CN115908619A

CN115908619A - Difference analysis method and device for high-precision map and storage medium

Info

Publication number: CN115908619A
Application number: CN202111157579.9A
Authority: CN
Inventors: 李奕乐; 肖良富; 王辉; 范争光; 张海洋
Original assignee: Hefei Siweitu New Technology Co ltd
Current assignee: Hefei Siweitu New Technology Co ltd
Priority date: 2021-09-30
Filing date: 2021-09-30
Publication date: 2023-04-04

Abstract

The embodiment of the application provides a difference analysis method, a device and a storage medium for a high-precision map, wherein the method comprises the following steps: respectively acquiring mutually corresponding map sheets in a first high-precision map and a second high-precision map, wherein the first high-precision map and the second high-precision map are respectively segmented into a plurality of map sheets with the same specification, and each map sheet comprises a road link and elements associated with the road link; and aiming at each pair of mutually corresponding map sheets, sequentially combining the map sheet corresponding to the first high-precision map and the map sheet corresponding to the second high-precision map with the high-precision map data of different levels and the element attributes of the elements according to the road link in the map sheets and the elements associated with the road link, and determining the difference result of the first high-precision map and the second high-precision map for correcting or making the high-precision map. The method provided by the embodiment of the application can solve the problem that the prior art cannot comprehensively and effectively realize the difference analysis of the high-precision map.

Description

Difference analysis method and device for high-precision map and storage medium

Technical Field

The embodiment of the application relates to the technical field of high-precision maps, in particular to a difference analysis method and device for a high-precision map and a storage medium.

Background

In the field of unmanned driving, a high-precision map is taken as a service provider of prior environment information, has a very important role, and particularly plays a vital role in the processes of high-precision positioning, environment perception assistance, planning and decision making. High-precision maps need to contain detailed road models including lane models, road components, road attributes, and other various dynamic information in addition to what conventional maps contain.

Based on the requirement of unmanned driving on a high-precision map, the absolute and relative coordinate precision of the high-precision map must be higher, and the information elements contained in the high-precision map are richer and more detailed, so that high-freshness, high-precision and multi-dimensional roads and additional information must be provided for an automatic driving system. Because of this, the high-precision map provides new challenges in terms of process, quality, update period, road attribute richness, and the like, compared with the traditional electronic map.

At present, differential identification is generally adopted to perform multi-version high-precision map differentiation, but the method strongly depends on the differential identification, has special requirements on a map manufacturing process, and has limitations, so that the prior art cannot comprehensively and effectively realize the differential analysis of a high-precision map.

Disclosure of Invention

The embodiment of the application provides a method and a device for analyzing the difference of a high-precision map and a storage medium, so as to solve the problem that the prior art cannot comprehensively and effectively realize the difference analysis of the high-precision map.

In a first aspect, an embodiment of the present application provides a difference analysis method for a high-precision map, including:

respectively obtaining mutually corresponding map sheets in a first high-precision map and a second high-precision map, wherein the first high-precision map and the second high-precision map are respectively segmented into a plurality of map sheets with the same specification, and each map sheet comprises a road link and elements associated with the road link;

and for each pair of corresponding maps, according to the road link in the maps and the element associated with the road link, comparing the maps corresponding to the first high-precision map and the maps corresponding to the second high-precision map in sequence by combining high-precision map data of different levels and the element attribute of the element, and determining the difference result of the first high-precision map and the second high-precision map for correcting or manufacturing the high-precision maps.

In a second aspect, an embodiment of the present application provides a difference analysis apparatus for a high-precision map, including:

the system comprises an acquisition module, a processing module and a display module, wherein the acquisition module is used for respectively acquiring corresponding map sheets in a first high-precision map and a second high-precision map, the first high-precision map and the second high-precision map are respectively segmented into a plurality of map sheets with the same specification, and each map sheet comprises a road link and elements associated with the road link;

and the difference analysis module is used for comparing the map sheet corresponding to the first high-precision map and the map sheet corresponding to the second high-precision map sequentially by combining high-precision map data of different levels and element attributes of the elements according to the road link in the map sheets and the elements associated with the road link, and determining a difference result between the first high-precision map and the second high-precision map so as to correct or make the high-precision maps.

In a third aspect, an embodiment of the present application provides a computer-readable storage medium, where computer-executable instructions are stored, and when a processor executes the computer-executable instructions, a difference analysis method for a high-precision map is implemented as described in the first aspect and various possible designs of the first aspect.

In a fourth aspect, embodiments of the present application provide a computer program product, which includes a computer program, and when the computer program is executed by a processor, the method for analyzing a difference of a high-precision map according to the first aspect and various possible designs of the first aspect is implemented.

In the method, the apparatus, and the storage medium for analyzing the difference of the high-precision map provided in this embodiment, first, corresponding frames in a first high-precision map and a second high-precision map are respectively obtained, where the first high-precision map and the second high-precision map have been divided into a plurality of frames with the same specification, and each frame includes a road link and an element associated with the road link; then, for each pair of corresponding maps, according to the road link in the maps and the element associated with the road link, comparing the map corresponding to the first high-precision map with the map corresponding to the second high-precision map in sequence by combining high-precision map data of different levels and the element attribute of the element, and determining the difference result between the first high-precision map and the second high-precision map for correcting or making the high-precision map. The method has the advantages that the special requirements on the map making process are eliminated by a brand new difference mode, difference is carried out without strong dependence on difference identification, changes can be found actively by means of the step matching, such as element attribute comparison, difference analysis of the high-precision map can be achieved comprehensively and effectively, a more comprehensive error correction mechanism can be provided for the high-precision map, and the accuracy of high-precision map making is improved.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to these drawings without inventive exercise.

Fig. 1 is a scene schematic diagram of a difference analysis method for a high-precision map according to an embodiment of the present application;

fig. 2 is a schematic flowchart of a difference analysis method for a high-precision map according to an embodiment of the present disclosure;

fig. 3 is a schematic flowchart of a difference analysis method for a high-precision map according to another embodiment of the present application;

FIG. 4 is a schematic view of a diagram provided by an embodiment of the present application;

fig. 5 is a schematic diagram of an association relationship between LINKs and elements provided in the embodiment of the present application;

fig. 6 is a schematic view of a scene of a difference analysis method for a high-precision map according to yet another embodiment of the present application;

fig. 7 is a schematic view of a scene of a difference analysis method for a high-precision map according to yet another embodiment of the present application;

fig. 8 is a schematic structural diagram of a difference analysis apparatus for a high-precision map according to an embodiment of the present application;

fig. 9 is a schematic structural diagram of a difference analysis device for a high-precision map according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The terms "first," "second," "third," "fourth," and the like in the description and in the claims of the present application and in the above-described drawings (if any) are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are, for example, capable of operation in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

At present, differential identification is generally adopted to carry out multi-version high-precision map differential, but the mode strongly depends on the differential identification, has special requirements on a map manufacturing process and has limitations, so that the prior art cannot comprehensively and effectively realize the differential analysis of a high-precision map.

Aiming at the problems in the prior art, the technical idea of the application is that a brand-new difference mode is adopted to carry out difference comparison on two input maps, namely grading matching, attribute similarity is adopted in the difference method, so that the special requirements on the map manufacturing process are abandoned, difference does not need to be carried out by relying on difference identification strongly, element similarity is adopted to replace identification difference, change can be found actively, the difference analysis of the high-precision map can be realized comprehensively and effectively, the difference analysis of the high-precision map can be further provided for the high-precision map, and the accuracy of the high-precision map manufacturing is improved.

In practical application, referring to fig. 1, fig. 1 is a scene schematic diagram of a difference analysis method of a high-precision map provided in the embodiment of the present application. The execution subject of the embodiment may be a difference analysis device of a high-precision map, such as a terminal device, a fixed terminal, a mobile terminal, a computer device (e.g., a kiosk, etc.), or an electronic device with image and data analysis capabilities, or a server, a processor, etc. Taking the terminal device 10 as an example, two versions of maps, such as two versions of high-precision maps (a first high-precision map and a second high-precision map), are collected by the collecting device 20, the collecting device 20 transmits the two versions of high-precision maps to the terminal device 10, and the terminal device 10 performs differential comparison on the two versions of input maps (i.e., the high-precision maps) and outputs a differential result.

Specifically, firstly, two versions of high-precision maps are divided into a plurality of map sheets (namely MESH) with the same specification (such as the same size), then a brand-new difference mode is used for carrying out hierarchical matching on the two versions of high-precision maps, and a flow diagram of difference analysis of the high-precision maps shown in fig. 2 is combined, such as map sheet level matching, road network matching and lane level matching, wherein in the process of lane level matching, element attribute matching is required, and finally, a result, namely a difference result of the two versions of high-precision maps is extracted.

The technical solution of the present application will be described in detail below with specific examples. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments.

Fig. 3 is a schematic flowchart of a difference analysis method for a high-precision map according to another embodiment of the present application, where the method may include:

s101, obtaining corresponding map sheets in a first high-precision map and a second high-precision map respectively, wherein the first high-precision map and the second high-precision map are divided into a plurality of map sheets with the same specification.

Wherein each map sheet comprises a road link and an element associated with the road link. The high-accuracy map, such as the first high-accuracy map and/or the second high-accuracy map, may or may not be a finished map, i.e., intermediate data, such as data used to create the high-accuracy map. For example, the same specification may be the same size, and is not limited specifically herein.

In the present embodiment, the execution subject may be a difference analysis device of a high-precision map. The method comprises the steps that two high-precision maps, namely a first high-precision map and a second high-precision map, are obtained through collection equipment, and difference analysis equipment of the high-precision maps divides the two input high-precision maps, namely, the complete map is organized into a plurality of MESHs (namely, map sheets, see a diagram sheet schematic diagram shown in FIG. 4) with the same size. The subsequent difference analysis device of the high-precision map performs difference processing on the map data in units of MESH to reduce the data processing scale.

Specifically, inside the MESH, LINKs (a road segment in a road network, that is, LINKs are signs representing roads in a mapping process and may be referred to as road LINKs hereinafter) and surface features (elements) associated with the LINKs are provided, see fig. 5 for a schematic diagram of the association relationship between the LINKs and the elements. The LINK-associated elements are all point-like elements and line-like elements which are extended from the road range of the LINK by a certain distance in both directions, and the element illustrated in fig. 5 is a point-like element and its associated LINK.

For the same map, linkids (i.e., identification numbers of road LINKs) of LINKs are persistent, objectids (i.e., identification numbers of elements) of elements are persistent, and each LINK and element correspond to a unique ID (i.e., an identification number). Corresponding map sheets, LINKs and the like can be selected from the two versions of high-precision maps through the identity identification numbers.

S102, aiming at each pair of corresponding map sheets, comparing the map sheet corresponding to the first high-precision map with the map sheet corresponding to the second high-precision map sequentially by combining high-precision map data of different levels and element attributes of elements according to the road link in the map sheets and the elements related to the road link, and determining a difference result between the first high-precision map and the second high-precision map so as to correct or make the high-precision map.

In this embodiment, a brand-new difference mode is used to perform hierarchical matching, such as map-level matching, road network matching, lane-level matching, and element attribute comparison, on the two versions of high-precision maps in combination with high-precision map data of different levels, and finally a result, that is, a difference result of the two versions of high-precision maps is extracted. The difference analysis mode (such as the difference mode) does not need to rely on difference identification to carry out difference, element attribute comparison is adopted to replace identification difference, change can be found actively, the difference analysis method is different from the existing mode of using difference identification to verify difference reliability, difference analysis of a high-precision map can be achieved comprehensively and effectively, a more comprehensive error correction mechanism can be provided for the high-precision map based on a difference result, and the accuracy of high-precision map manufacturing is improved.

In a possible design, the present embodiment provides a detailed description of S102 on the basis of the above embodiments. According to the road link in the map sheets and the element associated with the road link, comparing the map sheet corresponding to the first high-precision map with the map sheet corresponding to the second high-precision map in sequence by combining high-precision map data of different levels and the element attribute of the element, and determining the difference result between the first high-precision map and the second high-precision map, which can be realized by the following steps:

step a1, according to a road link in the map sheet and elements associated with the road link, map sheet matching comparison, road network matching comparison and lane matching comparison are sequentially carried out on the map sheet corresponding to the first high-precision map and the map sheet corresponding to the second high-precision map, and a lane matching result is obtained.

Step a2, if the lane matching comparison is consistent, performing attribute comparison on elements with consistent lane matching comparison, if the attribute comparison is inconsistent, outputting an attribute matching result, wherein the attribute matching result is used for representing element attribute change, and if the attribute comparison is completely consistent, outputting the attribute matching result as an element unchanged.

In this embodiment, the high-precision map data at different levels may include map data, road network data, and lane data. And according to the road link in the map and the elements associated with the road link, sequentially carrying out map matching comparison, road network matching comparison and lane matching comparison by combining the map data, the road network data and the lane data respectively to obtain a lane matching result.

The map sheet corresponding to the first high-precision map and the map sheet corresponding to the second high-precision map are subjected to map sheet matching comparison, road network matching comparison and lane matching comparison in sequence, and the method can be realized through the following steps:

step a11, according to a road link in each map sheet and elements associated with the road link, map sheet matching comparison is performed on the map sheet corresponding to the first high-precision map and the map sheet corresponding to the second high-precision map, if the map sheet matching comparison is not consistent, a map sheet matching result is output, and the map sheet matching result is used for indicating that the map sheet is newly added or deleted.

Step a12, if the map matching comparison is consistent, performing road network matching comparison on target maps which are matched and compared with the map, and if the road network matching comparison is inconsistent, outputting a road network matching result which is used for indicating whether roads are added or deleted.

Step a13, if the road network matching comparison is consistent, performing lane matching comparison on the road links with consistent road network matching comparison, and if the lane matching comparison is inconsistent, outputting a lane matching result, wherein the lane matching result is used for indicating that the element is newly added or deleted.

In the embodiment, MESH-level matching is firstly carried out on the map sheets into which the two high-precision maps are divided, and if the map sheets are not completely consistent, a MESH-level difference result is firstly output; and if the road network matching is consistent with the road network matching, the next step is carried out.

Specifically, the map is divided into MESH in order to reduce the data processing scale of the map, and since the difference processing is performed, two different versions of the map can be selected for comparison. The MESH list of the first version of map and the MESH list of the second version of map are selected for comparison (see table 1 below), if the two lists are not completely consistent, the MESH-level differential result is output first, and the MESH-level differential result outputs the result of all the associated elements of the associated LINKs inside the whole MESH, which indicates that all the elements inside the whole MESH do not match.

TABLE 1

And if the road network matching is consistent with the road network matching, the next step is carried out. Selecting an MESH shared by a first version map (namely a first high-precision map) and a second version map (namely a second high-precision map), selecting a Link of the MESH in the second version map, marking the Link as Link1, performing road network matching with the first version map, if the Link and the Link-associated land feature are not matched, deleting the road, selecting the Link of the MESH in the first version map, marking the Link2, performing road network matching with the second version map, if the Link and the Link-associated land feature are not matched, performing lane level matching by selecting the shared Link, if the elements in the second version map are successfully matched in the first version map, performing element attribute comparison, namely judging the attributes of the LINKs, if one of the attributes is inconsistent, indicating that the attribute of the element is changed, outputting that the attribute of the element is changed, and if all the attributes are consistent, indicating that the element is not changed.

According to the difference analysis method of the high-precision map, the continuous frame images are obtained, and feature extraction is carried out on each frame of image in the continuous frame images to obtain image feature points of each frame of image; then detecting dynamic characteristic points corresponding to the continuous frame images according to the image characteristic points of each frame image; performing super-pixel segmentation on each frame of image to determine a dynamic target area; and according to the detected dynamic feature points and the dynamic target area, removing the feature points of the dynamic area from the image feature points of each frame of image to obtain the feature points of the static area, wherein the feature points of the static area are used for providing a data source for executing the operations of camera positioning and environment mapping, detecting and segmenting the dynamic area target, and then removing the dynamic area target, so that the method is simple to realize, can improve the positioning precision of the binocular vision SLAM algorithm in the dynamic environment, and further can obtain accurate positioning and mapping results by utilizing a pure static environment.

In a possible design, on the basis of the above-mentioned embodiments, how to implement road network matching is described in detail. The target map sheets with the consistent map sheet matching and comparing are subjected to road network matching and comparing, if the road network matching and comparing are inconsistent, a road network matching result is output, and the road network matching method can be realized through the following steps:

step b1, obtaining the map sheets with the consistent map sheet matching and comparison from the first high-precision map and the second high-precision map as the target map sheets, and obtaining map data corresponding to the target map sheets, wherein the map data comprises the positions of all road links in the target map sheets.

And b2, if the first high-precision map is taken as a reference map, acquiring a target road link from the target map sheet in the second high-precision map, and acquiring all road links within a preset distance threshold range of the target road link from the first high-precision map according to the position of the target road link to form a road link set.

And b3, determining whether the road network matching comparison is consistent or not according to the road link set and the target road link, and if not, determining that the road network matching result is the target road link and the factor associated with the target road link is the road addition.

And b4, if the second high-precision map is taken as a reference map, acquiring a target road link from the target map sheet in the first high-precision map, and acquiring all road links within a preset distance threshold range of the target road link from the second high-precision map according to the position of the target road link to form a road link set.

And b5, determining whether the road network matching comparison is consistent or not according to the road link set and the target road link, and if not, determining that the road network matching result is the target road link and the factor associated with the target road link is road deletion.

According to the road link set and the target road link, whether the road network matching comparison is consistent or not is determined, and the method can be realized through the following steps:

and c1, grouping the road links in the road link set according to the topological relation among the road links in the road link set to obtain a plurality of continuous road link sets.

And c2, aiming at each road link group, calculating the intersection length and the distance between the road link group and the target road link to obtain the optimal road link group.

And c3, if the optimal road link group is empty, determining that the road network matching comparison is inconsistent.

In this embodiment, a MESH (as a target map) common to a first map and a second map is selected, and if the first map is used as a reference map, a Link of the MESH in the second map is selected and marked as Link1 (i.e., a target road Link), and is subjected to road network matching with the first map.

The matching method comprises the following steps:

firstly, according to the Link1 absolute position, all links within a range of x meters near Link1 (namely within a preset distance threshold value range) are selected from a first version map and are marked as LinkSet (namely a road Link set). And according to the topological relation among the links, including the direction and the Link connection relation, the links in the Link set are grouped, so that a plurality of continuous Link groups are obtained.

Then, for a plurality of continuous Link groups, the intersection length and the distance between the plurality of continuous Link groups and Link1 are respectively calculated, and then the optimal Link group is selected. Specifically, referring to fig. 6, fig. 6 is a schematic scene diagram of a difference analysis method for a high-precision map according to still another embodiment of the present application. Analyzing all possible links in the range around Link1 as shown in fig. 6, based on the permutation combination and the continuity between the Link groups, a set of linksets can be found, for example:

{ L1, L3, L4, L5, L6, L10, L11, L12}, { L2, L3, L4, L5, L6, L10, L13, L14}, and the like. And selecting the Link group which has the longest intersection with the Link1, the shortest distance and the closest topological result according to the Link groups in the linkSet as the final Link group which is output by road network matching. Segment1, segment2, … and segment6 respectively represent segment1, segment2, … and segment 6. The data structure of each segment can be represented as: < now, list < L1, L2> >; < none, L3>; < now, list < (L4, L5, L6), (L4, L5, L6) >; < none, L10> table; < none, list < (L11, L12), (L13, L14) >. The data structure can indicate whether each segment needs to be subjected to incremental selection in the subsequent calculation process, if the segment is displayed as 'available', the optimal item needs to be selected from the list options, and if the segment is displayed as 'unavailable', the incremental calculation is not needed.

If the selected optimal Link group is empty, link1 and the land feature associated with Link1 are indicated to be newly added.

Conversely, selecting the LINK of the first version map to match the second version map, that is, taking the second version map as the reference map, if the LINK of the first version map is not matched with the second version map, the LINK1 and the LINK1 associated feature are indicated to be the road deletion, that is, the LINK and the LINK associated feature are indicated to be the road deletion. And finally, outputting new addition and deletion of road levels, and performing lane level matching on the LINKs of both road levels.

In one possible design, how lane-level matching is performed may be achieved by:

and d1, taking any road link with consistent road network matching comparison as a shared road link, and extracting element attributes from the shared road link in the first high-precision map and the second high-precision map respectively.

And d2, performing attribute similarity calculation on each element attribute to obtain the similarity of the elements corresponding to the first high-precision map and the second high-precision map on the element attributes.

And d3, comparing the elements according to the similarity of each element on the element attribute, and calculating to obtain the similarity of the elements.

And d4, matching the elements according to the similarity of the elements, determining that the lane matching is inconsistent if the preset conditions are not met, and outputting lane matching results.

In this embodiment, first, the element attributes are extracted: a common LINK is selected, and features (elements) associated with the LINK on the two versions of the map are selected, including but not limited to signs, road markings, road borders, etc., as shown in table 2, the present embodiment is illustrated with signs and road markings and the following attributes as examples, but the scope of the present application includes but is not limited to the following elements and attributes.

TABLE 2

Then calculating the similarity of the single attributes: in order to match two elements, a similarity function (different dimensions of a plurality of attributes) needs to be defined for each attribute first.

And the similarity of certain elements of the two versions of maps on the attribute i is shown.

Specifically, for the continuous type attribute (the continuous type attribute means that the attribute value is a continuous numerical value), for example, the line width, the value can be continuously taken within a certain range. The difference can be used to define:

wherein +>

A numerical value representing an attribute i of a certain element in the first map; />

A numerical value representing an attribute i of an element in the second map. max represents the maximum value of this attribute for all elements, and is mainly used to normalize the difference to 0 to 1.

For the attribute of the three-dimensional coordinate, the coordinate can be divided into three continuous attributes in the x, y and z directions to calculate the similarity of the attributes respectively.

For discrete attributes (the value of discrete attributes is limited, for example, in the shape of a sign, there are three types, rectangular, circular and triangular). A two-dimensional matrix look-up table can be defined, different attributes are assigned according to values in the look-up table, and the similarity can be defined as follows:

wherein x is a number between 0 and 0.5 and y is a number between 0.5 and 1. There are different values of x and y for different attributes.

For the character type attribute, the similarity of character type attributes is usually compared with the similarity of character strings, taking the character type on a sign as an example, the higher the similarity of characters on the sign is, the higher the similarity of character type attributes is, and conversely, the lower the similarity of characters on the sign is, the lower the similarity of character type attributes is. Wherein, the similarity of character type attribute can be defined by edit distance: the edit distance represents the minimum number of edits required to convert from one string to another. For the two strings a and b, the formula for the edit distance is defined as follows:

wherein, lev _a,b (i, j) represents the edit distance of the two character strings a and b; i and j represent subscripts of the character string a and the character string b, respectively. Subscripts start with 1, e.g., abel is 1 from adel edit distance. Then, the edit distance is divided by the Max length of the two character strings, and the result can be summarized as 0-1 as the character string difference degree. The language difference may also have an influence on the algorithm, such as the following formula:

here, the edge-dis (String 1, string 2) is an edit distance between the character String1 and the character String2, and max (legthofstring 1, legthofstring 2) is the longest character String length between the character String1 and the character String 2. The above formula converts the edit distance of the strings String1 and String2 into a matching probability value in the range of 0-1, and is used for measuring the similarity of the two strings.

Then, calculating the element similarity: comparing each element in the second map with the associated surface feature in the LINK in the first map, and calculating a matching degree (attribute similarity):

where i denotes an attribute index. Omega _i And representing the weight value of the attribute, wherein the weight value of the attribute is comprehensively defined according to the importance degree and the precision of the attribute. For example, an attribute has a high importance, and can largely determine the similarity of the attributes, so that the attribute we give a relatively high weight value. For another example, the three-dimensional coordinates of an element have good precision in the x and y directions and poor precision in the z direction, so that in order to reduce the influence of the precision in the z direction, we give higher weight values in the x and y directions and lower weight values in the z direction.

And (4) representing the similarity of the two versions of map elements on the attribute i, wherein the algorithm refers to a single attribute similarity function.

And finally, element matching: and if the preset condition is not met, not performing element matching or not performing matching successfully. The preset condition here may be that the distance threshold α is not exceeded and/or the similarity threshold β is not exceeded.

Specifically, first, a threshold value of absolute accuracy is set: a distance threshold alpha is set to indicate the GPS absolute position accuracy of the elements, and if the distance threshold alpha is exceeded, the elements cannot be the same element, and matching between the elements is not performed. The absolute position accuracy is set based on statistical knowledge of the map accuracy, which varies from map to map. According to the accuracy of the map, three dimensions of horizontal direction, vertical direction and height are considered. In addition, the accuracy thresholds of the dot elements and the line elements are also set separately. The calculation of absolute accuracy is illustrated in fig. 7, where the following are exemplified by a dotted element and a linear element, respectively:

as shown in fig. 7, in calculating the absolute accuracy, the linear elements only consider the lateral distance, and the point elements consider not only the lateral distance but also the longitudinal distance. It is not an accurate absolute point-to-point distance because of the different lateral and longitudinal accuracies.

Then setting a threshold value of the similarity: and setting a similarity threshold value beta to indicate the minimum similarity of the elements, wherein if the similarity is less than the threshold value, the elements cannot be the same element and are not matched.

Element matching: and respectively calculating the similarity of each element in the second version of map and the element in the absolute precision threshold range in the first version of map, wherein if the similarity is less than beta, the elements are not the same element and are not matched. Otherwise, selecting the element with the highest similarity from the first map to indicate that the matching is successful.

For elements in the second version of map, if there is no match in the first version of map, it indicates a new addition. Conversely, for an element in the first version of the map, if there is no match in the second version of the map, it indicates a deletion. And (5) successfully matching other elements, and entering the following steps: and judging attribute change (namely element attribute comparison).

Specifically, if the lane matching comparison is consistent, the attribute comparison is performed on the elements with consistent lane matching comparison, and the following steps can be implemented:

and e1, if the elements in the second high-precision map are successfully matched in the first high-precision map, determining that the lane matching comparison is consistent, and judging whether the attributes of the elements in the second high-precision map are consistent with the attributes of the elements in the first high-precision map.

And e2, if the elements in the first high-precision map are successfully matched in the second high-precision map, determining that the lane matching comparison is consistent, and judging whether the attributes of the elements in the first high-precision map are consistent with the attributes of the elements in the second high-precision map.

In this embodiment, if the elements in the second version map are successfully matched with the elements in the first version map, the attributes of the two are determined. If one attribute is inconsistent, the attribute of the element is changed, and the attribute of the element is output; if all attributes are consistent, the explanation element is unchanged. Similarly, if the elements in the first map are successfully matched in the second map, the attributes of the elements are judged. If one attribute is inconsistent, the attribute of the element is changed, and the attribute of the element is output to be changed; if all attributes are consistent, the explanation element is unchanged.

By hierarchical matching, the difference result is output, as shown in table 3.

TABLE 3

The method and the device mainly use the attribute similarity to carry out difference on the elements with the same ID, so that the method and the device do not depend on difference identification strongly, and can carry out difference on any two versions of common maps. Specifically, because the elements of the difference between the two versions of maps without change can maintain the consistency of the ID, the matching and the difference of the same ID elements of the two versions of maps can be carried out by comparing the attribute similarity, the requirement on the manufacturing process of the maps is not high, any two versions of maps without difference identification can also be subjected to difference, and the comprehensive and effective difference analysis is realized. The difference can check whether the changes and the invariants are not in accordance with the expectation, thereby improving the quality of the map and simultaneously verifying the accuracy of the tool.

Therefore, potential problems existing in the whole high-precision map manufacturing process can be found in advance by comparing the new version map with the old version map, and unchanged and changed elements in the map can be found. For elements that do not change, consistency of the ID may be maintained, and for elements that change, it may be checked subsequently whether the changes are as expected. By the active discovery of the problems, the accuracy of map making is improved, difference is not strongly dependent on difference identification, and the requirement on a map process is low.

In order to implement the difference analysis method for the high-precision map, the embodiment provides a difference analysis device for the high-precision map. Referring to fig. 8, fig. 8 is a schematic structural diagram of a difference analysis apparatus for a high-precision map provided in an embodiment of the present application; the high-precision map difference analysis device 80 includes: an acquisition module 801 and a difference analysis module 802; an obtaining module 801, configured to obtain corresponding map sheets in a first high-precision map and a second high-precision map, where the first high-precision map and the second high-precision map have been split into multiple map sheets of the same specification, and each map sheet includes a road link and an element associated with the road link; and a difference analysis module 802, configured to compare, for each pair of corresponding maps, the map corresponding to the first high-precision map and the map corresponding to the second high-precision map sequentially according to a road link in the maps and an element associated with the road link, and determine a difference result between the first high-precision map and the second high-precision map, where the difference result is used to correct or create the high-precision map, by combining high-precision map data of different levels and element attributes of the element.

In this embodiment, an obtaining module 801 and a difference analysis module 802 are configured to first obtain corresponding map sheets in a first high-precision map and a second high-precision map, respectively, where the first high-precision map and the second high-precision map have been divided into multiple map sheets of the same specification, and each map sheet includes a road link and elements associated with the road link; then, for each pair of corresponding maps, according to the road link in the maps and the element associated with the road link, comparing the map corresponding to the first high-precision map with the map corresponding to the second high-precision map in sequence by combining high-precision map data of different levels and the element attribute of the element, and determining the difference result between the first high-precision map and the second high-precision map for correcting or making the high-precision map. The method has the advantages that the special requirements on the map making process are eliminated by a brand new difference mode, difference is carried out without strong dependence on difference identification, changes can be found actively by means of the step matching, such as element attribute comparison, difference analysis of the high-precision map can be achieved comprehensively and effectively, a more comprehensive error correction mechanism can be provided for the high-precision map, and the accuracy of high-precision map making is improved.

The apparatus provided in this embodiment may be configured to implement the technical solutions of the method embodiments, and the implementation principles and technical effects are similar, which are not described herein again.

In one possible design, the difference result includes an attribute matching result; the difference analysis module comprises a first analysis unit and a second analysis unit; the first analysis unit is used for sequentially carrying out map matching comparison, road network matching comparison and lane matching comparison on the map corresponding to the first high-precision map and the map corresponding to the second high-precision map according to a road link in the maps and elements associated with the road link to obtain a lane matching result; and the second analysis unit is used for comparing the attributes of the elements which are matched and compared with the lane when the lane is matched and compared with the lane, outputting an attribute matching result if the attribute comparison is inconsistent, wherein the attribute matching result is used for representing the attribute change of the elements, and outputting the attribute matching result as the element is unchanged if the attribute comparison is completely consistent.

In one possible design, the first analysis unit comprises a first analysis subunit, a second analysis subunit and a third analysis subunit; the first analysis subunit is configured to perform map matching comparison on the map corresponding to the first high-precision map and the map corresponding to the second high-precision map according to a road link in each map and an element associated with the road link, and if the map matching comparison is inconsistent, output a map matching result, where the map matching result is used to indicate that the map is newly added or deleted; the second analysis subunit is used for carrying out road network matching comparison on target maps which are matched and compared with the same map when the maps are matched and compared with the same map, and outputting a road network matching result if the road network matching comparison is inconsistent, wherein the road network matching result is used for indicating that a road is newly added or deleted; and the third analysis subunit is used for carrying out lane matching comparison on the road links with the consistent road network matching comparison when the road network matching comparison is consistent, and outputting a lane matching result if the lane matching comparison is inconsistent, wherein the lane matching result is used for indicating that the element is newly added or deleted.

In a possible design, the second analysis subunit is specifically configured to: acquiring a map sheet with consistent map sheet matching comparison from the first high-precision map and the second high-precision map as the target map sheet, and acquiring map data corresponding to the target map sheet, wherein the map data comprises the position of each road link in the target map sheet;

if the first high-precision map is taken as a reference map, acquiring a target road link from the target map sheet in the second high-precision map, and acquiring all road links within a preset distance threshold range of the target road link in the first high-precision map according to the position of the target road link to form a road link set;

determining whether the road network matching comparison is consistent or not according to the road link set and the target road link, and if not, determining that the road network matching result is the target road link and that the factor associated with the target road link is the road addition;

correspondingly, if the second high-precision map is taken as a reference map, whether the road network matching comparison is consistent or not is determined according to the road link set and the target road link, and if not, the road network matching result is determined to be the target road link and the factor associated with the target road link is road deletion.

In one possible design, the second analysis subunit is specifically configured to: grouping the road links in the road link set according to the topological relation among the road links in the road link set to obtain a plurality of continuous road link sets;

calculating the intersection length and the distance between the road link group and the target road link aiming at each road link group to obtain an optimal road link group;

and if the optimal road link group is empty, determining that the road network matching comparison is inconsistent. In one possible design, the third analysis subunit is specifically configured to:

taking any road link with consistent road network matching comparison as a common road link, and extracting element attributes from the common road link in the first high-precision map and the second high-precision map respectively;

performing attribute similarity calculation on each element attribute to obtain the similarity of elements corresponding to the first high-precision map and the second high-precision map on the element attributes;

comparing the elements according to the similarity of each element on the element attribute, and calculating to obtain the similarity of the elements;

and matching the elements according to the similarity of the elements, if the preset conditions are not met, determining that the lane matching comparison is inconsistent, and outputting lane matching results.

In one possible design, the second analysis unit is specifically configured to:

if the elements in the second high-precision map are successfully matched in the first high-precision map, determining that the lane matching comparison is consistent, and judging whether the attributes of the elements in the second high-precision map are consistent with the attributes of the elements in the first high-precision map;

correspondingly, if the elements in the first high-precision map are successfully matched in the second high-precision map, the lane matching is determined to be consistent, and whether the attributes of the elements in the first high-precision map are consistent with the attributes of the elements in the second high-precision map is judged.

In order to implement the method for analyzing the difference of the high-precision map, the embodiment provides a device for analyzing the difference of the high-precision map. Fig. 9 is a schematic structural diagram of a difference analysis device for a high-precision map according to an embodiment of the present application. As shown in fig. 9, the difference analysis device 90 of the high-precision map of the present embodiment includes: a processor 901 and a memory 902; a memory 902 for storing computer-executable instructions; a processor 901 for executing computer executable instructions stored in the memory to implement the steps performed in the above embodiments. Reference may be made in particular to the description relating to the method embodiments described above.

The embodiment of the present application further provides a computer-readable storage medium, in which computer-executable instructions are stored, and when a processor executes the computer-executable instructions, the method for analyzing the difference of the high-precision map is implemented.

Embodiments of the present application further provide a computer program product, which includes a computer program, and when the computer program is executed by a processor, the method for analyzing a difference of a high-precision map as described above is implemented.

In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described device embodiments are merely illustrative, and for example, the division of the modules is only one logical division, and other divisions may be realized in practice, for example, a plurality of modules may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or modules, and may be in an electrical, mechanical or other form. In addition, functional modules in the embodiments of the present application may be integrated into one processing unit, or each module may exist alone physically, or two or more modules are integrated into one unit. The unit formed by the modules can be realized in a hardware form, and can also be realized in a form of hardware and a software functional unit.

The integrated module implemented in the form of a software functional module may be stored in a computer-readable storage medium. The software functional module is stored in a storage medium and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) or a processor (processor) to execute some steps of the methods according to the embodiments of the present application. It should be understood that the Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the present invention may be embodied directly in a hardware processor, or in a combination of the hardware and software modules within the processor.

The memory may comprise a high-speed RAM memory, and may further comprise a non-volatile storage NVM, such as at least one disk memory, and may also be a usb disk, a removable hard disk, a read-only memory, a magnetic or optical disk, etc. The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended ISA (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, the buses in the figures of the present application are not limited to only one bus or one type of bus. The storage medium may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks. A storage media may be any available media that can be accessed by a general purpose or special purpose computer.

An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. Of course, the storage medium may also be integral to the processor. The processor and the storage medium may reside in an Application Specific Integrated Circuits (ASIC). Of course, the processor and the storage medium may reside as discrete components in an electronic device or host device.

Those of ordinary skill in the art will understand that: all or a portion of the steps of implementing the above-described method embodiments may be performed by hardware associated with program instructions. The program may be stored in a computer-readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.

Finally, it should be noted that: the above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and these modifications or substitutions do not depart from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A difference analysis method of a high-precision map is characterized by comprising the following steps:

respectively acquiring corresponding map sheets in a first high-precision map and a second high-precision map, wherein the first high-precision map and the second high-precision map are respectively segmented into a plurality of map sheets with the same specification, and each map sheet comprises a road link and elements associated with the road link;

2. The method of claim 1, wherein the difference result comprises an attribute matching result; the determining, according to a road link in the map sheet and an element associated with the road link, a difference result between the first high-precision map and the second high-precision map by comparing the map sheet corresponding to the first high-precision map and the map sheet corresponding to the second high-precision map in combination with high-precision map data of different levels and element attributes of the element in sequence includes: according to the road link in the map sheet and the elements related to the road link, sequentially carrying out map sheet matching comparison, road network matching comparison and lane matching comparison on the map sheet corresponding to the first high-precision map and the map sheet corresponding to the second high-precision map to obtain a lane matching result;

if the lane matching comparison is consistent, performing attribute comparison on the elements with consistent lane matching comparison, if the attribute comparison is inconsistent, outputting an attribute matching result, wherein the attribute matching result is used for representing attribute change of the elements, and if the attribute comparison is completely consistent, outputting the attribute matching result as the elements are unchanged.

3. The method of claim 2, wherein the difference results further comprise at least one of an image frame matching result, a road network matching result, and a lane matching result;

according to the road link in the map sheet and the elements associated with the road link, sequentially performing map sheet matching comparison, road network matching comparison and lane matching comparison on the map sheet corresponding to the first high-precision map and the map sheet corresponding to the second high-precision map to obtain a lane matching result, and the method comprises the following steps of:

according to the road link in each map sheet and the elements associated with the road link, map sheet matching comparison is carried out on the map sheet corresponding to the first high-precision map and the map sheet corresponding to the second high-precision map, if the map sheet matching comparison is inconsistent, a map sheet matching result is output, and the map sheet matching result is used for indicating that the map sheet is added or deleted;

if the map sheet matching comparison is consistent, performing road network matching comparison on the target map sheet with the map sheet matching comparison being consistent, and if the road network matching comparison is inconsistent, outputting a road network matching result, wherein the road network matching result is used for indicating that a road is newly added or deleted;

and if the road network matching comparison is consistent, performing lane matching comparison on the road links with consistent road network matching comparison, and if the lane matching comparison is inconsistent, outputting a lane matching result, wherein the lane matching result is used for indicating that the element is newly added or deleted.

4. The method according to claim 3, wherein the performing road network matching comparison on the target map whose map matching comparison is consistent, and if the road network matching comparison is inconsistent, outputting a road network matching result, comprises:

obtaining a map sheet with consistent map sheet matching and comparison from the first high-precision map and the second high-precision map as the target map sheet, and obtaining map data corresponding to the target map sheet, wherein the map data comprises the position of each road link in the target map sheet;

if the first high-precision map is taken as a reference map, acquiring a target road link from the target map in the second high-precision map, and acquiring all road links within a preset distance threshold range of the target road link in the first high-precision map according to the position of the target road link to form a road link set;

determining whether the road network matching comparison is consistent according to the road link set and the target road link, and if not, determining that the road network matching result is the target road link and that the factor associated with the target road link is road addition;

5. The method of claim 4, wherein determining whether the road network match comparison is consistent from the set of road links and the target road link comprises:

grouping the road links in the road link set according to the topological relation among the road links in the road link set to obtain a plurality of continuous road link sets;

and if the optimal road link group is empty, determining that the road network matching comparison is inconsistent.

6. The method according to any one of claims 3 to 5, wherein the lane matching comparison is performed on the target road links with the consistent road network matching comparison, and if the lane matching comparison is not consistent, the lane matching result is output, including:

7. The method according to any one of claims 2 to 5, wherein if the lane matching matches, performing attribute matching on the elements matching the lane matching matches comprises:

8. A difference analysis device for a high-precision map, comprising:

9. A computer-readable storage medium, wherein a computer-executable instruction is stored in the computer-readable storage medium, and when a processor executes the computer-executable instruction, the method for analyzing the difference of the high-precision map according to any one of claims 1 to 7 is implemented.

10. A computer program product comprising a computer program, characterized in that the computer program, when executed by a processor, implements the method of difference analysis of high accuracy maps as claimed in any one of claims 1 to 7.