Map updating method and device based on lane marking, electronic equipment, storage medium and program product

Abstract

The embodiment of the invention provides a map updating method and device based on lane marking, electronic equipment, a storage medium and a program product. The method comprises the steps that a road reference line of a road to be updated and a collection lane marking line under each collection frequency are obtained, and the collection lane marking line is a collection lane marking line obtained when each real lane marking line is collected; then, aiming at the road to be updated, establishing each road cross section; wherein each acquisition lane marking is provided with a plurality of first intersection points between the acquisition lane marking and the road cross section, and the road reference line is provided with a plurality of second intersection points between the road reference line and the road cross section. For each first intersection point, determining absolute distance information of the first intersection point and relative distance information of the first intersection point; and determining the updated collected lane marking of each real lane marking of the road to be updated, and updating the high-precision map. Therefore, the updated acquired lane marking line can be accurately determined, and the high-precision map can be accurately updated.

Description

Technical Field

[0001] The present application relates to the field of traffic technology, and in particular to a lane marking-based map updating method, device, electronic device, storage medium and program product. Background Art

[0002] With the rapid development of autonomous driving technology, lane markings in high-precision maps need to be updated. Lane markings are used to indicate lane demarcation lines, curb markings, guardrail markings, etc. For real lane markings, collected lane markings can be obtained, that is, collected lane markings are data obtained by collecting real lane markings.

[0003] In the prior art, for each real lane marking, the average value of the position data represented by the collected lane markings collected multiple times is used as the updated position information of the collected lane markings corresponding to the real lane markings. Then, based on the updated position information of the collected lane markings, the high-precision map is updated.

[0004] However, in the above method, only the average value of the position data represented by the collected lane markings collected multiple times is used as the updated position information of the collected lane markings corresponding to the real lane markings. The accuracy of the position information obtained by such calculation method is low. Therefore, the updated position information of the collected lane markings cannot be accurately determined, and the high-precision map cannot be accurately updated. Summary of the invention

[0005] The embodiments of the present application provide a lane marking-based map updating method, device, electronic device, storage medium and program product, which are used to accurately determine the location information of updated lane markings, thereby accurately updating the high-precision map.

[0006] In a first aspect, an embodiment of the present application provides a map updating method based on lane markings, comprising:

[0007] Acquire road information of the road to be updated; wherein the road information includes a road reference line of the road to be updated and a collected lane marking of the road to be updated at each collection number, wherein the collected lane marking is a collected lane marking obtained when each real lane marking of the road to be updated is collected;

[0008] For the road to be updated, each road cross section of the road to be updated is established according to the road reference line of the road to be updated; wherein each collected lane marking line has a plurality of first intersection points, the first intersection point being the intersection point between the collected lane marking line and the road cross section; the road reference line has a plurality of second intersection points, the second intersection point being the intersection point between the road reference line and the road cross section;

[0009] For each first intersection of each collected lane marking, determine the absolute distance information of the first intersection; wherein the absolute distance information represents the distance information between the first intersection and a second intersection located in the same road cross section as the first intersection;

[0010] For each first intersection of each collected lane marking, determine the relative distance information of the first intersection; wherein the relative distance information represents the distance information between the first intersection and other first intersections located under the same road cross section during the same collection; or, the relative distance information represents the distance information between the first intersection and other first intersections located under the same collected lane marking as the first intersection;

[0011] According to the absolute distance information of the first intersection and the relative distance information of the first intersection, the updated collected lane marking of each real lane marking of the road to be updated is determined; and according to the updated collected lane marking of each real lane marking, the high-precision map is updated.

[0012] In a second aspect, an embodiment of the present application provides a map updating device based on lane markings, comprising:

[0013] An acquisition module, used to acquire road information of the road to be updated; wherein the road information includes a road reference line of the road to be updated and a collected lane marking of the road to be updated at each collection number, wherein the collected lane marking is a collected lane marking obtained when each real lane marking of the road to be updated is collected;

[0014] An establishing module is used to establish each road cross section of the road to be updated according to the road reference line of the road to be updated, wherein each collected lane marking line has a plurality of first intersection points, the first intersection point being the intersection point between the collected lane marking line and the road cross section; the road reference line has a plurality of second intersection points, the second intersection point being the intersection point between the road reference line and the road cross section;

[0015] A first determination module is used to determine, for each first intersection of each collected lane marking, absolute distance information of the first intersection; wherein the absolute distance information represents distance information between the first intersection and a second intersection located in the same road cross section as the first intersection;

[0016] A second determination module is used to determine, for each first intersection of each collected lane marking, relative distance information of the first intersection; wherein the relative distance information represents distance information between the first intersection and other first intersections located under the same road cross section during the same collection; or, the relative distance information represents distance information between the first intersection and other first intersections located under the same collected lane marking as the first intersection;

[0017] A third determination module is used to determine the updated collected lane marking of each real lane marking of the road to be updated according to the absolute distance information of the first intersection and the relative distance information of the first intersection;

[0018] The update module is used to update the high-precision map according to the updated collected lane markings of each real lane marking.

[0019] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0020] The memory stores computer-executable instructions;

[0021] The processor executes the computer-executable instructions stored in the memory to implement the method described in the first aspect above.

[0022] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method described in the first aspect above.

[0023] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the method described in the first aspect above.

[0024] The lane marking-based map updating method, device, electronic device, storage medium and program product provided in the embodiments of the present application first obtain the road reference line of the road to be updated and the collected lane markings of the road to be updated at each collection number, wherein the collected lane markings are the collected lane markings obtained when each real lane marking of the road to be updated is collected. Then, for the road to be updated, each road cross section of the road to be updated is established according to the road reference line of the road to be updated; wherein each collected lane marking has a plurality of first intersection points between the collected lane markings and the road cross section, and the road reference line has a plurality of second intersection points between the road reference line and the road cross section.

[0025] Then, for each first intersection of each collected lane marking, determine the absolute distance information of the first intersection; wherein the absolute distance information is the distance information between the second intersection located in the same road cross section as the first intersection and the first intersection. At the same time, for each first intersection of each collected lane marking, determine the relative distance information of the first intersection; wherein the relative distance information is the distance information between the first intersection and other first intersections located in the same road cross section as the first intersection during the same collection; or, the relative distance information is the distance information between the first intersection and other first intersections located under the same collected lane marking as the first intersection. Furthermore, based on the absolute distance information of the first intersection and the relative distance information of the first intersection, determine the updated collected lane marking of each real lane marking of the road to be updated; and update the high-precision map based on the updated collected lane marking of each real lane marking.

[0026] Since the second intersection is a point on the road reference line in the high-precision map, it has a higher accuracy. The absolute distance information of the first intersection calculated based on the second intersection with higher accuracy is also highly accurate. Because the two first intersections collected at the same time have the same source, for example, they are both extracted from the same photo, the relative positions of the two first intersections collected at the same time are relatively accurate. Therefore, the relative distance information of the first intersection determined based on the two first intersections collected at the same time is also highly accurate. Afterwards, based on the absolute distance information of the first intersection with higher accuracy and the relative distance information of the second intersection with higher accuracy, the absolute distance information of the position point of the updated collected lane marking of each real lane marking of the road to be updated is also highly accurate. Furthermore, the updated collected lane markings can be accurately determined, and the high-precision map can be accurately updated. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0028] Figure 1 A flowchart of a lane marking-based map updating method provided in an embodiment of the present application;

[0029] Figure 2 A schematic diagram of road information to be updated provided in an embodiment of the present application;

[0030] Figure 3 A schematic diagram of establishing a road cross section provided in an embodiment of the present application;

[0031] Figure 4 A schematic diagram of a first intersection and a second intersection provided in an embodiment of the present application;

[0032] Figure 5A schematic diagram of absolute distance information of a first intersection provided in an embodiment of the present application;

[0033] Figure 6 Schematic diagram of the relative distance information of the first intersection provided in the embodiment of the present application Figure 1 ;

[0034] Figure 7 Schematic diagram of the relative distance information of the first intersection provided in the embodiment of the present application Figure 2 ;

[0035] Figure 8 Schematic diagram of determining location information of a location point provided in an embodiment of the present application Figure 1 ;

[0036] Fig. 9 The updated collected lane marking schematic diagram provided in the embodiment of the present application;

[0037] Fig.10 Schematic diagram of determining location information of a location point provided in an embodiment of the present application Figure 2 ;

[0038] Fig.11 A flowchart of another lane marking-based map updating method provided in an embodiment of the present application;

[0039] Fig.12 A schematic diagram of the absolute longitudinal distance and the absolute lateral distance provided in an embodiment of the present application;

[0040] Fig.13 A schematic diagram of absolute distance of elevation provided in an embodiment of the present application;

[0041] Fig.14 A schematic diagram of the horizontal relative distance provided in an embodiment of the present application;

[0042] Fig.15 Schematic diagram of relative distance of elevation provided in the embodiment of the present application Figure 1 ;

[0043] Fig.16 A flowchart of another lane marking-based map updating method provided in an embodiment of the present application;

[0044] Fig.17 Schematic diagram of relative distance of elevation provided in the embodiment of the present application Figure 2 ;

[0045] Fig.18 A schematic diagram of the structure of a lane marking based map updating device provided in an embodiment of the present application;

[0046] Fig.19 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0047] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0048] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0049] First, the terms involved in this application are explained:

[0050] Road reference line: a line segment used to indicate a road in a high-precision map. The driving direction of each road reference line represents the actual direction of travel on the road.

[0051] Lane markings: In HD maps, lane markings are used to indicate lane boundaries, curb markings, guardrail markings, etc. Each lane marking is associated with a road reference line.

[0052] As the requirements for precision of autonomous driving technology become higher and higher, the lane markings in the high-precision map need to be updated. The collected lane markings can be obtained for the real lane markings, that is, the collected lane markings are the data obtained by collecting the real lane markings.

[0053] In one example, when updating the lane markings in a high-precision map, each road cross section perpendicular to the road reference line can be established along the road reference line of the road to be updated, every d meters (d is a natural number greater than 0, which can be determined based on empirical values, and this embodiment does not impose any restrictions on this). Among them, each collected lane marking has multiple first intersections, and the first intersection is the intersection between the collected lane marking and the road cross section. Then, for each road cross section, the average value of the coordinate values ​​of the first intersections collected multiple times for each real lane marking is used as the position information of the position point of the real lane marking under the road cross section. Then, for each real lane marking, the position information of the position points under each road cross section is sequentially connected in series to obtain the updated collected lane marking corresponding to the real lane marking. Then, according to the updated collected lane marking corresponding to each real lane marking, the lane markings in the high-precision map are updated.

[0054] However, in the above method, when collecting real lane markings, it may be affected by factors such as weak GPS signals, building obstruction, unstable vehicle speed, and bad weather, resulting in low accuracy of collected lane markings. As a result, for each road cross section, the average coordinate value of the first intersection of each real lane marking based on multiple collected collected lane markings is used as the position information of the position point of the real lane marking under the road cross section, and the accuracy is also low. Therefore, the updated collected lane markings cannot be accurately determined, and the high-precision map cannot be accurately updated.

[0055] In response to the above technical problems, the present application proposes the following technical concept: Referring to the above example, after obtaining multiple first intersections, for each road cross section, first determine the absolute distance information of an optimal real lane marking position point. Among them, the absolute distance information of the position point of the real lane marking represents the specific position of the position point of the real lane marking in the road cross section. Then, in the road cross section, determine the relative distance information between the position points of different real lane markings. Then, in the road cross section, based on the absolute distance information of the position point of the optimal real lane marking and the relative distance information between the position points of different real lane markings, determine the absolute distance information of all the real lane marking position points in the road cross section.

[0056] Alternatively, after obtaining multiple first intersections, for each real lane marking, first determine the absolute distance information of the position point of an optimal real lane marking. The absolute distance information of the position point of the real lane marking represents the specific position of the position point of the real lane marking under the road cross section. Then, under the real lane marking, determine the relative distance information between the position points under different road cross sections. Then, under the real lane marking, determine the absolute distance information of the position points under all road cross sections based on the absolute distance information of the position point of the optimal real lane marking and the relative distance information between the position points under different road cross sections.

[0057] After determining the absolute distance information of the position points under each road cross section of each real lane marking by one of the above two methods, for each real lane marking, the absolute distance information of each position point is sequentially connected in series to obtain the updated collected lane marking corresponding to the real lane marking. Then, according to the updated collected lane marking corresponding to each real lane marking, the lane marking in the high-precision map is updated.

[0058] This application can obtain the absolute distance information of a location point with higher accuracy by first determining the absolute distance information of an optimal real lane marking on the road to be updated. Then, through the absolute distance information of the location point with higher accuracy and the relative distance information with higher accuracy, the absolute distance information of other location points on the road to be updated can be derived with higher accuracy. In this way, the updated collected lane markings can be accurately determined, and the high-precision map can be accurately updated.

[0059] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0060] Figure 1 A flowchart of a lane marking-based map updating method provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the method includes:

[0061] S101. Acquire road information of a road to be updated; wherein the road information includes a road reference line of the road to be updated and a collected lane marking of the road to be updated at each collection number, wherein the collected lane marking is a collected lane marking obtained when each real lane marking of the road to be updated is collected.

[0062] Exemplarily, the execution subject of this embodiment can be a server, or a terminal device, or an electronic device, or a map update device or device based on lane markings, or other devices or equipment that can implement the solution of this application, without limitation. Among them, the server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server can also be a server of a distributed system, or a server combined with blockchain.

[0063] This embodiment is introduced with the execution subject being a server.

[0064] In reality, there may be many roads to be updated. For each road to be updated, the server may execute the solution of this embodiment.

[0065] For example, the server can obtain the road reference line of the road to be updated from the high-precision map data, and can obtain the marking combination of the road to be updated at each collection time from the crowdsourcing terminal device. The marking combination of the road to be updated at each collection time includes the collected lane markings obtained when each real lane marking of the road to be updated is collected.

[0066] Assume that the road to be updated contains real lane marking 1, real lane marking 2, real lane marking 3, and real lane marking 4, and the crowdsourcing terminal device collects a total of 3 marking combinations.

[0067] For example, you can combine Figure 2 Understand the road information of the road to be updated obtained by the server. Figure 2 This is a schematic diagram of the road information to be updated provided in the embodiment of the present application. Figure 2 As shown, the road information of the road to be updated obtained by the server is as follows Figure 2 As shown in M, M includes the road reference line L1 of the road to be updated from the high-precision map represented by L, and also includes the marking combination of the road to be updated collected for the first time from the crowdsourcing terminal device represented by M1, the marking combination of the road to be updated collected for the second time from the crowdsourcing terminal device represented by M2, and the marking combination of the road to be updated collected for the third time from the crowdsourcing terminal device represented by M3.

[0068] in, Figure 2 M1 in the example includes the real lane markings 1, 2, 3, and 4 of the road to be updated, and the collected lane markings M11, M12, M13, and M14 obtained during the collection. Similarly, Figure 2 M2 in the example includes real lane markings 1, 2, 3, and 4 of the road to be updated, and collected lane markings M21, M22, M23, and M24 obtained during collection. Figure 2 M3 in the figure includes the real lane marking 1, real lane marking 2, real lane marking 3, real lane marking 4 of the road to be updated, and the collected lane markings M31, collected lane markings M32, collected lane markings M33, and collected lane markings M34 obtained during the collection.

[0069] S102. For the road to be updated, establish each road cross section of the road to be updated according to the road reference line of the road to be updated; wherein each collected lane marking has a plurality of first intersection points, and the first intersection point is the intersection point between the collected lane marking line and the road cross section; the road reference line has a plurality of second intersection points, and the second intersection point is the intersection point between the road reference line and the road cross section.

[0070] Exemplarily, after acquiring the road information of the road to be updated, the server can, for the road to be updated, along the driving direction of the road reference line of the road to be updated, starting from the starting point of the driving direction of the road reference line, establish a road cross section perpendicular to the road reference line every d meters (d is a natural number greater than 0, which can be determined based on empirical values, and this embodiment does not impose any restrictions on this).

[0071] For example, you can combine Figure 3 To understand the road cross section, Figure 3 A schematic diagram of a road cross section provided in an embodiment of the present application is shown in FIG. Figure 3 As shown, the server can start from the starting point Ln1 of the driving direction 301 of the road reference line L1 of the road to be updated, and establish a road cross section A, road cross section B, and road cross section C perpendicular to the road reference line L1 every d meters on the road reference line.

[0072] For example, you can combine Figure 4 Understand the first intersection and the second intersection. Figure 4 A schematic diagram of the first intersection and the second intersection provided in an embodiment of the present application.

[0073] like Figure 4 As shown, the server refers to the example in step S101 to obtain the road information of the road to be updated, and refers to Figure 3 After establishing road cross sections A, B and C for the road to be updated, point An11, point An21 and point An31 are respectively the first intersection points between the collected lane markings M11, M21 and M31 obtained by collecting the real lane markings 1 for the first time, the second time and the third time and the road cross section A; point An12, point An22 and point An32 are respectively the first intersection points between the collected lane markings M12, M22 and M32 obtained by collecting the real lane markings 2 for the first time, the second time and the third time and the road cross section A; point An13, point An23, and point An33 are respectively the first intersection points between the collected lane marking M13, collected lane marking M23, and collected lane marking M33 obtained from the first, second, and third collection of the real lane marking 3 and the road cross-section A; point An14, point An24, and point An34 are respectively the first intersection points between the collected lane marking M14, collected lane marking M24, and collected lane marking M34 obtained from the first, second, and third collection of the real lane marking 4 and the road cross-section A; similarly, there are multiple first intersection points for the road cross-section B and the road cross-section C, which will not be repeated here.

[0074] Point ALn, point BLn, and point CLn are second intersection points between the road reference line L1 and the road cross section A, road cross section B, and road cross section C, respectively.

[0075] By establishing each road cross section, each collected lane marking corresponding to the road to be updated can be discretized into a plurality of first intersections, and the road reference line of the road to be updated can be offlined into a plurality of second intersections, thereby facilitating the subsequent determination of the updated collected lane marking of each real lane marking of the road to be updated based on the first intersections and the second intersections.

[0076] S103. For each first intersection of each collected lane marking, determine absolute distance information of the first intersection; wherein the absolute distance information represents distance information between the first intersection and the corresponding second intersection.

[0077] Exemplarily, after establishing each road cross-section of the road to be updated, the server can determine, for each first intersection of each collected lane marking, the distance information of the first intersection relative to the second intersection located in the same road cross-section as the first intersection, as the absolute distance information of the first intersection.

[0078] For example, you can combine Figure 5 To understand. Figure 5 This is a schematic diagram of the absolute distance information of the first intersection provided in the embodiment of the present application, such as Figure 5 As shown, under the road cross section A, the server can use the distance dh13 between the second intersection ALn between the road reference line L1 and the road cross section A and the first intersection An13 between the collected lane marking M13 and the road cross section A as the absolute distance information of the first intersection An13. Similarly, the server can determine the absolute distance information of each first intersection of each collected lane marking.

[0079] Since the second intersection is a point on the road reference line in the high-precision map, it has a higher accuracy. Therefore, the absolute distance information of the first intersection calculated based on the higher-precision second intersection also has a higher accuracy. This facilitates the subsequent accurate determination of the updated collected lane markings of each real lane marking of the road to be updated based on the higher-precision absolute distance information of the first intersection.

[0080] S104: for each first intersection of each collected lane marking, determine the relative distance information of the first intersection, wherein the distance information is the distance information between the first intersection and the corresponding other first intersections.

[0081] For example, after establishing each road cross section of the road to be updated, the server may determine the relative distance information of each first intersection of each collected lane marking. For example, the following two implementation methods may be included:

[0082] In a first possible implementation, the server may, for each first intersection of each collected lane marking, respectively obtain the distance information between the first intersection and each other first intersection located in the same road cross-section during the same collection as the first intersection, as the relative distance information of the first intersection.

[0083] For example, you can combine Figure 6 To understand. Figure 6 Schematic diagram of the relative distance information of the first intersection provided in the embodiment of the present application Figure 1 ,like Figure 6 As shown, for the first intersection An11 between the lane marking M11 and the road cross-section A, the other first intersections located in the same road cross-section A as the first intersection during the same collection are the first intersection An12 between the lane marking M12 and the road cross-section A, and the first intersection An13 between the lane marking M13 and the road cross-section A. The server can use the distance information dh12 between the first intersection An11 and the first intersection An12 as the relative distance information from the first intersection An11 to the first intersection An12.

[0084] At the same time, the server may use the distance information dh13 between the first intersection An11 and the first intersection An13 as the relative distance information from the first intersection An11 to the first intersection An13.

[0085] The relative distance information from the first intersection An11 to the first intersection An12 and the relative distance information from the first intersection An11 to the first intersection An13 are taken as the relative distance information of the first intersection An11.

[0086] By analogy, the server can determine the relative distance information of each first intersection point of each collected lane marking.

[0087] In a second possible implementation, the server may, for each first intersection of each collected lane marking, respectively obtain the distance information from the first intersection to each other first intersection located under the same collected lane marking as the first intersection, that is, use the difference between the absolute distance information of each other first intersection located under the same collected lane marking as the first intersection and the absolute distance information of the first intersection as the relative distance information of the first intersection.

[0088] For example, you can combine Figure 7 To understand. Figure 7 Schematic diagram of the relative distance information of the first intersection provided in the embodiment of the present application Figure 2 ,like Figure 7 As shown, it is assumed that the road to be updated has a total of 3 road cross sections, namely road cross section A, road cross section B, and road cross section C. It is assumed that the second intersection ALn, the second intersection BLn, and the second intersection CLn are the intersections of the road reference line L1 and the road cross section A, the road cross section B, and the road cross section C. It is assumed that the absolute distance information of the first intersection An13 is dha13, the absolute distance information of the first intersection Bn13 is dhb13, and the absolute distance information of the first intersection Cn13 is dhc13. For the first intersection An13 between the collection lane marking M13 and the road cross-section A, the other first intersections located under the same collection lane marking M13 are the first intersection Bn13 between the collection lane marking M13 and the road cross-section B, and the first intersection Cn13 between the collection lane marking M13 and the road cross-section C. The server can use the difference between the absolute distance information of the first intersection Bn13 and the absolute distance information of the first intersection An13, that is, the numerical value represented by the result of dhb13-dha13 as the relative distance information from the first intersection An13 to the first intersection Bn13.

[0089] At the same time, the server may use the difference between the absolute distance information of the first intersection Cn13 and the absolute distance information of the first intersection An13, that is, the value represented by the result of dhc13-dha13 as the relative distance information from the first intersection An13 to the first intersection Cn13.

[0090] The relative distance information from the first intersection An13 to the first intersection Bn13 and the relative distance information from the first intersection An13 to the first intersection Cn13 are taken as the relative distance information of the first intersection An13.

[0091] By analogy, the server can determine the relative distance information of each first intersection point of each collected lane marking.

[0092] Since the two first intersection points collected at the same time have the same source, for example, they are extracted from the same photo, the relative positions of the two first intersection points collected at the same time are relatively accurate, and therefore, the relative distance information of the first intersection points determined based on the two first intersection points collected at the same time is also relatively accurate. This facilitates the subsequent accurate determination of the updated collected lane markings of each real lane marking of the road to be updated based on the relatively accurate relative distance information of the first intersection points.

[0093] S105. Determine an updated collected lane marking of each real lane marking of the road to be updated according to the absolute distance information of the first intersection and the relative distance information of the first intersection.

[0094] Exemplarily, the following two implementations may be included:

[0095] In a first possible implementation, when the relative distance information of the first intersection is the distance information between the first intersection and other first intersections located in the same road cross section during the same collection, after determining the absolute distance information of the first intersection and the relative distance information of the first intersection, the server can determine the updated collected lane marking of each real lane marking of the road to be updated according to the two information. The specific process may include, for example:

[0096] The server determines, for each road cross section, a first absolute distance information set of the real lane marking in the road cross section based on the absolute distance information of the first intersection point collected multiple times for each real lane marking;

[0097] For each road cross section, the server uses the average value of each absolute distance information in the first absolute distance information set with the smallest standard deviation in the first absolute distance information set of each real lane marking in the road cross section as the absolute distance information of the position point of the real lane marking in the road cross section corresponding to the first absolute distance information set; or, for each road cross section, the server uses the average value of each absolute distance information in the first absolute distance information set with the smallest confidence interval of T distribution in the first absolute distance information set of each real lane marking in the road cross section (the confidence level used in calculating the T distribution can be determined based on experience, and this is not limited in this embodiment) as the absolute distance information of the position point of the real lane marking in the road cross section corresponding to the first absolute distance information set.

[0098] The server determines, for each road cross section, relative distance information of the first intersection of the real lane marking corresponding to the absolute distance information of the existing position point collected multiple times to the first intersection of the real lane marking corresponding to the absolute distance information of each non-existent position point located in the same road cross section at the same time of collection, and determines relative distance information of the real lane markings corresponding to two first intersections corresponding to the relative distance information as the first relative distance information set, which are the same;

[0099] For each road cross section, the server uses the average value of each relative distance information in each first relative distance information set as the first relative distance information corresponding to the first relative distance information set, from the position point of the real lane marking corresponding to the absolute distance information of the existing position point to the position point of the real lane marking corresponding to the absolute distance information of the non-existent position point. Then, the sum of each first relative distance information and the absolute distance information of the position point of the real lane marking corresponding to the absolute distance information of the existing position point corresponding to the first relative distance information is used as the absolute distance information of the position point of the real lane marking in the road cross section corresponding to the absolute distance information of the non-existent position point.

[0100] The server serially connects the absolute distance information of each real lane marking of the road to be updated at each road cross section along the driving direction of the road reference line of the road to be updated to obtain the updated collected lane markings of each real lane marking of the road to be updated.

[0101] For example, you can combine Figure 8 Understand the process of determining the absolute distance information of the position point of the actual lane marking in the above method. Figure 8 Schematic diagram of determining location information of a location point provided in an embodiment of the present application Figure 1 ,like Figure 8 As shown, it is assumed that the road to be updated has three road cross sections, namely road cross section A, road cross section B, and road cross section C. It is assumed that the real lane marking 1 is collected for the first, second, and third time respectively to obtain the collected lane marking M11, the collected lane marking M21, and the collected lane marking M31, then the first intersection point of the real lane marking 1 under the road cross section A is the first intersection point An11, the first intersection point An21, and the first intersection point An31. The absolute distance information of the first intersection point An11, the absolute distance information of the first intersection point An21, and the absolute distance information of the first intersection point An31 constitute the first absolute distance information set A1.

[0102] Assuming that the real lane marking 2 is collected for the first, second, and third time respectively to obtain the collected lane marking M12, the collected lane marking M22, and the collected lane marking M32, then the first intersection points of the real lane marking 2 under the road cross section A are the first intersection point An12, the first intersection point An22, and the first intersection point An32. The absolute distance information of the first intersection point An12, the absolute distance information of the first intersection point An22, and the absolute distance information of the first intersection point An32 constitute the first absolute distance information set A2.

[0103] Assuming that the real lane marking 3 is collected for the first, second, and third time respectively, the collected lane marking M13, the collected lane marking M23, and the collected lane marking M33 are obtained, then the first intersection points of the real lane marking 3 under the road cross section A are the first intersection point An13, the first intersection point An23, and the first intersection point An33. The absolute distance information of the first intersection point An13, the absolute distance information of the first intersection point An23, and the absolute distance information of the first intersection point An33 constitute the first absolute distance information set A3.

[0104] Assuming that the real lane marking 4 is collected for the first, second, and third time respectively to obtain the collected lane marking M14, the collected lane marking M24, and the collected lane marking M34, the first intersection points of the real lane marking 4 under the road cross section A are the first intersection point An14, the first intersection point An24, and the first intersection point An34. The absolute distance information of the first intersection point An14, the absolute distance information of the first intersection point An24, and the absolute distance information of the first intersection point An34 constitute the first absolute distance information set A4.

[0105] Under the road cross section A, assuming that among the first absolute distance information set A1, the first absolute distance information set A2, the first absolute distance information set A3, and the first absolute distance information set A4, the standard deviation of the first absolute distance information set A1 is the smallest, then the average of the absolute distance information of the first intersection An11, the absolute distance information of the first intersection An21, and the absolute distance information of the first intersection An31 in the first absolute distance information set A1 is used as the absolute distance information of the position point of the real lane marking 1 under the road cross section A. Alternatively, under the road cross section A, assuming that among the first absolute distance information set A1, the first absolute distance information set A2, the first absolute distance information set A3, and the first absolute distance information set A4, the confidence interval of the T distribution of the first absolute distance information set A1 is the smallest, then the average of the absolute distance information of the first intersection An11, the absolute distance information of the first intersection An21, and the absolute distance information of the first intersection An31 in the first absolute distance information set A1 is used as the absolute distance information of the position point of the real lane marking 1 under the road cross section A.

[0106] Since the standard deviation of the first absolute distance information set is used to indicate the degree of discreteness of the distribution of the absolute distance information in the set, the smaller the standard deviation, the more concentrated the distribution of the absolute distance information collected multiple times in the set, that is, the absolute distance information collected multiple times in the set is closer to the true absolute distance information. Therefore, at each road cross section, the average value of each absolute distance information in the first absolute distance information set with the smallest standard deviation is closer to the true absolute distance information. It can also be understood that at this time, at this road cross section, the absolute distance information has a higher accuracy.

[0107] Since the confidence interval of the T distribution of the first absolute distance information set is used to indicate the credibility of the absolute distance information in the set, the smaller the confidence interval of the T distribution is, the closer the absolute distance information collected multiple times in the set is to the true absolute distance information. Therefore, under each road cross section, the average value of each absolute distance information in the first absolute distance information set with the smallest confidence interval of the T distribution is closest to the true absolute distance information. It can also be understood that at this time, under this road cross section, the absolute distance information has a higher accuracy.

[0108] After obtaining the absolute distance information of the position point of the real lane marking 1 under the road cross-section A, the server can form a first relative distance information set including the relative distance information between the first intersection point An11 of the real lane marking 1 and the first intersection point An12 of the real lane marking 2 collected at the same time, the relative distance information between the first intersection point An21 of the real lane marking 1 and the first intersection point An22 of the real lane marking 2 collected at the same time, and the relative distance information between the first intersection point An31 of the real lane marking 1 and the first intersection point An32 of the real lane marking 2 collected at the same time.

[0109] Afterwards, the server uses the average value of the relative distance information in the first relative distance information set as the first relative distance information from the position point of the real lane marking 1 to the position point of the real lane marking 2 under the road cross section A. Then, the sum of the first relative distance information and the absolute distance information of the position point of the real lane marking 1 under the road cross section A is used as the absolute distance information of the position point of the real lane marking 2 under the road cross section A.

[0110] The server can use a similar implementation method to obtain the absolute distance information of the position point of the real lane marking 3 under the road cross section A and the absolute distance information of the position point of the real lane marking 4 under the road cross section A. The server can also Figure 8 A similar method as that used on road cross section A is used on road cross section B and road cross section C to obtain the absolute distance information of the position points of all real lane markings under road cross section B and road cross section C, respectively, which will not be repeated here.

[0111] Then, for example, you can combine Fig. 9 Understand the process of the server obtaining the updated collected lane markings of the road to be updated based on the absolute distance information of each location point, Fig. 9 The updated lane marking collection schematic diagram provided in the embodiment of the present application is as follows: Fig. 9As shown, assuming that the server determines that the position points of the real lane marking 1 of the road to be updated under road cross section A, road cross section B, and road cross section C are point Ak1, point Bk1, and point Ck1, respectively, the server can connect the absolute distance information of the above three points in sequence along the driving direction 901 of the road reference line of the road to be updated to obtain the updated collected lane marking S1 of the real lane marking 1 of the road to be updated.

[0112] like Fig. 9 As shown, assuming that the server determines that the position points of the real lane marking 2 of the road to be updated are respectively Ak2, Bk2, and Ck2 under the road cross section A, the road cross section B, and the road cross section C; determines that the position points of the real lane marking 3 of the road to be updated are respectively Ak3, Bk3, and Ck3 under the road cross section A, the road cross section B, and the road cross section C; determines that the position points of the real lane marking 4 of the road to be updated are respectively Ak4, Bk4, and Ck4 under the road cross section A, the road cross section B, and the road cross section C. Then, a method similar to the real lane marking 1 can be used to obtain the updated collected lane marking S2 of the real lane marking 2 of the road to be updated, the updated collected lane marking S3 of the real lane marking 3 of the road to be updated, and the updated collected lane marking S4 of the real lane marking 4 of the road to be updated, which will not be repeated here.

[0113] In a second possible implementation, when the relative distance information of the first intersection is the distance information between the first intersection and other first intersections located under the same collected lane marking line as the first intersection, after determining the absolute distance information of the first intersection and the relative distance information of the first intersection, the server can determine the updated collected lane marking line of each real lane marking line of the road to be updated according to the two information. The specific process may include, for example:

[0114] The server determines, for each real lane marking, a second absolute distance information set of the road cross section under the real lane marking based on the absolute distance information of the first intersection points collected multiple times in each road cross section;

[0115] For each real lane marking, the server uses the average value of each absolute distance information in the second absolute distance information set with the smallest standard deviation in the second absolute distance information set for each road cross section under the real lane marking as the absolute distance information of the position point on the road cross section corresponding to the real lane marking in the second absolute distance information set; or, for each real lane marking, uses the average value of each absolute distance information in the second absolute distance information set with the smallest confidence interval of T distribution in the second absolute distance information set for each road cross section under the real lane marking as the absolute distance information of the position point on the road cross section corresponding to the real lane marking in the second absolute distance information set;

[0116] The server determines, for each real lane marking, relative distance information collected multiple times from each first intersection in the road cross section corresponding to the absolute distance information of the existing position point to each first intersection in the road cross section corresponding to the absolute distance information of each non-existent position point located under the same collected lane marking as the first intersection, relative distance information corresponding to two first intersections that have the same road cross section as the second relative distance information set;

[0117] For each real lane marking, the server takes the average value of each relative distance information in each second relative distance information set as the second relative distance information corresponding to the second relative distance information set, from the position point under the road cross section corresponding to the absolute distance information of the existing position point to the position point under the road cross section corresponding to the absolute distance information of the non-existent position point.

[0118] For each real lane marking, the server uses the sum of each second relative distance information and the absolute distance information of the position point corresponding to the road cross section corresponding to the lateral absolute distance of the existing position point corresponding to the second relative distance information as the absolute distance information of the position point in the road cross section where the lateral absolute distance of the real lane marking does not exist.

[0119] The server serially connects the absolute distance information of each real lane marking of the road to be updated at each road cross section along the driving direction of the road reference line of the road to be updated to obtain the updated collected lane markings of each real lane marking of the road to be updated.

[0120] For example, you can combine Fig.10 Understand the process of determining the absolute distance information of the position point of the actual lane marking in the above method. Fig.10 Schematic diagram of determining location information of a location point provided in an embodiment of the present application Figure 2 ,in, Fig.10 and Figure 8 The lane markings and the first intersection point are assumed to be the same. For details, please refer to the above description. Figure 8 The description of the assumed collection lane markings and the first intersection point will not be repeated here.

[0121] like Fig.10 As shown, the server can combine the absolute distance information of the first intersection An11, the absolute distance information of the first intersection An21, and the absolute distance information of the first intersection An31 of the real lane marking 1 under the road cross section A to form a second absolute distance information set A1; combine the absolute distance information of the first intersection Bn11, the absolute distance information of the first intersection Bn21, and the absolute distance information of the first intersection Bn31 of the real lane marking 1 under the road cross section B to form a second absolute distance information set B1; combine the absolute distance information of the first intersection Cn11, the absolute distance information of the first intersection Cn21, and the absolute distance information of the first intersection Cn31 of the real lane marking 1 under the road cross section C to form a second absolute distance information set C1.

[0122] Under the real lane marking 1, assuming that among the second absolute distance information set A1, the second absolute distance information set B1, and the second absolute distance information set C1, the standard deviation of the second absolute distance information set B1 is the smallest, then the average of the absolute distance information of the first intersection Bn11, the absolute distance information of the first intersection Bn21, and the absolute distance information of the first intersection Bn31 in the second absolute distance information set B1 is used as the absolute distance information of the position point of the real lane marking 1 under the road cross section B. Alternatively, under the real lane marking 1, assuming that among the second absolute distance information set A1, the second absolute distance information set B1, and the second absolute distance information set C1, the confidence interval of the T distribution of the second absolute distance information set B1 is the smallest, then the average of the absolute distance information of the first intersection Bn11, the absolute distance information of the first intersection Bn21, and the absolute distance information of the first intersection Bn31 in the second absolute distance information set B1 is used as the absolute distance information of the position point of the real lane marking 1 under the road cross section B.

[0123] Since the standard deviation of the second absolute distance information set is used to indicate the degree of discreteness of the distribution of the absolute distance information in the set, a smaller standard deviation indicates that the distribution of the absolute distance information collected multiple times in the set is more concentrated, that is, the absolute distance information collected multiple times in the set is closer to the real absolute distance information. Therefore, under each real lane marking line, the average value of the absolute distance information in the second absolute distance information set with the smallest standard deviation is closer to the real absolute distance information. It can also be understood that at this time, under the real lane marking line, the absolute distance information has a higher accuracy.

[0124] Since the confidence interval of the T distribution of the second absolute distance information set is used to indicate the credibility of the absolute distance information in the set, the smaller the confidence interval of the T distribution is, the closer the absolute distance information collected multiple times in the set is to the true absolute distance information. Therefore, under each real lane marking line, the average value of each absolute distance information in the second absolute distance information set with the smallest confidence interval of the T distribution is closest to the true absolute distance information. It can also be understood that at this time, under the real lane marking line, the absolute distance information has a higher accuracy.

[0125] After obtaining the absolute distance information of the position point of the real lane marking 1 under the road cross-section B, the server can form a second relative distance information set by including the relative distance information between the first intersection Bn11 under the road cross-section B and the first intersection An11 under the road cross-section A collected at the same time, the relative distance information between the first intersection Bn21 under the road cross-section B and the first intersection An21 under the road cross-section A collected at the same time, and the relative distance information between the first intersection Bn31 under the road cross-section B and the first intersection An31 under the road cross-section A collected at the same time.

[0126] Afterwards, the server uses the average value of the relative distance information in the second relative distance information set as the second relative distance information from the position point under the road cross section B to the position point under the road cross section A under the real lane marking 1. Then, the sum of the second relative distance information and the absolute distance information of the position point of the real lane marking 1 under the road cross section B is used as the absolute distance information of the position point of the real lane marking 1 under the road cross section A.

[0127] The server can use a similar method to obtain the absolute distance information of the position point of the real lane marking 1 under the road cross section C. The server can also use a similar method on the real lane marking 1 on the real lane marking 2, the real lane marking 3, and the real lane marking 4 to obtain the absolute distance information of all the position points of the real lane marking 2, the real lane marking 3, and the real lane marking 4, which will not be repeated here.

[0128] Afterwards, the server can connect the absolute distance information of each real lane marking of the road to be updated at each road cross section in sequence along the driving direction of the road reference line of the road to be updated to obtain the updated collected lane markings of each real lane marking of the road to be updated. Fig. 9 The implementation method of the described examples is the same, and the specific implementation method can be referred to the above Fig. 9 The description is not repeated here.

[0129] In the first method described above, for each road cross section, the absolute distance information of the position point determined by the average value of each absolute distance information in the first absolute distance information set selected by the minimum standard deviation or the minimum T-distribution confidence interval has a high accuracy. The first relative distance information of the position point determined by the average value of each relative distance information in the first relative distance information set of two first intersections collected multiple times at the same time also has a high accuracy. Furthermore, based on the absolute distance information of the position point with higher accuracy and the first relative distance information of the position point with higher accuracy, the absolute distance information of the position point of each real lane marking at each road cross section is also determined to have a high accuracy. Therefore, based on the absolute distance information of the position point of each real lane marking at each road cross section with higher accuracy, the updated collected lane marking is also determined to have a high accuracy.

[0130] In the second method described above, for each real lane marking, the absolute distance information of the position point determined by the average value of each absolute distance information in the second absolute distance information set selected by the minimum standard deviation or the minimum T-distribution confidence interval has a high accuracy. The second relative distance information of the position point determined by taking the average value of each relative distance information in the second relative distance information set of two first intersections collected at the same time under two road cross sections collected multiple times also has a high accuracy. Furthermore, based on the absolute distance information of the position point with higher accuracy and the second relative distance information of the position point with higher accuracy, the absolute distance information of the position point of each real lane marking under each road cross section determined is also high in accuracy. Therefore, based on the absolute distance information of the position point of each real lane marking under each road cross section with higher accuracy, the updated collected lane marking determined is also high in accuracy.

[0131] S106. Update the high-precision map according to the updated collected lane markings of each real lane marking.

[0132] For example, after obtaining the updated collected lane markings of each real lane marking, the server may update the high-precision map according to the following process, for example:

[0133] The server converts the coordinate information of the position point of the updated collected lane marking of each real lane marking of the road to be updated into the coordinate information corresponding to the coordinate system of the high-precision map, and obtains the converted collected lane marking of the road to be updated based on the coordinate information.

[0134] The server obtains lane markings in the high-precision map of the road to be updated, wherein the lane markings in the high-precision map include lane marking identifiers in the high-precision map. And obtains lane marking identifiers in the high-precision map of the converted collected lane markings of the road to be updated.

[0135] If, for the road to be updated, the lane marking symbol in the high-precision map of the converted collected lane marking exists in the lane marking in the high-precision map, and the distance between the converted collected lane marking corresponding to the lane marking symbol and the lane marking in the high-precision map is greater than a preset threshold, the server deletes the lane marking in the high-precision map corresponding to the lane marking symbol in the high-precision map, and adds a new lane marking in the high-precision map based on the converted collected lane marking.

[0136] If, for the road to be updated, the lane marking logo in the high-precision map of the converted collected lane marking exists in the lane marking in the high-precision map, and the distance between the converted collected lane marking corresponding to the lane marking logo and the lane marking in the high-precision map is less than or equal to a preset threshold, the server updates the lane marking in the high-precision map corresponding to the lane marking logo according to the coordinate information of the converted collected lane marking in the high-precision map.

[0137] If the lane marking in the high-precision map of the converted collected lane marking of the road to be updated does not exist in the lane marking in the high-precision map, the server adds a lane marking in the high-precision map in the high-precision map according to the converted collected lane marking.

[0138] For example, referring to step S105 Fig. 9 In the example of , after the server obtains the updated collected lane markings S1, the updated collected lane markings S2, the updated collected lane markings S3, and the updated collected lane markings S4 of the road to be updated, it converts the coordinate information of the position points of these updated collected lane markings into the coordinate information corresponding to the coordinate system of the high-precision map, and obtains the converted collected lane markings S1, the converted collected lane markings S2, the converted collected lane markings S3, and the converted collected lane markings S4 of the road to be updated based on the coordinate information.

[0139] Assuming that the lane marking icon in the high-precision map of the converted collected lane marking S1 exists in the lane marking in the high-precision map, and the distance between the converted collected lane marking corresponding to the lane marking icon and the lane marking in the high-precision map is greater than a preset threshold, the server deletes the lane marking in the high-precision map corresponding to the lane marking icon in the high-precision map, and adds a lane marking in the high-precision map according to the converted collected lane marking S1.

[0140] Assuming that the lane marking identifier in the high-precision map of the converted collected lane marking S2 exists in the lane marking in the high-precision map, and the distance between the converted collected lane marking corresponding to the lane marking identifier and the lane marking in the high-precision map is less than or equal to a preset threshold, the server updates the lane marking in the high-precision map corresponding to the lane marking identifier according to the coordinate information of the converted collected lane marking S2 in the high-precision map.

[0141] Assuming that the lane marking mark in the high-precision map in the converted collected lane marking S3 does not exist in the lane marking in the high-precision map, the server adds a lane marking in the high-precision map in the high-precision map according to the converted collected lane marking S3.

[0142] The lane marking-based map updating method provided in the embodiment of the present application first obtains the road reference line of the road to be updated and the marking combination of the road to be updated at each acquisition number, wherein the marking combination of the road to be updated at each acquisition number includes the collected lane marking obtained when each real lane marking of the road to be updated is collected. Then, for the road to be updated, each road cross section of the road to be updated is established according to the road reference line of the road to be updated; wherein each collected lane marking has a plurality of first intersection points between the collected lane marking and the road cross section, and the road reference line has a plurality of second intersection points between the road reference line and the road cross section.

[0143] Then, for each first intersection of each collected lane marking, determine the absolute distance information of the first intersection; wherein the absolute distance information is the distance information between the second intersection located in the same road cross section as the first intersection and the first intersection. At the same time, for each first intersection of each collected lane marking, determine the relative distance information of the first intersection; wherein the relative distance information is the distance information between the first intersection and other first intersections located in the same road cross section as the first intersection during the same collection; or, the relative distance information is the distance information between the first intersection and other first intersections located under the same collected lane marking as the first intersection. Furthermore, based on the absolute distance information of the first intersection and the relative distance information of the first intersection, determine the updated collected lane marking of each real lane marking of the road to be updated; and update the high-precision map based on the updated collected lane marking of each real lane marking.

[0144] Since the second intersection is a point on the road reference line in the high-precision map, it has a higher accuracy. The absolute distance information of the first intersection calculated based on the second intersection with higher accuracy is also highly accurate. Because the two first intersections collected at the same time have the same source, for example, they are both extracted from the same photo, the relative positions of the two first intersections collected at the same time are relatively accurate. Therefore, the relative distance information of the first intersection determined based on the two first intersections collected at the same time is also highly accurate. Afterwards, based on the absolute distance information of the first intersection with higher accuracy and the relative distance information of the second intersection with higher accuracy, the absolute distance information of the position point of the updated collected lane marking of each real lane marking of the road to be updated is also highly accurate. Furthermore, the updated collected lane markings can be accurately determined, and the high-precision map can be accurately updated.

[0145] Fig.11 A flowchart of another lane marking-based map updating method provided in an embodiment of the present application is shown in FIG. Fig.11 As shown, in this embodiment Figure 1 Based on the embodiment, a map updating method based on lane markings is described in detail, and the method includes:

[0146] S1101. Obtain road information of the road to be updated; wherein the road information includes a road reference line of the road to be updated and a collected lane marking of the road to be updated at each collection number, wherein the collected lane marking is a collected lane marking obtained when each real lane marking of the road to be updated is collected.

[0147] Exemplarily, the execution subject of this embodiment can be a server, or a terminal device, or an electronic device, or a map update device or device based on lane markings, or other devices or equipment that can implement the solution of this application, without limitation. Among them, the server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server can also be a server of a distributed system, or a server combined with blockchain.

[0148] This embodiment is introduced with the execution subject being a server.

[0149] In reality, there may be many roads to be updated. For each road to be updated, the server may execute the solution of this embodiment.

[0150] The implementation methods of step 1101 and step S101 are similar. For detailed introduction, please refer to the description of step S101, which will not be repeated here.

[0151] S1102. For the road to be updated, establish each road cross section of the road to be updated according to the road reference line of the road to be updated; wherein each collected lane marking has a plurality of first intersection points, and the first intersection point is the intersection point between the collected lane marking line and the road cross section; the road reference line has a plurality of second intersection points, and the second intersection point is the intersection point between the road reference line and the road cross section.

[0152] The implementation methods of step 1102 and step S102 are similar. For detailed description, please refer to the description of step S102, which will not be repeated here.

[0153] S1103. For each first intersection of each collected lane marking, determine the absolute distance information of the first intersection; wherein the absolute distance information includes: a lateral absolute distance, an elevation absolute distance, and a longitudinal absolute distance.

[0154] In one example, the lateral absolute distance represents the lateral distance information between a first intersection and a second intersection located in the same road cross-section as the first intersection; the elevation absolute distance represents the elevation distance information between the first intersection and the second intersection located in the same road cross-section as the first intersection; the longitudinal absolute distance represents the distance information between the starting point of the road reference line and the second intersection located in the same road cross-section as the first intersection.

[0155] In one example, step S1103 includes the following steps:

[0156] The first step of step S1103: for each first intersection of each collected lane marking, if it is determined that the first intersection is located on the left side of the second intersection in the same road cross section in the driving direction along the road reference line, a negative value is assigned to the absolute value of the distance between the second intersection in the same road cross section as the first intersection and the first intersection to obtain the lateral absolute distance of the first intersection.

[0157] The second step of step S1103: for each first intersection of each collected lane marking, if it is determined that the first intersection is located to the right of the second intersection in the same road cross section in the driving direction along the road reference line, then the numerical value represented by the absolute value of the distance between the second intersection in the same road cross section as the first intersection and the first intersection is used as the lateral absolute distance of the first intersection.

[0158] The third step of step S1103: for each first intersection of each collected lane marking, the difference between the elevation coordinate value of the first intersection and the elevation coordinate value of the second intersection located in the same road cross section as the first intersection is used as the absolute elevation distance of the first intersection. The coordinate system in which the elevation coordinate value is located is the coordinate system used to establish the high-precision map.

[0159] The fourth step of step S1103: for each first intersection of each collected lane marking, the value represented by the length between the starting point of the road reference line and the second intersection located in the same road cross section as the first intersection on the road reference line is used as the longitudinal absolute distance of the first intersection.

[0160] For example, after establishing each road cross section of the road to be updated, the server may determine the absolute distance information of the first intersection for each first intersection of each collected lane marking.

[0161] The lateral absolute distance of the first intersection, for example, can be understood as the distance of the first intersection relative to its corresponding road reference line (i.e., the second intersection) in the road cross section where it is located. Since in the same road cross section, when the second intersection is at the same distance from the first intersections of two different real lane markings, but along the driving direction of the corresponding road reference line, the two first intersections are located on the left and right of the second intersection respectively, the positions of the two first intersections are different, therefore, it is necessary to distinguish the lateral absolute distances of the two first intersections so as to accurately express the lateral absolute distances of the two first intersections.

[0162] Specifically, if the server determines that the first intersection is located to the left of the second intersection in the same road cross-section in the driving direction along the road reference line, the lateral absolute distance of the first intersection can be assigned a negative value to the absolute value of the distance between the second intersection located in the same road cross-section as the first intersection and the first intersection.

[0163] If the server determines that the first intersection is located to the right of the second intersection in the same road cross-section in the driving direction along the road reference line, then the lateral absolute distance of the first intersection can be the absolute value of the distance between the second intersection located in the same road cross-section as the first intersection and the first intersection.

[0164] For example, you can combine Fig.12 To understand the absolute horizontal distance, such as Fig.12 As shown, assuming that the absolute value of the distance between the first intersection An11 and the second intersection ALn is dh11. Along the driving direction 1201 of the road reference line L1, the first intersection An11 is located on the left side of the second intersection ALn, and the server uses -dh11 as the lateral absolute distance of the first intersection An11. Assuming that the absolute value of the distance between the first intersection An13 and the second intersection ALn is dh13, along the driving direction 1201 of the road reference line L1, the first intersection An13 is located on the right side of the second intersection ALn, and the server uses dh13 as the lateral absolute distance of the first intersection An13.

[0165] The absolute distance of the elevation of the first intersection can be understood as, for example, the distance of the elevation of the first intersection relative to the elevation of the second intersection located in the same road cross section as the first intersection. Specifically, the server can use the difference between the elevation coordinate value of the first intersection and the elevation coordinate value of the second intersection located in the same road cross section as the first intersection as the absolute distance of the elevation of the first intersection. The coordinate system in which the elevation coordinate value is located is the coordinate system used to establish the high-precision map.

[0166] For example, you can combine Fig.13 To understand the absolute distance of elevation, Fig.13 The elevation absolute distance diagram provided in the embodiment of the present application is as follows: Fig.13 As shown, the X-axis, Y-axis, and Z-axis represent the X-axis, Y-axis, and Z-axis of the coordinate system corresponding to the high-precision map, respectively, where the Z-axis represents the elevation coordinate value. Assuming that the first intersection An11, the first intersection An13, and the second intersection ALn are located in the same road cross section, the elevation coordinate value of the first intersection An11 on the Z axis is z11, the elevation coordinate value of the second intersection ALn on the Z axis is zc, and the absolute value of the distance between z11 and zc on the Z axis is dg11, then the server uses z11-zc=dg11 as the absolute elevation distance of the first intersection An11.

[0167] Assuming that the elevation coordinate value of the first intersection point An13 on the Z axis is z13, and the absolute value of the distance between z13 and zc on the Z axis is dg13, the server uses z13-zc=-dg13 as the absolute elevation distance of the first intersection point An13.

[0168] The longitudinal absolute distance of the first intersection, for example, can be understood as the length of the road cross section where the first intersection is located, from the starting point of the driving direction of the road reference line corresponding to the first intersection, along the driving direction of the road reference line. Specifically, the server can use the value represented by the connection length between the starting point of the driving direction of the road reference line corresponding to the first intersection and the second intersection in the road cross section where the first intersection is located, on the road reference line, as the longitudinal absolute distance of the first intersection. It can be understood that the longitudinal absolute distances of the first intersections in the same road cross section are the same.

[0169] For example, you can combine Fig.12 To understand the absolute vertical distance, Fig.12 A schematic diagram of the absolute longitudinal distance and the absolute lateral distance provided in the embodiment of the present application is shown in FIG. Fig.12As shown, the longitudinal absolute distance between the first intersection An11 of the lane marking M11 and the road cross section A, and the longitudinal absolute distance between the first intersection An13 of the lane marking M13 and the road cross section A, are both the length dz11 on the road reference line between the starting point Ln1 of the driving direction 1201 of the road reference line L1 corresponding to the first intersection and the second intersection ALn of the road cross section A where the first intersection is located.

[0170] By analogy, the server can determine the absolute horizontal distance of each first intersection of each collected lane marking, the absolute elevation distance of each first intersection, and the absolute longitudinal distance of each first intersection.

[0171] Since the second intersection is a point on the road reference line in the high-precision map, it has a higher accuracy. Therefore, the lateral absolute distance, elevation absolute distance, and longitudinal absolute distance of the first intersection calculated based on the higher-precision second intersection are also highly accurate. This facilitates the subsequent accurate determination of the updated collected lane markings of each real lane marking of the road to be updated based on the lateral absolute distance, elevation absolute distance, and longitudinal absolute distance of the first intersection with higher accuracy.

[0172] S1104. For each first intersection of each collected lane marking, determine relative distance information of the first intersection; wherein the relative distance information includes: a lateral relative distance located at the same road cross section and an elevation relative distance located at the same road cross section.

[0173] In an example, the lateral relative distance located in the same road cross-section represents the lateral distance information between the first intersection and other first intersections located in the same road cross-section during the same collection; the elevation relative distance located in the same road cross-section represents the elevation distance information between the first intersection and other first intersections located in the same road cross-section during the same collection.

[0174] In one example, step S1104 includes the following steps:

[0175] The first step of step S1104: for each first intersection of each collected lane marking, if it is determined that in the driving direction along the road reference line, other first intersections that are located in the same road cross-section as the first intersection at the same time of collection are located on the left side of the first intersection, then a negative value is assigned to the absolute value of the distance between the other first intersections that are located in the same road cross-section as the first intersection at the same time of collection, so as to obtain the lateral relative distance from the first intersection to the other first intersections that are located in the same road cross-section as the first intersection at the same time of collection.

[0176] The second step of step S1104: for each first intersection of each collected lane marking, if it is determined that in the driving direction along the road reference line, other first intersections that are located in the same road cross-section as the first intersection at the same time of collection are located on the right side of the first intersection, then a numerical value represented by the absolute value of the distance between the other first intersections that are located in the same road cross-section as the first intersection at the same time of collection and the first intersection is used as the lateral relative distance from the first intersection to the other first intersections that are located in the same road cross-section as the first intersection at the same time of collection.

[0177] The third step of step S1104: for each first intersection of each collected lane marking, the difference between the elevation coordinate values ​​of other first intersections located in the same road cross-section when the first intersection is collected at the same time and the elevation coordinate value of the first intersection is used as the relative elevation distance from the first intersection to the other first intersection located in the same road cross-section when the first intersection is collected at the same time.

[0178] For example, any two first intersections of different real lane markings collected at the same time and located in the same road cross section have a higher accuracy of relative distance information between the two first intersections because they are from the same source, for example, extracted from the same collected photo. For each road cross section and each first intersection of each collected lane marking, the server can determine the relative distance information from the first intersection to each first intersection of different real lane markings collected at the same time. The relative distance information of the first intersection includes the lateral relative distance of the first intersection and the elevation relative distance of the first intersection.

[0179] Specifically, the lateral relative distance of the first intersection, for example, can be understood as the distance of the first intersection, in the road cross section where the first intersection is located, relative to each first intersection collected at the same time and located on different real lane markings from the first intersection.

[0180] Because in the same road cross-section, when the distances from the first intersection to two different first intersections collected at the same time as the first intersection and located on different real lane markings are the same, but along the driving direction of the corresponding road reference line, two different first intersections collected at the same time as the first intersection and located on different real lane markings are respectively located on the left and right side of the first intersection, the positions of these two different first intersections are different. Therefore, it is necessary to distinguish the lateral relative distances from the first intersection to the two different first intersections, so that the lateral relative distance of the first intersection can be accurately expressed.

[0181] If the server determines that, in the driving direction along the road reference line, the first intersection that is located in the same road cross-section as the first intersection at the same time of collection and is located on a different real lane marking from the first intersection is located to the left of the first intersection, then the absolute value of the distance between the first intersection that is located in the same road cross-section as the first intersection at the same time of collection and is located on a different real lane marking from the first intersection and the first intersection is assigned a negative value as the lateral relative distance from the first intersection to the first intersection that is located in the same road cross-section as the first intersection at the same time of collection and is located on a different real lane marking from the first intersection.

[0182] If the server determines that, in the driving direction along the road reference line, the first intersection that is located in the same road cross-section as the first intersection at the same time of collection and is located on a different real lane marking from the first intersection is located on the right side of the first intersection, then the absolute value of the distance between the first intersection that is located in the same road cross-section as the first intersection at the same time of collection and is located on a different real lane marking from the first intersection and the first intersection is used as the lateral relative distance from the first intersection to the first intersection that is located in the same road cross-section as the first intersection at the same time of collection and is located on a different real lane marking from the first intersection.

[0183] In one example, it is assumed that the road to be updated contains real lane markings 1, 2, and 3. For example, Fig.14 Understand the horizontal relative distance, Fig.14 A schematic diagram of the horizontal relative distance provided in the embodiment of the present application is shown in FIG. Fig.14 As shown, the absolute value of the distance between the first intersection An11 between the collected lane marking M11 and the road cross section A for the real lane marking 1 and the first intersection An12 between the collected lane marking M12 and the road cross section A for the real lane marking 2 collected at the same time is dh21. In the driving direction 1401 along the road reference line, the first intersection An11 is located on the left side of the first intersection An12, and the server uses -dh21 as the lateral relative distance from the first intersection An12 to the first intersection An11.

[0184] Assume that the absolute value of the distance between the first intersection An13 and the first intersection An12 between the collected lane marking M13 and the road cross section A collected for the real lane marking 3 is dh23. In the driving direction 1401 along the road reference line, the first intersection An13 is located on the right side of the first intersection An12, and the server uses dh23 as the lateral relative distance from the first intersection An12 to the first intersection An13.

[0185] Then, the server takes the lateral relative distance -dh21 from the first intersection An12 to the first intersection An11 and the lateral relative distance dh23 from the first intersection An12 to the first intersection An13 as the lateral relative distance of the first intersection An12.

[0186] The relative distance of the elevation of the first intersection, for example, can be understood as the distance of the elevation of each first intersection located on a different real lane marking line from the first intersection and located in the same road cross section at the same time of collection, relative to the elevation of the first intersection. Specifically, the server can use the difference between the elevation coordinate value of each first intersection located on a different real lane marking line from the first intersection and located in the same road cross section at the same time of collection and the elevation coordinate value of the first intersection as the relative distance of the elevation from the first intersection to the first intersection located on a different real lane marking line from the first intersection and located in the same road cross section at the same time of collection. The coordinate system in which the elevation coordinate value is located is the coordinate system used to establish the high-precision map.

[0187] In an example, it is assumed that the road to be updated contains real lane markings 1, real lane markings 2, and real lane markings 3. For example, Fig.15 Understand the relative distance of elevation, Fig.15 Schematic diagram of relative distance of elevation provided in the embodiment of the present application Figure 1 ,like Fig.15 As shown, the X-axis, Y-axis, and Z-axis represent the X-axis, Y-axis, and Z-axis of the coordinate system corresponding to the high-precision map, respectively, where the Z-axis represents the elevation coordinate value. Assuming that the first intersection point An11 of the real lane marking 1, the first intersection point An12 of the real lane marking 2, and the first intersection point An13 of the real lane marking 3 collected at the same time are located in the same road cross section, the elevation coordinate value of the first intersection point An11 on the Z axis is z11, the elevation coordinate value of the first intersection point An12 on the Z axis is z12, and the absolute value of the distance between z11 and z12 on the Z axis is dg12, then the server uses z11-z12=dg12 as the relative elevation distance from the first intersection point An12 to the first intersection point An11.

[0188] Assuming that the elevation coordinate value of the first intersection An13 on the Z axis is z13, and the absolute value of the distance between z13 and z12 on the Z axis is dg23, the server uses z13-z12=-dg23 as the relative elevation distance from the first intersection An12 to the first intersection An13.

[0189] Then, the server uses the relative elevation distance dg12 from the first intersection point An12 to the first intersection point An11 and the relative elevation distance −dg23 from the first intersection point An12 to the first intersection point An13 as the relative elevation distance of the first intersection point An12 .

[0190] By analogy, the server can determine the lateral relative distance of each first intersection point of each collected lane marking and the elevation relative distance of the first intersection point.

[0191] For each road cross section, from each first intersection, the lateral relative distance and elevation relative distance of the first intersection to each real lane marking collected at the same time as the first intersection and located at a different location from the first intersection are determined to have high accuracy, so that it is convenient to subsequently determine the absolute distance information of the position point of the real lane marking with high accuracy based on the high-precision lateral relative distance and elevation relative distance. In this way, the position information of the position point of the updated collected lane marking can be accurately determined, and the high-precision map can be accurately updated.

[0192] S1105. For each road cross section, determine a first set of absolute lateral distances of the real lane markings in the road cross section based on the multiple collected absolute lateral distances of the first intersection of each real lane marking.

[0193] Exemplarily, after determining the lateral absolute distance of each first intersection collected multiple times in each road cross-section, the server can group the lateral absolute distances of the first intersection collected multiple times in the road cross-section according to their corresponding real lane markings, and the lateral absolute distances of each first intersection in each group constitute a first lateral absolute distance set.

[0194] For example, you can combine Figure 8 To understand, such as Figure 8 As shown, the server collects the lateral absolute distance of the first intersection An11, the lateral absolute distance of the first intersection An21, and the lateral absolute distance of the first intersection An31 of the real lane marking 1 under the road cross section A for multiple times to form a first lateral absolute distance set A1.

[0195] By analogy, the server determines the first lateral absolute distance set A2 of the real lane marking 2 in the road cross section A, the first lateral absolute distance set A3 of the real lane marking 3 in the road cross section A, and the first lateral absolute distance set A4 of the real lane marking 4 in the road cross section A.

[0196] The server can also Figure 8A similar method as that on road cross section A is used on road cross section B and road cross section C to obtain the first lateral absolute distance sets of all real lane markings under road cross section B and road cross section C, respectively, which will not be repeated here.

[0197] S1106. For each road cross section, the average value of each lateral absolute distance in the first lateral absolute distance set with the smallest standard deviation in the first lateral absolute distance set of each real lane marking line in the road cross section is used as the lateral absolute distance of the position point of the real lane marking line in the road cross section corresponding to the first lateral absolute distance set; or, for each road cross section, the average value of each lateral absolute distance in the first lateral absolute distance set with the smallest confidence interval of T distribution in the first lateral absolute distance set of each real lane marking line in the road cross section is used as the lateral absolute distance of the position point of the real lane marking line in the road cross section corresponding to the first lateral absolute distance set.

[0198] For example, after determining the first lateral absolute distance set of each real lane marking line in the road cross section for each road cross section, the server first selects the first lateral absolute distance set whose lateral absolute distances collected multiple times are closest to the real lateral absolute distance. Then, the average value of each lateral absolute distance in the first lateral absolute distance set is used as the lateral absolute distance of the position point of the real lane marking line in the road cross section corresponding to the first lateral absolute distance set.

[0199] Among them, a first lateral absolute distance set whose lateral absolute distances collected multiple times are closest to the true lateral absolute distance is selected from each first lateral absolute distance set. For example, a set with the smallest standard deviation can be selected from each first lateral absolute distance set, or a set with the smallest confidence interval of the T distribution can be selected from each first lateral absolute distance set (the confidence level used in calculating the T distribution can be determined based on experience, and this embodiment does not impose any restrictions on this).

[0200] For example, you can combine Figure 8For understanding, assuming that the server refers to the example in step S1105 and determines that under road cross section A, among the first lateral absolute distance set A1 of real lane marking 1, the first lateral absolute distance set A2 of real lane marking 2, the first lateral absolute distance set A3 of real lane marking 3, and the first lateral absolute distance set A4 of real lane marking 4, the standard deviation of the first lateral absolute distance set A1 is the smallest, then the average of the lateral absolute distances of the first intersection An11, the first intersection An21, and the first intersection An31 in the first lateral absolute distance set A1 is used as the lateral absolute distance of the position point of the real lane marking 1 under road cross section A. Alternatively, under the road cross section A, assuming that among the first lateral absolute distance set A1, the first lateral absolute distance set A2, the first lateral absolute distance set A3, and the first lateral absolute distance set A4, the average of the lateral absolute distances of the first intersection An11, the lateral absolute distance of the first intersection An21, and the lateral absolute distance of the first intersection An31 in the first lateral absolute distance set A1 is taken as the lateral absolute distance of the position point of the real lane marking 1 under the road cross section A.

[0201] The server can also Figure 8 A similar method as that used on road cross section A is used on road cross section B and road cross section C to obtain the lateral absolute distance of a position point of a real lane marking under road cross section B and road cross section C, respectively, which will not be repeated here.

[0202] Since the standard deviation of the first lateral absolute distance set is used to indicate the degree of discreteness of the lateral absolute distance distribution in the set, a smaller standard deviation indicates that the distribution of the lateral absolute distances collected multiple times in the set is more concentrated, that is, the lateral absolute distances collected multiple times in the set are closer to the true lateral absolute distances. Therefore, at each road cross section, the average value of each lateral absolute distance in the first lateral absolute distance set with the smallest standard deviation is closer to the true lateral absolute distance. It can also be understood that at this time, at this road cross section, the lateral absolute distance accuracy of the position point under the real lane marking corresponding to the second lateral absolute distance set is higher.

[0203] Since the confidence interval of the T distribution of the first lateral absolute distance set is used to indicate the credibility of the lateral absolute distances in the set, the smaller the confidence interval of the T distribution is, the closer the lateral absolute distances collected multiple times in the set are to the true lateral absolute distances. Therefore, under each road cross section, the average value of each lateral absolute distance in the first lateral absolute distance set with the smallest confidence interval of the T distribution is closest to the true lateral absolute distance. It can also be understood that at this time, under this road cross section, the lateral absolute distance accuracy of the position point under the real lane marking corresponding to the second lateral absolute distance set is higher.

[0204] S1107. For each road cross section, determine the absolute lateral distance of the position point of the real lane marking on the road cross section based on the absolute lateral distance of the position point of the real lane marking on the road cross section and the relative lateral distance from each first intersection of the real lane marking to each other first intersection of the real lane marking on the same road cross section during the same data collection.

[0205] In one example, step S1107 includes the following process: for each road cross section, repeat the following steps until the lateral absolute distance of the position point of each real lane marking line under the road cross section is determined:

[0206] For each road cross section, based on multiple acquisitions, from each first intersection point of the real lane marking corresponding to the lateral absolute distance of each existing position point, to the first intersection point of the real lane marking corresponding to each lateral absolute distance of the non-existent position point located in the same road cross section as the first intersection point during the same acquisition, the lateral relative distances of the real lane markings corresponding to the two first intersection points corresponding to the lateral relative distances are the same, and are determined as the first lateral relative distance set;

[0207] For each road cross section, the average value of each lateral relative distance in the first lateral relative distance set with the smallest standard deviation in each first lateral relative distance set is used as the first optimal lateral relative distance from the position point of the real lane marking corresponding to the lateral absolute distance of the existing position point to the position point of the real lane marking corresponding to the lateral absolute distance of the non-existing position point, corresponding to the first lateral relative distance set; or, for each road cross section, the average value of each lateral relative distance in the first lateral relative distance set with the smallest confidence interval of T distribution in each first lateral relative distance set is used as the first optimal lateral relative distance from the position point of the real lane marking corresponding to the lateral absolute distance of the existing position point to the position point of the real lane marking corresponding to the lateral absolute distance of the non-existing position point, corresponding to the first lateral relative distance set;

[0208] For each road cross-section, the sum of the first optimal lateral relative distance and the lateral absolute distance of the position point of the real lane marking corresponding to the lateral absolute distance of the existing position point corresponding to the first optimal lateral relative distance is taken as the lateral absolute distance of the position point of the real lane marking in the road cross-section corresponding to the non-existent position point of the first optimal lateral relative distance.

[0209] Exemplarily, the server determines, at each road cross-section, the absolute lateral distance of a position point of a real lane marking at the road cross-section, and determines the lateral relative distances of each first intersection point of the real lane marking corresponding to the lateral absolute distances of the existing position points collected multiple times to the first intersection point of the real lane marking corresponding to the lateral absolute distances of the non-existent position points collected at the same time as the first intersection point. The following steps may be repeated until the absolute lateral distances of the position points of each real lane marking at the road cross-section are determined.

[0210] The server groups the lateral relative distances from each first intersection of the real lane markings corresponding to the lateral absolute distance of each existing position point collected multiple times under the road cross-section to the first intersection of the real lane markings corresponding to each lateral absolute distance of each non-existent position point collected at the same time as the first intersection according to a pair of real lane markings they correspond to, and the lateral relative distances in each group constitute a first lateral relative distance set.

[0211] Then, the server first selects a first lateral relative distance set in which the lateral relative distances collected multiple times are closest to the real lateral relative distance from each first lateral relative distance set. Then, the average value of each lateral relative distance in the first lateral relative distance set is used as the first optimal lateral relative distance from the position point of the real lane marking corresponding to the lateral absolute distance of the existing position point to the position point of the real lane marking corresponding to the lateral absolute distance of the non-existent position point.

[0212] Among them, a first lateral relative distance set whose lateral relative distances collected multiple times are closest to the true lateral relative distance is selected from each first lateral relative distance set. For example, a set with the smallest standard deviation can be selected from each first lateral relative distance set, or a set with the smallest confidence interval of the T distribution can be selected from each first lateral relative distance set (the confidence level used in calculating the T distribution can be determined based on experience, and this embodiment does not impose any restrictions on this).

[0213] Afterwards, the server can add the first optimal lateral relative distance to the lateral absolute distance of the position point of the real lane marking corresponding to the lateral absolute distance of the existing position point corresponding to the first optimal lateral relative distance, and obtain the lateral absolute distance of the position point of the real lane marking corresponding to the lateral absolute distance of the non-existent position point corresponding to the first optimal lateral relative distance.

[0214] For example, you can combine Figure 8 For understanding, assuming that the server refers to the example in step S1106 to determine the lateral absolute distance between the position points of the real lane marking 1 under the road cross section A, thereafter, the lateral relative distance between the first intersection point An11 of the real lane marking 1 and the first intersection point An12 of the real lane marking 2 collected at the same time, the lateral relative distance between the first intersection point An21 of the real lane marking 1 and the first intersection point An22 of the real lane marking 2 collected at the same time, and the lateral relative distance between the first intersection point An31 of the real lane marking 1 and the first intersection point An32 of the real lane marking 2 collected at the same time under the road cross section A can constitute a first lateral relative distance set.

[0215] At the same time, the server forms a first lateral relative distance set by the lateral relative distance between the first intersection point An11 of the real lane marking 1 and the first intersection point An13 of the real lane marking 3 collected at the same time, the lateral relative distance between the first intersection point An21 of the real lane marking 1 and the first intersection point An23 of the real lane marking 3 collected at the same time, and the lateral relative distance between the first intersection point An31 of the real lane marking 1 and the first intersection point An33 of the real lane marking 3 collected at the same time.

[0216] In addition, the server forms a first lateral relative distance set by the lateral relative distance between the first intersection point An11 of the real lane marking 1 and the first intersection point An14 of the real lane marking 4 collected at the same time, the lateral relative distance between the first intersection point An21 of the real lane marking 1 and the first intersection point An24 of the real lane marking 4 collected at the same time, and the lateral relative distance between the first intersection point An31 of the real lane marking 1 and the first intersection point An34 of the real lane marking 4 collected at the same time.

[0217] Under the road cross section A, assuming that among the above three first lateral relative distance sets, the standard deviation of the first lateral relative distance set corresponding to the first intersection point of the real lane marking 1 to the first intersection point of the real lane marking 3 collected at the same time is the smallest, then the average value of each lateral relative distance in the set is taken as the first optimal lateral relative distance from the position point of the real lane marking 1 to the position point of the real lane marking 3. Then the sum of the first optimal lateral relative distance and the lateral absolute distance of the position point of the real lane marking 1 is taken as the lateral absolute distance of the position point of the real lane marking 3. Alternatively, assuming that among the above three first lateral relative distance sets, the confidence interval of the T distribution of the first lateral relative distance set corresponding to the first intersection point of the real lane marking 1 to the first intersection point of the real lane marking 3 collected at the same time is the smallest, then the average value of each lateral relative distance in the set is taken as the first optimal lateral relative distance from the position point of the real lane marking 1 to the position point of the real lane marking 3. Then the sum of the first optimal lateral relative distance and the lateral absolute distance of the position point of the real lane marking 1 is taken as the lateral absolute distance of the position point of the real lane marking 3.

[0218] Then, under the road cross section A, the server determines the first lateral relative distance set corresponding to the first intersection point of the real lane marking 3 to the first intersection point of the real lane marking 2 collected at the same time, and the first lateral relative distance set corresponding to the first intersection point of the real lane marking 3 to the first intersection point of the real lane marking 4 collected at the same time. The specific implementation method is similar to the above method and will not be repeated here.

[0219] Next, under the road cross section A, among the first lateral relative distance set corresponding to the first intersection point of the real lane marking 3 and the first intersection point of the real lane marking 2 collected at the same time, the first lateral relative distance set corresponding to the first intersection point of the real lane marking 3 and the first intersection point of the real lane marking 4 collected at the same time, the first lateral relative distance set corresponding to the first intersection point of the real lane marking 1 and the first lateral relative distance set corresponding to the first intersection point of the real lane marking 2 collected at the same time, and the first lateral relative distance set corresponding to the first intersection point of the real lane marking 1 and the first lateral relative distance set corresponding to the first intersection point of the real lane marking 4 collected at the same time. Assuming that the standard deviation of the first lateral relative distance set corresponding to the first intersection point of the real lane marking 1 and the first intersection point of the real lane marking 4 collected at the same time is the smallest, the average value of each lateral relative distance in the set is taken as the first optimal lateral relative distance from the position point of the real lane marking 1 to the position point of the real lane marking 4. Then, the sum of the first optimal lateral relative distance and the lateral absolute distance of the position point of the real lane marking 1 is taken as the lateral absolute distance of the position point of the real lane marking 4. Alternatively, assuming that the confidence interval of the T distribution of the first lateral relative distance set corresponding to the first intersection point of the real lane marking 1 to the first intersection point of the real lane marking 4 collected at the same time is the smallest, the average value of each lateral relative distance in the set is used as the first optimal lateral relative distance from the position point of the real lane marking 1 to the position point of the real lane marking 4. Then, the sum of the first optimal lateral relative distance and the lateral absolute distance of the position point of the real lane marking 1 is used as the lateral absolute distance of the position point of the real lane marking 4.

[0220] Afterwards, by analogy, the server can determine the lateral absolute distance of the position point of the actual lane marking 2 under the road cross section A.

[0221] Furthermore, the server can Figure 8 A similar implementation method as that on the road cross section A is used on the road cross sections B and C in the figure to obtain the lateral absolute distances of the position points of all the real lane markings under the road cross sections B and C, respectively, which will not be repeated here.

[0222] Since the first intersection points collected at the same time under the same road cross section have the same source, for example, they are all extracted from the same photo, the lateral relative distance accuracy of the first intersection points collected at the same time is relatively high, and then the average value of each lateral relative distance in the set with the smallest standard deviation or the smallest confidence interval of the T distribution in the lateral relative distance set is used as the lateral relative distance between the position points. Then, the lateral absolute distance is corrected by the lateral relative distance between the position points, so that the lateral absolute distance accuracy of the position points of the real lane markings is relatively high.

[0223] S1108. For each road cross section, determine a first absolute elevation distance set of the real lane marking in the road cross section based on the absolute elevation distances of the first intersection points collected multiple times for each real lane marking.

[0224] Exemplarily, after determining the absolute elevation distance of each first intersection collected multiple times under each road cross-section, the server can group the absolute elevation distances of the first intersection collected multiple times under the road cross-section according to their corresponding real lane markings, and the absolute elevation distances of each first intersection in each group constitute a first absolute elevation distance set.

[0225] For example, you can combine Figure 8 To understand, such as Figure 8 As shown, the server collects the absolute elevation distances of the first intersection An11, the first intersection An21, and the first intersection An31 of the real lane marking 1 under the road cross section A for multiple times to form a first absolute elevation distance set A1.

[0226] By analogy, the server determines the first elevation absolute distance set A2 of the real lane marking 2 at the road cross section A, the first elevation absolute distance set A3 of the real lane marking 3 at the road cross section A, and the first elevation absolute distance set A4 of the real lane marking 4 at the road cross section A.

[0227] The server can also Figure 8 A similar method as that on road cross section A is used on road cross section B and road cross section C to obtain the first elevation absolute distance sets of all real lane markings under road cross section B and road cross section C, respectively, which will not be repeated here.

[0228] S1109. For each road cross section, the average value of the absolute elevation distances in the first absolute elevation distance set with the smallest standard deviation in the first absolute elevation distance set of each real lane marking line in the road cross section is used as the absolute elevation distance of the position point of the real lane marking line in the road cross section corresponding to the first absolute elevation distance set; or, for each road cross section, the average value of the absolute elevation distances in the first absolute elevation distance set with the smallest confidence interval of T distribution in the first absolute elevation distance set of each real lane marking line in the road cross section is used as the absolute elevation distance of the position point of the real lane marking line in the road cross section corresponding to the first absolute elevation distance set.

[0229] For example, after determining the first absolute elevation distance set of each real lane marking line in the road cross section for each road cross section, the server first selects the first absolute elevation distance set whose absolute elevation distances collected multiple times are closest to the real absolute elevation distance. Then, the average value of the absolute elevation distances in the first absolute elevation distance set is used as the absolute elevation distance of the position point of the real lane marking line in the road cross section corresponding to the first absolute elevation distance set.

[0230] Among them, the first absolute elevation distance set whose absolute elevation distances collected multiple times are closest to the true absolute elevation distance is selected from each first absolute elevation distance set. For example, the set with the smallest standard deviation can be selected from each first absolute elevation distance set, or the set with the smallest confidence interval of the T distribution can be selected from each first absolute elevation distance set (the confidence level used in the T distribution calculation can be determined based on experience, and this embodiment does not impose any restrictions on this).

[0231] For example, you can combine Figure 8 For understanding, assuming that the server refers to the example in step S1108 and determines that under road cross section A, among the first elevation absolute distance set A1 of real lane marking 1, the first elevation absolute distance set A2 of real lane marking 2, the first elevation absolute distance set A3 of real lane marking 3, and the first elevation absolute distance set A4 of real lane marking 4, the standard deviation of the first elevation absolute distance set A3 is the smallest, then the average of the absolute elevation distances of the first intersection An13, the first intersection An23, and the first intersection An33 in the first elevation absolute distance set A3 is used as the absolute elevation distance of the position point of the real lane marking 3 under road cross section A. Alternatively, under the road cross section A, assuming that among the first elevation absolute distance set A1, the first elevation absolute distance set A2, the first elevation absolute distance set A3, and the first elevation absolute distance set A4, the average of the absolute elevation distances of the first intersection An13, the absolute elevation distance of the first intersection An23, and the absolute elevation distance of the first intersection An33 in the first elevation absolute distance set A3 is taken as the absolute elevation distance of the position point of the real lane marking 3 under the road cross section A.

[0232] The server can also Figure 8 A similar method as that used on road cross section A is used on road cross section B and road cross section C to obtain the absolute elevation distance of a position point of a real lane marking under road cross section B and road cross section C, respectively, which will not be repeated here.

[0233] Since the standard deviation of the first elevation absolute distance set is used to indicate the degree of discreteness of the distribution of the absolute elevation distances in the set, the smaller the standard deviation, the more concentrated the distribution of the absolute elevation distances collected multiple times in the set. In other words, the absolute elevation distances collected multiple times in the set are closer to the true absolute elevation distances. Therefore, at each road cross section, the average value of the absolute elevation distances in the first elevation absolute distance set with the smallest standard deviation is closer to the true absolute elevation distance. It can also be understood that at this time, at this road cross section, the absolute elevation distance accuracy is higher.

[0234] Since the confidence interval of the T distribution of the first absolute elevation distance set is used to indicate the credibility of the absolute elevation distances in the set, the smaller the confidence interval of the T distribution is, the closer the absolute elevation distances collected multiple times in the set are to the true absolute elevation distances. Therefore, under each road cross section, the average value of the absolute elevation distances in the first absolute elevation distance set with the smallest confidence interval of the T distribution is closest to the true absolute elevation distance. It can also be understood that at this time, under this road cross section, the absolute elevation distance has a higher accuracy.

[0235] S1110. For each road cross section, the absolute elevation distance of the position point of the real lane marking on the road cross section is determined based on the absolute elevation distance of the position point of the real lane marking on the road cross section and the relative elevation distance of each first intersection point of the real lane marking to each other first intersection point located on the same road cross section during the same data collection.

[0236] In one example, step S1110 includes the following process:

[0237] For each road cross section, repeat the following steps until the absolute elevation distance of each real lane marking position point under the road cross section is determined:

[0238] For each road cross section, based on multiple acquisitions, from each first intersection point of the real lane marking corresponding to the absolute elevation distance of each existing position point, to the first intersection point of the real lane marking corresponding to the absolute elevation distance of each non-existent position point located in the same road cross section during the same acquisition, the relative elevation distances of the real lane markings corresponding to the two first intersection points corresponding to the relative elevation distances are the same, and are determined as the first relative elevation distance set;

[0239] For each road cross section, the average value of each elevation relative distance in the first elevation relative distance set with the smallest standard deviation in each first elevation relative distance set is used as the first optimal elevation relative distance from the position point of the real lane marking corresponding to the absolute elevation distance of the existing position point to the position point of the real lane marking corresponding to the absolute elevation distance of the non-existing position point, corresponding to the first elevation relative distance set; or, for each road cross section, the average value of each elevation relative distance in the first elevation relative distance set with the smallest confidence interval of the T distribution in each first elevation relative distance set is used as the first optimal elevation relative distance from the position point of the real lane marking corresponding to the absolute elevation distance of the existing position point to the position point of the real lane marking corresponding to the absolute elevation distance of the non-existing position point, corresponding to the first elevation relative distance set;

[0240] For each road cross-section, the sum of the first optimal elevation relative distance and the absolute elevation distance of the position point of the real lane marking corresponding to the absolute elevation distance of the existing position point corresponding to the first optimal elevation relative distance is taken as the absolute elevation distance of the position point of the real lane marking in the road cross-section corresponding to the first optimal elevation relative distance and the absolute elevation distance of the non-existent position point.

[0241] Exemplarily, the server determines, under each road cross-section, the absolute elevation distance of a position point of a real lane marking under the road cross-section, and determines the relative elevation distances of each first intersection point of the real lane marking corresponding to the absolute elevation distances of the existing position points collected multiple times to the first intersection point of the real lane marking corresponding to the absolute elevation distances of the non-existent position points collected at the same time as the first intersection point. The following steps may be repeated until the absolute elevation distances of the position points of each real lane marking under the road cross-section are determined.

[0242] The server groups the relative elevation distances from each first intersection of the real lane markings corresponding to the absolute elevation distance of each existing position point collected multiple times under the road cross-section to the first intersection of the real lane markings corresponding to the absolute elevation distance of each non-existent position point collected at the same time as the first intersection according to the pair of real lane markings they correspond to, and the relative elevation distances in each group constitute a first relative elevation distance set.

[0243] Then, the server first selects the first relative elevation distance set whose relative elevation distances collected multiple times are closest to the real relative elevation distance from each first relative elevation distance set. Then, the average value of each relative elevation distance in the first relative elevation distance set is used as the first optimal relative elevation distance from the position point of the real lane marking corresponding to the absolute elevation distance of the existing position point to the position point of the real lane marking corresponding to the absolute elevation distance of the non-existent position point.

[0244] Among them, the first elevation relative distance set whose elevation relative distances collected multiple times are closest to the true elevation relative distance is selected from each first elevation relative distance set. For example, the set with the smallest standard deviation can be selected from each first elevation relative distance set, or the set with the smallest confidence interval of the T distribution can be selected from each first elevation relative distance set (the confidence level used in the T distribution calculation can be determined based on experience, and this embodiment does not impose any restrictions on this).

[0245] Afterwards, the server can add the first optimal relative elevation distance to the absolute elevation distance of the real lane marking position point corresponding to the absolute elevation distance of the existing position point corresponding to the first optimal relative elevation distance, and obtain the absolute elevation distance of the real lane marking position point corresponding to the absolute elevation distance of the non-existent position point corresponding to the first optimal relative elevation distance.

[0246] For example, you can combine Figure 8 For understanding, assuming that the server refers to the example in step S1109 to determine the absolute elevation distance of the position point of the real lane marking 3 under the road cross section A, the relative elevation distance between the first intersection point An13 of the real lane marking 3 and the first intersection point An11 of the real lane marking 1 collected at the same time, the relative elevation distance between the first intersection point An23 of the real lane marking 3 and the first intersection point An21 of the real lane marking 1 collected at the same time, and the relative elevation distance between the first intersection point An33 of the real lane marking 3 and the first intersection point An31 of the real lane marking 1 collected at the same time under the road cross section A can constitute a first relative elevation distance set.

[0247] At the same time, the server forms a first relative elevation distance set by the relative elevation distance between the first intersection point An13 of the real lane marking 3 and the first intersection point An12 of the real lane marking 2 collected at the same time, the relative elevation distance between the first intersection point An23 of the real lane marking 3 and the first intersection point An22 of the real lane marking 2 collected at the same time, and the relative elevation distance between the first intersection point An33 of the real lane marking 3 and the first intersection point An32 of the real lane marking 2 collected at the same time under the road cross section A.

[0248] In addition, the server forms a first relative elevation distance set by the relative elevation distance between the first intersection point An13 of the real lane marking 3 and the first intersection point An14 of the real lane marking 4 collected at the same time, the relative elevation distance between the first intersection point An23 of the real lane marking 3 and the first intersection point An24 of the real lane marking 4 collected at the same time, and the relative elevation distance between the first intersection point An33 of the real lane marking 3 and the first intersection point An34 of the real lane marking 4 collected at the same time under the road cross section A.

[0249] Under the road cross section A, assuming that among the above three first relative elevation distance sets, the standard deviation of the first relative elevation distance set corresponding to the first intersection point of the real lane marking 3 to the first intersection point of the real lane marking 2 collected at the same time is the smallest, then the average value of each relative elevation distance in the set is used as the first optimal relative elevation distance from the position point of the real lane marking 3 to the position point of the real lane marking 2. Then, the sum of the first optimal relative elevation distance and the absolute elevation distance of the position point of the real lane marking 3 is used as the absolute elevation distance of the position point of the real lane marking 2. Alternatively, assuming that among the above three first relative elevation distance sets, the confidence interval of the T distribution of the first relative elevation distance set corresponding to the first intersection point of the real lane marking 3 to the first intersection point of the real lane marking 2 collected at the same time is the smallest, then the average value of each relative elevation distance in the set is used as the first optimal relative elevation distance from the position point of the real lane marking 3 to the position point of the real lane marking 2. Then, the sum of the first optimal relative elevation distance and the absolute elevation distance of the position point of the real lane marking line 3 is taken as the absolute elevation distance of the position point of the real lane marking line 2 .

[0250] Then, under the road cross section A, the server determines the first elevation relative distance set corresponding to the first intersection point of the real lane marking 2 to the first intersection point of the real lane marking 1 collected at the same time, and the first elevation relative distance set corresponding to the first intersection point of the real lane marking 2 to the first intersection point of the real lane marking 4 collected at the same time. The specific implementation method is similar to the above method and will not be repeated here.

[0251] Next, under the road cross section A, in the first relative elevation distance set corresponding to the first intersection point of the real lane marking 2 and the first intersection point of the real lane marking 1 collected at the same time, the first relative elevation distance set corresponding to the first intersection point of the real lane marking 2 and the first intersection point of the real lane marking 4 collected at the same time, the first relative elevation distance set corresponding to the first intersection point of the real lane marking 3 and the first relative elevation distance set corresponding to the first intersection point of the real lane marking 1 collected at the same time. Assuming that the standard deviation of the first relative elevation distance set corresponding to the first intersection point of the real lane marking 2 and the real lane marking 1 collected at the same time is the smallest, the average value of each relative elevation distance in the set is taken as the first optimal relative elevation distance from the position point of the real lane marking 2 to the position point of the real lane marking 1. Then, the sum of the first optimal relative elevation distance and the absolute elevation distance of the position point of the real lane marking 2 is taken as the absolute elevation distance of the position point of the real lane marking 1. Alternatively, assuming that the confidence interval of the T distribution of the first relative elevation distance set corresponding to the first intersection point of the real lane marking 2 to the first intersection point of the real lane marking 1 collected at the same time is the smallest, the average value of each relative elevation distance in the set is used as the first optimal relative elevation distance from the position point of the real lane marking 2 to the position point of the real lane marking 1. Then, the sum of the first optimal relative elevation distance and the absolute elevation distance of the position point of the real lane marking 2 is used as the absolute elevation distance of the position point of the real lane marking 1.

[0252] Afterwards, by analogy, the server can determine the absolute elevation distance of the position point of the actual lane marking 4 under the road cross section A.

[0253] Furthermore, the server can Figure 8 A similar implementation method as that on the road cross section A is used on the road cross sections B and C in the figure to obtain the absolute elevation distances of the position points of all the real lane markings under the road cross sections B and C, respectively, which will not be repeated here.

[0254] Since the first intersection points under the same road cross section have the same source, for example, they are all extracted from the same photo, and thus the relative distance accuracy of the first intersection points collected at the same time is relatively high, then the average value of each relative distance in the set with the smallest standard deviation or the smallest confidence interval of the T distribution in the relative distance set is taken as the relative distance between the position points. Then, the relative distance between the position points is used to correct the absolute distance, so that the absolute distance accuracy of the position points of the real lane marking is relatively high.

[0255] It is worth noting that, since the absolute horizontal distance and absolute elevation distance of the real lane marking position point are calculated independently, the relative horizontal distance and relative elevation distance of the real lane marking position point are also calculated independently. Therefore, under the same road cross section, the order of the real lane marking corresponding to the absolute horizontal distance of the position point determined by the server and the order of the real lane marking corresponding to the absolute elevation distance of the position point are not necessarily the same. The purpose of this processing is to ensure that the absolute horizontal distance and absolute elevation distance of the position point do not affect each other when calculating the absolute horizontal distance and absolute elevation distance of the position point, so as to make their respective accuracy higher.

[0256] S1111. Use the longitudinal absolute distance in the absolute distance information of any first intersection in each road cross section as the longitudinal absolute distance of the position point of each real lane marking in the road cross section.

[0257] Exemplarily, the longitudinal absolute distance of the position point of the real lane marking is the length of the road cross section corresponding to the position point from the starting point of the driving direction of the road reference line corresponding to the position point on the road reference line along the driving direction of the road reference line. The longitudinal absolute distance of the first intersection is the length of the road cross section where the first intersection is located from the starting point of the driving direction of the road reference line corresponding to the first intersection on the road reference line along the driving direction of the road reference line. Furthermore, under the same road cross section, the longitudinal absolute distance of each first intersection and the longitudinal absolute distance of each position point of the real lane marking are the same.

[0258] Therefore, the server may use the longitudinal absolute distance in the absolute distance information of any first intersection in each road cross section as the longitudinal absolute distance of the position point of each real lane marking in the road cross section.

[0259] For example, assuming that the longitudinal absolute distance in the absolute distance information of the first intersection An11 is dz11, the longitudinal absolute distance of each position point of the real lane marking corresponding to the same road cross section as the first intersection An11 is dz11.

[0260] S1112. Use the absolute lateral distance, absolute elevation distance, and absolute longitudinal distance of each real lane marking position point in each road cross section as the position information of each real lane marking position point in each road cross section of the road to be updated.

[0261] For example, after the above steps, the server determines that the absolute horizontal distance, absolute vertical distance, and absolute vertical distance of the position point of each real lane marking on the road to be updated are also separate distance information, without forming a point. Therefore, the server can form the position information of the position point of the real lane marking on the road to be updated in each road cross section by the absolute horizontal distance, absolute vertical distance, and absolute vertical distance of the position point of each real lane marking on the road to be updated in each road cross section.

[0262] For example, assuming that the server determines that the absolute lateral distance of the position point of the real lane marking 1 under the road cross-section A is H1, the absolute longitudinal distance is Z1, and the absolute elevation distance is G1, then (H1, Z1, G1) can be used as the position information of the position point of the real lane marking 1 under the road cross-section A.

[0263] S1113. Determine an updated collected lane marking of each real lane marking of the road to be updated according to the position information of the position point of each real lane marking of the road to be updated in each road cross section.

[0264] Exemplarily, after determining the position information of each real lane marking of the road to be updated at each road cross section, the server can sequentially connect them along the driving direction of the road reference line of the road to be updated to obtain the updated collected lane markings of each real lane marking of the road to be updated.

[0265] For example, you can combine Fig. 9 Understanding, such as Fig. 9 As shown, assuming that the server determines that the position points of the real lane marking 1 of the road to be updated under road cross section A, road cross section B, and road cross section C are point Ak1, point Bk1, and point Ck1, respectively, the server can connect the position information of the above three points in sequence along the driving direction 901 of the road reference line of the road to be updated to obtain the updated collected lane marking S1 of the real lane marking 1 of the road to be updated.

[0266] like Fig. 9As shown, assuming that the server determines that the position points of the real lane marking 2 of the road to be updated are respectively Ak2, Bk2, and Ck2 under the road cross section A, the road cross section B, and the road cross section C; determines that the position points of the real lane marking 3 of the road to be updated are respectively Ak3, Bk3, and Ck3 under the road cross section A, the road cross section B, and the road cross section C; determines that the position points of the real lane marking 4 of the road to be updated are respectively Ak4, Bk4, and Ck4 under the road cross section A, the road cross section B, and the road cross section C. Then, a method similar to the real lane marking 1 can be used to obtain the updated collected lane marking S2 of the real lane marking 2 of the road to be updated, the updated collected lane marking S3 of the real lane marking 3 of the road to be updated, and the updated collected lane marking S4 of the real lane marking 4 of the road to be updated, which will not be repeated here.

[0267] S1114. Update the high-precision map according to the updated collected lane markings of each real lane marking.

[0268] The implementation methods of step S1114 and step S106 are similar. For detailed introduction, please refer to the description of step S106, which will not be repeated here.

[0269] The lane marking-based map updating method provided in the embodiment of the present application has a high accuracy of the lateral relative distances of the first intersections collected at the same time under the same road cross section, because the sources are the same, for example, they are all extracted from the same photo. Then, the average value of each lateral relative distance in the set with the smallest standard deviation in the lateral relative distance set or the smallest confidence interval of the T distribution is used as the lateral relative distance between the position points. Then, the lateral absolute distance is corrected by the lateral relative distance between the position points, so that the lateral absolute distance accuracy of the position points of the real lane markings is high. Similarly, the absolute distance accuracy of the elevation of the position points of the real lane markings determined in a similar manner is also high.

[0270] Since the lateral absolute distance and elevation absolute distance of the location point of the real lane marking are calculated independently, the lateral relative distance and elevation relative distance of the location point of the real lane marking are also calculated independently. This processing can ensure that when calculating the lateral absolute distance and elevation absolute distance of the location point, the two will not affect each other, thereby making the calculated accuracy of each higher.

[0271] Furthermore, based on the horizontal absolute distance, elevation absolute distance, and longitudinal absolute distance of the position points of the real lane markings with high accuracy, the position information of the position points of the real lane markings is also determined with high accuracy. Thus, the updated collected lane markings can be accurately determined, and the high-precision map can be accurately updated.

[0272] Fig.16 A flowchart of another lane marking-based map updating method provided in an embodiment of the present application is shown in FIG. Fig.16 As shown, in this embodiment Figure 1 Based on the embodiment, a map updating method based on lane markings is described in detail, and the method includes:

[0273] S1601. Obtain road information of the road to be updated; wherein the road information includes a road reference line of the road to be updated and a collected lane marking of the road to be updated at each collection number, wherein the collected lane marking is a collected lane marking obtained when each real lane marking of the road to be updated is collected.

[0274] Exemplarily, the execution subject of this embodiment can be a server, or a terminal device, or an electronic device, or a map update device or device based on lane markings, or other devices or equipment that can implement the solution of this application, without limitation. Among them, the server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server can also be a server of a distributed system, or a server combined with blockchain.

[0275] This embodiment is introduced with the execution subject being a server.

[0276] In reality, there may be many roads to be updated. For each road to be updated, the server may execute the solution of this embodiment.

[0277] The implementation of step 1601 is similar to that of step S101. For detailed description, please refer to the description of step S101, which will not be repeated here.

[0278] S1602. For the road to be updated, establish each road cross section of the road to be updated according to the road reference line of the road to be updated; wherein each collected lane marking has a plurality of first intersection points, and the first intersection point is the intersection point between the collected lane marking line and the road cross section; the road reference line has a plurality of second intersection points, and the second intersection point is the intersection point between the road reference line and the road cross section.

[0279] The implementation of step 1602 is similar to that of step S102. For detailed description, please refer to the description of step S102, which will not be repeated here.

[0280] S1603. For each first intersection of each collected lane marking, determine the absolute distance information of the first intersection; wherein the absolute distance information includes: a lateral absolute distance, an elevation absolute distance, and a longitudinal absolute distance.

[0281] In one example, the lateral absolute distance represents the lateral distance information between a first intersection and a second intersection located in the same road cross-section as the first intersection; the elevation absolute distance represents the elevation distance information between the first intersection and the second intersection located in the same road cross-section as the first intersection; the longitudinal absolute distance represents the distance information between the starting point of the road reference line and the second intersection located in the same road cross-section as the first intersection.

[0282] In one example, step S1603 includes the following steps:

[0283] The first step of step S1603: for each first intersection of each collected lane marking, if it is determined that the first intersection is located on the left side of the second intersection in the same road cross section in the driving direction along the road reference line, a negative value is assigned to the absolute value of the distance between the second intersection in the same road cross section as the first intersection and the first intersection to obtain the lateral absolute distance of the first intersection.

[0284] The second step of step S1603: for each first intersection of each collected lane marking, if it is determined that the first intersection is located to the right of the second intersection in the same road cross section in the driving direction along the road reference line, then the numerical value represented by the absolute value of the distance between the second intersection in the same road cross section as the first intersection and the first intersection is used as the lateral absolute distance of the first intersection.

[0285] The third step of step S1603: for each first intersection of each collected lane marking, the difference between the elevation coordinate value of the first intersection and the elevation coordinate value of the second intersection located in the same road cross section as the first intersection is used as the absolute elevation distance of the first intersection.

[0286] The fourth step of step S1603: for each first intersection of each collected lane marking, the value represented by the length between the starting point of the road reference line and the second intersection located on the same road cross section as the first intersection on the road reference line is used as the longitudinal absolute distance of the first intersection.

[0287] The implementation methods of step 1603 and step S1103 are similar. For detailed introduction, please refer to the description of step S1103, which will not be repeated here.

[0288] S1604. For each first intersection of each collected lane marking, determine relative distance information of the first intersection; wherein the relative distance information includes: a lateral relative distance under the same collected lane marking and an elevation relative distance under the same collected lane marking.

[0289] In one example, the lateral relative distance under the same acquisition lane marking line represents the lateral distance information between the first intersection and other first intersections under the same acquisition lane marking line; the elevation relative distance under the same acquisition lane marking line represents the elevation distance information between the first intersection and other first intersections under the same acquisition lane marking line.

[0290] In one example, step S1604 includes the following steps:

[0291] The first step of step S1604: for each first intersection of each collected lane marking, the difference between the lateral absolute distances to other first intersections located under the same collected lane marking as the first intersection and the lateral absolute distance to the first intersection is used as the lateral relative distance from the first intersection to the other first intersections located under the same collected lane marking as the first intersection.

[0292] The second step of step S1604: for each first intersection of each collected lane marking, the difference between the elevation coordinate values ​​of other first intersections located under the same collected lane marking as the first intersection and the elevation coordinate value of the first intersection is used as the relative elevation distance from the first intersection to the other first intersections located under the same collected lane marking as the first intersection.

[0293] For example, any two first intersections under the same real lane marking and located in different road cross sections under the same collected lane marking have a higher accuracy of relative distance information between the two first intersections because they have the same source, for example, they are extracted from the same collected photo. The server can determine, for each real lane marking and each first intersection of each collected lane marking, the relative distance information from the first intersection to each first intersection under the same collected lane marking and located in a different road cross section from the first intersection. The relative distance information of the first intersection includes the lateral relative distance of the first intersection and the elevation relative distance of the first intersection.

[0294] Specifically, the lateral relative distance of the first intersection can be understood as, for example, the lateral absolute distance of each first intersection located at a different road cross section from the first intersection under the same acquisition lane marking as the first intersection, relative to the lateral absolute distance of the first intersection. Specifically, the server can use the difference between the lateral absolute distances of each first intersection located at a different road cross section from the first intersection under the same acquisition lane marking as the first intersection and the lateral absolute distance of the first intersection as the lateral relative distance from the first intersection to the first intersection located at a different road cross section from the first intersection under the same acquisition lane marking as the first intersection.

[0295] For example, you can combine Figure 7 Understand. Figure 7 As shown, assuming that the lateral absolute distance of the first intersection An13 is dha13, the lateral absolute distance of the first intersection Bn13 is dhb13, and the lateral absolute distance of the first intersection Cn13 is dhc13. For the first intersection An13, the first intersections located under the same collection lane marking M13 as the first intersection An13 and located in different road cross sections from the first intersection An13 are the first intersection Bn13 and the first intersection Cn13, then the server can use the difference between the lateral absolute distance of the first intersection Bn13 and the lateral absolute distance of the first intersection An13, that is, the value represented by the result of dhb13-dha13 as the lateral relative distance from the first intersection An13 to the first intersection Bn13.

[0296] At the same time, the server may use the difference between the lateral absolute distance of the first intersection Cn13 and the lateral absolute distance of the first intersection An13, that is, the value represented by the result of dhc13-dha13 as the lateral relative distance from the first intersection An13 to the first intersection Cn13.

[0297] The lateral relative distance from the first intersection An13 to the first intersection Bn13 and the lateral relative distance from the first intersection An13 to the first intersection Cn13 are referred to as the lateral relative distance of the first intersection An13.

[0298] The relative distance of the elevation of the first intersection can be understood as, for example, the distance of the elevation of each first intersection located at a different road cross section from the first intersection under the same acquisition lane marking as the first intersection relative to the elevation of the first intersection. Specifically, the server can use the difference between the elevation coordinate values ​​of each first intersection located at a different road cross section from the first intersection under the same acquisition lane marking as the first intersection and the elevation coordinate value of the first intersection as the relative distance of the elevation from the first intersection to the first intersection located at a different road cross section from the first intersection under the same acquisition lane marking as the first intersection. The coordinate system in which the elevation coordinate values ​​are located is the coordinate system used to establish a high-precision map.

[0299] In one example, it is assumed that the road to be updated includes road cross section A, road cross section B, and road cross section C. For example, Fig.17 To understand, Fig.17 Schematic diagram of relative distance of elevation provided in the embodiment of the present application Figure 2 ,like Fig.17As shown, the X-axis, Y-axis, and Z-axis represent the X-axis, Y-axis, and Z-axis of the coordinate system corresponding to the high-precision map, respectively, where the Z-axis represents the elevation coordinate value. Assuming that the first intersection An13 under the road cross section A, the first intersection Bn13 under the road cross section B, and the first intersection Cn13 under the road cross section C are located on the same acquisition lane marking, the elevation coordinate value of the first intersection An13 on the Z axis is za13, the elevation coordinate value of the first intersection Bn13 on the Z axis is zb13, and the absolute value of the distance between za13 and zb13 on the Z axis is dgab, then the server uses zb13-za13=dgab as the relative elevation distance from the first intersection An13 to the first intersection Bn13.

[0300] Assuming that the elevation coordinate value of the first intersection Cn13 on the Z axis is zc13, and the absolute value of the distance between za13 and zc13 on the Z axis is dgac, the server uses zc13-za13=-dgac as the relative elevation distance from the first intersection An13 to the first intersection Cn13.

[0301] Then, the server uses the relative elevation distance dgab from the first intersection point An13 to the first intersection point Bn13 and the relative elevation distance -dgac from the first intersection point An13 to the first intersection point Cn13 as the relative elevation distance of the first intersection point An13.

[0302] By analogy, the server can determine the lateral relative distance of each first intersection point of each collected lane marking and the elevation relative distance of the first intersection point.

[0303] For each real lane marking, from each first intersection, the lateral relative distance and elevation relative distance to the first intersection located on the same collected lane marking as the first intersection and located at a different road cross section from the first intersection are determined with high accuracy, so as to facilitate the subsequent determination of the absolute distance information of the position point of the real lane marking with high accuracy based on the high-precision lateral relative distance and elevation relative distance. In this way, the position information of the position point of the updated collected lane marking can be accurately determined, and the high-precision map can be accurately updated.

[0304] S1605. For each real lane marking, determine a second set of absolute lateral distances of the road cross section under the real lane marking based on the absolute lateral distances of the first intersection points collected multiple times under each road cross section.

[0305] Exemplarily, after determining the lateral absolute distance of each first intersection point collected multiple times under each real lane marking, the server can group the lateral absolute distances of the first intersection points collected multiple times under the real lane marking according to their corresponding road cross-sections, and the lateral absolute distances of each first intersection point in each group constitute a second lateral absolute distance set.

[0306] For example, you can combine Fig.10 To understand, such as Fig.10 As shown, the server may form a second lateral absolute distance set A1 by combining the lateral absolute distance of the first intersection An11, the lateral absolute distance of the first intersection An21, and the lateral absolute distance of the first intersection An31 of the real lane marking 1 under the road cross section A.

[0307] By analogy, the server determines a second lateral absolute distance set B1 of the real lane marking 1 in the road cross section B and a second lateral absolute distance set C1 of the real lane marking 1 in the road cross section C.

[0308] The server can also use a similar method as on real lane marking 1 on real lane marking 2, real lane marking 3, and real lane marking 4 at the same time to obtain the second lateral absolute distance set of all road cross sections under real lane marking 2, real lane marking 3, and real lane marking 4, respectively, which will not be repeated here.

[0309] S1606. For each real lane marking, the average of each lateral absolute distance in the second lateral absolute distance set with the smallest standard deviation in the second lateral absolute distance set for each road cross section under the real lane marking is used as the lateral absolute distance of the position point on the road cross section corresponding to the real lane marking in the second lateral absolute distance set; or, for each real lane marking, the average of each lateral absolute distance in the second lateral absolute distance set with the smallest confidence interval of T distribution in the second lateral absolute distance set for each road cross section under the real lane marking is used as the lateral absolute distance of the position point on the road cross section corresponding to the real lane marking in the second lateral absolute distance set.

[0310] For example, after determining the second lateral absolute distance set of each road cross section under each real lane marking for each real lane marking, the server first selects the second lateral absolute distance set whose lateral absolute distances collected multiple times are closest to the real lateral absolute distance. Then, the average of each lateral absolute distance in the second lateral absolute distance set is used as the lateral absolute distance of the position point of the real lane marking on the road cross section corresponding to the second lateral absolute distance set.

[0311] Among them, a second lateral absolute distance set whose lateral absolute distances collected multiple times are closest to the true lateral absolute distance is selected from each second lateral absolute distance set. For example, a set with the smallest standard deviation can be selected from each second lateral absolute distance set, or a set with the smallest confidence interval of the T distribution can be selected from each second lateral absolute distance set (the confidence level used in calculating the T distribution can be determined based on experience, and this embodiment does not impose any restrictions on this).

[0312] For example, you can combine Fig.10 For understanding, assuming that the server refers to the example in step S1605 and determines that among the second lateral absolute distance set A1 under road cross section A, the second lateral absolute distance set B1 under road cross section B, and the second lateral absolute distance set C1 under road cross section C under the real lane marking 1, the standard deviation of the second lateral absolute distance set B1 is the smallest, then the average of the lateral absolute distances of the first intersection Bn11, the first intersection Bn21, and the first intersection Bn31 in the second lateral absolute distance set B1 is used as the lateral absolute distance of the position point of the real lane marking 1 under road cross section B. Alternatively, among the second lateral absolute distance set A1 under road cross section A, the second lateral absolute distance set B1 under road cross section B, and the second lateral absolute distance set C1 under road cross section C under the real lane marking 1, the confidence interval of the T distribution of the second lateral absolute distance set B1 is the smallest, then the average of the lateral absolute distances of the first intersection Bn11, the first intersection Bn21, and the first intersection Bn31 in the second lateral absolute distance set B1 is taken as the lateral absolute distance of the position point of the real lane marking 1 under road cross section B.

[0313] The server can also use a similar method as on real lane marking 1 on real lane marking 2, real lane marking 3, and real lane marking 4 at the same time to obtain the lateral absolute distance of the position point in a road cross section under real lane marking 2, real lane marking 3, and real lane marking 4, respectively, which will not be repeated here.

[0314] Since the standard deviation of the second lateral absolute distance set is used to indicate the degree of discreteness of the lateral absolute distance distribution in the set, a smaller standard deviation indicates that the distribution of the lateral absolute distances collected multiple times in the set is more concentrated, that is, the lateral absolute distances collected multiple times in the set are closer to the true lateral absolute distances. Therefore, under each real lane marking line, the average value of each lateral absolute distance in the second lateral absolute distance set with the smallest standard deviation is closer to the true lateral absolute distance. It can also be understood that at this time, under the real lane marking line, the lateral absolute distance accuracy of the position point in the road cross section corresponding to the second lateral absolute distance set is higher.

[0315] Since the confidence interval of the T distribution of the second lateral absolute distance set is used to indicate the credibility of the lateral absolute distances in the set, the smaller the confidence interval of the T distribution is, the closer the lateral absolute distances collected multiple times in the set are to the true lateral absolute distances. Therefore, under each real lane marking line, the average value of each lateral absolute distance in the second lateral absolute distance set with the smallest confidence interval of the T distribution is closest to the true lateral absolute distance. It can also be understood that at this time, under the real lane marking line, the lateral absolute distance accuracy of the position point in the road cross section corresponding to the second lateral absolute distance set is higher.

[0316] S1607. For each real lane marking, determine the absolute lateral distance of the position point of the real lane marking in each road cross section based on the absolute lateral distance of the position point of the real lane marking and the relative lateral distance from each first intersection of the real lane marking to each other first intersection under the same collected lane marking line as the first intersection.

[0317] In one example, step S1607 includes the following process:

[0318] For each real lane marking, repeat the following steps until the absolute lateral distance of the position point of the real lane marking in each road cross section is determined:

[0319] For each real lane marking, based on multiple acquisitions, from the first intersection point in the road cross section corresponding to the lateral absolute distance of each existing position point to the first intersection point in the road cross section corresponding to the lateral absolute distance of each non-existent position point located under the same acquired lane marking as the first intersection point, the lateral relative distances of the road cross sections corresponding to the two first intersection points corresponding to the lateral relative distances are the same, and are determined as the second lateral relative distance set;

[0320] For each real lane marking, the average value of each lateral relative distance in the second lateral relative distance set with the smallest standard deviation in each second lateral relative distance set is used as the second optimal lateral relative distance from the position point corresponding to the road cross section corresponding to the lateral absolute distance of the existing position point to the position point corresponding to the road cross section corresponding to the lateral absolute distance of the non-existing position point; or, for each real lane marking, the average value of each lateral relative distance in the second lateral relative distance set with the smallest confidence interval of T distribution in each second lateral relative distance set is used as the second optimal lateral relative distance from the position point corresponding to the road cross section corresponding to the lateral absolute distance of the non-existing position point to the position point corresponding to the road cross section corresponding to the lateral absolute distance of the non-existing position point;

[0321] For each real lane marking, the sum of the second optimal lateral relative distance and the lateral absolute distance of the position point corresponding to the road cross section at the lateral absolute distance of the existing position point corresponding to the second optimal lateral relative distance is taken as the lateral absolute distance of the position point of the real lane marking in the road cross section at the lateral absolute distance of the non-existing position point corresponding to the second optimal lateral relative distance.

[0322] Exemplarily, the server determines the absolute lateral distance of a position point under each real lane marking in a road cross section, and determines the lateral relative distances from each first intersection in the road cross section corresponding to the lateral absolute distances from the existing position points collected multiple times to the first intersection in the road cross section corresponding to the lateral absolute distances from the non-existent position points collected at the same time as the first intersection. The following steps may be repeated until the absolute lateral distances of the position points of the real lane marking in each road cross section are determined.

[0323] The server groups the lateral relative distances from each first intersection in the road cross section corresponding to the lateral absolute distance of each existing position point, which are collected multiple times under the real lane marking, to the first intersection in the road cross section corresponding to the lateral absolute distance of each non-existent position point, which are collected at the same time as the first intersection, according to a pair of road cross sections they correspond to, and the lateral relative distances in each group constitute a second lateral relative distance set.

[0324] Then, the server first selects a second lateral relative distance set whose lateral relative distances collected multiple times are closest to the real lateral relative distance from each second lateral relative distance set. Then, the average value of each lateral relative distance in the second lateral relative distance set is used as the second optimal lateral relative distance from the position point of the real lane marking in the road cross section corresponding to the lateral absolute distance of the existing position point to the position point of the real lane marking in the road cross section corresponding to the lateral absolute distance of the non-existent position point.

[0325] Among them, a second lateral relative distance set whose lateral relative distances collected multiple times are closest to the true lateral relative distance is selected from each second lateral relative distance set. For example, a set with the smallest standard deviation can be selected from each second lateral relative distance set, or a set with the smallest confidence interval of the T distribution can be selected from each second lateral relative distance set (the confidence level used in calculating the T distribution can be determined based on experience, and this embodiment does not impose any restrictions on this).

[0326] Afterwards, the server can add the second optimal lateral relative distance to the lateral absolute distance of the position point of the real lane marking in the road cross-section corresponding to the lateral absolute distance of the existing position point, and obtain the lateral absolute distance of the position point of the real lane marking in the road cross-section corresponding to the lateral absolute distance of the existing position point, which corresponds to the second optimal lateral relative distance.

[0327] For example, you can combine Fig.10 For understanding, it is assumed that the server refers to the example in step S1606 to determine the absolute lateral distance of the position point of the real lane marking 1 under the road cross section B. After that, the lateral relative distance between the first intersection Bn11 under the road cross section B and the first intersection An11 under the road cross section A collected at the same time, the lateral relative distance between the first intersection Bn21 under the road cross section B and the first intersection An21 under the road cross section A collected at the same time, and the lateral relative distance between the first intersection Bn31 under the road cross section B and the first intersection An31 under the road cross section A collected at the same time under the real lane marking 1 can constitute a second lateral relative distance set.

[0328] At the same time, the server forms a second lateral relative distance set by the lateral relative distance between the first intersection Bn11 under the road cross section B and the first intersection Cn11 under the road cross section C collected at the same time, the lateral relative distance between the first intersection Bn21 under the road cross section B and the first intersection Cn21 under the road cross section C collected at the same time, and the lateral relative distance between the first intersection Bn31 under the road cross section B and the first intersection Cn31 under the road cross section C collected at the same time.

[0329] Under the real lane marking 1, assuming that the standard deviation of the second lateral relative distance set corresponding to the first intersection under the road cross section B and the first intersection under the road cross section A collected at the same time is the smallest among the above two second lateral relative distance sets, then the average value of each lateral relative distance in the set is used as the second optimal lateral relative distance from the position point of the real lane marking 1 corresponding to the road cross section B to the position point of the real lane marking 1 corresponding to the road cross section A. Then, the sum of the second optimal lateral relative distance and the lateral absolute distance of the position point of the real lane marking 1 corresponding to the road cross section B is used as the lateral absolute distance of the position point of the real lane marking 1 corresponding to the road cross section A. Alternatively, under the real lane marking 1, assuming that the confidence interval of the T distribution of the second lateral relative distance set corresponding to the first intersection under the road cross section B and the first intersection under the road cross section A collected at the same time is the smallest among the above two second lateral relative distance sets, then the average value of each lateral relative distance in the set is used as the second optimal lateral relative distance from the position point of the real lane marking 1 corresponding to the road cross section B to the position point of the real lane marking 1 corresponding to the road cross section A. Then, the sum of the second optimal lateral relative distance and the lateral absolute distance of the position point of the real lane marking 1 corresponding to the road cross section B is taken as the lateral absolute distance of the position point of the real lane marking 1 corresponding to the road cross section A.

[0330] Then, the server determines the second lateral relative distance set corresponding to the first intersection under the road cross section A to the first intersection under the road cross section C collected at the same time under the real lane marking 1. The specific implementation method is similar to the above method and will not be repeated here.

[0331] Next, under the real lane marking 1, in the second lateral relative distance set corresponding to the first intersection under the road cross section A to the first intersection under the road cross section C collected at the same time, and the second lateral relative distance set corresponding to the first intersection under the road cross section B to the first intersection under the road cross section C collected at the same time. Assuming that the standard deviation of the second lateral relative distance set corresponding to the first intersection under the road cross section A to the first intersection under the road cross section C collected at the same time is the smallest, the average value of each lateral relative distance in the set is used as the second optimal lateral relative distance from the position point of the real lane marking 1 under the road cross section A to the position point of the real lane marking 1 under the road cross section C. Then, the sum of the second optimal lateral relative distance and the lateral absolute distance of the position point of the real lane marking 1 under the road cross section A is used as the lateral absolute distance of the position point of the real lane marking 1 under the road cross section C. Alternatively, assuming that the confidence interval of the T distribution of the second lateral relative distance set corresponding to the first intersection point under the road cross section A to the first intersection point under the road cross section C collected at the same time is the smallest, the average value of each lateral relative distance in the set is used as the second optimal lateral relative distance from the position point of the real lane marking 1 under the road cross section A to the position point of the real lane marking 1 under the road cross section C. Then, the sum of the second optimal lateral relative distance and the lateral absolute distance of the position point of the real lane marking 1 under the road cross section A is used as the lateral absolute distance of the position point of the real lane marking 1 under the road cross section C.

[0332] At the same time, the server can use a similar implementation method as on real lane marking 1 on real lane marking 2, real lane marking 3, and real lane marking 4 to obtain the lateral absolute distances of the position points corresponding to all road cross sections of real lane marking 2, real lane marking 3, and real lane marking 4, respectively, which will not be repeated here.

[0333] Since the first intersection points collected at the same time under the same real lane marking have the same source, for example, they are all extracted from the same photo, the lateral relative distance accuracy of the first intersection points collected at the same time is relatively high, and then the average value of each lateral relative distance in the set with the smallest standard deviation or the smallest confidence interval of the T distribution in the lateral relative distance set is used as the lateral relative distance between the position points. Then, the lateral absolute distance is corrected by the lateral relative distance between the position points, so that the lateral absolute distance accuracy of the position points of the real lane marking is relatively high.

[0334] S1608. For each real lane marking, determine a second absolute elevation distance set of the road cross section under the real lane marking based on the absolute elevation distances of the first intersection points collected multiple times under each road cross section.

[0335] Exemplarily, after determining the absolute elevation distance of each first intersection point collected multiple times under each real lane marking, the server can group the absolute elevation distances of the first intersection points collected multiple times under the real lane marking according to their corresponding road cross-sections, and the absolute elevation distances of each first intersection point in each group constitute a second absolute elevation distance set.

[0336] For example, you can combine Fig.10 To understand, such as Fig.10 As shown, the server may form a second absolute elevation distance set A1 with the absolute elevation distance of the first intersection point An11, the absolute elevation distance of the first intersection point An21, and the absolute elevation distance of the first intersection point An31 of the real lane marking 1 under the road cross section A.

[0337] By analogy, the server determines a second elevation absolute distance set B1 of the real lane marking line 1 in the road cross section B and a second elevation absolute distance set C1 of the real lane marking line 1 in the road cross section C.

[0338] The server can also use a similar method as on real lane marking 1 on real lane marking 2, real lane marking 3, and real lane marking 4 at the same time to obtain the second elevation absolute distance set of all road cross sections under real lane marking 2, real lane marking 3, and real lane marking 4, respectively, which will not be repeated here.

[0339] S1609. For each real lane marking, the average of the absolute elevation distances in the second absolute elevation distance set with the smallest standard deviation in the second absolute elevation distance set for each road cross section under the real lane marking is used as the absolute elevation distance of the position point on the road cross section corresponding to the real lane marking in the second absolute elevation distance set; or, for each real lane marking, the average of the absolute elevation distances in the second absolute elevation distance set with the smallest confidence interval of the T distribution in the second absolute elevation distance set for each road cross section under the real lane marking is used as the absolute elevation distance of the position point on the road cross section corresponding to the real lane marking in the second absolute elevation distance set.

[0340] For example, after determining the second absolute elevation distance set of each road cross section under the real lane marking for each real lane marking, the server first selects the second absolute elevation distance set whose absolute elevation distances collected multiple times are closest to the real absolute elevation distance. Then, the average of the absolute elevation distances in the second absolute elevation distance set is used as the absolute elevation distance of the position point of the real lane marking under the road cross section corresponding to the second absolute elevation distance set.

[0341] Among them, a second elevation absolute distance set whose absolute elevation distances collected multiple times are closest to the true elevation absolute distance is selected from each second elevation absolute distance set. For example, a set with the smallest standard deviation can be selected from each second elevation absolute distance set, or a set with the smallest confidence interval of the T distribution can be selected from each second elevation absolute distance set (the confidence level used in calculating the T distribution can be determined based on experience, and this embodiment does not impose any restrictions on this).

[0342] For example, you can combine Fig.10 For understanding, assuming that the server refers to the example in step S1608 and determines that among the second elevation absolute distance set A1 under road cross section A, the second elevation absolute distance set B1 under road cross section B, and the second elevation absolute distance set C1 under road cross section C under the real lane marking 1, the standard deviation of the second elevation absolute distance set C1 is the smallest, then the average of the absolute elevation distances of the first intersection Cn11, the first intersection Cn21, and the first intersection Cn31 in the second elevation absolute distance set C1 is used as the absolute elevation distance of the position point of the real lane marking 1 under the road cross section C. Alternatively, among the second absolute elevation distance set A1 under road cross section A, the second absolute elevation distance set B1 under road cross section B, and the second absolute elevation distance set C1 under road cross section C under the real lane marking 1, the confidence interval of the T distribution of the second absolute elevation distance set C1 is the smallest, then the average of the absolute elevation distances of the first intersection Cn11, the absolute elevation distance of the first intersection Cn21, and the absolute elevation distance of the first intersection Cn31 in the second absolute elevation distance set C1 is taken as the absolute elevation distance of the position point of the real lane marking 1 under road cross section C.

[0343] The server can also use a similar method as on real lane marking 1 on real lane marking 2, real lane marking 3, and real lane marking 4 at the same time to obtain the absolute elevation distance of the position point in a road cross section under real lane marking 2, real lane marking 3, and real lane marking 4, respectively, which will not be repeated here.

[0344] Since the standard deviation of the second elevation absolute distance set is used to indicate the degree of discreteness of the distribution of the absolute elevation distances in the set, a smaller standard deviation indicates that the distribution of the multiple collected absolute elevation distances in the set is more concentrated. In other words, the multiple collected absolute elevation distances in the set are closer to the true absolute elevation distances. Therefore, under each real lane marking line, the average value of the absolute elevation distances in the second elevation absolute distance set with the smallest standard deviation is closer to the true absolute elevation distance. It can also be understood that at this time, under the real lane marking line, the absolute elevation distance accuracy of the position points in the road cross section corresponding to the second elevation absolute distance set is higher.

[0345] Since the confidence interval of the T distribution of the second elevation absolute distance set is used to indicate the credibility of the absolute elevation distances in the set, the smaller the confidence interval of the T distribution is, the closer the absolute elevation distances collected multiple times in the set are to the true absolute elevation distances. Therefore, under each real lane marking line, the average value of the absolute elevation distances in the second elevation absolute distance set with the smallest confidence interval of the T distribution is closest to the true absolute elevation distance. It can also be understood that at this time, under the real lane marking line, the absolute elevation distance accuracy of the position points in the road cross section corresponding to the second elevation absolute distance set is higher.

[0346] S1610. For each real lane marking, determine the absolute elevation distance of the position point of the real lane marking in each road cross section based on the absolute elevation distance of the position point of the real lane marking and the relative elevation distance of each first intersection point of the real lane marking to each other first intersection point under the same collected lane marking line as the first intersection point.

[0347] In one example, step S1610 includes the following process:

[0348] For each real lane marking, repeat the following steps until the absolute elevation distance of the position point of the real lane marking in each road cross section is determined:

[0349] For each real lane marking, based on multiple acquisitions, from the first intersection point under the road cross section corresponding to the absolute elevation distance of each existing position point, to the first intersection point under the road cross section corresponding to the absolute elevation distance of each non-existent position point located under the same acquired lane marking as the first intersection point, the relative elevation distances of the road cross sections corresponding to the two first intersection points corresponding to the relative elevation distances are the same, and are determined as the second relative elevation distance set;

[0350] For each real lane marking, the average value of each elevation relative distance in the second elevation relative distance set with the smallest standard deviation in each second elevation relative distance set is used as the second optimal elevation relative distance from the position point corresponding to the road cross section corresponding to the absolute elevation distance of the existing position point to the position point corresponding to the road cross section corresponding to the absolute elevation distance of the non-existent position point, corresponding to the second elevation relative distance set; or, for each real lane marking, the average value of each elevation relative distance in the second elevation relative distance set with the smallest confidence interval of the T distribution in each second elevation relative distance set is used as the second optimal elevation relative distance from the position point corresponding to the road cross section corresponding to the absolute elevation distance of the non-existent position point to the position point corresponding to the road cross section corresponding to the absolute elevation distance of the non-existent position point, corresponding to the second elevation relative distance set;

[0351] For each real lane marking, the sum of the second optimal elevation relative distance and the absolute elevation distance of the location point corresponding to the absolute elevation distance of the existing location point corresponding to the second optimal elevation relative distance and the absolute elevation distance of the location point corresponding to the road cross section is taken as the absolute elevation distance of the location point of the real lane marking in the road cross section where the absolute elevation distance of the location point corresponding to the second optimal elevation relative distance does not exist.

[0352] Exemplarily, the server determines the absolute elevation distance of the position points under each real lane marking in a road cross section, and determines the relative elevation distances of each first intersection point in the road cross section corresponding to the absolute elevation distances of the existing position points collected multiple times to the first intersection point in the road cross section corresponding to the absolute elevation distances of the non-existent position points collected at the same time as the first intersection point. After that, the server may repeat the following steps until the absolute elevation distances of the position points of the real lane marking in each road cross section are determined.

[0353] The server groups the relative elevation distances from each first intersection in the road cross section corresponding to the absolute elevation distance of each existing position point collected multiple times under the real lane marking to the first intersection in the road cross section corresponding to the absolute elevation distance of each non-existent position point collected at the same time as the first intersection according to a pair of road cross sections they correspond to, and the relative elevation distances in each group constitute a second relative elevation distance set.

[0354] Then, the server first selects the second relative elevation distance set whose relative elevation distances collected multiple times are closest to the real relative elevation distance from each second relative elevation distance set. Then, the average value of each relative elevation distance in the second relative elevation distance set is used as the second optimal relative elevation distance of the position point of the real lane marking line in the road cross section corresponding to the absolute elevation distance of the existing position point to the position point of the real lane marking line in the road cross section corresponding to the absolute elevation distance of the non-existent position point.

[0355] Among them, a second elevation relative distance set whose elevation relative distances collected multiple times are closest to the true elevation relative distance is selected from each second elevation relative distance set. For example, a set with the smallest standard deviation can be selected from each second elevation relative distance set, or a set with the smallest confidence interval of the T distribution can be selected from each second elevation relative distance set (the confidence level used in the T distribution calculation can be determined based on experience, and this embodiment does not impose any restrictions on this).

[0356] Afterwards, the server can add the second optimal relative elevation distance to the absolute elevation distance of the location point of the real lane marking in the road cross-section corresponding to the absolute elevation distance of the existing location point corresponding to the second optimal relative elevation distance, and obtain the absolute elevation distance of the location point of the real lane marking in the road cross-section corresponding to the absolute elevation distance of the existing location point corresponding to the second optimal relative elevation distance.

[0357] For example, you can combine Fig.10 For understanding, it is assumed that the server refers to the example in step S1609 to determine the absolute elevation distance of the position point of the real lane marking 1 under the road cross section C. After that, the relative elevation distance between the first intersection Cn11 under the road cross section C and the first intersection An11 under the road cross section A collected at the same time, the relative elevation distance between the first intersection Cn21 under the road cross section C and the first intersection An21 under the road cross section A collected at the same time, and the relative elevation distance between the first intersection Cn31 under the road cross section C and the first intersection An31 under the road cross section A collected at the same time can be formed into a second relative elevation distance set.

[0358] At the same time, the server forms a second relative elevation distance set by the relative elevation distance between the first intersection Cn11 under the road cross section C and the first intersection Bn11 under the road cross section B collected at the same time, the relative elevation distance between the first intersection Cn21 under the road cross section C and the first intersection Bn21 under the road cross section B collected at the same time, and the relative elevation distance between the first intersection Cn31 under the road cross section C and the first intersection Bn31 under the road cross section B collected at the same time.

[0359] Under the real lane marking 1, assuming that the standard deviation of the second relative elevation distance set corresponding to the first intersection under the road cross section C and the first intersection under the road cross section A collected at the same time is the smallest among the above two second relative elevation distance sets, then the average value of each relative elevation distance in the set is used as the second optimal relative elevation distance from the position point of the real lane marking 1 corresponding to the road cross section C to the position point of the real lane marking 1 corresponding to the road cross section A. Then, the sum of the second optimal relative elevation distance and the absolute elevation distance of the position point of the real lane marking 1 corresponding to the road cross section C is used as the absolute elevation distance of the position point of the real lane marking 1 corresponding to the road cross section A. Alternatively, under the real lane marking 1, assuming that the confidence interval of the T distribution of the second relative elevation distance set corresponding to the first intersection under the road cross section C and the first intersection under the road cross section A collected at the same time is the smallest among the above two second relative elevation distance sets, then the average value of each relative elevation distance in the set is used as the second optimal relative elevation distance from the position point of the real lane marking 1 corresponding to the road cross section C to the position point of the real lane marking 1 corresponding to the road cross section A. Then, the sum of the second optimal relative elevation distance and the absolute elevation distance of the position point of the real lane marking 1 corresponding to the road cross section C is taken as the absolute elevation distance of the position point of the real lane marking 1 corresponding to the road cross section A.

[0360] Then, the server determines the second elevation relative distance set corresponding to the first intersection under the road cross section A to the first intersection under the road cross section B collected at the same time under the real lane marking 1. The specific implementation method is similar to the above method and will not be repeated here.

[0361] Next, under the real lane marking 1, in the set of second relative elevation distances corresponding to the first intersection under the road cross section A and the first intersection under the road cross section B collected at the same time, and the set of second relative elevation distances corresponding to the first intersection under the road cross section C and the first intersection under the road cross section B collected at the same time. Assuming that the standard deviation of the set of second relative elevation distances corresponding to the first intersection under the road cross section C and the first intersection under the road cross section B collected at the same time is the smallest, the average value of each relative elevation distance in the set is used as the second optimal relative elevation distance from the position point of the real lane marking 1 under the road cross section C to the position point of the real lane marking 1 under the road cross section B. Then, the sum of the second optimal relative elevation distance and the absolute elevation distance of the position point of the real lane marking 1 under the road cross section C is used as the absolute elevation distance of the position point of the real lane marking 1 under the road cross section B. Alternatively, assuming that the confidence interval of the T distribution of the second relative elevation distance set corresponding to the first intersection point under the road cross section C to the first intersection point under the road cross section B collected at the same time is the smallest, the average value of each relative elevation distance in the set is used as the second optimal relative elevation distance from the position point of the real lane marking 1 under the road cross section C to the position point of the real lane marking 1 under the road cross section B. Then, the sum of the second optimal relative elevation distance and the absolute elevation distance of the position point of the real lane marking 1 under the road cross section C is used as the absolute elevation distance of the position point of the real lane marking 1 under the road cross section B.

[0362] At the same time, the server can use a similar implementation method as on real lane marking 1 on real lane marking 2, real lane marking 3, and real lane marking 4 to obtain the absolute elevation distances of the position points corresponding to all road cross sections of real lane marking 2, real lane marking 3, and real lane marking 4, respectively, which will not be repeated here.

[0363] Since the first intersection points collected at the same time under the same real lane marking have the same source, for example, they are all extracted from the same photo, and thus the relative distance accuracy of the first intersection points collected at the same time is relatively high, then the average of the relative distances in the set with the smallest standard deviation or the smallest confidence interval of the T distribution in the relative distance set is taken as the relative distance between the positions. Then, the relative distance between the positions is used to correct the absolute distance, so that the absolute distance accuracy of the position points of the real lane marking is relatively high.

[0364] It is worth noting that, since the absolute horizontal distance and absolute elevation distance of the real lane marking position point are calculated independently, the relative horizontal distance and relative elevation distance of the real lane marking position point are also calculated independently. Therefore, under the same road cross section, the order of the real lane marking corresponding to the absolute horizontal distance of the position point determined by the server and the order of the real lane marking corresponding to the absolute elevation distance of the position point are not necessarily the same. The purpose of this processing is to ensure that the absolute horizontal distance and absolute elevation distance of the position point do not affect each other when calculating the absolute horizontal distance and absolute elevation distance of the position point, so as to make their respective accuracy higher.

[0365] S1611. Use the longitudinal absolute distance in the absolute distance information of any first intersection in each road cross section as the longitudinal absolute distance of the position point of each real lane marking in the road cross section.

[0366] The implementation methods of step S1611 and step S1111 are similar. For detailed introduction, please refer to the description of step S1111, which will not be repeated here.

[0367] S1612. Use the absolute horizontal distance, absolute elevation distance, and absolute vertical distance of each real lane marking position point in each road cross section as the position information of each real lane marking position point in each road cross section of the road to be updated.

[0368] The implementation methods of step S1612 and step S1112 are similar. For detailed introduction, please refer to the description of step S1112, which will not be repeated here.

[0369] S1613. Determine an updated collected lane marking of each real lane marking of the road to be updated according to the position information of the position point of each real lane marking of the road to be updated in each road cross section.

[0370] The implementation methods of step S1613 and step S1113 are similar. For detailed introduction, please refer to the description of step S1113, which will not be repeated here.

[0371] S1614. Update the high-precision map according to the updated collected lane markings of each real lane marking.

[0372] The implementation of step S1614 is similar to that of step S106. For detailed description, please refer to the description of step S106, which will not be repeated here.

[0373] The lane marking-based map updating method provided in the embodiment of the present application has a high accuracy of the lateral relative distances of the first intersection points collected at the same time under the same real lane marking, because the sources of the first intersection points collected at the same time are the same, for example, they are all extracted from the same photo. Then, the average value of each lateral relative distance in the set with the smallest standard deviation in the lateral relative distance set or the smallest confidence interval of the T distribution is used as the lateral relative distance between the position points. Then, the lateral absolute distance is corrected by the lateral relative distance between the position points, so that the lateral absolute distance accuracy of the position points of the real lane marking is high. Similarly, the absolute distance accuracy of the elevation of the position points of the real lane marking determined in a similar manner is also high.

[0374] Since the lateral absolute distance and elevation absolute distance of the location point of the real lane marking are calculated independently, the lateral relative distance and elevation relative distance of the location point of the real lane marking are also calculated independently. This processing can ensure that when calculating the lateral absolute distance and elevation absolute distance of the location point, the two will not affect each other, thereby making the calculated accuracy of each higher.

[0375] Furthermore, based on the horizontal absolute distance, elevation absolute distance, and longitudinal absolute distance of the position points of the real lane markings with high accuracy, the position information of the position points of the real lane markings is also determined with high accuracy. Thus, the updated collected lane markings can be accurately determined, and the high-precision map can be accurately updated.

[0376] Fig.16 The implementation method described in the embodiment of the invention uses the relative distance information of the first intersection points under the same lane marking line in different road cross sections when determining the updated lane marking line. Therefore, when the relative positions of the first intersection points of the lane marking line are relatively accurate, the relative distance information of the first intersection points under the same lane marking line is used. Fig.16 The implementation method introduced in the embodiment can more accurately determine the updated collected lane markings and can more accurately update the high-precision map.

[0377] Fig.11 The implementation method described in the embodiment uses the relative distance information of the first intersection points of different real lane markings collected at the same time in the same road cross section when determining the updated collected lane markings. Therefore, when the relative positions of the first intersection points of different real lane markings collected at the same time are relatively accurate, use Fig.11 The implementation method introduced in the embodiment can more accurately determine the updated collected lane markings and can more accurately update the high-precision map.

[0378] Fig.18 A schematic diagram of the structure of a map updating device based on lane markings provided in an embodiment of the present application is shown in FIG. Fig.18As shown, the lane marking-based map updating device 180 provided in this embodiment includes:

[0379] The acquisition module 1801 is used to acquire the road information of the road to be updated; wherein the road information includes the road reference line of the road to be updated and the collected lane markings of the road to be updated at each collection number, and the collected lane markings are the collected lane markings obtained when collecting each real lane marking of the road to be updated;

[0380] Establishing module 1802, for establishing each road cross section of the road to be updated according to the road reference line of the road to be updated; wherein each collected lane marking line has a plurality of first intersection points, the first intersection point being the intersection point between the collected lane marking line and the road cross section; the road reference line has a plurality of second intersection points, the second intersection point being the intersection point between the road reference line and the road cross section;

[0381] The first determination module 1803 is used to determine, for each first intersection of each collected lane marking, absolute distance information of the first intersection; wherein the absolute distance information represents distance information between the first intersection and a second intersection located in the same road cross section as the first intersection;

[0382] The second determination module 1804 is used to determine the relative distance information of the first intersection for each first intersection of each collected lane marking; wherein the relative distance information represents the distance information between the first intersection and other first intersections located under the same road cross section during the same collection; or the relative distance information represents the distance information between the first intersection and other first intersections located under the same collected lane marking as the first intersection;

[0383] The third determination module 1805 is used to determine the updated collected lane marking of each real lane marking of the road to be updated according to the absolute distance information of the first intersection and the relative distance information of the first intersection;

[0384] The updating module 1806 is used to update the high-precision map according to the updated collected lane markings of each real lane marking.

[0385] The lane marking-based map updating device provided in this embodiment can execute the method provided in the above method embodiment, and its implementation principle and technical effect are similar, which will not be described in detail in this embodiment.

[0386] Fig.19 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Fig.19As shown, the electronic device 190 provided in this embodiment includes: at least one processor 1901 and a memory 1902. Optionally, the device 190 further includes a communication component 1903. The processor 1901, the memory 1902 and the communication component 1903 are connected via a bus 1904.

[0387] In the specific implementation process, at least one processor 1901 executes the computer execution instructions stored in the memory 1902, so that at least one processor 1901 executes the above method.

[0388] The specific implementation process of processor 1901 can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.

[0389] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the invention may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0390] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (Non-volatile Memory, NVM), such as at least one disk memory.

[0391] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of the present application is not limited to only one bus or one type of bus.

[0392] The present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0393] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.

[0394] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, 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 disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0395] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (Application Specific Integrated Circuits, referred to as: ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0396] The division of units is only a logical function division, and there may be other divisions in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0397] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0398] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0399] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0400] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other media that can store program codes.

[0401] Finally, it should be noted that those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention, which follow the general principles of the present invention and include common knowledge or customary technical means in the art not disclosed by the present invention, are not limited to the precise structure described above and shown in the drawings, and may be modified and changed in various ways without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A map updating method based on lane markings, It is characterized in that The method comprises: Acquire road information of the road to be updated; wherein the road information includes a road reference line of the road to be updated and a collected lane marking of the road to be updated at each collection number, wherein the collected lane marking is a collected lane marking obtained when each real lane marking of the road to be updated is collected; For the road to be updated, each road cross section of the road to be updated is established according to the road reference line of the road to be updated; wherein each collected lane marking line has a plurality of first intersection points, the first intersection point being the intersection point between the collected lane marking line and the road cross section; the road reference line has a plurality of second intersection points, the second intersection point being the intersection point between the road reference line and the road cross section; For each first intersection of each collected lane marking, determine the absolute distance information of the first intersection; wherein the absolute distance information represents the distance information between the first intersection and the corresponding second intersection; For each first intersection of each collected lane marking, determine relative distance information of the first intersection; wherein the relative distance information represents distance information between the first intersection and corresponding other first intersections; According to the absolute distance information of the first intersection and the relative distance information of the first intersection, the updated collected lane marking of each real lane marking of the road to be updated is determined; and according to the updated collected lane marking of each real lane marking, the high-precision map is updated.

2. The method according to claim 1, It is characterized in that The absolute distance information includes: lateral absolute distance, elevation absolute distance, and longitudinal absolute distance; the relative distance information includes: lateral relative distance at the same road cross section and elevation relative distance at the same road cross section; or, lateral relative distance at the same collected lane marking line and elevation relative distance at the same collected lane marking line; The lateral absolute distance represents the lateral distance information between the first intersection and the second intersection located in the same road cross section as the first intersection; the elevation absolute distance represents the elevation distance information between the first intersection and the second intersection located in the same road cross section as the first intersection; the longitudinal absolute distance represents the distance information between the starting point of the road reference line and the second intersection located in the same road cross section as the first intersection; The lateral relative distance located in the same road cross section represents the lateral distance information between the first intersection and other first intersections located in the same road cross section during the same collection; the elevation relative distance located in the same road cross section represents the elevation distance information between the first intersection and other first intersections located in the same road cross section during the same collection; The lateral relative distance located under the same collection lane marking line represents the lateral distance information between the first intersection and other first intersections located under the same collection lane marking line; the elevation relative distance located under the same collection lane marking line represents the elevation distance information between the first intersection and other first intersections located under the same collection lane marking line.

3. The method according to claim 2, It is characterized in that When the relative distance information includes: a lateral relative distance located at the same road cross section and an elevation relative distance located at the same road cross section, determining the updated collected lane marking of each real lane marking of the road to be updated according to the absolute distance information of the first intersection and the relative distance information of the first intersection, comprises: For each road cross section, determine the lateral absolute distance of each real lane marking line position point in the road cross section according to the lateral absolute distance in the absolute distance information of the first intersection and the lateral relative distance in the relative distance information of the first intersection collected multiple times in the road cross section; For each road cross section, the absolute elevation distance of the position point of each real lane marking line under the road cross section is determined according to the absolute elevation distance in the absolute distance information of the first intersection and the relative elevation distance in the relative distance information of the first intersection collected multiple times under the road cross section; The longitudinal absolute distance in the absolute distance information of any first intersection in each road cross section is used as the longitudinal absolute distance of the position point of each real lane marking in the road cross section; Taking the lateral absolute distance, the elevation absolute distance, and the longitudinal absolute distance of the position point of each real lane marking line in each road cross section as the position information of the position point of each real lane marking line in each road cross section of the road to be updated; According to the position information of the position point of each real lane marking of the road to be updated under each road cross section, the updated collected lane marking of each real lane marking of the road to be updated is determined.

4. The method according to claim 3, It is characterized in that For each road cross section, the lateral absolute distance of each real lane marking position point under the road cross section is determined according to the lateral absolute distance in the absolute distance information of the first intersection and the lateral relative distance in the relative distance information of the first intersection collected multiple times under the road cross section, including: For each road cross section, determining a first lateral absolute distance set of the real lane marking in the road cross section based on the lateral absolute distances of the first intersection points collected multiple times for each real lane marking; For each road cross section, the average value of each lateral absolute distance in the first lateral absolute distance set with the smallest standard deviation in the first lateral absolute distance set of each real lane marking line in the road cross section is used as the lateral absolute distance of the position point of the real lane marking line in the road cross section corresponding to the first lateral absolute distance set; or, for each road cross section, the average value of each lateral absolute distance in the first lateral absolute distance set with the smallest confidence interval of T distribution in the first lateral absolute distance set of each real lane marking line in the road cross section is used as the lateral absolute distance of the position point of the real lane marking line in the road cross section corresponding to the first lateral absolute distance set; For each road cross section, the absolute lateral distance of the position point of the real lane marking in the road cross section is determined based on the absolute lateral distance of the position point of the real lane marking in the road cross section, and the lateral relative distance from each first intersection of the real lane marking to each other first intersection point located in the same road cross section during the same collection as the first intersection.

5. The method according to claim 3, It is characterized in that For each road cross section, the absolute elevation distance of the position point of each real lane marking line under the road cross section is determined according to the absolute elevation distance in the absolute distance information of the first intersection and the relative elevation distance in the relative distance information of the first intersection collected multiple times under the road cross section, including: For each road cross section, determining a first absolute distance set of elevations of the real lane markings in the road cross section based on the absolute distances of elevations of the first intersections collected multiple times for each real lane marking; For each road cross section, the average value of each absolute elevation distance in the first absolute elevation distance set with the smallest standard deviation among the first absolute elevation distance set of each real lane marking line under the road cross section is used as the absolute elevation distance of the position point of the real lane marking line under the road cross section corresponding to the first absolute elevation distance set; or, for each road cross section, the average value of each absolute elevation distance in the first absolute elevation distance set with the smallest confidence interval of the T distribution among the first absolute elevation distance set of each real lane marking line under the road cross section is used as the absolute elevation distance of the position point of the real lane marking line under the road cross section corresponding to the first absolute elevation distance set; For each road cross-section, the absolute elevation distance of the position point of the real lane marking in the road cross-section is determined based on the absolute elevation distance of the position point of the real lane marking in the road cross-section and the relative elevation distance of each first intersection of the real lane marking to each other first intersection point located in the same road cross-section during the same collection.

6. The method according to claim 5, It is characterized in that For each road cross section, the absolute elevation distance of the position point of the real lane marking line under the road cross section is determined according to the absolute elevation distance of the position point of the real lane marking line under the road cross section and the relative elevation distance of each first intersection point of the real lane marking line to each other first intersection point under the same road cross section during the same acquisition, including: For each road cross section, repeat the following steps until the absolute elevation distance of each real lane marking position point under the road cross section is determined: For each road cross section, based on multiple acquisitions, from each first intersection point of the real lane marking corresponding to the absolute elevation distance of each existing position point, to the first intersection point of the real lane marking corresponding to the absolute elevation distance of each non-existent position point located in the same road cross section during the same acquisition, the relative elevation distances of the real lane markings corresponding to the two first intersection points corresponding to the relative elevation distances are the same, and are determined as the first relative elevation distance set; For each road cross section, the average value of each elevation relative distance in the first elevation relative distance set with the smallest standard deviation in each first elevation relative distance set is used as the first optimal elevation relative distance from the position point of the real lane marking corresponding to the absolute elevation distance of the existing position point to the position point of the real lane marking corresponding to the absolute elevation distance of the non-existing position point, corresponding to the first elevation relative distance set; or, for each road cross section, the average value of each elevation relative distance in the first elevation relative distance set with the smallest confidence interval of the T distribution in each first elevation relative distance set is used as the first optimal elevation relative distance from the position point of the real lane marking corresponding to the absolute elevation distance of the existing position point to the position point of the real lane marking corresponding to the absolute elevation distance of the non-existing position point, corresponding to the first elevation relative distance set; For each road cross-section, the sum of the first optimal elevation relative distance and the absolute elevation distance of the position point of the real lane marking corresponding to the absolute elevation distance of the existing position point corresponding to the first optimal elevation relative distance is taken as the absolute elevation distance of the position point of the real lane marking in the road cross-section corresponding to the first optimal elevation relative distance and the absolute elevation distance of the non-existent position point.

7. The method according to claim 2, It is characterized in that When the relative distance information includes: a lateral relative distance under the same collected lane marking line and an elevation relative distance under the same collected lane marking line, determining an updated collected lane marking line of each real lane marking line of the road to be updated according to the absolute distance information of the first intersection and the relative distance information of the first intersection, comprising: For each real lane marking, determine the lateral absolute distance of a position point of the real lane marking in each road cross section according to the lateral absolute distance in the absolute distance information of the first intersection and the lateral relative distance in the relative distance information of the first intersection collected multiple times under the real lane marking; For each real lane marking, determine the absolute elevation distance of a position point of the real lane marking in each road cross section based on the absolute elevation distance in the absolute distance information of the first intersection and the relative elevation distance in the relative distance information of the first intersection collected multiple times under the real lane marking; The longitudinal absolute distance in the absolute distance information of any first intersection in each road cross section is used as the longitudinal absolute distance of the position point of each real lane marking in the road cross section; Taking the lateral absolute distance, the elevation absolute distance, and the longitudinal absolute distance of the position point of each real lane marking line in each road cross section as the position information of the position point of each real lane marking line in each road cross section of the road to be updated; According to the position information of the position point of each real lane marking of the road to be updated under each road cross section, the updated collected lane marking of each real lane marking of the road to be updated is determined.

8. A map updating device based on lane markings, It is characterized in that include: An acquisition module, used to acquire road information of the road to be updated; wherein the road information includes a road reference line of the road to be updated and a collected lane marking of the road to be updated at each collection number, wherein the collected lane marking is a collected lane marking obtained when each real lane marking of the road to be updated is collected; An establishing module is used to establish each road cross section of the road to be updated according to the road reference line of the road to be updated, wherein each collected lane marking line has a plurality of first intersection points, the first intersection point being the intersection point between the collected lane marking line and the road cross section; the road reference line has a plurality of second intersection points, the second intersection point being the intersection point between the road reference line and the road cross section; A first determination module is used to determine, for each first intersection of each collected lane marking, absolute distance information of the first intersection; wherein the absolute distance information represents distance information between the first intersection and a second intersection located in the same road cross section as the first intersection; A second determination module is used to determine, for each first intersection of each collected lane marking, relative distance information of the first intersection; wherein the relative distance information represents distance information between the first intersection and other first intersections located under the same road cross section during the same collection; or, the relative distance information represents distance information between the first intersection and other first intersections located under the same collected lane marking as the first intersection; A third determination module is used to determine the updated collected lane marking of each real lane marking of the road to be updated according to the absolute distance information of the first intersection and the relative distance information of the first intersection; The update module is used to update the high-precision map according to the updated collected lane markings of each real lane marking.

9. An electronic device, It is characterized in that The electronic device comprises: a processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium / computer program product, It is characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 7 when executed by a processor; and / or, The computer program product comprises a computer program, which implements the method according to any one of claims 1 to 7 when executed by a processor.

Images