Map data updating method and electronic device

Abstract

The disclosed embodiments disclose a map data update method and electronic device. The method comprises: obtaining geographic data of a target area, the geographic data comprising geographic point cloud data; extracting vectors of map elements to be updated within the target area based on the geographic data; aligning the target area with a corresponding area in a pre-stored high-precision map, performing element differentiation between the vectors of map elements to be updated and the vectors of existing map elements within the corresponding area to determine whether map elements within the target area have changed; and if map elements within the target area have changed, performing a map data update. This technical solution can improve the accuracy and timeliness of high-precision map updates.

Description

Technical Field

[0001] The present disclosure relates to the technical field of map production, and more particularly to a map data updating method and electronic device. Background Art

[0002] With the advancement of technology, advanced driver assistance and autonomous driving have become hot topics in research because they can help drivers drive safely. Currently, one of the technical implementation paths for these two technologies relies on high-precision maps. Compared to standard maps, HD maps offer a more detailed and accurate representation of the real world. Large-scale, fresh, and high-quality updates of HD maps are crucial in this field, determining the functional safety of maps used during driving. Due to the high cost and limited availability of specialized HD map collection vehicles, large-scale geographic data collection and updates are currently unavailable. Therefore, HD map updates primarily utilize front-mounted or rear-mounted vision and positioning sensors on crowdsourcing vehicles to capture road images and corresponding location information. Image processing techniques are then used to identify and extract HD map features from the captured road images and calculate their 3D coordinates. These features are then differentially identified with existing HD map data to update the geographic data. However, this type of method has the following shortcomings: 1) The ranging accuracy of monocular images collected by a single camera is not high, and the accuracy of extracted map elements is poor. Fine differentiation cannot be achieved, making it difficult to meet the accuracy requirements of high-precision map updates. Small changes cannot be detected, which impairs the quality of map updates; 2) Visual sensors are sensitive to light. Insufficient light (such as at night) causes the system to fail, making it difficult to meet the freshness requirements of high-precision maps (such as daily updates). Summary of the Invention

[0003] Embodiments of the present disclosure provide a map data updating method and an electronic device.

[0004] In a first aspect, an embodiment of the present disclosure provides a method for updating map data.

[0005] Specifically, the map data updating method includes:

[0006] Acquiring geographic data of a target area, wherein the geographic data includes geographic point cloud data;

[0007] Extracting the map element vector to be updated within the target area based on the geographic data;

[0008] After aligning the target area with the corresponding area in the pre-stored high-precision map, performing element difference between the map element vector to be updated and the existing map element vector in the corresponding area to determine whether the map element in the target area has changed;

[0009] If the map elements in the target area change, map data updating processing is performed.

[0010] In a possible implementation, the method is applied to an update device, and if the map elements in the target area change, performing map data update processing includes:

[0011] If a map element within the target area changes, determining a change confidence level of the changed element;

[0012] If the change confidence is greater than a preset threshold, the map element vector to be updated corresponding to the changed element is sent to the server;

[0013] If the change confidence is less than or equal to a preset threshold, the map element vector and geographic data to be updated corresponding to the changed element are sent to the server.

[0014] In a possible implementation, the method is applied to a server, and if a map element in the target area changes, performing map data update processing includes:

[0015] If a map element within the target area changes, determining a change confidence level of the changed element;

[0016] If the change confidence is greater than a preset threshold, updating the map based on the to-be-updated map element vector corresponding to the changed element;

[0017] If the change confidence is less than or equal to a preset threshold, the map element vector to be updated and the original data corresponding to the changed element are output; in response to receiving an input update instruction, the map is updated based on the update instruction, where the update instruction is used to instruct to update the map based on the map element vector to be updated corresponding to the changed element or to update the map based on the changed element vector input by the user.

[0018] In a possible implementation, obtaining geographic data of the target area includes:

[0019] Obtain the original point cloud data of the target area;

[0020] Points far from roads and dynamic object points in the original point cloud data are filtered to obtain geographic point cloud data of the target area.

[0021] In a possible implementation, the raw data further includes real-time trajectory height, and filtering the off-road points and dynamic object points in the raw point cloud data to obtain geographic point cloud data includes:

[0022] Determining the ground elevation based on the actual trajectory height and a pre-stored device height above the ground;

[0023] Determine the points in the original point cloud data whose heights are within a preset range from the ground elevation as ground points, and determine the remaining points as non-ground points;

[0024] Determining the road boundaries of the target area based on pre-stored high-precision map data;

[0025] Extending the road boundary outward by a first preset distance as a lateral effective boundary, and extending the ground elevation upward by a second preset distance as a longitudinal effective boundary;

[0026] Filtering out useless points above the longitudinal effective boundary and useless points outside the transverse effective boundary;

[0027] Dynamic object points whose heights exceed a third preset distance from the ground elevation among the non-ground points are filtered out.

[0028] In a possible implementation, the geographic data further includes image data, and extracting the map element vector to be updated within the target area based on the geographic data includes:

[0029] Extracting map elements based on the geographic point cloud data to obtain a first map element vector within the target area;

[0030] Extracting map elements based on the geographic point cloud data and the image data to obtain a second map element vector within the target area;

[0031] The first map element vector and the second map element vector are fused to obtain the map element vector to be updated.

[0032] In a possible implementation, extracting map elements based on the geographic point cloud data to obtain a first map element vector within the target area includes:

[0033] generating a point cloud intensity map based on intensity information of ground points in the geographic point cloud data;

[0034] generating a point cloud density map based on density information corresponding to non-ground points in the geographic point cloud data, wherein the density information includes density information of the non-ground points when projected onto the ground;

[0035] Extracting a first map element of the target area from the point cloud intensity map and the point cloud density map;

[0036] Boundary fitting of the first map element is performed based on geographic point cloud data corresponding to the first map element to obtain the first map element vector.

[0037] In a possible implementation, extracting map elements based on the geographic point cloud data and the image data to obtain a second map element vector within the target area includes:

[0038] performing image segmentation and recognition on the image data to obtain a second map element;

[0039] Based on the correspondence between the image data and the geographic point cloud data, obtaining the geographic point cloud data corresponding to the second map element;

[0040] Boundary fitting of the second map element is performed based on geographic point cloud data corresponding to the second map element to obtain the second map element vector.

[0041] In a possible implementation, the fusing the first map element vector and the second map element vector to obtain the map element vector to be updated includes:

[0042] Calculating a first fitting residual of the first map element vector and a second fitting residual of the second map element vector;

[0043] determining a first confidence level of the first map element vector based on the first fitting residual, and determining a second confidence level of the second map element vector based on the second fitting residual;

[0044] Based on the first confidence level and the second confidence level, the first map element vector and the second map element vector are fused to obtain the map element vector to be updated.

[0045] In a possible implementation, determining the change confidence of the change factor includes:

[0046] The change confidence of the changed element is determined based on the confidence of the to-be-updated map element vector corresponding to the changed element, the type of the changed element, and the change type, where the change type includes addition and deletion.

[0047] In a second aspect, an embodiment of the present disclosure provides a method for updating map data.

[0048] Specifically, the map data updating method includes:

[0049] Receive map information of changed elements sent by the update device;

[0050] If the map information includes a map element vector to be updated, performing a map update based on the map element vector to be updated corresponding to the changed element;

[0051] If the map information includes map element vectors to be updated, geographic point cloud data, and image data, the map element vectors, geographic point cloud data, and image data to be updated are displayed; in response to receiving an input update instruction, the map is updated based on the update instruction, and the update instruction is used to instruct the map to be updated based on the map element vectors to be updated corresponding to the changed elements or based on the changed element vectors input by the user.

[0052] In a third aspect, an embodiment of the present disclosure provides a method for updating map data.

[0053] Specifically, the map data updating method includes:

[0054] The updating device obtains geographic data of a target area, extracts map element vectors to be updated within the target area based on the geographic data, aligns the target area with a corresponding area in a pre-stored high-precision map, performs element difference between the map element vectors to be updated and existing map element vectors within the corresponding area, and determines whether map elements within the target area have changed; if a map element within the target area has changed, transmits the map element vectors to be updated of the changed elements to a server;

[0055] The server performs map data update processing based on the map element vector to be updated of the changed element.

[0056] In a third aspect, an embodiment of the present disclosure provides a map data updating device.

[0057] Specifically, the map data updating device includes:

[0058] A first acquisition module is configured to acquire geographic data of a target area, wherein the geographic data includes geographic point cloud data;

[0059] an extraction module configured to extract the map element vectors to be updated within the target area based on the geographic data;

[0060] a difference module configured to align the target area with a corresponding area in a pre-stored high-precision map, and then perform element difference between the map element vector to be updated and the existing map element vector in the corresponding area to determine whether the map element in the target area has changed;

[0061] The first updating module is configured to perform map data updating processing if the map elements in the target area change.

[0062] In a possible implementation, the apparatus is applied to an update device, and the first update module is configured to:

[0063] If a map element within the target area changes, determining a change confidence level of the changed element;

[0064] If the change confidence is greater than a preset threshold, the map element vector to be updated corresponding to the changed element is sent to the server;

[0065] If the change confidence is less than or equal to a preset threshold, the map element vector and geographic data to be updated corresponding to the changed element are sent to the server.

[0066] In a possible implementation, the apparatus is applied to a server, and the first update module is configured to:

[0067] If a map element within the target area changes, determining a change confidence level of the changed element;

[0068] If the change confidence is greater than a preset threshold, updating the map based on the to-be-updated map element vector corresponding to the changed element;

[0069] If the change confidence is less than or equal to a preset threshold, the map element vector to be updated and the geographic data corresponding to the changed element are output; in response to receiving an input update instruction, the map is updated based on the update instruction, and the update instruction is used to instruct to update the map based on the map element vector to be updated corresponding to the changed element or to update the map based on the changed element vector input by the user.

[0070] In a possible implementation, the acquisition module is configured to:

[0071] Obtain the original point cloud data of the target area;

[0072] Points far from roads and dynamic object points in the original point cloud data are filtered to obtain geographic point cloud data.

[0073] In a possible implementation, the raw data further includes real-time trajectory height, and the portion of the acquisition module that filters out points far from roads and dynamic object points in the raw point cloud data to obtain geographic point cloud data is configured as follows:

[0074] Determining the ground elevation based on the actual trajectory height and a pre-stored device height above the ground;

[0075] Determine the points in the original point cloud data whose heights are within a preset range from the ground elevation as ground points, and determine the remaining points as non-ground points;

[0076] Determining the road boundaries of the target area based on pre-stored high-precision map data;

[0077] Extending the road boundary outward by a first preset distance as a lateral effective boundary, and extending the ground elevation upward by a second preset distance as a longitudinal effective boundary;

[0078] Filtering out useless points above the longitudinal effective boundary and useless points outside the transverse effective boundary;

[0079] Dynamic object points whose heights exceed a third preset distance from the ground elevation among the non-ground points are filtered out.

[0080] In a possible implementation, the geographic data further includes image data, and the extraction module is configured to:

[0081] Extracting map elements based on the geographic point cloud data to obtain a first map element vector within the target area;

[0082] Extracting map elements based on the geographic point cloud data and the image data to obtain a second map element vector within the target area;

[0083] The first map element vector and the second map element vector are fused to obtain the map element vector to be updated.

[0084] In a possible implementation, the part of the extraction module that extracts map elements based on the geographic point cloud data to obtain the first map element vector within the target area is configured as follows:

[0085] generating a point cloud intensity map based on intensity information of ground points in the geographic point cloud data;

[0086] generating a point cloud density map based on density information corresponding to non-ground points in the geographic point cloud data, wherein the density information includes density information of the non-ground points when projected onto the ground;

[0087] Extracting a first map element of the target area from the point cloud intensity map and the point cloud density map;

[0088] Boundary fitting of the first map element is performed based on geographic point cloud data corresponding to the first map element to obtain the first map element vector.

[0089] In a possible implementation, the portion of the extraction module that extracts map elements based on the geographic point cloud data and the image data to obtain the second map element vector within the target area is configured as follows:

[0090] performing image segmentation and recognition on the image data to obtain a second map element;

[0091] Based on the correspondence between the image data and the geographic point cloud data, obtaining the geographic point cloud data corresponding to the second map element;

[0092] Boundary fitting of the second map element is performed based on geographic point cloud data corresponding to the second map element to obtain the second map element vector.

[0093] In a possible implementation, the part of the extraction module that fuses the first map element vector and the second map element vector to obtain the map element vector to be updated is configured as follows:

[0094] Calculating a first fitting residual of the first map element vector and a second fitting residual of the second map element vector;

[0095] determining a first confidence level of the first map element vector based on the first fitting residual, and determining a second confidence level of the second map element vector based on the second fitting residual;

[0096] Based on the first confidence level and the second confidence level, the first map element vector and the second map element vector are fused to obtain the map element vector to be updated.

[0097] In a possible implementation, the part of the first updating module that determines the change confidence of the change element is configured as follows:

[0098] The change confidence of the changed element is determined based on the confidence of the to-be-updated map element vector corresponding to the changed element, the type of the changed element, and the change type, where the change type includes addition and deletion.

[0099] In a fourth aspect, an embodiment of the present disclosure provides a map data updating device.

[0100] Specifically, the map data updating device includes:

[0101] a second acquisition module configured to acquire map information of changed elements sent by the update device;

[0102] The second update module is configured to, if the map information includes a map element vector to be updated, perform a map update based on the map element vector to be updated corresponding to the changed element; if the map information includes a map element vector to be updated, geographic point cloud data, and image data, display the map element vector to be updated, geographic point cloud data, and image data; and in response to receiving an input update instruction, perform a map update based on the update instruction, wherein the update instruction is used to instruct a map update based on the map element vector to be updated corresponding to the changed element or based on a changed element vector input by a user.

[0103] In a fifth aspect, an embodiment of the present disclosure provides an electronic device, comprising a memory and a processor, wherein the memory is used to store one or more computer instructions supporting the above-mentioned device to execute the above-mentioned method, and the processor is configured to execute the computer instructions stored in the memory.

[0104] In a sixth aspect, an embodiment of the present disclosure provides a computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the method steps described in any of the above aspects.

[0105] In a seventh aspect, an embodiment of the present disclosure provides a computer program product, comprising a computer program / instruction, wherein the computer program / instruction, when executed by a processor, implements the method steps described in any of the above aspects.

[0106] In an eighth aspect, an embodiment of the present disclosure provides a navigation method, wherein a navigation route calculated at least based on a starting point, an end point and road conditions is obtained based on an electronic map, and navigation guidance is performed on the carrier based on the navigation route, and the electronic map is implemented based on any one of the above methods.

[0107] The technical solutions provided by the embodiments of the present disclosure may have the following beneficial effects:

[0108] The above technical solution can use lidar as the main sensor for data collection for high-precision map updates, and can obtain high-precision point cloud data. Compared with the measurement method based on monocular vision, map feature vector extraction based on point cloud data can greatly improve the accuracy of map feature vectors, so that when the map is updated based on this, the accuracy of the high-precision map will not be reduced; at the same time, the fine feature vector extraction can also help to achieve fine feature differentiation, which can discover detailed changes in the map and improve the quality of map updates; and the lidar measurement does not depend on the lighting environment. This feature improves the availability of the update system and significantly improves the timeliness of updates for night changes.

[0109] It should be understood that the foregoing general description and the following detailed description are merely exemplary and explanatory and are not restrictive of the embodiments of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0110] Other features, objectives and advantages of the embodiments of the present disclosure will become more apparent through the following detailed description of non-limiting embodiments in conjunction with the accompanying drawings. In the accompanying drawings:

[0111] Figure 1 The following is a schematic diagram of a map data update scenario.

[0112] Figure 2A schematic diagram illustrating a map data updating system according to an embodiment of the present disclosure is shown.

[0113] Figure 3 A schematic structural diagram of an updating device according to an embodiment of the present disclosure is shown.

[0114] Figure 4 A flowchart of a map data updating method according to an embodiment of the present disclosure is shown.

[0115] Figure 5 A flowchart of a map data updating method according to an embodiment of the present disclosure is shown.

[0116] Figure 6 A flowchart of a map data updating method according to an embodiment of the present disclosure is shown.

[0117] Figure 7 A structural block diagram of a map data updating device according to an embodiment of the present disclosure is shown.

[0118] Figure 8 A structural block diagram of a map data updating device according to an embodiment of the present disclosure is shown.

[0119] Figure 9 A structural block diagram of an electronic device according to an embodiment of the present disclosure is shown.

[0120] Figure 10 It is a structural diagram of a computer system suitable for implementing the method according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0121] Hereinafter, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily implement them. In addition, for the sake of clarity, parts not related to the description of the exemplary embodiments are omitted in the accompanying drawings.

[0122] In the embodiments of the present disclosure, it should be understood that terms such as "including" or "having" are intended to indicate the existence of features, numbers, steps, behaviors, components, parts, or a combination thereof disclosed in this specification, and are not intended to exclude the possibility of one or more other features, numbers, steps, behaviors, components, parts, or a combination thereof existing or being added.

[0123] It should also be noted that, in the absence of conflict, the embodiments and features of the embodiments of the present disclosure can be combined with each other. The embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0124] Figure 1 This is a schematic diagram of a map data update scenario. Figure 1As shown, vehicles 101, 102, and 103 are crowdsourcing vehicles, which are equipped with corresponding collection equipment to collect corresponding geographic data. For example, the three vehicles are driving along their respective planned routes on the road, while collecting corresponding road environment data, such as lane line information and road sign information. It should be understood that the three vehicles are only exemplary, and there may be more or fewer vehicles, and the above driving routes are also exemplary. Figure 1 As shown, when the real-world environment changes (for example, newly added lane lines, changed lane lines, newly added road signs or changed road signs, etc.), collecting corresponding road environment data through crowdsourcing vehicles helps to quickly update geographic data.

[0125] In the prior art, crowdsourcing vehicles are equipped with front-mounted or rear-mounted visual sensors (such as driving recorders) and positioning sensors to collect images of the road and corresponding location information. Image processing technology is then used to identify and extract high-precision map elements from the collected road images and calculate the three-dimensional coordinates of the elements, and then to perform differential analysis with the existing high-precision map data and update the geographic data. However, this type of method has the following shortcomings: 1) The ranging accuracy of the monocular image collected by a camera is not high, and the accuracy of the extracted map elements is poor, making it impossible to achieve fine differentiation, making it difficult to meet the accuracy requirements of high-precision map updates, and small changes cannot be detected, which impairs the quality of map updates; 2) Visual sensors are sensitive to light, and insufficient light (such as at night) causes system failure, making it difficult to meet the freshness requirements of high-precision maps (such as daily updates). In order to solve this problem, the present application provides a map data update solution to form a high-freshness, high-quality map data update.

[0126] Figure 2 A schematic diagram of a map data updating system according to an embodiment of the present disclosure is shown. Figure 2As shown, the system includes an update device 201 and a cloud server 202, with data transmitted between the update device 201 and the cloud server 202 via a network. LiDAR technology has developed rapidly in recent years, with breakthroughs in mass production of vehicle-grade LiDAR. LiDAR offers the advantages of low cost, high reliability, and compact portability. Furthermore, the rapid development of AI (artificial intelligence) algorithms and AI computing chips has made it possible to process multi-source data collected on-device in real time. Therefore, the update device 201 provided in this application can be configured with multiple sensors mainly composed of lidar, and the update device 201 can be configured on a crowdsourcing vehicle. During the driving process of the crowdsourcing vehicle, the lidar in the update device 201 can collect point cloud data. An AI computing chip can be configured on the update device 201, and the multi-source perception and computing capabilities of the AI computing chip can be used for data processing or the collected point cloud data can be directly sent to the cloud server for data processing. The specific processing process can be to process the collected point cloud data to obtain the map element vector to be updated, match and align the map element vector to be updated with the pre-stored existing map element vector and perform element difference, determine whether the map element has changed, and perform map data update processing when the map element has changed. This solution uses lidar as the main data collection sensor for high-precision map updates, which can obtain high-precision point cloud data. Compared with the measurement method based on monocular vision, map feature vector extraction based on point cloud data can greatly improve the accuracy of map feature vectors, so that when the map is updated based on this data, the accuracy of the high-precision map will not be reduced. At the same time, the fine feature vector extraction also helps to achieve fine feature differentiation, which can detect detailed changes in the map and improve the quality of map updates. Moreover, lidar measurement does not rely on the lighting environment. This feature improves the availability of the update system and significantly improves the timeliness of updates for nighttime changes.

[0127] Figure 3 A schematic structural diagram of an updating device according to an embodiment of the present disclosure is shown as follows: Figure 3As shown, the update device may include the following three modules: an acquisition module 301, a software module 302, and an auxiliary module 303. The acquisition module 301 includes multiple sensors for collecting road data, such as lidar, cameras, GNSS (Global Navigation Satellite System) (supporting RTK (Real Time Kinematic)), and IMU (Inertial Measurement Unit). Lidar and cameras are measurement sensors used to collect laser point cloud data and image data of road features. GNSS / IMUs are positioning and attitude sensors used to obtain the spatial position and attitude angle of the collected data. GNSS systems refer to all satellite navigation systems and can provide positioning information. GNSS systems incorporate RTK technology, which significantly improves GNSS measurement accuracy and is a measurement method capable of achieving centimeter-level positioning accuracy in real time in the field. IMUs are sensors that can measure angles and attitudes. Software module 302 primarily performs sensor control and data analysis, data quality checks and status monitoring, logging, and exception handling. That is, through the software module 302, the sensor can be controlled and data can be collected. The collected data can then be quality checked (for example, data integrity). At the same time, the operating status of each sensor is also monitored at all times. When the data quality is abnormal or the sensor status is abnormal, the log system will record and transmit it back to the data center. At the same time, the abnormal level will be followed up and the driver will be reminded in an appropriate form. The auxiliary module 303 includes a control system, a computing unit, a communication unit, a storage unit, a power supply unit, cables, a mechanical structure, etc. These three modules together constitute a complete update device. Among them, the control system includes a time synchronization subsystem, which can ensure the time synchronization between the sensors in the acquisition module 301, and after the equipment integration is completed, the internal calibration parameters and external calibration parameters of the sensor can be obtained through sensor calibration, and the update equipment height can be updated to achieve the temporal and spatial consistency of multiple sensors. The computing unit includes the computing power of a CPU (central processing unit) and a GPU (Graphics Processing Unit), and a built-in ARM (Advanced RISC Machines, RISC (Reduced Instruction Set Computer) microprocessor) architecture operating system. It can perform real-time combined navigation solutions on data collected by GNSS / IMU to obtain high-precision trajectory files; the communication unit includes 4G / 5G and Wi-Fi communication functions.The updated equipment can also be rainproof, dustproof, and earthquake-resistant to ensure the safe and stable operation of the equipment in complex environments.

[0128] Figure 4 A flowchart of a map data updating method according to an embodiment of the present disclosure is shown. Figure 4 As shown, the map data updating method includes the following steps S401-S404:

[0129] In step S401, geographic data of a target area is obtained, wherein the geographic data includes geographic point cloud data;

[0130] In step S402, the map element vectors to be updated in the target area are extracted based on the geographic data;

[0131] In step S403, after the target area is aligned with the corresponding area in the pre-stored high-precision map, the map element vector to be updated is subjected to element difference with the existing map element vector in the corresponding area to determine whether the map element in the target area has changed;

[0132] In step S404, if the map elements in the target area are changed, map data update processing is performed.

[0133] In one possible implementation, the map data update method may be executed by an update device or a server installed on a collection vehicle. The collection vehicle may be a crowdsourcing vehicle, a professional collection vehicle, a collection robot, or other road-drivable equipment.

[0134] In one possible implementation, the target area refers to the area captured in real time by the updating device during the travel of the acquisition vehicle. The updating device is equipped with a laser radar (LiDAR) that can collect geographic point cloud data of the target area. Point cloud data is generally collected by 3D scanning equipment such as LiDAR and contains information on several points within the target area, including XYZ position information, RGB color information, and intensity information. It is a complex, multi-dimensional data set. The point cloud data of the target area refers to a massive collection of points representing the surface characteristics of objects within the target area.

[0135] In one possible implementation, map elements refer to the basic content that constitutes a map, such as buildings, roads, road signs, and street signs. Map element vectors refer to the spatial coordinates and attribute information of map elements. Using artificial intelligence technology to perform semantic segmentation on geographic point cloud data, the outline, location, and corresponding coordinates of each map element in the geographic point cloud data can be obtained. This allows the vectors of each map element within the target area to be obtained. These map element vectors, derived from real-time collected geographic data, are referred to as map element vectors to be updated.

[0136] In one possible implementation, since the position of the map element vector to be updated may deviate from that of the map element on the pre-stored high-precision map, it cannot be used directly for updating and needs to be matched and aligned with the area in the pre-stored high-precision map. For example, assuming that the target area refers to the middle 50m area of section A, the target area needs to be matched with the pre-stored high-precision map, and the target area is matched and aligned with the middle 50m area of section A on the high-precision map. Then, the map element vector to be updated is differentially compared with the existing map element vectors in the matching area to compare whether the map elements have changed, such as whether a new lane has been added, the markings have been repainted, the signs have been added or removed, etc. The element differential here refers to performing a differential operation on the map elements corresponding to the map element vectors to be updated in the target area and the map elements corresponding to the existing map element vectors in the matching area. If the operation result is 0, it indicates that the elements in the current target area are exactly the same as the existing elements, and the map elements in the target area have not changed; if the operation result is not 0, it indicates that the elements in the current target area are different from the existing elements, and the map elements in the target area have changed. In this case, it may be that the elements in the current target area have at least one more or less map element than the existing elements. For example, the map element vector to be updated in the target area has two road signs, while the existing map element vector of the corresponding area in the high-precision map has only one road sign. After the target area is aligned with the corresponding area, one road sign in the corresponding area completely overlaps with one road sign in the map element vector to be updated. After element difference is performed, it is determined that there is one more road sign in the current target area, indicating that the map element vector has changed and a new road sign has been added.

[0137] In one possible implementation, the map data update method is performed by an update device. When a map element within the target area changes, the update device performs a map data update process, including sending the map element vectors of the changed elements to be updated to a server, so that the server can update the map data accordingly. In this case, the update device may only send the map element vectors of the changed elements to the server, reducing the amount of data transmitted back and saving backhaul traffic fees.

[0138] In a possible implementation, the execution subject of the map data update is a server. When a map element in the target area changes, the server can directly update the map data based on the to-be-updated map element vector of the changed element.

[0139] It should be noted here that the changed element can be a newly added or modified map element, or a deleted map element. When the changed element is a deleted map element, the map element vector to be updated corresponding to the changed element is a null value.

[0140] This embodiment can use lidar as the main sensor for data collection for high-precision map updates, so that high-precision point cloud data can be obtained. Compared with the measurement method based on monocular vision, map feature vector extraction based on point cloud data can greatly improve the accuracy of map feature vectors, so that when the map is updated based on this, the accuracy of the high-precision map will not be reduced; at the same time, the fine feature vector extraction also helps to achieve fine feature differentiation, which can detect detailed changes in the map and improve the quality of map updates; and the lidar measurement does not depend on the lighting environment. This feature improves the availability of the update system and significantly improves the timeliness of updates for nighttime changes.

[0141] In a possible implementation, when the map data update is applied to an update device, if the map elements in the target area change in step S104, performing the map data update process includes:

[0142] If a map element within the target area changes, determining a change confidence level of the changed element;

[0143] If the change confidence is greater than a preset threshold, the map element vector to be updated corresponding to the changed element is sent to the server;

[0144] If the change confidence is less than or equal to a preset threshold, the map element vector to be updated and the original data corresponding to the changed element are sent to the server.

[0145] In this embodiment, the change confidence of the change element refers to the reliability of the change element, which can be configured according to preset rules. Under normal circumstances, road signs on the road are easily blocked by vehicles traveling on the road, and road signs beside the road are easily blocked by trees. Therefore, the change element can be set to deletion change, and the change element type is that the change confidence of these easily blocked elements is less than or equal to the lower value of the preset threshold. The change confidence of other change elements can be the extraction accuracy of these change elements.

[0146] In this embodiment, when the change confidence is greater than a preset threshold, it indicates that the changed element has a high probability of actually changing, and a map update is required at this time. In order to reduce the amount of data transmission, the update device can only send the updated map element vector corresponding to the changed element to the server. The server can directly add, delete, and modify the high-precision base map based on the updated map element vector, and then incrementally compile the high-precision map and republish it.

[0147] In this embodiment, when the change confidence is less than or equal to a preset threshold, in order to avoid erroneous updates, the map element vector to be updated and the geographic data corresponding to the changed element can be sent to the server, and the server can display the map element vector to be updated and the original data of the changed element for the update personnel. The update personnel can manually confirm whether the changed element is a real change. If it is not a real change, the map update will not be performed. If it is a real change and the map element vector to be updated of the changed element is accurate, an update instruction can be entered to perform a map update, and the update instruction instructs to perform a map update based on the map element vector to be updated corresponding to the changed element; if it is a real change and the map element vector to be updated of the changed element is inaccurate, the update personnel can enter the accurate changed element vector by themselves, and then enter the update instruction, and the update instruction instructs to perform an update based on the changed element vector entered by them.

[0148] In a possible implementation, when the map data update is applied to the server, if the map elements in the target area change in step S104, performing the map data update process includes:

[0149] If a map element within the target area changes, determining a change confidence level of the changed element;

[0150] If the change confidence is greater than a preset threshold, updating the map based on the to-be-updated map element vector corresponding to the changed element;

[0151] If the change confidence is less than or equal to a preset threshold, the map element vector to be updated and the original data corresponding to the changed element are output; and in response to receiving an input update instruction, the map is updated based on the update instruction, where the update instruction indicates that the map is updated based on the map element vector to be updated corresponding to the changed element or based on the changed element vector carried in the update instruction.

[0152] In this embodiment, with the popularization of 5G networks, when network transmission speed and traffic fees are not an issue, data processing does not need to be performed on the update device. The acquired geographic data can be transmitted back to the server. With the help of the server's more powerful computing power and historical data reserves, more sophisticated and complex data processing can be performed, which can further improve the quality of map updates.

[0153] In this embodiment, after the server obtains the geographic data of the target area from the acquisition device, it executes the above steps. When the map elements in the target area change, the change confidence of the changed elements is determined; when the change confidence is greater than the preset threshold, it indicates that the changed elements have a high probability of changing and the map needs to be updated. At this time, the server can directly add, delete, and modify the high-precision base map based on the map element vector to be updated, and then incrementally compile the high-precision map and republish it.

[0154] In this embodiment, when the change confidence is less than or equal to a preset threshold, in order to avoid erroneous updates, the server can display the map element vector to be updated and the original data of the changed element to the update personnel, and the update personnel can manually confirm whether the changed element is a real change. If it is not a real change, the map update will not be performed. If it is a real change and the map element vector to be updated of the changed element is accurate, an update instruction can be entered to perform a map update, and the update instruction instructs to perform a map update based on the map element vector to be updated corresponding to the changed element; if it is a real change and the map element vector to be updated of the changed element is inaccurate, the update personnel can enter the accurate changed element vector by themselves, and then enter the update instruction, and the update instruction instructs to perform an update based on the changed element vector entered by them.

[0155] In a possible implementation, step S101 in the above map data updating method, i.e., obtaining geographic data of the target area, may include the following steps:

[0156] Obtain the original point cloud data of the target area;

[0157] Points far from roads and dynamic object points in the original point cloud data are filtered to obtain geographic point cloud data.

[0158] In this embodiment, when the above method is applied to the update device, in order to reduce the amount of data processed by the point cloud, the requirements for computing power are lowered, the real-time requirements are met, and the quality of the point cloud data is improved, so that the update device can better process the collected point cloud data in real time, the collected original point cloud data can be filtered to filter out points far from the road and dynamic object points in the original point cloud data.

[0159] In this embodiment, if the above method is applied to the server, in order to reduce the amount of data processed by the point cloud, make data processing more lightweight, meet real-time requirements, improve the quality of point cloud data, reduce the amount of return data, and save return traffic fees, the update device can also filter out the away-road points and dynamic object points in the original point cloud data and return them to the server. At this time, the geographic power data obtained by the server does not include away-road points and dynamic object points.

[0160] In a possible implementation, the raw data further includes actual trajectory height, and filtering the off-road points and dynamic object points in the raw point cloud data to obtain geographic point cloud data includes:

[0161] In this embodiment, the update device is equipped with surveying and mapping sensors such as GNSS, RTK, and IMU. After combined navigation processing based on the RTK, IMU, and positioning data, a real-time trajectory file can be obtained. The specific processing process is conventional and will not be repeated here. The real-time trajectory file includes a real-time trajectory altitude, which refers to the real-time trajectory altitude of the update device. The pre-stored device ground elevation refers to the height of the update device from the ground. It can be pre-measured and stored in the update device, or calculated by the update device. The real-time ground elevation is obtained by subtracting the device ground elevation from the real-time trajectory altitude of the update device.

[0162] In this embodiment, for the original point cloud data collected in real time, points with heights near the ground elevation are ground points, and the remaining points are non-ground points. For example, the pre-stored equipment is 2m above the ground, and the real-time trajectory height of the target area collected at a certain moment is 40m. Then the ground elevation at the current moment can be obtained as 40m-2m=38m. This elevation is floated upward by 0.05m as the elevation h of the ground point, and the elevation h is used to classify the point cloud in the target area. If the height of the point in the point cloud is greater than the elevation h, it is a non-ground point, and if it is less than or equal to the elevation h, it is a ground point.

[0163] In this embodiment, when the above method is applied to the update device, the server can periodically synchronize the latest high-precision map data in the server to the update device through the network.

[0164] In this embodiment, the update device or server can use the existing high-precision map data as prior information to quickly filter out points far from the road. High-precision map updates mainly focus on the map elements of the road and its surroundings. Points far from the road are useless points. The existing high-precision map provides road boundary information, which can be used to set the effective boundary range of the point cloud data. In the horizontal direction, the road boundary is expanded by a first preset distance, such as 3m, as the horizontal effective boundary. In the vertical direction, the ground elevation can be expanded upward by a second preset distance, such as 12m, as the longitudinal effective boundary. These two effective boundaries are used to filter the original point cloud data in the target area, filtering out useless points above the longitudinal effective boundary and useless points outside the lateral effective boundary. At this time, the remaining point cloud data is the point cloud data of the road and objects around the road, which can simply and effectively reduce the amount of point cloud data.

[0165] In this embodiment, dynamic objects in the point cloud (such as vehicles and people traveling on the road) will obviously interfere with the extraction of map elements. In order to improve the quality of point cloud data, dynamic object points need to be filtered out in advance. Usually, within the road range, non-ground points with a height of less than 3m from the ground are dynamic objects. Based on this rule, dynamic object points among the non-ground points whose height exceeds the third preset distance of the ground elevation, such as 3m, can be filtered out. In this way, dynamic object points can be filtered out simply and effectively.

[0166] In a possible implementation, the geographic data further includes image data, and extracting the map element vector to be updated within the target area based on the geographic data includes:

[0167] Extracting map elements based on the geographic point cloud data to obtain a first map element vector within the target area;

[0168] Extracting map elements based on the geographic point cloud data and the image data to obtain a second map element vector within the target area;

[0169] The first map element vector and the second map element vector are fused to obtain the map element vector to be updated.

[0170] In this embodiment, in order to make the extracted map element vectors more accurate, the map element vectors can be first extracted using different methods, and then the extracted map element vectors are fused to obtain the map element vectors to be updated. The map element vectors extracted using different methods have certain duplication and complementarity. The fusion here refers to merging the duplicate vectors, and the different vectors complement each other.

[0171] In this embodiment, map feature extraction can be performed in two different ways: one is to extract features based solely on geographic point cloud data, obtaining a first map feature vector; the other is to extract map features using point-to-graph fusion, obtaining a second map feature vector. These two map feature vectors are then fused to obtain the map feature vector to be updated.

[0172] This embodiment can combine the advantages of laser point cloud and visual image perception extraction schemes to extract element vectors, which can improve the adaptability of the system while ensuring high accuracy of elements.

[0173] In one embodiment of the present disclosure, the step of extracting map elements based on the geographic point cloud data to obtain the first map element vector within the target area may include the following steps:

[0174] In this embodiment, the geographic point cloud data includes point cloud data of ground points and point cloud data of non-ground points. The specific distinction method can refer to the other embodiments mentioned above and will not be described in detail here.

[0175] In this embodiment, the reflection intensity difference between the road and road markings on the ground is typically large, with the road markings having a higher reflection intensity. Therefore, when extracting map features from the point cloud data of ground points, the three-dimensional point cloud data can be first converted into a two-dimensional point cloud intensity map based on the intensity information of each point in the point cloud data. This point cloud intensity map refers to an image that uses the intensity value of each ground point as the pixel value. Due to the large difference in reflection intensity between the road and road markings on the ground, this point cloud intensity map can effectively distinguish different map features on the ground. Here, the conversion process of the two-dimensional point cloud intensity map can be to project the three-dimensional ground points onto the ground and calculate the intensity value of each ground grid point (such as the average of the intensity values of all projected points within the ground grid point) as the image pixel value, thereby obtaining the corresponding point cloud intensity map.

[0176] In this embodiment, the difference in reflection intensity of non-ground points is not large, so using a point cloud intensity map to extract them is not effective. However, the density of the point clouds of different map elements projected onto the ground is very different. Different density characteristics can reflect the spatial characteristics of map elements. For example, a pole has only a circular projection, but the point cloud data corresponding to the pole is a cylindrical point cloud data. After projecting it onto the ground, the density is very large in a local range. Therefore, the density characteristic can be used to identify different non-ground elements. Therefore, when extracting map elements from the point cloud data of non-ground points, the non-ground points can be projected onto the ground, and the density value of each ground grid point (such as the number of all projected points within the ground grid point) is counted as the image pixel value, thereby obtaining the corresponding point cloud density map.

[0177] In this embodiment, image recognition and segmentation technology is already a very mature image processing technology. Here, a trained image segmentation model can be used to identify and segment the first map elements in the target area, such as lane lines, ground markings, road poles, road signs, etc., on the point cloud intensity map and point cloud density map respectively.

[0178] In this embodiment, after the first map element is segmented, the three-dimensional geographic point cloud data corresponding to the first map element can be obtained. For example, assuming that the first map element is a lane line on the ground, the position of the lane line in the point cloud intensity map can be obtained after the lane line is segmented. Based on the geographic point cloud data at each position in the point cloud intensity map, the geographic point cloud data corresponding to the lane line can be obtained.

[0179] In this embodiment, boundary fitting can be performed on the geographic point cloud data corresponding to the first map element to obtain the boundary points of the first map element. The boundary points are used to represent the map element, thereby completing the extraction of the first map element vector.

[0180] In one embodiment of the present disclosure, extracting map elements based on the geographic point cloud data and the image data to obtain a second map element vector within the target area includes:

[0181] performing image segmentation and recognition on the image data to obtain a second map element;

[0182] Obtaining geographic point cloud data corresponding to the second map element;

[0183] Boundary fitting of the second map element is performed based on geographic point cloud data corresponding to the second map element to obtain the second map element vector.

[0184] In this embodiment, a point-graph fusion approach can be used to extract map element vectors. First, a trained image segmentation and extraction model can be used to perform image segmentation and recognition on the image data to obtain the second map element. Combining the internal and external calibration parameters of each sensor and the height of the device above the ground, the image data and geographic point cloud data (point cloud projection) can be synchronized in time and space to achieve a one-to-one correspondence (this is prior art and will not be described in detail here). In this way, the geographic point cloud data corresponding to the second map element in the image data can be obtained. Boundary fitting can be performed on the geographic point cloud data corresponding to the second map element to obtain the boundary points of the second map element. The boundary points are used to represent the map element, thus completing the extraction of the second map element vector.

[0185] In one embodiment of the present disclosure, the fusing of the first map element vector and the second map element vector to obtain the map element vector to be updated includes:

[0186] Calculating a first fitting residual of the first map element vector and a second fitting residual of the second map element vector;

[0187] determining a first confidence level of the first map element vector based on the first fitting residual, and determining a second confidence level of the second map element vector based on the second fitting residual;

[0188] Based on the first confidence level and the second confidence level, the first map element vector and the second map element vector are fused to obtain the map element vector to be updated.

[0189] In this embodiment, the fitting residual refers to the difference between the actual observation vector of the map element and the fitted map element vector, which is used to indicate the quality of the fitting of the map element vector. The specific calculation process is well known to those skilled in the art and will not be repeated here.

[0190] In this embodiment, when fitting the boundary of the first map element based on the geographic point cloud data corresponding to the first map element, a first fitting residual of the first map element vector is also calculated. When fitting the boundary of the second map element based on the geographic point cloud data corresponding to the second map element, a second fitting residual of the second map element vector is also calculated. Following the rule that the larger the fitting residual, the lower the confidence level, a first confidence level of the first map element vector is determined based on the first fitting residual, and a second confidence level of the second map element vector is determined based on the second fitting residual. These confidence levels represent the reliability of the map element vectors.

[0191] In this embodiment, the first map element vector and the second map element vector may be fused based on the first confidence level and the second confidence level to obtain the map element vector to be updated; for example, when merging overlapping first map element vectors and second map element vectors, the map element vector with a lower confidence level may be deleted, and the map element vector with a higher confidence level may be retained.

[0192] In a possible implementation, determining the change confidence of the change factor includes:

[0193] The change confidence of the changed element is determined based on the confidence of the to-be-updated map element vector corresponding to the changed element, the type of the changed element, and the change type, where the change type includes addition and deletion.

[0194] In this embodiment, under normal circumstances, road signs on the road are easily blocked by vehicles traveling on the road, and road signs beside the road are easily blocked by trees. Therefore, the change element can be set to deletion change, and the change element type is that the change confidence of these easily blocked elements is less than or equal to the lower value of the preset threshold, and the change confidence of other change elements can be the confidence of these change elements.

[0195] Figure 5 A flowchart of a map data updating method according to an embodiment of the present disclosure is shown. Figure 5 As shown, the map data updating method includes the following steps S501-S503:

[0196] In step S501, map information of a changed element sent by an update device is received, wherein the map information includes a map element vector to be updated corresponding to the changed element;

[0197] In step S502, if the map information includes a map element vector to be updated, then the map is updated based on the map element vector to be updated corresponding to the changed element;

[0198] In step S503, if the map information includes map element vectors to be updated, geographic point cloud data and image data, the map element vectors to be updated, geographic point cloud data and image data are displayed; in response to receiving an input update instruction, the map is updated based on the update instruction, and the update instruction is used to instruct the map to be updated based on the map element vectors to be updated corresponding to the changed elements or based on the changed element vectors input by the user.

[0199] In a possible implementation, the map data updating method is executed by a server.

[0200] In one possible implementation, the update device is mounted on a collection vehicle that can normally travel on a road, and the update device can perform steps S401 to S404 above. The update device extracts the vectors of the map elements to be updated within the target area based on the geographic data collected by the device, performs element differences with the existing map element vectors, and compares whether the map elements within the target area have changed. The update device then sends map information of the changed elements to a server. After receiving the map information of the changed elements, the server can perform a map update based on the geographic information of the changed elements.

[0201] In one possible implementation, when a map element within the target area changes, the update device determines the change confidence of the changed element; the change confidence of the changed element refers to the reliability of the changed element, and can be configured according to preset rules, such as determining the change confidence of the changed element based on the confidence of the map element vector to be updated corresponding to the changed element, the type of the changed element, and the change type, where the change type includes addition and deletion.

[0202] In one possible implementation, when the change confidence is greater than a preset threshold, it indicates that the changed element is very likely to be truly changed. At this time, a map update is required. In order to reduce the amount of data transmission, the update device can only send the updated map element vector corresponding to the changed element to the server. The server can directly add, delete, and modify the high-precision base map based on the updated map element vector, and then incrementally compile the high-precision map and republish it.

[0203] In a possible implementation, when the change confidence is less than or equal to a preset threshold, in order to avoid erroneous updates, the map element vector to be updated and geographic data corresponding to the changed element can be sent to the server, and the server can display the map element vector to be updated and the original data of the changed element for the update personnel. The update personnel can manually confirm whether the changed element is a real change. If it is not a real change, the map update will not be performed. If it is a real change and the map element vector to be updated of the changed element is accurate, an update instruction can be entered to perform a map update, and the update instruction instructs to perform a map update based on the map element vector to be updated corresponding to the changed element; if it is a real change and the map element vector to be updated of the changed element is inaccurate, the update personnel can enter the accurate changed element vector by themselves, and then enter the update instruction, and the update instruction instructs to perform an update based on the changed element vector entered by them.

[0204] In this embodiment, the update device may only send map information of changed elements to the server, thereby reducing the amount of transmitted data.

[0205] Figure 6 A flowchart of a map data updating method according to an embodiment of the present disclosure is shown. Figure 6 As shown, the map data updating method includes the following steps S601-S604:

[0206] In step S601, the updating device obtains geographic data of a target area, and extracts vectors of map elements to be updated within the target area based on the geographic data;

[0207] In step S602, after aligning the target area with the corresponding area in the pre-stored high-precision map, the updating device performs element difference between the map element vector to be updated and the existing map element vector in the corresponding area to determine whether the map element in the target area has changed;

[0208] If the map elements in the target area change, the following steps S603 are executed;

[0209] In step S603, the map element vector to be updated of the changed element is sent to the server;

[0210] In step S604, the server performs map data update processing based on the map element vector to be updated of the changed element.

[0211] For example, a map data update scenario disclosed in the present invention can be described as follows: the update device on the crowdsourcing vehicle can collect the original data of the target area during the normal driving of the crowdsourcing vehicle, and the original data includes original point cloud data, image data and real-time trajectory height; the update device can filter out useless points (far away from road points and dynamic object points) in the original point cloud data based on the real-time trajectory height to obtain geographic point cloud data, and then extract the map element vector to be updated in the target area based on the geographic point cloud data and image data, match and align the target area with the corresponding area in the pre-stored high-precision map, and perform element difference between the map element vector to be updated and the existing map element vector in the corresponding area to determine whether the map element in the target area has changed; in the target area, the map element vector to be updated is matched with the existing map element vector in the corresponding area. When a map element in an area changes, the change confidence of the changed element is determined; when the change confidence is greater than a preset threshold, the map element vector to be updated corresponding to the changed element is sent to the server, so that the server performs a map update based on the map element vector to be updated corresponding to the changed element; when the change confidence is less than or equal to the preset threshold, the map element vector to be updated and the geographic data corresponding to the changed element are sent to the server, and the server displays the map element vector to be updated and the original data corresponding to the changed element, and then responds to receiving an update instruction input by the user, performs a map update based on the update instruction, and the update instruction is used to indicate that a map update is performed based on the map element vector to be updated corresponding to the changed element or based on the changed element vector input by the user. In this scenario, the server can also send the latest high-precision map and the upgraded image segmentation algorithm to the update device, and can also obtain the log information sent back by the update device.

[0212] In other possible implementations, another map data update scenario disclosed herein can be described as follows: the update device on the crowdsourcing vehicle can collect raw data of the target area during the normal driving of the crowdsourcing vehicle, and the raw data includes raw point cloud data, image data, and real-time trajectory height; the update device can filter out useless points (far away from road points and dynamic object points) in the raw point cloud data based on the real-time trajectory height to obtain geographic point cloud data, and then send the geographic cloud data and the corresponding image data to the server. The server can extract the map element vector to be updated in the target area based on the geographic point cloud data and the image data, and after matching and aligning the target area with the corresponding area in the pre-stored high-precision map, the map element vector to be updated is sent to the server. Performing element differences with existing map element vectors in the corresponding area to determine whether map elements in the target area have changed; when a map element in the target area has changed, determining a change confidence of the changed element; when the change confidence is greater than a preset threshold, performing a map update based on the to-be-updated map element vector corresponding to the changed element; when the change confidence is less than or equal to the preset threshold, displaying the to-be-updated map element vector and original data corresponding to the changed element; and then, in response to receiving an update instruction input by a user, performing a map update based on the update instruction, wherein the update instruction is used to instruct to perform a map update based on the to-be-updated map element vector corresponding to the changed element or based on the changed element vector input by the user.

[0213] The following are embodiments of the apparatus disclosed herein, which can be used to execute embodiments of the method disclosed herein.

[0214] Figure 7 FIG1 shows a structural block diagram of a map data updating device according to an embodiment of the present disclosure. The device can be implemented as part or all of an electronic device through software, hardware, or a combination of both. Figure 7 As shown, the map data updating device includes:

[0215] A first acquisition module 701 is configured to acquire geographic data of a target area, wherein the geographic data includes geographic point cloud data;

[0216] An extraction module 702 is configured to extract the map element vectors to be updated within the target area based on the geographic data;

[0217] The difference module 703 is configured to align the target area with the corresponding area in the pre-stored high-precision map, and then perform element difference between the map element vector to be updated and the existing map element vector in the corresponding area to determine whether the map element in the target area has changed;

[0218] The first updating module 704 is configured to perform map data update processing if the map elements in the target area change.

[0219] In a possible implementation, the apparatus is applied to update a device, and the first updating module 704 is configured to:

[0220] If a map element within the target area changes, determining a change confidence level of the changed element;

[0221] If the change confidence is greater than a preset threshold, the map element vector to be updated corresponding to the changed element is sent to the server;

[0222] If the change confidence is less than or equal to a preset threshold, the map element vector and geographic data to be updated corresponding to the changed element are sent to the server.

[0223] In a possible implementation, the apparatus is applied to a server, and the first updating module 704 is configured to:

[0224] If a map element within the target area changes, determining a change confidence level of the changed element;

[0225] If the change confidence is greater than a preset threshold, updating the map based on the to-be-updated map element vector corresponding to the changed element;

[0226] If the change confidence is less than or equal to a preset threshold, the map element vector to be updated and the geographic data corresponding to the changed element are output; in response to receiving an input update instruction, the map is updated based on the update instruction, and the update instruction is used to instruct to update the map based on the map element vector to be updated corresponding to the changed element or to update the map based on the changed element vector input by the user.

[0227] In a possible implementation, the first acquisition module 701 is configured to:

[0228] Obtain the original point cloud data of the target area;

[0229] Points far from roads and dynamic object points in the original point cloud data are filtered to obtain geographic point cloud data.

[0230] In a possible implementation, the raw data further includes real-time trajectory height, and the portion of the first acquisition module 701 that filters out points far from roads and dynamic object points in the raw point cloud data to obtain geographic point cloud data is configured as follows:

[0231] Determining the ground elevation based on the actual trajectory height and a pre-stored device height above the ground;

[0232] Determine the points in the original point cloud data whose heights are within a preset range from the ground elevation as ground points, and determine the remaining points as non-ground points;

[0233] Determining the road boundaries of the target area based on pre-stored high-precision map data;

[0234] Extending the road boundary outward by a first preset distance as a lateral effective boundary, and extending the ground elevation upward by a second preset distance as a longitudinal effective boundary;

[0235] Filtering out useless points above the longitudinal effective boundary and useless points outside the transverse effective boundary;

[0236] Dynamic object points whose heights exceed a third preset distance from the ground elevation among the non-ground points are filtered out.

[0237] In a possible implementation, the geographic data further includes image data, and the extraction module 702 is configured to:

[0238] Extracting map elements based on the geographic point cloud data to obtain a first map element vector within the target area;

[0239] Extracting map elements based on the geographic point cloud data and the image data to obtain a second map element vector within the target area;

[0240] The first map element vector and the second map element vector are fused to obtain the map element vector to be updated.

[0241] In a possible implementation, the portion of the extraction module 702 that extracts map elements based on the geographic point cloud data to obtain the first map element vector within the target area is configured as follows:

[0242] generating a point cloud intensity map based on intensity information of ground points in the geographic point cloud data;

[0243] generating a point cloud density map based on density information corresponding to non-ground points in the geographic point cloud data, wherein the density information includes density information of the non-ground points when projected onto the ground;

[0244] Extracting a first map element of the target area from the point cloud intensity map and the point cloud density map;

[0245] Boundary fitting of the first map element is performed based on geographic point cloud data corresponding to the first map element to obtain the first map element vector.

[0246] In a possible implementation, the portion of the extraction module 702 that extracts map elements based on the geographic point cloud data and the image data to obtain the second map element vector within the target area is configured as follows:

[0247] performing image segmentation and recognition on the image data to obtain a second map element;

[0248] Based on the correspondence between the image data and the geographic point cloud data, obtaining the geographic point cloud data corresponding to the second map element;

[0249] Boundary fitting of the second map element is performed based on geographic point cloud data corresponding to the second map element to obtain the second map element vector.

[0250] In a possible implementation, the part of the extraction module 702 that fuses the first map element vector and the second map element vector to obtain the map element vector to be updated is configured as follows:

[0251] Calculating a first fitting residual of the first map element vector and a second fitting residual of the second map element vector;

[0252] determining a first confidence level of the first map element vector based on the first fitting residual, and determining a second confidence level of the second map element vector based on the second fitting residual;

[0253] Based on the first confidence level and the second confidence level, the first map element vector and the second map element vector are fused to obtain the map element vector to be updated.

[0254] In a possible implementation, the portion of the first updating module 704 that determines the change confidence of the change element is configured as follows:

[0255] The change confidence of the changed element is determined based on the confidence of the to-be-updated map element vector corresponding to the changed element, the type of the changed element, and the change type, where the change type includes addition and deletion.

[0256] Figure 7 The technical terms and technical features involved in the embodiments shown and related Figures 1 to 5 The technical terms and technical features mentioned in the embodiments shown and related are the same or similar. Figure 7 The explanation and description of the technical terms and technical features involved in the embodiments shown and related can refer to the above Figures 1 to 5 The explanations of the illustrated and related embodiments will not be repeated here.

[0257] Figure 8FIG1 shows a structural block diagram of a map data updating device according to an embodiment of the present disclosure. The device can be implemented as part or all of an electronic device through software, hardware, or a combination of both. Figure 8 As shown, the map data updating device includes:

[0258] A second acquisition module 801 is configured to acquire map information of changed elements sent by an update device, wherein the map information includes a vector of a map element to be updated;

[0259] The second update module 802 is configured to, if the map information includes a map element vector to be updated, perform a map update based on the map element vector to be updated corresponding to the changed element; if the map information includes a map element vector to be updated, geographic point cloud data and image data, display the map element vector to be updated, geographic point cloud data and image data; and in response to receiving an input update instruction, perform a map update based on the update instruction, wherein the update instruction is used to instruct to perform a map update based on the map element vector to be updated corresponding to the changed element or based on the changed element vector input by the user.

[0260] Figure 8 The technical terms and technical features involved in the embodiments shown and related Figures 1 to 5 The technical terms and technical features mentioned in the embodiments shown and related are the same or similar. Figure 8 The explanation and description of the technical terms and technical features involved in the embodiments shown and related can refer to the above Figures 1 to 5 The explanations of the illustrated and related embodiments will not be repeated here.

[0261] The present disclosure also discloses a navigation service, wherein the carrier's location is determined based on the above-mentioned carrier positioning method, and a navigation guidance service for a corresponding scenario is provided to the carrier based on the carrier's location. The corresponding scenario is one or a combination of AR navigation, elevated navigation, or main and auxiliary road navigation.

[0262] The embodiment of the present disclosure also discloses a navigation method, wherein a navigation route calculated based on at least a starting point, an end point and road conditions is obtained based on an electronic map, and navigation guidance is performed on the carrier based on the navigation route and the positioning of the carrier, and the positioning of the carrier is achieved by the above-mentioned carrier positioning method.

[0263] The present disclosure also discloses an electronic device, Figure 9 A structural block diagram of an electronic device according to an embodiment of the present disclosure is shown. Figure 9 As shown, the electronic device 900 includes a memory 901 and a processor 902; wherein,

[0264] The memory 901 is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor 902 to implement the above method steps.

[0265] Figure 10 It is a structural diagram of a computer system suitable for implementing the method according to an embodiment of the present disclosure.

[0266] like Figure 10 As shown, the computer system 1000 includes a processing unit 1001, which can execute various processes in the above-mentioned embodiments according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage unit 1008 into a random access memory (RAM) 1003. Various programs and data required for the operation of the system 1000 are also stored in the RAM 1003. The processing unit 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0267] The following components are connected to the I / O interface 1005: an input section 1006 including a keyboard, a mouse, etc.; an output section 1007 including a cathode ray tube (CRT), a liquid crystal display (LCD), a speaker, etc.; a storage section 1008 including a hard disk, etc.; and a communication section 1009 including a network interface card such as a LAN card, a modem, etc. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the I / O interface 1005 as needed. A removable medium 1011, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1010 as needed so that a computer program read therefrom can be installed into the storage section 1008 as needed. Among them, the processing unit 1001 can be implemented as a processing unit such as a CPU, a GPU, a TPU, an FPGA, or an NPU.

[0268] In particular, according to embodiments of the present disclosure, the methods described above can be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program tangibly embodied on a computer-readable medium, the computer program comprising program code for executing the methods described above. In such embodiments, the computer program can be downloaded and installed from a network via the communication portion 1009 and / or installed from the removable medium 1011.

[0269] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the diagram or block diagram can represent a module, program segment or part of the code, and the module, program segment or part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, as well as the combination of boxes in the block diagram and / or flow chart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or can be implemented using a combination of dedicated hardware and computer instructions.

[0270] The units or modules involved in the embodiments described in this disclosure may be implemented in software or hardware. The units or modules described may also be provided in a processor, and the names of these units or modules do not, in certain circumstances, constitute limitations on the units or modules themselves.

[0271] As another aspect, embodiments of the present disclosure further provide a computer-readable storage medium. This computer-readable storage medium may be included in the apparatus described in the above embodiments, or may be a standalone computer-readable storage medium not incorporated into the apparatus. The computer-readable storage medium stores one or more programs, which are used by one or more processors to execute the methods described in the embodiments of the present disclosure.

[0272] The above description is merely a preferred embodiment of the present disclosure and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also encompass other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the inventive concept. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in the embodiments of the present disclosure.

Claims

1. A method for updating map data, wherein: include: Acquiring geographic data of a target area, wherein the geographic data includes geographic point cloud data and image data collected by vehicle-mounted sensors; Extracting map elements based on the geographic point cloud data to obtain a first map element vector within the target area; Extracting map elements based on the geographic point cloud data and the image data to obtain a second map element vector within the target area; fusing the first map element vector and the second map element vector to obtain a map element vector to be updated in the target area; After aligning the target area with the corresponding area in the pre-stored high-precision map, performing element difference between the map element vector to be updated and the existing map element vector in the corresponding area to determine whether the map element in the target area has changed; If the map elements in the target area are changed, map data update processing is performed based on the to-be-updated map element vector corresponding to the changed element or based on the changed element vector input by the user.

2. The method according to claim 1, wherein The method is applied to an update device, and if a map element in the target area changes, performing map data update processing based on a to-be-updated map element vector corresponding to the changed element or based on a changed element vector input by a user includes: If a map element within the target area changes, determining a change confidence level of the changed element; If the change confidence is greater than a preset threshold, the map element vector to be updated corresponding to the changed element is sent to the server; If the change confidence is less than or equal to a preset threshold, the map element vector and geographic data to be updated corresponding to the changed element are sent to the server.

3. The method according to claim 1, wherein The method is applied to a server, and if a map element in the target area changes, performing map data update processing based on a to-be-updated map element vector corresponding to the changed element or based on a changed element vector input by a user includes: If a map element within the target area changes, determining a change confidence level of the changed element; If the change confidence is greater than a preset threshold, updating the map based on the to-be-updated map element vector corresponding to the changed element; If the change confidence is less than or equal to a preset threshold, the map element vector to be updated and the geographic data corresponding to the changed element are output; in response to receiving an input update instruction, the map is updated based on the update instruction, and the update instruction is used to instruct to update the map based on the map element vector to be updated corresponding to the changed element or to update the map based on the changed element vector input by the user.

4. The method according to claim 1, wherein The obtaining of geographic data of the target area includes: Obtain the original point cloud data of the target area; Points far from roads and dynamic object points in the original point cloud data are filtered to obtain geographic point cloud data of the target area.

5. The method according to claim 4, wherein The original point cloud data also includes real-time trajectory height. The filtering of the points far from the road and the dynamic object points in the original point cloud data to obtain the geographic point cloud data includes: Determining the ground elevation based on the real-time trajectory height and a pre-stored device height above the ground; Determine the points in the original point cloud data whose heights are within a preset range from the ground elevation as ground points, and determine the remaining points as non-ground points; Determining the road boundaries of the target area based on pre-stored high-precision map data; Extending the road boundary outward by a first preset distance as a lateral effective boundary, and extending the ground elevation upward by a second preset distance as a longitudinal effective boundary; Filtering out useless points above the longitudinal effective boundary and useless points outside the transverse effective boundary; Dynamic object points whose heights exceed a third preset distance from the ground elevation among the non-ground points are filtered out.

6. The method according to claim 1, wherein The extracting map elements based on the geographic point cloud data to obtain a first map element vector in the target area includes: generating a point cloud intensity map based on intensity information of ground points in the geographic point cloud data; generating a point cloud density map based on density information corresponding to non-ground points in the geographic point cloud data, wherein the density information includes density information of the non-ground points when projected onto the ground; Extracting a first map element of the target area from the point cloud intensity map and the point cloud density map; Boundary fitting of the first map element is performed based on geographic point cloud data corresponding to the first map element to obtain the first map element vector.

7. The method according to claim 1 or 6, wherein: The extracting map elements based on the geographic point cloud data and the image data to obtain a second map element vector in the target area includes: performing image segmentation and recognition on the image data to obtain a second map element; Based on the correspondence between the image data and the geographic point cloud data, obtaining the geographic point cloud data corresponding to the second map element; Boundary fitting of the second map element is performed based on geographic point cloud data corresponding to the second map element to obtain the second map element vector.

8. The method according to claim 7, wherein: The fusing the first map element vector and the second map element vector to obtain the map element vector to be updated in the target area includes: Calculating a first fitting residual of the first map element vector and a second fitting residual of the second map element vector; determining a first confidence level of the first map element vector based on the first fitting residual, and determining a second confidence level of the second map element vector based on the second fitting residual; Based on the first confidence level and the second confidence level, the first map element vector and the second map element vector are fused to obtain the map element vector to be updated.

9. The method according to claim 2 or 3, wherein: Determining the change confidence of the change factor includes: The change confidence of the changed element is determined based on the confidence of the to-be-updated map element vector corresponding to the changed element, the type of the changed element, and the change type, where the change type includes addition and deletion.

10. A method for updating map data, wherein: include: Get map information of changed elements sent by the update device; If the map information includes a to-be-updated map element vector, updating the map based on the to-be-updated map element vector corresponding to the changed element, wherein the to-be-updated map element vector is obtained by fusing a first map element vector within a target area and a second map element vector within the target area, the first map element vector being obtained by extracting map elements based on geographic point cloud data collected by an on-board sensor, and the second map element vector being obtained by extracting map elements based on geographic point cloud data and image data collected by the on-board sensor; If the map information includes map element vectors to be updated, geographic point cloud data, and image data, the map element vectors, geographic point cloud data, and image data to be updated are displayed; in response to receiving an input update instruction, the map is updated based on the update instruction, and the update instruction is used to instruct the map to be updated based on the map element vectors to be updated corresponding to the changed elements or based on the changed element vectors input by the user.

11. A method for updating map data, wherein: include: The updating device acquires geographic data of the target area, wherein the geographic data includes geographic point cloud data and image data collected by the vehicle-mounted sensor; Extracting map elements based on the geographic point cloud data to obtain a first map element vector within the target area; Extracting map elements based on the geographic point cloud data and the image data to obtain a second map element vector within the target area; fusing the first map element vector and the second map element vector to obtain a map element vector to be updated in the target area; After aligning the target area with the corresponding area in the pre-stored high-precision map, performing element difference between the map element vector to be updated and the existing map element vector in the corresponding area to determine whether the map element in the target area has changed; if the map element in the target area has changed, sending the map element vector to be updated of the changed element to the server; The server performs map data update processing based on the map element vector to be updated of the changed element.

12. An electronic device comprising a memory and a processor; wherein: The memory is configured to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method steps according to any one of claims 1 to 11.

13. A navigation method, wherein: Obtain a navigation route calculated based on at least a starting point, an end point, and road conditions based on an electronic map, and perform navigation guidance on the carrier based on the navigation route, wherein the electronic map is implemented based on any one of the methods described in claims 1 to 11.

Images