Map updating method and device, server and storage medium

Abstract

The application provides a map updating method and device, a server and a storage medium, and relates to the technical field of electronic maps. An environment image of a target road collected by a vehicle each time the vehicle travels on the target road is received, and a set of environment images is generated; the coordinates of a first road sign element in each environment image in the set of environment images in a target coordinate system are determined respectively; the server aggregates each first road sign element according to the coordinates of each first road sign element in the target coordinate system and a set threshold value to obtain at least one point of interest category; the confidence interval of the distribution of the coordinates of the first road sign element of the corresponding point of interest category is determined according to the Gaussian distribution of the coordinates of the first road sign element of each point of interest category; and the coordinates of a second road sign element of any type in the pre-stored map are updated when the coordinates of the second road sign element are not within the confidence interval associated with the first road sign element of the corresponding type.

Description

Technical Field

[0001] This application relates to the field of electronic map technology, and in particular to a map updating method, apparatus, server and storage medium. Background Technology

[0002] Typically, road signs are used to guide vehicle movement when driving on a road. These signs include lane markings, speed limit signs, and directional signs. Electronic maps usually include these road sign elements so drivers can navigate using them. For example, when a vehicle passes a speed limit sign, the map might announce, "Speed ​​limit 80 ahead, you are speeding." When a vehicle passes a directional sign indicating a U-turn, the map might announce, "Please turn around," making navigation easier. However, road repairs and other factors can cause changes in the positions of road signs. If electronic maps continue to navigate based on the original positions of these road signs, navigation errors may occur, necessitating map updates.

[0003] One current method for updating maps is as follows: when multiple vehicles pass the target road, each vehicle collects environmental images of the target road, and the ratio of the number of environmental images containing a first road sign feature to the total number of environmental images is used as the confidence level of the presence of the first road sign feature. When the confidence level of the presence of the first road sign feature is greater than a set threshold, the distance between the coordinates of the first road sign feature and the coordinates of a second road sign feature of the same type on the target road in a preset map is determined. When the distance is greater than a set threshold, the coordinates of the second road sign feature are updated based on the coordinates of the first road sign feature. However, the accuracy of the coordinates of the first road sign feature determined in the above method is low, resulting in low accuracy in updating the coordinates of the second road sign feature, thus, misnavigation can still occur. Summary of the Invention

[0004] This application provides a map updating method, apparatus, server, and storage medium to address the problem that in the prior art, the accuracy of the coordinates of the first road sign element is low, which leads to the low accuracy of the updated coordinates of the second road surface element, thus still causing the problem of misnavigation.

[0005] In a first aspect, this application provides a map updating method, comprising: receiving environmental images of a target road collected each time a vehicle travels on a target road, and generating a set of environmental images; determining the coordinates of a first road sign element in each environmental image in the set of environmental images in a target coordinate system; performing aggregation processing on each first road sign element according to the coordinates of each first road sign element in the target coordinate system and a set threshold to obtain at least one point of interest category; wherein the distance between any two aggregated first road sign elements in each point of interest category is less than a set threshold; determining a confidence interval for the distribution of the coordinates of the first road sign elements of the corresponding point of interest category according to the Gaussian distribution of the coordinates of the first road sign elements of each point of interest category; updating the coordinates of the second road sign element when the coordinates of a second road sign element in any point of interest category in a pre-stored map are not within the confidence interval associated with the first road sign element of the corresponding point of interest category, so that the updated coordinates of the second road sign element are within the confidence interval.

[0006] In one possible implementation, determining the confidence interval of the distribution of the coordinates of the first landmark elements of each point of interest category based on the Gaussian distribution of their coordinates includes: using a Gaussian distribution model to determine the mean e and standard deviation σ of the coordinates of the first landmark elements of each point of interest category; determining the minimum value m of x in the candidate interval (μ-xσ, μ+xσ) such that the distribution of the coordinates of the first landmark elements of each point of interest category reaches a preset proportion, thereby determining the confidence range (μ-mσ, μ+mσ), where μ is the mathematical expectation of the Gaussian distribution model; and determining the confidence interval (e-[μ-mσ], e+[μ+mσ]) of the distribution of the coordinates of the first landmark elements of the corresponding point of interest category based on the confidence range (μ-mσ, μ+mσ).

[0007] Understandably, the confidence intervals (e-[μ-xσ], e+[μ+xσ]) determined by using the mean e and standard deviation σ of the coordinates of the first landmark feature of each point of interest category obtained by the Gaussian distribution model are highly reliable.

[0008] In one possible implementation, the target coordinate system is a local coordinate system, and the collection of acquired environmental images is located in a latitude and longitude coordinate system. Identifying the coordinates of the first road sign element in each environmental image within the target coordinate system includes: identifying multiple lane lines of the target road in each environmental image, and determining one lane line from these multiple lane lines as a road reference line; establishing a local coordinate system in each environmental image with the starting point of the road reference line as the origin, and determining the coordinates of the first road sign element in the local coordinate system, wherein the eastward direction along the starting point of the road reference line is the X-axis of the local coordinate system, the northward direction along the starting point of the road reference line is the Y-axis of the local coordinate system, and the vertically downward direction along the starting point of the road reference line is the Z-axis of the local coordinate system.

[0009] Since the origin of the local coordinate system is the starting point of the road reference line, the X-axis of the local coordinate system is located due east of the starting point of the road reference line, the Y-axis is located due north of the starting point of the road reference line, and the Z-axis is located vertically downward from the starting point of the road reference line. In this way, determining the coordinates of the first road sign element in the local coordinate system is more in line with the road application scenario and has high accuracy.

[0010] In one possible implementation, the target coordinate system is a reference coordinate system, and the collection of acquired environmental images is located in a latitude and longitude coordinate system. Identifying the coordinates of a first road sign element in each environmental image within the target coordinate system includes: identifying multiple lane lines of a target road in each environmental image, and determining one lane line from these multiple lane lines as a road reference line; establishing a local coordinate system in each environmental image with the origin of the road reference line, and determining the coordinates of the first road sign element in the local coordinate system; projecting the first road sign element onto the road reference line in each environmental image to obtain a projection point; determining the distance from the first road sign element to the projection point as the abscissa of the first road sign element in the reference coordinate system; determining the distance from the projection point to the origin of the road reference line as the ordinate of the first road sign element in the reference coordinate system; and determining the elevation difference between the first road sign element and the projection point as the vertical coordinate of the first road sign element in the reference coordinate system, thereby obtaining the coordinates of the first road sign element in the reference coordinate system.

[0011] Since the distance from the first road sign feature to the projection point is the abscissa of the first road sign feature in the reference coordinate system; the distance from the projection point to the starting point of the road reference line is the ordinate of the first road sign feature in the reference coordinate system; and the elevation difference between the first road sign feature and the projection point is the vertical coordinate of the first road sign feature in the reference coordinate system, determining the coordinates of the first road sign feature in the local coordinate system is more in line with the road application scenario and has high accuracy.

[0012] In one possible implementation, based on the coordinates of each first landmark element in the target coordinate system and a set threshold, the first landmark elements are aggregated to obtain at least one point of interest category, including: in each environmental image, when a linear element is identified, the environmental image is divided into multiple cells at preset steps along the extension direction of the road reference line; in each environmental image, the sub-linear elements in each cell are identified as first landmark elements; and the first landmark elements located in the same cell in the set of environmental images are aggregated into one category.

[0013] Since the first landmark feature is a sub-line feature within each cell, this makes it easier to determine the location where the coordinates of the entire line feature change, resulting in high accuracy.

[0014] In one possible implementation, determining the sub-line feature within each cell as the first landmark feature in each environmental image includes: in each environmental image, when a cell contains multiple sub-line features, randomly selecting one sub-line feature as the first landmark feature; when a cell does not contain any sub-line features, determining the coordinates of the first landmark feature in that cell in the reference coordinate system as follows: Where m is the cell number, L is the preset step size, (la1, lo1, al1) is the coordinate of the first landmark element in the previous cell in the reference coordinate system, and (la2, lo2, al2) is the coordinate of the first landmark element in the next cell in the reference coordinate system.

[0015] Since linear features are continuous, if a cell does not include a child linear feature, it may be due to an error causing the child linear feature within the cell to be missing. Therefore, the coordinates of the child linear features within the cell in the reference coordinate system can be added. High reliability; when a cell contains multiple sub-linear features, randomly select one sub-linear feature as the first landmark feature to remove redundant data.

[0016] In one possible implementation, when the coordinates of a second landmark element in any point of interest category in a pre-stored map are not within the confidence interval associated with a first landmark element of the corresponding point of interest category, the coordinates of the second landmark element are updated so that the updated coordinates of the second landmark element are within the confidence interval. This includes: expanding the confidence interval by a preset margin to obtain an expanded confidence interval; and updating the coordinates of a second landmark element of any type in the pre-stored map when the coordinates of the second landmark element are not within the expanded confidence interval associated with a first landmark element of the corresponding type, so that the updated coordinates of the second landmark element are within the expanded confidence interval.

[0017] Understandably, the expanded confidence interval is more robust.

[0018] In one possible implementation, after aggregating the first landmark elements in the set of environmental images that are less than a set threshold into a class, the method further includes: denoising the outliers in each class of first landmark elements after aggregation.

[0019] Understandably, the reliability of the first landmark element is higher after denoising.

[0020] Secondly, this application provides a map updating apparatus, comprising: an information receiving unit for receiving a set of environmental images of a target road collected when a vehicle travels multiple times on the target road; a coordinate determining unit for determining the coordinates of a first road sign element in each environmental image in a target coordinate system; a data aggregation unit for aggregating each first road sign element according to its coordinates in the target coordinate system and a set threshold to obtain at least one point of interest category; wherein the distance between any two aggregated first road sign elements in each point of interest category is less than a set threshold; and a confidence interval determining unit for determining a confidence interval for the distribution of the coordinates of the first road sign elements in each point of interest category according to the Gaussian distribution of their coordinates. The map updating unit is used to update the coordinates of a second road sign element in any point of interest category in a pre-stored map when the coordinates of that second road sign element are not within the confidence interval associated with the first road sign element in the corresponding point of interest category, so that the updated coordinates of the second road sign element are within the confidence interval.

[0021] Thirdly, this application provides a server including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the server performs the method provided in the first aspect.

[0022] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the computer to perform the method provided in the first aspect.

[0023] This application provides a map updating method, apparatus, server, and storage medium. The server can aggregate the coordinates of each first road sign element collected multiple times on a target road in a target coordinate system, based on the coordinates of each first road sign element in the target coordinate system and a set threshold, to obtain at least one point of interest (POI) category. The distance between any two aggregated first road sign elements within each POI category is less than the set threshold. This allows for the determination of the coordinate set of the same first target element collected multiple times in the target coordinate system. Furthermore, based on the Gaussian distribution of the coordinates of the first road sign elements in each POI category, a confidence interval for the distribution of the coordinates of the first road sign elements in that POI category is determined. When the coordinates of any type of second road sign element in the pre-stored map are not within the confidence interval associated with the corresponding type of first road sign element, it indicates that the coordinates of this type of second road sign element have changed. Therefore, the server updates the coordinates of the second road sign element so that the updated coordinates are within the confidence interval, improving the accuracy of the coordinates of the second road sign elements in the pre-stored map. This allows vehicles to navigate using a pre-stored map containing updated road sign locations, thus reducing the likelihood of misnavigation. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 A flowchart of the map update method provided in the embodiments of this application;

[0026] Figure 2 A schematic diagram illustrating the determination of the coordinates (la, lo, al) of the first landmark element P in the reference coordinate system, provided for an embodiment of this application;

[0027] Figure 3 A schematic diagram showing the distribution of the first road sign element associated with the confidence interval M1, the confidence interval M2 obtained by expanding the confidence interval M1, and the road reference line R in the local coordinate system, provided for embodiments of this application.

[0028] Figure 4 for Figure 1 The detailed flowchart of S104 in the document;

[0029] Figure 5 for Figure 1 The detailed flowchart of S103 in the document;

[0030] Figure 6 A schematic diagram illustrating the division of the road reference line R into cells, provided in an embodiment of this application;

[0031] Figure 7 A functional block diagram of the map updating device provided in the embodiments of this application. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments made by those skilled in the art under the guidance of these embodiments are within the scope of protection of this application.

[0033] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0034] One current method for updating maps is as follows: when multiple vehicles pass the target road, each vehicle collects environmental images of the target road, and the ratio of the number of environmental images containing a first road sign feature to the total number of environmental images is used as the confidence level of the presence of the first road sign feature. When the confidence level of the presence of the first road sign feature is greater than a set threshold, the distance between the coordinates of the first road sign feature and the coordinates of a second road sign feature of the same type on the target road in a preset map is determined. When the distance is greater than a set threshold, the coordinates of the second road sign feature are updated based on the coordinates of the first road sign feature. However, the accuracy of the coordinates of the first road sign feature determined in the above method is low, resulting in low accuracy in updating the coordinates of the second road sign feature, thus, misnavigation can still occur.

[0035] Based on the aforementioned technical problems, the inventive concept of this application is as follows: the server can determine the confidence interval of the distribution of the coordinates of the first landmark elements of each interest category on the target road based on the Gaussian distribution of the coordinates of the first landmark elements of each interest category collected multiple times. When the coordinates of any type of second landmark element in the pre-stored map are not within the confidence interval associated with the first landmark element of the corresponding type, it indicates that the coordinates of this type of second landmark element have changed. The server updates the coordinates of the second landmark element so that the updated coordinates of the second landmark element are within the confidence interval, thereby improving the accuracy of the coordinates of the second landmark elements in the pre-stored map.

[0036] The technical solutions of this application and how they solve the aforementioned technical problems will be described in detail below with specific embodiments. These 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 this application will now be described with reference to the accompanying drawings.

[0037] Figure 1 This application provides a flowchart of a map update method, which is applied to a server. The server has a communicative connection with vehicles to perform data interaction. This application provides a map update method, including:

[0038] S101: The server receives environmental images of the target road each time the vehicle travels on the target road, and generates a set of environmental images.

[0039] For example, if vehicle A travels on road A (i.e., the target road) 50 times, vehicle A can capture 50 environmental images of road A, resulting in a set of environmental images. Similarly, if vehicle A travels on road B 100 times, vehicle A can capture 100 environmental images of road B, resulting in a set of environmental images.

[0040] S102: The server determines the coordinates of the first landmark element in each environmental image in the target coordinate system within the set of environmental images.

[0041] Specifically, the specific implementation of S102 includes, but is not limited to, the following two methods:

[0042] The first method involves the server identifying multiple lane lines of the target road in each environmental image and selecting one lane line from these lane lines as the road reference line. For example, the server can choose the lane line with the best integrity and longest length from the multiple lane lines of the target road as the road reference line.

[0043] In each environmental image, the server establishes a local coordinate system with the starting point of the road reference line as the origin, and determines the coordinates of the first road sign element in the local coordinate system.

[0044] In this system, the direction due east of the starting point of the road reference line is the X-axis of the local coordinate system, the direction due north of the starting point of the road reference line is the Y-axis of the local coordinate system, and the vertical downward direction of the starting point of the road reference line is the Z-axis of the local coordinate system.

[0045] Since the origin of the local coordinate system is the starting point of the road reference line, the X-axis of the local coordinate system is located due east of the starting point of the road reference line, the Y-axis is located due north of the starting point of the road reference line, and the Z-axis is located vertically downward from the starting point of the road reference line. In this way, determining the coordinates of the first road sign element in the local coordinate system is more in line with the road application scenario and has high accuracy.

[0046] The second method uses the target coordinate system as the reference coordinate system. The server identifies multiple lane lines of the target road in each environmental image and determines one lane line as the road reference line from among them. For example, the lane line with the best integrity and longest length can be selected from the multiple lane lines of the target road as the road reference line.

[0047] In each environmental image, the server establishes a local coordinate system with the starting point of the road reference line as the origin and determines the coordinates of the first road sign feature in the local coordinate system. In each environmental image, the server projects the first road sign feature onto the road reference line to obtain a projection point. The server determines the distance from the first road sign feature to the projection point as the abscissa of the first road sign feature in the reference coordinate system. The distance from the projection point to the starting point of the road reference line determines the ordinate of the first road sign feature in the reference coordinate system, and the elevation difference between the first road sign feature and the projection point determines the vertical coordinate of the first road sign feature in the reference coordinate system, thus obtaining the coordinates of the first road sign feature in the reference coordinate system.

[0048] like Figure 2 As shown, the server establishes a local coordinate system XOY with the origin O of the road reference line R, and determines the coordinates (x1, y1, z1) of the first road sign element P in the local coordinate system. The first road sign element P is then projected onto the road reference line R to obtain the projection point Pt. The server determines the distance la from the first road sign element P to the projection point Pt as the abscissa of the first road sign element P in the reference coordinate system. The distance lo from the projection point Pt to the origin O of the road reference line R is determined as the ordinate of the first road sign element P in the reference coordinate system. The elevation difference between the first road sign element P and the projection point Pt is determined as the vertical coordinate of the first road sign element P in the reference coordinate system, thus obtaining the coordinates (la, lo, al) of the first road sign element P in the reference coordinate system.

[0049] Understandably, the distance from the first road sign feature to the projection point is the abscissa of the first road sign feature in the reference coordinate system; the distance from the projection point to the starting point of the road reference line is the ordinate of the first road sign feature in the reference coordinate system; and the elevation difference between the first road sign feature and the projection point is the vertical coordinate of the first road sign feature in the reference coordinate system. Therefore, determining the coordinates of the first road sign feature in the local coordinate system is more consistent with the road surface application scenario and has high accuracy.

[0050] S103: The server aggregates each first landmark element according to its coordinates in the target coordinate system and a set threshold to obtain at least one point of interest category; wherein the distance between any two first landmark elements aggregated in each point of interest category is less than the set threshold.

[0051] Understandably, the set of first landmark features in the set of environmental images that are less than a set threshold may be the same first landmark feature in environmental images collected multiple times from the target road.

[0052] Furthermore, the server can denoise outliers in each class of first landmark features after aggregation, making the reliability of the first landmark features in the denoised class higher.

[0053] S104: The server determines the confidence interval of the distribution of the coordinates of the first landmark element of each point of interest category based on the Gaussian distribution of the coordinates of the first landmark element of each point of interest category.

[0054] For example, the confidence interval [edw, e+d+w] is determined for the distribution of the coordinates of the first landmark element corresponding to the point of interest category.

[0055] Understandably, the confidence interval of the distribution of the coordinates of the first landmark element is used to indicate the interval in which the distribution of the coordinates of the first landmark element reaches a preset proportion, wherein the preset proportion can be 95%, 99.7%, etc., and is not limited here.

[0056] Specifically, the server expands the confidence interval by a preset margin to obtain an expanded confidence interval. For example, the expanded confidence interval could be [edw, e+d+w]. Understandably, the expanded confidence interval is more robust.

[0057] S105: When the coordinates of the second landmark element in any point of interest category in the pre-stored map are not within the confidence interval associated with the first landmark element of the corresponding point of interest category, the server updates the coordinates of the second landmark element so that the updated coordinates of the second landmark element are within the confidence interval.

[0058] Specifically, when the confidence interval is expanded, S105 can be implemented as follows: when the coordinates of any type of second landmark feature in the pre-stored map are not within the expanded confidence interval associated with the corresponding type of first landmark feature, the server updates the coordinates of the second landmark feature so that the updated coordinates of the second landmark feature are within the expanded confidence interval.

[0059] like Figure 3 As shown, the average coordinates of multiple first landmark features N1 of the same specified type after aggregation are e. The first landmark features are associated with a confidence interval M1 (i.e., [ed, e+d]). Confidence interval M1 is expanded to obtain confidence interval M2 (i.e., [edw, e+d+w]). The coordinates of second landmark features N2 of the specified type in the pre-stored map are also included. Figure 3 As can be seen from the data, the coordinates of the second landmark element N2 of the specified type in the pre-stored map are not within the expanded confidence interval M2 associated with the first landmark element. Therefore, the coordinates of the second landmark element N2 are updated so that the updated coordinates of the second landmark element are within the expanded confidence interval.

[0060] In summary, the map updating method provided in this application allows a server to aggregate first landmark elements based on their coordinates in the target coordinate system and a set threshold to obtain at least one point of interest (POI) category. The distance between any two aggregated first landmark elements within each POI category is less than the set threshold. This allows for the determination of the coordinate set of the same first target element collected multiple times in the target coordinate system. Furthermore, based on the Gaussian distribution of the coordinates of the first landmark elements in each POI category, a confidence interval for the distribution of coordinates of the first landmark elements in that POI category is determined. If the coordinates of any type of second landmark element in the pre-stored map are not within the confidence interval associated with the corresponding type of first landmark element, it indicates that the coordinates of this type of second landmark element have changed. Therefore, the server updates the coordinates of the second landmark element to ensure that the updated coordinates are within the confidence interval, thus improving the accuracy of the second landmark element coordinates in the pre-stored map. This allows vehicles to navigate using a pre-stored map containing the updated landmark element locations, reducing the likelihood of misnavigation.

[0061] In one possible implementation, such as Figure 4 As shown, S104 can be specifically implemented as follows:

[0062] S401: The server uses a Gaussian distribution model to determine the mean e and standard deviation σ of the coordinates of the first landmark feature for each point of interest category.

[0063] S402: The server determines the minimum value m of x in the candidate interval (μ-xσ, μ+xσ) such that the distribution of the coordinates of the first landmark element of each interest category reaches a preset proportion, and determines the confidence range (μ-mσ, μ+mσ), where μ is the mathematical expectation of the Gaussian distribution model.

[0064] For example, when the preset ratio is 95% and x = 1, the candidate interval is (μ-σ, μ+σ); when the preset ratio is 99.7% and x = 3, the candidate interval is (μ-3σ, μ+3σ).

[0065] S403: The server determines the confidence interval (e-[μ-mσ], e+[μ+mσ]) of the distribution of coordinates of the first landmark element corresponding to the point of interest category based on the confidence range (μ-mσ, μ+mσ).

[0066] Understandably, the confidence intervals (e-[μ-mσ], e+[μ+mσ]) determined by the server using the mean e and standard deviation σ of the coordinates of the first landmark feature of each point of interest category obtained by the Gaussian distribution model are highly reliable.

[0067] Figure 4 A flowchart of another map update method is provided for embodiments of this application, in the above... Figure 1 Based on the illustrated embodiments, as Figure 5 As shown, one specific implementation of S103 above is as follows:

[0068] S501: In each environmental image, when the server identifies linear features, it divides the environmental image into multiple cells at preset steps along the extension direction of the road reference line.

[0069] like Figure 6 As shown, the environmental image is divided into multiple cells at preset step lengths L along the extension direction of the road reference line R. The preset step length L is between 1m and 10m to avoid information loss and distortion of linear features when the preset step length is too large; or to avoid information overfitting leading to uneven linear features and high computational load when the preset step length is too small.

[0070] S502: In each environment image, the server determines the sub-linear features within each cell as the first landmark feature.

[0071] Specifically, for any sub-linear feature, the cell number n to which it belongs is determined based on the vertical coordinate lo of the sub-linear feature and its formula n = lo / L.

[0072] S503: The server aggregates the first landmark features located in the same cell in the collection of environmental images into one category.

[0073] Based on S501-S503, since the first landmark element is a sub-line element within each cell, this facilitates the determination of the location where the coordinates of the entire line element change, resulting in high accuracy.

[0074] Specifically, S503 can be implemented as follows:

[0075] In each environment image, when a cell contains multiple sub-linear features, a sub-linear feature is randomly selected as the first landmark feature.

[0076] When one of the cells does not include sub-line features, the coordinates of the first landmark feature within that cell in the reference coordinate system are determined to be...

[0077] Where m is the cell number, L is the preset step size, (la1, lo1, al1) is the coordinate of the first landmark element in the previous cell in the reference coordinate system, and (la2, lo2, al2) is the coordinate of the first landmark element in the next cell in the reference coordinate system.

[0078] Since linear features are continuous, if a cell does not include a child linear feature, it may be due to an error causing the child linear feature within the cell to be missing. Therefore, the coordinates of the child linear features within the cell in the reference coordinate system can be added. High reliability; when a cell contains multiple sub-linear features, randomly select one sub-linear feature as the first landmark feature to remove redundant data.

[0079] Please see Figure 7 This application provides a map updating device 700. It should be noted that the basic principle and technical effects of the map updating device 700 provided in this application are the same as those in the above embodiments. For the sake of brevity, any parts not mentioned in this application can be referred to the corresponding content in the above embodiments. The map updating device 700 includes an information receiving unit 701, a coordinate determining unit 702, a data aggregation unit 703, a confidence interval determining unit 704, and a map updating unit 705.

[0080] The information receiving unit 701 is used to receive a collection of environmental images of the target road collected when the vehicle travels on the target road multiple times.

[0081] The coordinate determination unit 702 is used to determine the coordinates of the first landmark element in each environmental image in the target coordinate system.

[0082] The data aggregation unit 703 is used to aggregate each first landmark element according to the coordinates of each first landmark element in the target coordinate system and a set threshold to obtain at least one interest point category; wherein the distance between any two first landmark elements aggregated in each interest point category is less than the set threshold.

[0083] The confidence interval determination unit 704 is used to determine the confidence interval of the distribution of the coordinates of the first landmark element of the corresponding interest point category based on the Gaussian distribution of the coordinates of the first landmark element of each interest point category.

[0084] Map update unit 705 is used to update the coordinates of the second landmark element in any point of interest category in the pre-stored map when the coordinates of the second landmark element are not within the confidence interval associated with the first landmark element of the corresponding point of interest category, so that the updated coordinates of the second landmark element are within the confidence interval.

[0085] In one possible implementation, the confidence interval determination unit 704 is specifically used for the server to determine the mean e and standard deviation σ of the coordinates of the first landmark element of each point of interest category using a Gaussian distribution model; the server to determine the minimum value m of x in the candidate interval (μ-xσ, μ+xσ) such that the distribution of the coordinates of the first landmark element of each point of interest category reaches a preset proportion, thereby determining the confidence range (μ-mσ, μ+mσ), where μ is the mathematical expectation of the Gaussian distribution model; and the server to determine the confidence interval (e-[μ-mσ], e+[μ+mσ]) of the distribution of the coordinates of the first landmark element of the corresponding point of interest category based on the confidence range (μ-mσ, μ+mσ).

[0086] In one possible implementation, the target coordinate system is a local coordinate system, and the collection of acquired environmental images is located in a latitude and longitude coordinate system. The coordinate determination unit 702 is specifically used to identify multiple lane lines of the target road in each environmental image, and determine one lane line from the multiple lane lines of the target road as a road reference line; in each environmental image, a local coordinate system is established with the starting point of the road reference line as the origin, and the coordinates of the first road sign element in the local coordinate system are determined, wherein the eastward direction along the starting point of the road reference line is the X-axis of the local coordinate system, the northward direction along the starting point of the road reference line is the Y-axis of the local coordinate system, and the vertically downward direction along the starting point of the road reference line is the Z-axis of the local coordinate system.

[0087] In one possible implementation, the target coordinate system is a reference coordinate system, and the collection of acquired environmental images is in a latitude and longitude coordinate system. The coordinate determination unit 702 is specifically used to identify multiple lane lines of the target road in each environmental image, and determine one lane line from the multiple lane lines of the target road as a road reference line; in each environmental image, a local coordinate system is established with the starting point of the road reference line as the origin, and the coordinates of the first road sign element in the local coordinate system are determined; in each environmental image, the server projects the first road sign element onto the road reference line to obtain a projection point; the distance from the first road sign element to the projection point is determined as the abscissa of the first road sign element in the reference coordinate system; the distance from the projection point to the starting point of the road reference line is determined as the ordinate of the first road sign element in the reference coordinate system; and the elevation difference between the first road sign element and the projection point is determined as the vertical coordinate of the first road sign element in the reference coordinate system, thereby obtaining the coordinates of the first road sign element in the reference coordinate system.

[0088] In one possible implementation, the data aggregation unit 703 is specifically used to divide the environmental image into multiple cells at preset steps along the extension direction of the road reference line when a linear feature is identified in each environmental image; in each environmental image, the sub-linear feature in each cell is determined as a first road sign feature; and the first road sign features located in the same cell in the set of environmental images are aggregated into one category.

[0089] In one possible implementation, the data aggregation unit 703 is specifically configured to, in each environmental image, randomly select one sub-linear feature as the first landmark feature when one cell contains multiple sub-linear features; and when one cell does not contain any sub-linear features, determine the coordinates of the first landmark feature in that cell in the reference coordinate system as follows: Where m is the cell number, L is the preset step size, (la1, lo1, al1) is the coordinate of the first landmark element in the previous cell in the reference coordinate system, and (la2, lo2, al2) is the coordinate of the first landmark element in the next cell in the reference coordinate system.

[0090] In one possible implementation, the apparatus 700 provided in this application embodiment may further include: a confidence interval expansion unit, configured to expand the confidence interval by a preset range to obtain an expanded confidence interval; when the coordinates of any type of second landmark element in the pre-stored map are not within the expanded confidence interval associated with the corresponding type of first landmark element, the coordinates of the second landmark element are updated so that the updated coordinates of the second landmark element are within the expanded confidence interval.

[0091] In one possible implementation, the apparatus 700 provided in this application embodiment may further include: a data denoising unit, used to denoise outliers in each type of first landmark element after aggregation.

[0092] This application also provides a server, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it causes the server to perform the method provided in the above embodiments of this application.

[0093] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the computer to perform the method provided in the above embodiments of this application.

[0094] This application also provides a computer program product, including a computer program that, when run, causes a computer to perform the methods provided in the above embodiments.

[0095] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0096] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A map updating method, characterized in that, The method includes: Receive environmental images of the target road acquired each time the vehicle travels on the target road, and generate a set of environmental images; The coordinates of the first road sign element in each of the environmental images in the set of environmental images are determined in the target coordinate system; the target coordinate system is a coordinate system constructed based on road reference lines; Based on the coordinates of each first landmark element in the target coordinate system and a set threshold, the first landmark elements are aggregated to obtain at least one point of interest category; wherein the distance between any two first landmark elements aggregated in each point of interest category is less than the set threshold. Based on the Gaussian distribution of the coordinates of the first landmark element of each point of interest category, determine the confidence interval of the distribution of the coordinates of the first landmark element of the corresponding point of interest category; If the coordinates of a second landmark element in any point of interest category in the pre-stored map are not within the confidence interval associated with the first landmark element of the corresponding point of interest category, the coordinates of the second landmark element are updated so that the updated coordinates of the second landmark element are within the confidence interval.

2. The method according to claim 1, characterized in that, The step of determining the confidence interval of the distribution of the coordinates of the first landmark element of each point of interest category based on the Gaussian distribution of the coordinates of the first landmark element of each point of interest category includes: Using a Gaussian distribution model, the mean e and standard deviation σ of the coordinates of the first landmark feature for each point of interest category are determined; In the candidate interval (μ-xσ, μ+xσ), the minimum value m of x is determined such that the distribution of the coordinates of the first landmark element of each interest point category reaches a preset proportion, so as to determine the confidence range (μ-mσ, μ+mσ), where μ is the mathematical expectation of the Gaussian distribution model; Based on the confidence range (μ-mσ, μ+mσ), determine the confidence interval (e-[μ-mσ], e+[μ+mσ]) for the distribution of coordinates of the first landmark feature corresponding to the point of interest category.

3. The method according to claim 1, characterized in that, The target coordinate system is a local coordinate system, and the collection of acquired environmental images is located in a latitude and longitude coordinate system. The coordinates of the first landmark element in each environmental image in the target coordinate system are identified, including: Identify multiple lane lines of the target road in each of the environmental images, and determine one lane line from the multiple lane lines of the target road as a road reference line; In each environmental image, a local coordinate system is established with the starting point of the road reference line as the origin, and the coordinates of the first road sign element in the local coordinate system are determined. The X-axis of the local coordinate system is the eastward direction along the starting point of the road reference line, the Y-axis is the northward direction along the starting point of the road reference line, and the Z-axis is the vertically downward direction along the starting point of the road reference line.

4. The method according to claim 1, characterized in that, The target coordinate system is the reference coordinate system, and the collection of acquired environmental images is located in a latitude and longitude coordinate system. The coordinates of the first landmark element in each environmental image in the target coordinate system are identified, including: Identify multiple lane lines of the target road in each of the environmental images, and determine one lane line from the multiple lane lines of the target road as a road reference line; In each of the environmental images, a local coordinate system is established with the starting point of the road reference line as the origin, and the coordinates of the first road sign element in the local coordinate system are determined. In each of the environmental images, the first road sign element is projected onto the road reference line to obtain a projection point; The distance from the first road sign element to the projection point is determined as the abscissa of the first road sign element in the reference coordinate system; the distance from the projection point to the starting point of the road reference line is determined as the ordinate of the first road sign element in the reference coordinate system; and the elevation difference between the first road sign element and the projection point is determined as the vertical coordinate of the first road sign element in the reference coordinate system, so as to obtain the coordinates of the first road sign element in the reference coordinate system.

5. The method according to claim 4, characterized in that, The step of aggregating each first landmark element according to its coordinates in the target coordinate system and a set threshold to obtain at least one point of interest category includes: In each of the environmental images, when a linear feature is identified, the environmental image is divided into multiple cells at preset steps along the extension direction of the road reference line; In each of the environmental images, the sub-linear features within each cell are identified as first landmark features; The first landmark elements located in the same cell within the set of environmental images are grouped into one category.

6. The method according to claim 5, characterized in that, In each environment image, the sub-linear features within each cell are identified as the first waypoint features, including: In each of the environmental images, when one cell contains multiple sub-line features, a sub-line feature is randomly selected as the first landmark feature; when one cell does not contain any sub-line features, the coordinates of the first landmark feature in that cell in the reference coordinate system are determined to be ( ). , L, ); Where m is the cell number, L is the preset step size, (la1, lo1, al1) are the coordinates of the first landmark element in the previous cell in the reference coordinate system, and (la2, lo2, al2) are the coordinates of the first landmark element in the next cell in the reference coordinate system.

7. The method according to any one of claims 1-6, characterized in that, When the coordinates of a second landmark feature in any point of interest category in the pre-stored map are not within the confidence interval associated with the first landmark feature of the corresponding point of interest category, the coordinates of the second landmark feature are updated so that the updated coordinates of the second landmark feature are within the confidence interval, including: The confidence interval is expanded by a preset range to obtain the expanded confidence interval; When the coordinates of a second landmark feature of any type in the pre-stored map are not within the expanded confidence interval associated with the first landmark feature of the corresponding type, the coordinates of the second landmark feature are updated so that the updated coordinates of the second landmark feature are within the expanded confidence interval.

8. The method according to any one of claims 1-6, characterized in that, After grouping the first landmark features in the set of environmental images whose distance is less than a set threshold into one category, the method further includes: Denoise out of the outliers in each category of first landmark features after aggregation.

9. A map updating device, characterized in that, The device includes: The information receiving unit is used to receive a collection of environmental images of the target road collected when the vehicle travels on the target road multiple times. A coordinate determination unit is used to determine the coordinates of the first road sign element in each of the environmental images in the target coordinate system; the target coordinate system is a coordinate system constructed based on road reference lines; The data aggregation unit is used to aggregate each of the first landmark elements according to the coordinates of each of the first landmark elements in the target coordinate system and a set threshold, so as to obtain at least one point of interest category; wherein the distance between any two first landmark elements aggregated in each point of interest category is less than the set threshold. The confidence interval determination unit is used to determine the confidence interval of the distribution of the coordinates of the first landmark element of each interest point category based on the Gaussian distribution of the coordinates of the first landmark element of each interest point category. The map update unit is used to update the coordinates of the second landmark element when the coordinates of the second landmark element in any point of interest category in the pre-stored map are not within the confidence interval associated with the first landmark element of the corresponding point of interest category, so that the updated coordinates of the second landmark element are within the confidence interval.

10. A server comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it causes the server to perform the method as described in any one of claims 1 to 8.

11. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it causes the computer to perform the method as described in any one of claims 1 to 8.

Images