first map features within the accuracy benchmark have one true distance value in the distance truth library. For example, first map features A and B within the accuracy benchmark have one true distance value in the distance truth library.

[0104] In another example, if the collection device performs M collection trips (M is a positive integer greater than 1) on the entire road section, a distance truth value sub-library can be constructed based on the map data collected during each collection trip, resulting in M ​​distance truth value sub-libraries. These M distance truth value sub-libraries constitute the distance truth value library. In this case, two first map elements within the accuracy benchmark have M distance truth values ​​in the distance truth value library. For example, first map elements A and B within the accuracy benchmark have M distance truth values ​​in the distance truth value library. It is understandable that since the error between the measured distance and the true value between two first map elements collected during the same collection trip is solely due to the relative accuracy of the collection device, the error between the measured distance and the true value between one first map element collected during one collection trip and another first map element collected during another collection trip is also affected by the absolute accuracy of the collection device. Therefore, the distance truth value is established between two first map elements collected during a single collection trip. That is, the distance truth value can be calculated between two first map elements collected during the same collection trip. Two first map elements collected during different collection trips cannot form a first map element pair for calculating the distance truth value and cannot be used to calculate the distance truth value.

[0105] Specifically, each distance truth value sub-library may include at least one distance truth value network, each distance truth value network includes at least two first map elements, and the length of the connection line between the two first map elements is used to represent the distance truth value between the two first map elements. Figure 2 This is a structural diagram of a distance truth network provided by an embodiment of the present disclosure. Figure 2 The distance truth network includes five first map elements 210, and every two first map elements 210 are connected by a line representing the distance truth value between them.

[0106] Specifically, when storing each first map element in the distance truth library, the corresponding map data and / or reference location information may be stored, but the present invention is not limited thereto. The reference location information will be explained in detail later and will not be elaborated here.

[0107] Specifically, each first map element may also correspond to a unique identity document (ID), and each distance truth value also corresponds to a unique ID. The association between the ID of the first map element and the ID of the distance truth value associated with the first map element is also stored in the distance truth value library.

[0108] It can be understood that since the collection equipment can collect map data for the entire road, the first map elements obtained based on semantic segmentation of the map data cover every first map element on the entire road section, so that the distance truth library covers the distance truth value between every two first map elements within the accuracy benchmark range on the entire road section, which is conducive to covering all second map elements in the high-precision map when performing map accuracy detection on the high-precision map, rather than one that can only cover part of the second map elements as in the existing technology, thereby improving the accuracy of the map accuracy assessment.

[0109] It can also be understood that when the collection equipment uses a device with collection and mobile functions such as a RIEGL collection vehicle, compared to the existing technology of obtaining the true value of the distance between two first map elements based on surveying and mapping equipment such as RTK measuring instruments and total stations, it can not only obtain the true value of the distance between every two first map elements within the accuracy benchmark range on the entire road section, but also help save time and labor costs.

[0110] In the embodiment of the present disclosure, map data collected by a collection device that meets the set accuracy conditions for the entire road can be obtained, and semantic segmentation can be performed based on the map data to obtain at least two first map elements; the true distance value between two first map elements located within the accuracy reference range among the at least two first map elements is obtained, and a distance truth value library is constructed based on the true distance value between the two first map elements located within the accuracy reference range, so that the distance truth value library can include the true distance value between any two first map elements located within the accuracy reference range on the entire road section, so that when the map accuracy of the high-precision map is subsequently tested based on the distance truth value library, two second map elements located within the accuracy reference range on the entire road section in the high-precision map can be covered. Compared with the sampling detection method in the prior art, the number of covered second map elements is greater, which is conducive to improving the problem in the prior art that the evaluation result of map accuracy is greatly different from the actual result due to uneven sampling, and improving the accuracy of the evaluation of map accuracy.

[0111] In some other embodiments of the present disclosure, the map data includes image data and / or point cloud data, and performing semantic segmentation based on the map data includes: performing image segmentation based on the image data; and / or performing point cloud segmentation based on the point cloud data.

[0112] Specifically, when map data includes image data, semantic segmentation can be performed on the image data to segment pixels based on semantics, achieving pixel-level classification. For example, if the image data includes a lane line and a traffic sign, semantic segmentation can be performed on the image data to group the pixels corresponding to the lane line into one pixel subset and the traffic sign into another pixel subset.

[0113] In this case, the semantic segmentation network may include an image semantic segmentation network. Image segmentation can be achieved by inputting image data into the image semantic segmentation network. The image semantic segmentation network may be any network capable of performing semantic segmentation on image data, and is not limited here.

[0114] Specifically, when map data includes point cloud data, semantic segmentation can be performed on the point cloud data to semantically segment the 3D points, thereby forming multiple independent 3D point sets. For example, if the point cloud data includes a lane line and a guardrail, semantic segmentation can be performed on the point cloud data to separate the 3D points corresponding to the lane line into one 3D point subset and the guardrail into another 3D point subset.

[0115] In this case, the semantic segmentation network may include a point cloud semantic segmentation network. Point cloud data is input into the point cloud semantic segmentation network to achieve point cloud segmentation. The point cloud semantic segmentation network can be any network that can achieve semantic segmentation of point cloud data, and is not limited here.

[0116] It should be noted that when both image segmentation and point cloud segmentation are performed, the map elements obtained by image segmentation and the map elements obtained by point cloud segmentation can be taken as a union or intersection, which is not limited here.

[0117] It can be understood that by performing image segmentation on image data and / or point cloud segmentation on point cloud data, semantic segmentation of map data is achieved, which makes the segmentation method simple and easy to operate, and is conducive to improving the efficiency of constructing the distance truth library.

[0118] Figure 5 A flowchart of another method for constructing a distance truth library provided in an embodiment of the present disclosure is further optimized and expanded based on the above technical solution and can be combined with the above various optional implementation methods.

[0119] like Figure 5 As shown, the method for constructing a distance truth library provided by the embodiment of the present disclosure may include:

[0120] S510: Acquire map data.

[0121] Among them, the map data is collected along the entire road by collection equipment that meets the set accuracy conditions.

[0122] Specifically, S510 is similar to S110 and will not be described in detail here.

[0123] S520: Perform semantic segmentation based on the map data to obtain at least two first map elements.

[0124] Specifically, S520 is similar to S120 and will not be described in detail here.

[0125] S530: Obtain reference position information of at least two first map elements.

[0126] Specifically, the reference location information refers to information corresponding to the reference location on the first map element. The reference location may include a reference point, i.e., a point (3D point or pixel) in the first map element; a reference line segment, i.e., a line segment containing multiple discrete points in the first map element, or a line segment consisting of multiple continuous points, but is not limited thereto. Furthermore, the reference location information includes at least the 3D coordinates of the reference location, and the reference location information may be determined based on the map data of the first map element.

[0127] For example, when the first map element is a traffic sign, the reference location information may include the three-dimensional coordinates of the center point of the traffic sign and the orientation information (ie, vector) of the surface of the traffic sign marked with traffic information, but is not limited thereto.

[0128] For another example, when the first map element is a rod-shaped object, the reference position information may include the three-dimensional coordinates of each point on the central axis of the rod-shaped object, but is not limited thereto.

[0129] For another example, when the first map element is a lane line, the reference position information may include the three-dimensional coordinates of the beginning, center, and end of the lane line, but is not limited thereto.

[0130] It should be noted that, for other types of first map elements, the selection of reference positions can be set by those skilled in the art according to actual conditions and is not limited here.

[0131] S540: Determine a true distance between any two map elements of the at least two first map elements based on the reference position information.

[0132] Specifically, when the reference position information of two first map elements is information corresponding to line segments, the distance between the lines can be used as the true value of the distance between the two first map elements.

[0133] For example, when the two first map elements are both rod-shaped objects, the distance between the central axes of the two rod-shaped objects can be used as the true value of the distance between the two rod-shaped objects. The distance between the two central axes can be the distance from a point at 1 / 2 the height of one central axis to a point at 1 / 2 the height of the other central axis, or it can be the distance between any two points of the same height on the two central axes, but is not limited to this.

[0134] Specifically, when the reference position information of two first map elements is information corresponding to a point and a line segment, respectively, the distance between the point and the line segment may be used as the true value of the distance between the two first map elements.

[0135] For example, when the two first map elements are a pole and a traffic sign, the distance between the center point of the traffic sign and the central axis of the pole can be used as the true value of the distance between the traffic sign and the pole.

[0136] For another example, when the two first map elements are a pole and a lane line, the distance between the center point, the beginning, or the end of the lane line and the central axis of the pole can be used as the true distance between the traffic sign and the pole.

[0137] Specifically, when the reference position information of two first map elements is information corresponding to a point and a point, respectively, the distance between the points can be used as the true distance value between the two first map elements.

[0138] For example, when the two first map elements are two traffic signs, the distance between the center points of the two traffic signs may be used as the true value of the distance between the two traffic signs.

[0139] For another example, when the two first map elements are a traffic sign and a lane line, the distance between the center point of the traffic sign and the center point, the beginning, or the end of the lane line can be used as the true value of the distance between the traffic sign and the lane line.

[0140] For another example, when the two first map elements are two lane lines, the distance between the center points of the two lane lines, the distance between the beginnings of the two lane lines, or the distance between the ends of the two lane lines can be used as the true value of the distance between the two lane lines.

[0141] Specifically, for every two first map elements, the true value of the distance between them is calculated to obtain the true value of the distance corresponding to every two first map elements.

[0142] S550: Obtain the true distance value between two first map elements within the accuracy reference range based on the true distance value between any two first map elements.

[0143] Specifically, from the true distance values ​​obtained in S540, the true distance values ​​within the accuracy reference range are screened out. For example, when the accuracy reference range is 80 meters to 120 meters, the true distance values ​​within the range of 80 meters to 120 meters are screened out.

[0144] S560: Construct a distance truth value library based on the distance truth values ​​between the two first map elements within the accuracy reference range.

[0145] Specifically, S560 is similar to S140 and will not be described in detail here.

[0146] In the disclosed embodiment, by determining the true distance value between any two map elements among at least two first map elements based on reference position information, the calculation method of the true distance value between the two first map elements is simplified and convenient, which is conducive to quickly determining the true distance value between any two first map elements, thereby improving the construction speed of the truth value library.

[0147] Figure 6 This is a flowchart of a method for detecting high-precision maps, provided in an embodiment of the present disclosure. This method can be applied to detecting the accuracy of high-precision maps. This method can be performed by a high-precision map detection device, which can be implemented using software and / or hardware and integrated into any electronic device with computing capabilities.

[0148] like Figure 6 As shown, the detection method of the high-precision map provided by the embodiment of the present disclosure may include:

[0149] S610: Acquire two second map elements within the accuracy reference range in the high-precision map, where the two second map elements constitute a second map element pair.

[0150] Specifically, the high-precision map may be a high-precision map of at least part of the entire road section corresponding to the distance truth library.

[0151] Specifically, for the description of the accuracy benchmark range, please refer to the previous text and will not be repeated here.

[0152] Specifically, the second map element may be any map element in the high-precision map. Optionally, the second map element includes at least one of a lane line, a pole, a traffic sign, a guardrail, a curb, and a gantry.

[0153] In some embodiments, see Figure 7 , S610 may include:

[0154] S611a, obtaining the distance mark value between two second map elements to obtain at least one distance mark value;

[0155] S612a: Filter out distance mark values ​​within a reference accuracy range from the at least one distance mark value, and determine two second map elements corresponding to each distance mark value within the reference accuracy range as a second map element pair.

[0156] In some other embodiments, S610 may include:

[0157] S611b, obtaining a rough distance between every two second map elements to obtain at least one rough distance value;

[0158] S612b, obtaining a distance mark value between two first map elements whose coarse-measured distance values ​​are within a coarse-measured accuracy reference range, to obtain at least one distance mark value;

[0159] S613b: Filter out distance mark values ​​within the accuracy reference range from the at least one distance mark value, and determine two second map elements corresponding to each distance mark value within the accuracy reference range as a second map element pair.

[0160] Specifically, for the description of the coarse distance value, please refer to the previous text and will not be repeated here.

[0161] Specifically, each second map element pair in the high-precision map can be obtained through S610.

[0162] S620: Match the two second map elements with the first map elements in the distance truth library to obtain two first map elements corresponding to the two second map elements, and the corresponding two first map elements constitute a first map element pair.

[0163] The distance truth value library stores the distance truth values ​​of the two first map elements within the accuracy reference range. The distance truth value library can be constructed using the distance truth value library construction method described in any embodiment of the present disclosure.

[0164] Specifically, the two first map elements (i.e., first map element pair) corresponding to the two second map elements (i.e., second map element pair) are the same pair of map elements in the road environment. For example, a lane line and a pole in the road environment constitute a second map element pair in the HD map and also constitute a first map element pair in the distance truth database. In this case, the first map element pair corresponds to the second map element pair.

[0165] In some embodiments, see Figure 8 , S620 may include:

[0166] S621a, matching one of the second map elements in the second map element pair with the first map element in the distance truth database to obtain a first map element corresponding to the second map element;

[0167] S622a: Match the other second map element in the second map element pair with the first map element in the distance truth library to obtain a first map element corresponding to the second map element.

[0168] Specifically, when the second map element is matched with the first map element, the matching can be based on the map data and / or reference location information of the second map, and the map data and / or reference location information of the first map element in the distance truth library, which is not limited here.

[0169] The map data of the second map may include image data and / or point cloud data, and the reference position information of the second map data is similar to the interpretation of the reference position information of the first map element, which will not be repeated here.

[0170] In some other embodiments, S620 may include:

[0171] S621b, matching one of the second map elements in the second map element pair with the first map element in the distance truth database to obtain a first map element corresponding to the second map element;

[0172] S622b, searching the distance truth value library for a first map element whose distance truth value to the first map element matched in S621b is within the accuracy reference range, to obtain at least one first map element;

[0173] S623b: Match the other second map element in the second map element pair with the at least one first map element found in S622b to obtain a first map element corresponding to the second map element.

[0174] Specifically, the first map element pair corresponding to each second map element pair can be obtained through S620. Moreover, it can be understood that when the distance truth value library includes at least two distance truth value sub-libraries, each second map element pair may correspond to at least two first map element pairs.

[0175] S630: Obtain the distance mark value of the second map element pair from the high-precision map, and obtain the distance true value corresponding to the first map element pair from the distance true value library.

[0176] Specifically, the high-precision map can be read to obtain the distance mark value of the second map element pair in the second high-precision map, but is not limited to this.

[0177] Specifically, the true distance value corresponding to the first map element pair in the distance truth value library can be read. For example, based on the association between the ID of the first map element and the ID of the true distance value, the ID of the true distance value corresponding to the first map element pair can be determined, and the true distance value can be read based on the ID of the true distance value, but the present invention is not limited to this.

[0178] Specifically, S430 can be used to obtain the distance tag value corresponding to each second map element pair and the distance true value corresponding to each first map element pair. Furthermore, it is understood that when a second map element pair corresponds to at least two first map element pairs, the distance tag value corresponding to the second map element pair corresponds to at least two distance tag values.

[0179] S640: Determine whether the map accuracy of the high-precision map meets the requirements based on the distance mark value and the distance true value.

[0180] In some embodiments, the distance mark value corresponds to a true distance value. In this case, S640 may include:

[0181] S641a. Based on each distance mark value and the true distance value corresponding to each distance mark value, determine whether the map accuracy of the high-precision map meets the requirements.

[0182] Specifically, the difference between each distance mark value and its corresponding distance true value is calculated to obtain at least one difference value. When the maximum value, minimum value, or average value of the at least one difference value is less than the preset difference threshold, it can be determined that the map accuracy of the high-precision map meets the requirements; otherwise, it can be determined that the map accuracy of the high-precision map does not meet the requirements.

[0183] In some other embodiments, the distance mark value corresponds to at least two true distance values. In this case, S640 may include:

[0184] S641b, for each distance mark value, determining its corresponding target distance mark value based on its corresponding distance mark value;

[0185] S642b. Based on each distance mark value and the target distance true value corresponding to each distance mark value, determine whether the map accuracy of the high-precision map meets the requirements.

[0186] In one example, S641b may include, for each distance mark value, selecting a maximum value, a minimum value, or any other distance mark value from its corresponding distance mark value as the target distance true value.

[0187] Specifically, S642b is similar to S641a and will not be repeated here.

[0188] In the disclosed embodiment, two second map elements within the accuracy reference range can be obtained from the high-precision map. The two second map elements constitute a second map element pair, and the two second map elements are matched with the first map elements in the distance truth library to obtain two first map elements corresponding to the two second map elements (i.e., a first map element pair). The distance mark values ​​of the second map element pair are obtained from the high-precision map, and the distance truth values ​​corresponding to the first map element pair are obtained from the distance truth library. Based on the distance mark values ​​and the distance truth values, it is determined whether the map accuracy of the high-precision map meets the requirements, so that the map accuracy of the high-precision map can be detected for the entire road section. Compared with the sampling map accuracy detection of the high-precision map in the prior art, the coverage is wide, and the problem of inaccurate accuracy detection caused by uneven sampling can be improved. The accuracy of the map accuracy assessment is improved so that the accuracy problems existing in the high-precision map can be discovered and corrected in a timely manner. In addition, automated full-road section detection can reduce the investment of manpower and material resources, which is conducive to reducing detection costs.

[0189] Figure 9 This is a flow chart of another high-precision map detection method provided in the embodiment of the present disclosure, which is further optimized and expanded based on the above technical solution and can be combined with the above optional implementation methods. Figure 9 As shown, the detection method of the high-precision map provided by the embodiment of the present disclosure may include:

[0190] S910: Acquire two second map elements within the accuracy reference range in the high-precision map, where the two second map elements constitute a second map element pair.

[0191] Specifically, S910 is similar to S610 and will not be described in detail here.

[0192] S920: Match the two second map elements with the first map elements in the distance truth library to obtain two first map elements corresponding to the two second map elements, and the corresponding two first map elements constitute a first map element pair.

[0193] The distance truth value library stores the distance truth values ​​of the two first map elements within the accuracy reference range, and the distance truth value library is constructed by the distance truth value library construction method described in any embodiment of the present disclosure.

[0194] Specifically, S920 is similar to S620 and will not be described in detail here.

[0195] S930. Obtain the distance mark value of the second map element pair from the high-precision map.

[0196] S940: Obtain at least two true distance values ​​corresponding to the first map element pair from a true distance value library.

[0197] In some embodiments, see Figure 10 , S940 may include:

[0198] S941a. Match one of the second map elements in the second map element pair with a first map element in the distance truth database to obtain at least two first map elements corresponding to the second map element.

[0199] S942a: Match the other second map element in the second map element pair with the first map element in the distance truth library to obtain at least two first map elements corresponding to the second map element.

[0200] At least two first map element pairs are matched through S941a and S942a, and the two first map elements in the first map element pairs are located in the same distance truth value sub-library.

[0201] In some other embodiments, S940 may include:

[0202] S941b. Match one of the second map elements in the second map element pair with a first map element in the distance truth database to obtain at least two first map elements corresponding to the second map element;

[0203] S942b. For each of the at least two first map elements obtained in S941b, search, from its corresponding distance truth value sub-library, for a first map element whose distance truth value is within the accuracy reference range, to obtain at least one first map element corresponding to each first map element.

[0204] S943b: Match the other second map element in the second map element pair with at least one first map element corresponding to each first map element found in S942b to obtain at least two first map elements corresponding to the second map element.

[0205] S950: Obtain an average of the true distance values ​​based on at least two true distance values.

[0206] Specifically, for each distance mark value, the average value of at least two distance true values ​​corresponding to the distance mark value is calculated, and the average value is the target distance true value.

[0207] S960: Determine whether the accuracy of the high-precision map meets the accuracy requirements based on the distance mark value and the average value of the distance true value.

[0208] Specifically, S960 is similar to S640 and will not be described in detail here.

[0209] In the disclosed embodiment, by setting a method for determining whether the accuracy of the high-precision map meets the accuracy requirements based on the distance mark value and the average distance true value, the average distance true value can be made closer to the true value of the first map element pair in the real world, thereby further improving the accuracy of the map accuracy assessment.

[0210] Figure 11 This is a schematic diagram of the structure of a distance truth library construction device provided by an embodiment of the present disclosure. The device can be implemented using software and / or hardware and can be integrated on any electronic device with computing capabilities.

[0211] like Figure 11 As shown, the apparatus 1100 for constructing a distance truth library provided by the embodiment of the present disclosure may include:

[0212] The first acquisition module 1110 is used to acquire map data, where the map data is acquired by an acquisition device that meets a set accuracy condition along the entire route;

[0213] a segmentation module 1120, configured to perform semantic segmentation based on the map data to obtain at least two first map elements;

[0214] The second acquisition module 1130 is configured to acquire a true distance value between two first map elements that are within the accuracy reference range among the at least two first map elements;

[0215] The construction module 1140 is configured to construct a distance truth value library based on the distance truth value between the two first map elements within the accuracy reference range.

[0216] In another embodiment of the present disclosure, the map data includes image data and / or point cloud data, and the segmentation module 1120 may include:

[0217] A first segmentation submodule is configured to perform image segmentation based on image data; and / or,

[0218] The second segmentation submodule is used to perform point cloud segmentation based on point cloud data.

[0219] In another embodiment of the present disclosure, the second acquisition module 1130 may include:

[0220] A first acquisition submodule, configured to acquire reference position information of at least two first map elements;

[0221] A first determining submodule, configured to determine a true distance between any two map elements of the at least two first map elements based on the reference position information;

[0222] The second acquisition submodule is configured to acquire the true distance value between two first map elements within the accuracy reference range based on the true distance value between any two first map elements.

[0223] In yet another embodiment of the present disclosure, the accuracy reference range is 80 meters to 120 meters.

[0224] In yet another embodiment of the present disclosure, the first map element includes at least one of a lane line, a pole, a traffic sign, a guardrail, a curb, and a gantry.

[0225] The apparatus for constructing a distance truth value library provided in the embodiments of the present disclosure can execute any distance truth value library construction method provided in the embodiments of the present disclosure, and has the corresponding functional modules and beneficial effects of the execution method. For any content not fully described in the embodiments of the apparatus of the present disclosure, reference can be made to the description of any method embodiment of the present disclosure.

[0226] Figure 12 This is a schematic structural diagram of a high-precision map detection device provided in an embodiment of the present disclosure. The device can be implemented using software and / or hardware and can be integrated on any electronic device with computing capabilities.

[0227] like Figure 12 As shown, the high-precision map detection device 1200 provided in the embodiment of the present disclosure may include:

[0228] The third acquisition module 1210 is configured to acquire two second map elements within the accuracy reference range in the high-precision map, where the two second map elements constitute a second map element pair.

[0229] A fourth acquisition module 1220 is configured to match the two second map elements with the first map elements in the distance truth library to obtain two first map elements corresponding to the two second map elements, wherein the two corresponding first map elements constitute a first map element pair, and the distance truth library stores the distance truth values ​​of the two first map elements within the accuracy reference range.

[0230] A fifth acquisition module 1230 is configured to acquire the distance mark value of the second map element pair from the high-precision map, and acquire the distance true value corresponding to the first map element pair from the distance true value library;

[0231] The determination module 1240 is used to determine whether the map accuracy of the high-precision map meets the requirements based on the distance mark value and the distance true value.

[0232] In another embodiment of the present disclosure, the fourth acquisition module 1220 includes:

[0233] A third acquisition submodule is configured to acquire at least two true distance values ​​corresponding to the first map element pair from a true distance value library;

[0234] The determination module 1240 includes: a fourth acquisition submodule, configured to acquire an average of the true distance values ​​based on at least two true distance values;

[0235] The second determination submodule is used to determine whether the accuracy of the high-precision map meets the accuracy requirements based on the distance mark value and the average value of the distance true value.

[0236] The high-precision map detection device provided in the embodiments of the present disclosure can execute any high-precision map detection method provided in the embodiments of the present disclosure, and has the corresponding functional modules and beneficial effects of the execution method. For any content not fully described in the embodiments of the device of the present disclosure, reference can be made to the description of any method embodiment of the present disclosure.

[0237] Figure 13 A structural schematic diagram of an electronic device provided in an embodiment of the present disclosure is used to exemplify an electronic device that implements a method for constructing an arbitrary distance truth value library or a method for detecting a high-precision map in an embodiment of the present disclosure, and should not be understood as a specific limitation on the embodiment of the present disclosure.

[0238] like Figure 13 As shown, the electronic device 1300 may include a processor (e.g., a central processing unit, a graphics processing unit, etc.) 1301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1302 or a program loaded from a storage device 1308 into a random access memory (RAM) 1303. Various programs and data required for the operation of the electronic device 1300 are also stored in the RAM 1303. The processor 1301, the ROM 1302, and the RAM 1303 are connected to each other via a bus 1304. An input / output (I / O) interface 1305 is also connected to the bus 1304.

[0239] Typically, the following devices may be connected to the I / O interface 1305: an input device 1306 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 1307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 1308 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1309. The communication device 1309 may allow the electronic device 1300 to communicate with other devices wirelessly or by wire to exchange data. Although the electronic device 1300 is shown as having various devices, it should be understood that it is not required to implement or have all of the devices shown. More or fewer devices may alternatively be implemented or have.

[0240] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 1309, or installed from the storage device 1308, or installed from the ROM 1302. When the computer program is executed by the processor 1301, the functions defined in the method for constructing an arbitrary distance truth library or the method for detecting a high-precision map provided in the embodiment of the present disclosure can be executed.

[0241] It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. Computer-readable storage media may be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or components, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, 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 disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or component. In the present disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof.

[0242] In some embodiments, the client and server can communicate using any currently known or future developed network protocol, such as HTTP (HyperText Transfer Protocol), and can be interconnected with 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"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0243] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0244] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: obtains map data, which is collected by a collection device that meets the set accuracy conditions; performs semantic segmentation based on the map data to obtain at least two first map elements; obtains the true distance value between two first map elements within the accuracy benchmark range among the at least two first map elements; and constructs a distance truth value library based on the true distance value between the two first map elements within the accuracy benchmark range. Or,

[0245] Acquire two second map elements within the accuracy reference range in the high-precision map, the two second map elements constituting a second map element pair; match the two second map elements with the first map elements in the distance truth value library to obtain two first map elements corresponding to the two second map elements, and the corresponding two first map elements constituting a first map element pair, the distance truth value library storing the distance truth values ​​of the two first map elements within the accuracy reference range; respectively acquire the distance mark values ​​of the second map element pair from the high-precision map, and acquire the distance truth values ​​corresponding to the first map element pair from the distance truth value library; determine whether the map accuracy of the high-precision map meets the requirements based on the distance mark values ​​and the distance truth values.

[0246] In embodiments of the present disclosure, computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the computer, partially on the computer, as a separate software package, partially on the computer and partially on a remote computer, or entirely on the remote computer or server. In cases involving a remote computer, the remote computer may be connected to the computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0247] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0248] The units involved in the embodiments described in this disclosure may be implemented in software or hardware, wherein the name of a unit does not necessarily limit the unit itself.

[0249] The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, and without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0250] In the context of the present disclosure, a computer-readable medium can be a tangible medium that can contain or store a program for use by an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a computer-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, 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 disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0251] The above description is merely a preferred embodiment of the present disclosure and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but also includes other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the above-mentioned disclosed concepts. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in this disclosure.

[0252] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details have been included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0253] Although the subject matter has been described in language specific to structural features and / or methodological logical acts, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are merely example forms of implementing the claims.

Claims

1. A method for constructing a distance truth library, comprising: Acquiring map data, wherein the map data is collected along the entire route by a collection device that meets set accuracy conditions; Performing semantic segmentation based on the map data to obtain at least two first map elements; Obtaining a true distance value between two first map elements among the at least two first map elements that are within the accuracy reference range; A distance truth value library is constructed based on the true distance value between the two first map elements within the accuracy reference range, and the distance truth value library is applied to map accuracy detection.

2. The method according to claim 1, wherein The map data includes image data and / or point cloud data, and performing semantic segmentation based on the map data includes: Perform image segmentation based on the image data; and / or, Point cloud segmentation is performed based on the point cloud data.

3. The method according to claim 1, wherein The obtaining of the true distance value between two first map elements that are within the accuracy reference range among the at least two first map elements includes: Acquiring reference position information of the at least two first map elements; Determining a true distance between any two map elements of the at least two first map elements based on the reference position information; The true value of the distance between the two first map elements that are within the accuracy reference range is obtained based on the true value of the distance between the arbitrary two first map elements.

4. The method according to claim 3, wherein: The accuracy reference range is 80 meters to 120 meters.

5. The method according to claim 3, wherein: The first map element includes at least one of a lane line, a pole, a traffic sign, a guardrail, a curb, and a gantry.

6. A method for detecting the accuracy of a high-precision map, comprising: Acquire two second map elements within an accuracy reference range in the high-precision map, wherein the two second map elements constitute a second map element pair; matching the two second map elements with first map elements in a distance truth library to obtain two first map elements corresponding to the two second map elements, wherein the corresponding two first map elements constitute a first map element pair, wherein the distance truth library is constructed using the method according to any one of claims 1 to 5, and wherein the distance truth library stores the distance truth values ​​of the two first map elements within an accuracy reference range; respectively obtaining the distance mark value of the second map element pair from the high-precision map, and obtaining the distance true value corresponding to the first map element pair from the distance true value library; Based on the distance mark value and the distance true value, determine whether the map accuracy of the high-precision map meets the requirements.

7. The method according to claim 6, wherein: The obtaining the true distance value of the first map element pair from the true distance value library includes: Obtaining at least two true distance values ​​corresponding to the first map element pair from the true distance value library; The determining whether the accuracy of the high-precision map meets the accuracy requirement based on the distance mark value and the distance true value includes: Obtaining a distance truth value average based on the at least two distance truth values; Determine whether the accuracy of the high-precision map meets the accuracy requirements based on the distance mark value and the average value of the distance true value.

8. A device for constructing a distance truth value library, comprising: A first acquisition module is used to acquire map data, wherein the map data is acquired by an acquisition device that meets a set accuracy condition along the entire road; a segmentation module, configured to perform semantic segmentation based on the map data to obtain at least two first map elements; A second acquisition module is configured to acquire a true distance value between two first map elements located within the accuracy reference range among the at least two first map elements; A construction module is used to construct a distance truth value library based on the distance truth value between the two first map elements within the accuracy reference range, and the distance truth value library is used in map accuracy detection.

9. A high-precision map accuracy detection device, comprising: A third acquisition module is configured to acquire two second map elements within an accuracy reference range in the high-precision map, wherein the two second map elements constitute a second map element pair; a fourth acquisition module, configured to match the two second map elements with first map elements in a distance truth value library to obtain two first map elements corresponding to the two second map elements, wherein the corresponding two first map elements constitute a first map element pair, wherein the distance truth value library is constructed using the method according to any one of claims 1 to 5, and the distance truth value library stores the distance truth values ​​of the two first map elements within a reference accuracy range; a fifth acquisition module, configured to respectively acquire the distance mark value of the second map element pair from the high-precision map, and acquire the distance true value corresponding to the first map element pair from the distance true value library; A determination module is used to determine whether the map accuracy of the high-precision map meets the requirements based on the distance mark value and the distance true value.

10. An electronic device comprising a memory and a processor, the memory being used to store executable instructions for the processor; the processor being used to read the executable instructions from the memory and execute the executable instructions to implement the method for constructing a distance truth library as described in any one of claims 1 to 5 or the method for detecting the accuracy of a high-precision map as described in any one of claims 6 to 7.

11. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for constructing a distance truth library as described in any one of claims 1 to 5 or the method for detecting the accuracy of a high-precision map as described in any one of claims 6 to 7 is implemented.

Images