WO2022007818A1 - 高精地图更新方法、车辆、服务器及存储介质 - Google Patents

高精地图更新方法、车辆、服务器及存储介质 Download PDF

Info

Publication number
WO2022007818A1
WO2022007818A1 PCT/CN2021/104861 CN2021104861W WO2022007818A1 WO 2022007818 A1 WO2022007818 A1 WO 2022007818A1 CN 2021104861 W CN2021104861 W CN 2021104861W WO 2022007818 A1 WO2022007818 A1 WO 2022007818A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
precision map
map
updated
information
Prior art date
Application number
PCT/CN2021/104861
Other languages
English (en)
French (fr)
Inventor
刘洋
孙连明
崔茂源
宋林桓
姜云鹏
Original Assignee
中国第一汽车股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国第一汽车股份有限公司 filed Critical 中国第一汽车股份有限公司
Publication of WO2022007818A1 publication Critical patent/WO2022007818A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/29—Geographical information databases
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/23—Updating
    • G06F16/2358—Change logging, detection, and notification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures

Definitions

  • the embodiments of the present application relate to the field of automatic driving, for example, to a method for updating a high-precision map, a vehicle, a server, and a storage medium.
  • Level 3, L3 conditional autonomous driving
  • High-precision map acquisition relies on sensors such as lidar, cameras, and high-precision combined inertial navigation, and requires the map collector to have certain surveying and mapping qualifications and professional surveying and mapping capabilities. Therefore, the cost of data acquisition, production and maintenance is extremely high. With the continuous construction of roads, the update of high-precision maps is more important than the production of initial maps, and the means of map acquisition cannot meet the requirements of data update due to factors such as cost.
  • the main process is: the camera installed on the autonomous vehicle collects the image, uploads the collected image to the server, and the server extracts the road feature elements from the image. , and compare the road feature elements with the high-precision map in real time. If the road feature elements extracted from the image are found to be inconsistent with the high-precision map, the high-precision map is updated.
  • the above method needs to upload the collected images to the server, consumes a large amount of traffic for data uploading, and requires high cost.
  • the present application provides a high-precision map update method, vehicle, server and storage medium, which solve the technical problem of high cost due to the consumption of a large amount of traffic for high-precision map update.
  • An embodiment of the present application provides a method for updating a high-precision map, including: determining whether the target semantic information can be successfully matched with the high-precision map according to target semantic information extracted from a collected image and a pre-stored high-precision map. The data in the map is matched; if it is determined that the target semantic information fails to match the data in the high-precision map, error information is sent to the server, so that the server can determine the high-precision map according to the error information. Whether update is required; receive incremental data of map update sent by the server, wherein the incremental data of map update is sent by the server when it is determined according to the error information that the high-precision map needs to be updated data; update the high-precision map according to the incremental data of the map update.
  • An embodiment of the present application provides a method for updating a high-precision map, including: receiving error information sent by a vehicle, wherein the error information is the target semantic information extracted by the vehicle from the collected image and pre-stored information.
  • High-precision map determine the information sent when the target semantic information fails to match the data in the high-precision map; according to the error information, determine whether the pre-stored high-precision map in the vehicle needs to be updated; When it is determined that the high-precision map pre-stored in the vehicle needs to be updated, the incremental data of the map update is determined; and the incremental data of the map update is sent to the vehicle.
  • Embodiments of the present application provide a vehicle, comprising: one or more processors; a memory configured to store one or more programs; when the one or more programs are executed by the one or more processors, all The one or more processors implement the above-mentioned first method for updating a high-precision map.
  • An embodiment of the present application further provides a server, including: one or more processors; a memory configured to store one or more programs; when the one or more programs are executed by the one or more processors, The one or more processors are caused to implement the above-mentioned second method for updating a high-precision map.
  • Embodiments of the present application further provide a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, implements the above-mentioned first or second method for updating a high-precision map.
  • FIG. 1 is a schematic flowchart of a method for updating a high-precision map according to an embodiment of the present application
  • FIG. 2 is a schematic diagram of the composition and structure of a vehicle according to an embodiment of the present application.
  • FIG. 3 is a schematic flowchart of a method for updating a high-precision map provided by another embodiment of the present application.
  • FIG. 4 is a schematic structural diagram of an apparatus for updating a high-precision map provided by an embodiment of the present application
  • FIG. 5 is a schematic structural diagram of an apparatus for updating a high-precision map provided by another embodiment of the present application.
  • FIG. 6 is a schematic structural diagram of a vehicle provided by the application.
  • FIG. 7 is a schematic structural diagram of a server provided by the present application.
  • FIG. 1 is a schematic flowchart of a method for updating a high-precision map according to an embodiment of the present application.
  • This embodiment is applicable to the scenario of updating the high-precision map while the autonomous vehicle is running.
  • This embodiment may be implemented by a high-precision map updating apparatus, which may be implemented by software and/or hardware, and may be integrated into an automatic driving controller of a vehicle.
  • the method for updating a high-precision map provided by this embodiment includes the following steps.
  • Step 101 According to the target semantic information extracted from the collected image and the pre-stored high-precision map, determine whether the target semantic information can be successfully matched with the data in the high-precision map.
  • the vehicle in this embodiment may be a vehicle with an automatic driving function.
  • FIG. 2 is a schematic structural diagram of a vehicle according to an embodiment of the present application.
  • the vehicle may include: a front-view camera 21 , a high-precision map storage module 22 , an automatic driving controller 23 and a communication module 24 .
  • the front-view camera 21 , the high-precision map storage module 22 and the communication module 24 are all connected to the automatic driving controller 23 .
  • the communication module 24 is configured to implement communication between the automatic driving controller 23 and the server 25 .
  • other sensors connected to the automatic driving controller 23 may also be included in the vehicle.
  • the communication module 24 in the vehicle may be a telematics box (Telematics BOX, T-BOX).
  • a high-precision map is stored in the high-precision map storage module 22 in the vehicle.
  • the front-view camera 21 in the vehicle collects an image of the surrounding environment, for example, an image of a road, and sends the image of the surrounding environment to the automatic driving controller 23 .
  • the forward-looking camera 21 may transmit the image to the automatic driving controller 23 through a low-voltage differential signaling (Low-Voltage Differential Signaling, LVDS) line.
  • the LVDS line uses LVDS technology to transmit images, which has the advantages of low power consumption, low bit error rate, low crosstalk and low radiation.
  • the high-precision map storage module 22 can transmit the high-precision map to the automatic driving controller 23 through Ethernet (Ethernet, ETH) communication.
  • Other sensors can exchange information with the automatic driving controller 23 through ETH communication or a controller area network (Controller Area Network, CAN) network.
  • the high-precision map storage module 22 can transmit the pre-stored high-precision map to the automatic driving controller 23 .
  • the automatic driving controller 23 can extract target semantic information from the image.
  • the automatic driving controller 23 can extract the target semantic information from the image by using an image processing algorithm.
  • the image processing algorithm here may be a semantic segmentation algorithm.
  • the target semantic information in this embodiment may include characteristics of road elements, such as characteristics of lane lines, guardrails, curbs, speed limit signs, ground signs, billboards, gantry and the like.
  • the characteristics of the lane line include color, line type, relative position to the vehicle, curvature, etc.; the characteristics of guardrails, curbs, billboards and gantry mainly include the relative position to the vehicle; the characteristics of speed limit signs and ground signs include content, relative position to the vehicle.
  • the automatic driving controller 23 can determine whether the relative positioning based on vision can be successful based on the target semantic information and the pre-stored high-precision map, that is, determine whether the target semantic information can be successfully compared with the high-precision map. data match.
  • the process of relative positioning based on vision may be: through the target semantic information in the image collected by the front-view camera, identify the relative position of the vehicle and the elements in the high-precision map, and obtain the vehicle on the map. For example, by identifying the positional relationship between the vehicle and the lane line, it can help to determine which lane the vehicle is positioned laterally in, and by identifying the longitudinal distance between the vehicle and the speed limit sign, it can help determine where the vehicle's longitudinal positioning is on the map. Location.
  • the target semantic information will not match the data in the high-precision map, that is, vision-based relative positioning.
  • the target semantic information extracted from the image collected by the front-view camera 21 will not include the speed limit sign, while the speed limit sign in the high-precision map is not included.
  • the corresponding position includes a speed limit sign, which causes the target semantic information to not match the data in the high-precision map.
  • the target semantic information extracted from the image collected by the front-view camera 21 is the information corresponding to the dashed line, and the corresponding position in the high-precision map is a solid line, which also causes the target semantic information to not match the data in the high-precision map.
  • the automatic driving controller 23 may determine whether the target semantic information can be successfully matched with the data in the high-precision map based on the image matching algorithm.
  • Step 102 when it is determined that the target semantic information fails to match the data in the high-precision map, send error information to the server, so that the server can determine whether the high-precision map needs to be updated according to the error information.
  • the target semantic information fails to match the data in the high-precision map, it means that the actual environment has changed, and the high-precision map pre-stored in the vehicle needs to be updated.
  • the autopilot controller 23 may send error information to the server.
  • the automatic driving controller 23 sends error information to the server through the communication module 24 . After the server receives the error message, it determines whether the high-precision map needs to be updated.
  • the error information in this embodiment may include: location information, error type, image, and target semantic information determined according to the image.
  • the error type is used to indicate the element in the HD map corresponding to the data that fails to match the target semantic information. Exemplarily, assuming that lane line matching fails, the error type may be lane line.
  • the communication module 24 in the vehicle can receive the position information of satellite positioning, and can send the position information to the automatic driving controller 23 .
  • the location information may be longitude and latitude information of the vehicle.
  • the automatic driving controller 23 packages the location information, error type, image and target semantic information determined according to the image into error information, and sends the error information to the server 25.
  • the server 25 can determine whether the actual environment has changed by means of actual survey, comparison of error information of multiple vehicles, manual comparison, etc., that is, whether the high-precision map needs to be updated. Various implementations for determining whether the HD map needs to be updated are described below.
  • the server 25 determines whether the high-precision map needs to be updated through an actual survey method.
  • the implementation method can be: the server sends the error information to the designated equipment of the map provider by means of active early warning, or the map provider queries the relevant data from the server from time to time to obtain the error information; the map provider arranges collection vehicles, staff, etc.
  • the server determines whether the target semantic information extraction is correct according to the elements in the actual environment, and the error type indicates Whether the element has changed; if it is determined that the target semantic information is extracted correctly, and the element indicated by the error type sends changes, it is determined that the high-precision map needs to be updated.
  • the server 25 determines whether the high-precision map needs to be updated by comparing the error information of multiple vehicles.
  • the implementation may be: determining a preset threshold of the total number of vehicles uploading error information included in the location information in the error information; when the total number of vehicles uploading error information at the location corresponding to the location information is greater than the preset threshold, then Make sure the HD map needs to be updated.
  • the server 25 determines whether the high-precision map needs to be updated by means of manual comparison.
  • the implementation can be as follows: manually determine a plurality of elements from the image, and then determine whether the element at the position corresponding to the position information in the error information in the high-precision map matches the element determined from the image, and if the element at the position matches the element determined from the image The element does not match the element determined from the image, and it is determined that the HD map needs to be updated.
  • the server 25 can check whether the environmental elements are updated in a targeted manner, which improves the high-precision map. Update efficiency.
  • the environmental elements in this embodiment may refer to road elements.
  • the method of determining the incremental data of the map update may be: obtaining the updated data of the element corresponding to the error type collected by the map collecting vehicle at the position corresponding to the position information; according to the updated data of the element corresponding to the error type, updating The position corresponding to the latter data and the element generate incremental data for the map update.
  • the map collection vehicle when it is determined that the high-precision map needs to be updated, can be arranged to perform on-site collection at the location corresponding to the location information included in the error message.
  • the map collecting vehicle sends the collected data of the elements corresponding to the error types to the server 25 . For example, if the error type of the error information indicates that the lane lines do not match, the map collecting vehicle sends updated data of the collected lane lines to the server 25 after collecting the actual lane lines on the spot.
  • the server 25 generates incremental data for map update according to the updated data, the position corresponding to the updated data, and the element.
  • the updated data in the incremental data of the map update is collected by the map collecting vehicle, compared with the way of updating the high-precision map based on the image collected by the camera of the vehicle, through this method
  • the incremental data of the map update determined by the method is more accurate, so that the high-precision map updated according to the incremental data updated by the map is more accurate, and the safety of automatic driving is improved.
  • the incremental data of the map update includes: the updated data, the position corresponding to the updated data, and the elements corresponding to the updated data.
  • the location corresponding to the updated data refers to the location information of the updated data in the high-precision map
  • the element corresponding to the updated data refers to the changed high-precision map.
  • the updated data in the incremental data of the map update refers to the data of the lane line after the change
  • the position corresponding to the updated data refers to the data of the lane line after the change.
  • the position of the lane line in the high-precision map, the element corresponding to the updated data refers to the lane line.
  • the updated data in the incremental data of the map update refers to the environmental data after removing the sign
  • the location corresponding to the updated data refers to the removed sign
  • the location of the location in the high-precision map, the element corresponding to the updated data refers to the sign.
  • Step 103 Receive incremental data of map update sent by the server.
  • the incremental data for map update is the data sent when the server determines that the high-precision map needs to be updated according to the error information.
  • Step 104 Update the high-precision map according to the incremental data of the map update.
  • the automatic driving controller 23 After receiving the incremental data of the map update sent by the server 25, the automatic driving controller 23 updates the high-precision map according to the incremental data of the map update.
  • the autopilot controller 23 may receive incremental data for map updates through the communication module 24 .
  • the automatic driving controller 23 determines the element to be updated in the pre-stored high-precision map according to the position corresponding to the updated data and the element corresponding to the updated data;
  • the data corresponding to the elements of is updated to the updated data to form an updated high-precision map, and the updated high-precision map is used as a pre-stored high-precision map.
  • the incremental data of the map update includes: the data of the lane line after the change, the target position of the lane line after the change in the high-precision map, and the lane line.
  • the automatic driving controller determines that the element to be updated is the lane line of the target position, and then updates the data of the lane line of the target position to the data included in the incremental data of the map update. updated data.
  • the high-precision map update method includes: determining whether the target semantic information can be successfully matched with the data in the high-precision map according to the target semantic information extracted from the collected images and the pre-stored high-precision map, When it is determined that the target semantic information fails to match the data in the high-precision map, an error message is sent to the server, so that the server can determine whether the high-precision map needs to be updated according to the error information, and receive the incremental data of the map update sent by the server.
  • the updated incremental data is the data sent when the server determines that the high-precision map needs to be updated according to the error information, and the high-precision map is updated according to the incremental data of the map update.
  • the high-precision map update method on the one hand, only when the vehicle determines that the target semantic information fails to match the data in the high-precision map, the error information is sent to the server. The amount of data transmitted is small, which reduces the cost. On the other hand, since only the incremental data of the map update determined by the server needs to be transmitted, the amount of data transmitted between the vehicle and the server is further reduced, and the cost is reduced. On the one hand, when updating, only the incremental data of the map update needs to be updated, which improves the update efficiency. Therefore, the high-precision map update method has lower cost and higher update efficiency.
  • FIG. 3 is a schematic flowchart of a method for updating a high-precision map provided by another embodiment of the present application. As shown in FIG. 3 , the method for updating a high-precision map in this embodiment includes the following steps.
  • Step 301 Receive error information sent by the vehicle.
  • the error information is the information sent when the vehicle determines that the target semantic information fails to match the data in the high-precision map according to the target semantic information extracted from the collected images and the pre-stored high-precision map.
  • Step 302 According to the error information, determine whether the pre-stored high-precision map in the vehicle needs to be updated.
  • the error information includes: location information, error type, image and target semantic information determined according to the image.
  • the error type is used to indicate the element in the HD map corresponding to the data that fails to match the target semantic information.
  • the server After the server receives the error information, it can determine whether the actual environment has changed by means of actual survey, comparison of error information of multiple vehicles, manual comparison, etc., that is, to determine whether the high-precision map needs to be updated. Various implementations for determining whether the HD map needs to be updated are described below.
  • the server determines whether the high-precision map needs to be updated through the actual survey method.
  • the implementation can be as follows: reaching the position corresponding to the position information in the error information, and finding the actual environment corresponding to the collected image; according to the elements in the actual environment, determining whether the extraction of the target semantic information is correct and whether the element indicated by the error type occurs Change; if it is determined that the target semantic information is extracted correctly, and the element indicated by the error type sends changes, it is determined that the high-precision map needs to be updated.
  • the server determines whether the high-precision map needs to be updated by comparing the error information of multiple vehicles.
  • the implementation manner may be: determining the total number of vehicles at the position corresponding to the position information in the error information; when the ratio of the number of vehicles with the error information uploaded at the position corresponding to the position information to the total number of vehicles is greater than a preset threshold , it is determined that the HD map needs to be updated.
  • the server determines whether the high-precision map needs to be updated through manual comparison.
  • the implementation method can be: manually determine a plurality of elements from the image, and then determine whether the element at the position corresponding to the position information in the error information in the high-precision map matches the element determined from the image. The element at this location does not match the element determined from the image, and it is determined that the HD map needs to be updated.
  • the server can check whether the environmental elements are updated in a targeted manner, which improves the high-precision map update. s efficiency.
  • the environmental elements in this embodiment may refer to road elements.
  • Step 303 When it is determined that the high-precision map pre-stored in the vehicle needs to be updated, determine the incremental data of the map update.
  • the server obtains the updated data of the element corresponding to the error type collected by the map collecting vehicle at the position corresponding to the location information; generates a map according to the updated data of the element corresponding to the error type, the position and the element corresponding to the updated data Updated incremental data.
  • Step 304 Send the incremental data of the map update to the vehicle.
  • the server After determining the incremental data of the map update, the server sends the incremental data of the map update to the vehicle, so that the vehicle can update the high-precision map according to the incremental data of the map update.
  • the method for updating the high-precision map further includes the following steps: when it is determined that the high-precision map pre-stored in the vehicle does not need to be updated, determining the target semantic information and the information in the actual environment corresponding to the image. Whether it is consistent; when it is determined that the target semantic information is inconsistent with the information in the actual environment corresponding to the image, it is determined that there is an error in the image processing algorithm in the vehicle; when it is determined that the target semantic information is consistent with the information in the actual environment corresponding to the image, it is determined that the An error occurred in the matching algorithm used to match the target semantic information with the data in the HD map.
  • the server can determine whether the semantic information of the target is consistent with the information in the actual environment corresponding to the image by means of actual survey, manual identification, and comparison of error information uploaded by multiple vehicles.
  • Image processing algorithms refer to algorithms that extract target semantic information from images.
  • the server in this embodiment may refer to a server of a high-precision map provider.
  • the server of the high-precision map provider may feed back the determined error information to the server of the vehicle manufacturer.
  • the server in this embodiment can evaluate the accuracy and reliability of the algorithm by counting the errors of the image processing algorithm and the errors of the matching algorithm.
  • the high-precision map update method provided in this embodiment includes: receiving error information sent by the vehicle, where the error information is that the vehicle determines the target semantics according to the target semantic information extracted from the collected images and the pre-stored high-precision map The information sent when the information fails to match the data in the high-precision map. According to the error information, it is determined whether the pre-stored high-precision map in the vehicle needs to be updated. When it is determined that the pre-stored high-precision map in the vehicle needs to be updated, it is determined whether the map is updated. Incremental data, which sends incremental data for map updates to the vehicle.
  • the high-precision map update method only sends error information to the server when the vehicle determines that the target semantic information fails to match the data in the high-precision map.
  • the amount of data transmitted between the vehicle and the server is further reduced, and the cost is reduced.
  • the high-precision map update method has lower cost and higher update efficiency.
  • FIG. 4 is a schematic structural diagram of an apparatus for updating a high-precision map according to an embodiment of the present application.
  • the device for updating a high-precision map provided in this embodiment includes the following modules: a determining module 41 , a sending module 42 , a receiving module 43 , and an updating module 44 .
  • the determining module 41 is configured to determine whether the target semantic information can be successfully matched with the data in the high-precision map according to the target semantic information extracted from the collected images and the pre-stored high-precision map.
  • the sending module 42 is configured to send error information to the server when it is determined that the target semantic information fails to match the data in the high-precision map, so that the server can determine whether the high-precision map needs to be updated according to the error information.
  • the receiving module 43 is configured to receive the incremental data of the map update sent by the server.
  • the incremental data for map update is the data sent when the server determines that the high-precision map needs to be updated according to the error information.
  • the update module 44 is configured to update the high-precision map according to the incremental data of the map update.
  • the error information includes: location information, error type, image, and target semantic information determined according to the image.
  • the error type is used to indicate the element in the HD map corresponding to the data that fails to match the target semantic information.
  • the device also includes an acquisition module, which is configured to acquire the position information of the vehicle.
  • the incremental data for map update includes: updated data, positions corresponding to the updated data, and elements corresponding to the updated data.
  • the update module 44 is set to: determine the element to be updated in the pre-stored high-precision map according to the position corresponding to the updated data and the element corresponding to the updated data; , the data corresponding to the element to be updated is updated to the updated data to form an updated high-precision map, and the updated high-precision map is used as a pre-stored high-precision map.
  • the high-precision map updating apparatus provided in the embodiment of the present application can execute the high-precision map updating method provided by the embodiment shown in FIG. 1 and various optional implementation manners of the present application, and has functional modules corresponding to the execution method.
  • FIG. 5 is a schematic structural diagram of an apparatus for updating a high-precision map according to another embodiment of the present application.
  • the apparatus for updating a high-precision map provided in this embodiment includes the following modules: a receiving module 51 , a first determining module 52 , a second determining module 53 , and a sending module 54 .
  • the receiving module 51 is configured to receive the error information sent by the vehicle.
  • the error information is the information sent when the vehicle determines that the target semantic information fails to match the data in the high-precision map according to the target semantic information extracted from the collected images and the pre-stored high-precision map.
  • the first determination module 52 is configured to determine whether the pre-stored high-precision map in the vehicle needs to be updated according to the error information.
  • the second determination module 53 is configured to determine incremental data for map update when it is determined that the high-precision map pre-stored in the vehicle needs to be updated.
  • the sending module 54 is configured to send the incremental data of the map update to the vehicle.
  • the error information includes: location information, error type, image, and target semantic information determined according to the image.
  • the error type is used to indicate the element in the HD map corresponding to the data that fails to match the target semantic information.
  • the second determination module 53 is configured to determine the incremental data of the map update in the following manner: acquiring the updated data of the element corresponding to the error type collected by the map collecting vehicle at the position corresponding to the position information; The updated data, the positions and elements corresponding to the updated data, generate incremental data for map update.
  • the apparatus further includes: a third determination module, a fourth determination module and a fifth determination module.
  • the third determining module is configured to determine whether the target semantic information is consistent with the information in the actual environment corresponding to the image when it is determined that the pre-stored high-precision map in the vehicle does not need to be updated.
  • the fourth determination module is configured to determine that an error occurs in the image processing algorithm in the vehicle when it is determined that the target semantic information is inconsistent with the information in the actual environment corresponding to the image.
  • the fifth determining module is configured to determine that an error occurs in the matching algorithm in the vehicle for matching the target semantic information and the data in the high-precision map when it is determined that the target semantic information is consistent with the information in the actual environment corresponding to the image.
  • the high-precision map updating apparatus provided in the embodiment of the present application can execute the high-precision map updating method provided in the embodiment shown in FIG. 3 and various optional implementation manners of the present application, and has functional modules corresponding to the execution method.
  • FIG. 6 is a schematic structural diagram of a vehicle provided by the present application. As shown in FIG. 6 , the vehicle includes a processor 60 and a memory 61 .
  • the number of processors 60 in the vehicle may be one or more, and one processor 60 is taken as an example in FIG. 6 ; the processor 60 and the memory 61 of the vehicle may be connected by a bus or in other ways, and in FIG. 6 it is connected by a bus For example.
  • the memory 61 can be configured to store software programs, computer-executable programs, and modules, such as program instructions and modules corresponding to the high-precision map update method in the embodiments of the present application (for example, high-precision map update methods).
  • the processor 60 executes various functional applications and data processing of the vehicle by running the software programs, instructions and modules stored in the memory 61 , that is, to implement the above-mentioned high-precision map update method.
  • the memory 61 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the vehicle, and the like.
  • the memory 61 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage device.
  • memory 61 may include memory located remotely from processor 60, which may be connected to the vehicle via a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
  • FIG. 7 is a schematic structural diagram of a server provided by the present application.
  • the server includes a processor 70 and a memory 71 .
  • the number of processors 70 in the server can be one or more, and one processor 70 is taken as an example in FIG. 7 ; the processor 70 and the memory 71 of the server can be connected by a bus or in other ways, and in FIG. 7, the connection is by a bus For example.
  • the memory 71 can be configured to store software programs, computer-executable programs, and modules, such as program instructions and modules corresponding to the high-precision map update method in the embodiments of the present application (for example, the high-precision map update method).
  • the processor 70 executes various functional applications and data processing of the server by running the software programs, instructions and modules stored in the memory 71 , that is, to implement the above-mentioned high-precision map update method.
  • the memory 71 may mainly include a stored program area and a stored data area, wherein the stored program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the server, and the like.
  • Memory 71 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage device.
  • memory 71 may include memory located remotely from processor 70, which may be connected to a server through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
  • the present application also provides a storage medium containing computer-executable instructions, when executed by a computer processor, the computer-executable instructions are used to execute a method for updating a high-precision map, the method comprising: The extracted target semantic information and the pre-stored high-precision map determine whether the target semantic information can be successfully matched with the data in the high-precision map; when it is determined that the target semantic information matches the data in the high-precision map When the matching fails, send error information to the server, so that the server can determine whether the high-precision map needs to be updated according to the error information; receive the incremental data of the map update sent by the server, wherein the map update The incremental data is the data sent when the server determines that the high-precision map needs to be updated according to the error information; the high-precision map is updated according to the incremental data of the map update.
  • the present application also provides a storage medium containing computer-executable instructions, where the computer-executable instructions are used to execute a method for updating a high-precision map when executed by a computer processor, the method comprising: receiving error information sent by a vehicle, The error information is sent when the vehicle determines that the target semantic information fails to match the data in the high-precision map according to the target semantic information extracted from the collected images and the pre-stored high-precision map information; according to the error information, determine whether the pre-stored high-precision map in the vehicle needs to be updated; when it is determined that the pre-stored high-precision map in the vehicle needs to be updated, determine the incremental data of the map update; The vehicle sends incremental data for the map update.
  • a storage medium containing computer-executable instructions provided by an embodiment of the present application the computer-executable instructions of the computer-executable instructions are not limited to the above-mentioned method operations, and can also execute any of the high-precision map updating methods provided by any embodiment of the present application. related operations.
  • the present application can be implemented by means of software and general hardware, and can also be implemented by hardware. Based on this understanding, the technical solution of the present application can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a floppy disk of a computer, a read-only memory (Read-Only Memory, ROM), Random access memory (Random Access Memory, RAM), flash memory (FLASH), hard disk or optical disk, etc., including several instructions to enable a computer device (which may be a personal computer, server, or network device, etc.) to execute multiple methods described in the examples.
  • a computer-readable storage medium such as a floppy disk of a computer, a read-only memory (Read-Only Memory, ROM), Random access memory (Random Access Memory, RAM), flash memory (FLASH), hard disk or optical disk, etc.
  • the multiple units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be realized;
  • the names are only for the convenience of distinguishing from each other, and are not used to limit the protection scope of this application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Remote Sensing (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Fuzzy Systems (AREA)
  • Mathematical Physics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Software Systems (AREA)
  • Computational Linguistics (AREA)
  • Traffic Control Systems (AREA)
  • Navigation (AREA)

Abstract

一种高精地图更新方法、车辆、服务器及存储介质，该方法包括：根据从采集到的图像中提取出的目标语义信息以及预先存储的高精地图，确定目标语义信息是否能成功与高精地图中的数据匹配（步骤101），在确定目标语义信息与高精地图中的数据匹配失败的情况下，向服务器发送错误信息，以使服务器根据错误信息确定高精地图是否需要更新（步骤102），接收服务器发送的地图更新的增量数据（步骤103），其中，地图更新的增量数据为服务器在根据错误信息确定需要更新高精地图的情况下发送的数据，根据地图更新的增量数据更新高精地图（步骤104）。

Description

高精地图更新方法、车辆、服务器及存储介质
本申请要求在2020年07月08日提交中国专利局、申请号为202010659658.9的中国专利申请的优先权，该申请的全部内容通过引用结合在本申请中。
技术领域
本申请实施例涉及自动驾驶领域，例如涉及一种高精地图更新方法、车辆、服务器及存储介质。
背景技术
第三级别(Level 3，L3)(有条件自动驾驶)自动驾驶车辆的功能的实现需要使用高精地图。高精度地图采集依赖激光雷达、摄像头、高精组合惯导等传感器，且需要采图方具备一定的测绘资质和专业的测绘能力，因此数据采集、制作及维护的成本极高。而随着道路不断修建，高精地图的更新比初始地图的制作更为重要，采图手段受限于成本等因素无法满足数据更新的要求。
行业中提出了基于众包的方式对高精地图进行更新，主要流程为：设置于自动驾驶车辆上的摄像头采集图像，将采集的图像上传至服务器中，服务器从该图像中提取出道路特征元素，并将道路特征元素与高精地图进行实时比对，如果发现从该图像中提取出的道路特征元素与高精地图不一致时，进行高精地图更新。
但是，上述方式需要将采集到的图像上传至服务器中，消耗了大量的流量用于数据上传，需要较高的成本。
发明内容
本申请提供一种高精地图更新方法、车辆、服务器及存储介质，解决了高精地图更新需消耗大量的流量而造成成本较高的技术问题。
本申请实施例提供一种高精地图更新方法，包括：根据从采集到的图像中提取出的目标语义信息以及预先存储的高精地图，确定所述目标语义信息是否能成功与所述高精地图中的数据匹配；在确定所述目标语义信息与所述高精地图中的数据匹配失败的情况下，向服务器发送错误信息，以使所述服务器根据所述错误信息确定所述高精地图是否需要更新；接收所述服务器发送的地图更新的增量数据，其中，所述地图更新的增量数据为所述服务器根在据所述错误信息确定需要更新所述高精地图的情况下发送的数据；根据所述地图更新的增 量数据更新所述高精地图。
本申请实施例提供一种高精地图更新方法，包括：接收车辆发送的错误信息，其中，所述错误信息为所述车辆在根据从采集到的图像中提取出的目标语义信息以及预先存储的高精地图，确定所述目标语义信息与所述高精地图中的数据匹配失败的情况下发送的信息；根据所述错误信息，确定所述车辆中预先存储的高精地图是否需要更新；在确定所述车辆中预先存储的高精地图需要更新的情况下，确定地图更新的增量数据；向所述车辆发送所述地图更新的增量数据。
本申请实施例提供一种车辆，包括：一个或多个处理器；存储器，设置为存储一个或多个程序；当所述一个或多个程序被所述一个或多个处理器执行，使得所述一个或多个处理器实现上述第一种高精地图更新方法。
本申请实施例还提供了一种服务器，包括：一个或多个处理器；存储器，设置为存储一个或多个程序；当所述一个或多个程序被所述一个或多个处理器执行，使得所述一个或多个处理器实现上述第二种高精地图更新方法。
本申请实施例还提供了一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现上述第一种或第二种高精地图更新方法。
附图说明
图1为本申请一实施例提供的高精地图更新方法的流程示意图；
图2为本申请一实施例提供的车辆的组成结构示意图；
图3为本申请另一实施例提供的高精地图更新方法的流程示意图；
图4为本申请一实施例提供的高精地图更新装置的结构示意图；
图5为本申请另一实施例提供的高精地图更新装置的结构示意图；
图6为本申请提供的车辆的结构示意图；
图7为本申请提供的服务器的结构示意图。
具体实施方式
下面结合附图和实施例对本申请进行说明。可以理解的是，此处所描述的实施例仅仅用于解释本申请，而非对本申请的限定。为了便于描述，附图中仅示出了与本申请相关的部分而非全部结构。
图1为本申请一实施例提供的高精地图更新方法的流程示意图。本实施例 适用于自动驾驶车辆行驶中，更新高精地图的场景。本实施例可以由高精地图更新装置来执行，该高精地图更新装置可以由软件和/或硬件的方式实现，该高精地图更新装置可以集成于车辆的自动驾驶控制器中。如图1所示，本实施例提供的高精地图更新方法包括如下步骤。
步骤101：根据从采集到的图像中提取出的目标语义信息以及预先存储的高精地图，确定目标语义信息是否能成功与高精地图中的数据匹配。
本实施例中的车辆可以为具有自动驾驶功能的车辆。图2为本申请一实施例提供的车辆的组成结构示意图。如图2所示，该车辆可以包括：前视摄像头21、高精地图存储模块22、自动驾驶控制器23以及通信模块24。前视摄像头21、高精地图存储模块22以及通信模块24均与自动驾驶控制器23连接。通信模块24设置为实现自动驾驶控制器23与服务器25之间的通信。可选地，该车辆中还可以包括与自动驾驶控制器23连接的其他传感器。
该车辆中的通信模块24可以为车载远程信息处理盒子(Telematics BOX，T-BOX)。该车辆中的高精地图存储模块22中存储有高精度地图。
本实施例中，在开启自动驾驶模式后，车辆中的前视摄像头21采集周围环境的图像，例如，道路的图像，并将周围环境的图像发送给自动驾驶控制器23。可选地，前视摄像头21可以通过低电压差分信号(Low-Voltage Differential Signaling，LVDS)线将图像传输至自动驾驶控制器23中。LVDS线采用了LVDS技术传输图像，具有低功耗、低误码率、低串扰和低辐射的优点。高精地图存储模块22可以通过以太网(Ethernet，ETH)通信方式将高精地图传输至自动驾驶控制器23。其他传感器可以通过ETH通信方式或者控制器局域网(Controller Area Network，CAN)网络与自动驾驶控制器23交互信息。
高精地图存储模块22可以将预先存储的高精地图传输至自动驾驶控制器23中。
自动驾驶控制器23接收到前视摄像头21采集到的图像后，可以从图像中提取出目标语义信息。自动驾驶控制器23可以采用图像处理算法从图像中提取出目标语义信息。这里的图像处理算法可以为语义分割算法。
本实施例中的目标语义信息可以包括道路元素的特征，如车道线、护栏、路缘石、限速牌、地面标识、广告牌、龙门架等的特征。车道线的特征包括颜色、线型、与车辆的相对位置、曲率等；护栏、路缘石、广告牌和龙门架的特征主要包括与车辆的相对位置；限速牌和地面标识的特征包括内容、与车辆的相对位置。
自动驾驶控制器23在提取出目标语义信息之后，可以基于目标语义信息以 及预先存储的高精地图，确定基于视觉的相对定位是否能成功，即，确定目标语义信息是否能成功与高精地图中的数据匹配。
一种实现方式中，基于视觉的相对定位的过程可以为：通过前视摄像头采集到的图像中的目标语义信息，识别出车辆与高精地图中的元素的相对位置，得出车辆在地图上的相对位置，例如通过识别车辆与车道线的位置关系可以辅助判断出车辆的横向定位在第几条车道内，通过识别车辆与限速牌的纵向距离辅助判断出车辆的纵向定位在地图的什么位置。
但是，如果前视摄像头采集到的图像对应的实际环境中的元素已经变更，而高精地图还未更新，就会导致目标语义信息与高精地图中的数据无法匹配，即无法实现基于视觉的相对定位。
举例来说，一种场景中，假设实际环境中原有的限速牌被拆除，则从前视摄像头21采集到的图像中提取出目标语义信息中不会包括限速牌，而高精地图中的相应位置处包括限速牌，这就导致目标语义信息与高精地图中的数据无法匹配。另一种场景中，假设实际环境中的车道线由实线变为虚线，则从前视摄像头21采集到的图像中提取出的目标语义信息为虚线对应的信息，而高精地图中的相应位置处为实线，这也会导致目标语义信息与高精地图中的数据无法匹配。
在步骤101中，自动驾驶控制器23可以基于图像匹配算法，确定目标语义信息是否能成功与高精地图中的数据匹配。
步骤102：当确定目标语义信息与高精地图中的数据匹配失败时，向服务器发送错误信息，以使服务器根据错误信息确定高精地图是否需要更新。
当确定目标语义信息与高精地图中的数据匹配失败时，说明实际环境已经发生了变化，车辆中预先存储的高精地图需要更新。
此时，自动驾驶控制器23可以向服务器发送错误信息。自动驾驶控制器23通过通信模块24向服务器发送错误信息。服务器接收到该错误信息之后，确定高精地图是否需要更新。
可选地，本实施例中的错误信息可以包括：位置信息、错误类型、图像以及根据图像确定出的目标语义信息。错误类型用于指示高精地图中与目标语义信息匹配失败的数据对应的元素。示例性地，假设车道线匹配失败，则错误类型可以为车道线。
一实施例中，车辆中的通信模块24可以接收卫星定位的位置信息，并可以将位置信息发送至自动驾驶控制器23中。示例性地，该位置信息可以为车辆的经度及纬度信息。
在确定目标语义信息与高精地图中的数据匹配失败时，自动驾驶控制器23将位置信息、错误类型、图像以及根据图像确定出的目标语义信息打包成错误信息，并将错误信息发送给服务器25。
服务器25在接收到错误信息之后，可以通过实际勘察、多个车辆的错误信息比对、人工对照等方式确定实际环境是否发生变化，即确定高精地图是否需要更新。以下对确定高精地图是否需要更新的多种实现方式进行描述。
第一种实现方式中，服务器25通过实际勘察方式，确定高精地图是否需要更新。该实现方式可以为：服务器通过主动预警的方式将错误信息发送地图提供方的指定设备，或地图提供方不定期从服务器中查询相关数据来获得错误信息；地图提供方安排采集车、工作人员等到达错误信息中的位置信息对应的位置处，找到采集到的图像对应的实际环境，并将实际环境反馈给服务器；服务器根据实际环境中的元素，确定目标语义信息提取是否正确，错误类型指示的元素是否发生变化；如果确定目标语义信息提取正确，且错误类型指示的元素发送变化，则确定高精地图需要更新。
第二种实现方式中，服务器25通过多个车辆的错误信息比对的方式，确定高精地图是否需要更新。该实现方式可以为：确定错误信息中的位置信息包括的上传错误信息的车辆的总数量的预设阈值；当该位置信息对应的位置处上传错误信息的车辆的总数量大于预设阈值，则确定高精地图需要更新。
第三种实现方式中，服务器25通过人工对照的方式，确定高精地图是否需要更新。该实现方式可以为：人工从图像中确定出多个元素，再确定高精地图中错误信息中的位置信息对应的位置处的元素是否与从图像中确定出的元素匹配，如果该位置处的元素与从图像中确定出的元素不匹配，确定高精地图需要更新。
可以看出，基于错误信息包括位置信息、错误类型、图像以及根据图像确定出的目标语义信息，服务器25在接收错误信息之后，可以有针对性地检查环境元素是否有更新，提高了高精地图更新的效率。本实施例中的环境元素可以指道路元素。
服务器25在确定高精地图需要更新时，需要确定地图更新的增量数据。确定地图更新的增量数据的方式可以为：获取地图采集车在位置信息对应的位置处采集到的错误类型对应的元素的更新后的数据；根据错误类型对应的元素的更新后的数据、更新后的数据对应的位置以及该元素，生成地图更新的增量数据。
在该实现方式中，当确定高精地图需要更新时，可以安排地图采集车到错 误信息中包括的位置信息对应的位置处进行现场采集。地图采集车将采集到的错误类型对应的元素的数据发送给服务器25。例如，错误信息的错误类型指示车道线不匹配，则地图采集车在现场采集到实际的车道线后，将采集到的车道线更新后的数据发送给服务器25。
服务器25根据该更新后的数据、更新后的数据对应的位置以及该元素，生成地图更新的增量数据。
在该实现方式中，由于地图更新的增量数据中的更新后的数据是通过地图采集车采集到的，相较于基于车辆的摄像头采集到的图像进行高精地图更新的方式，通过这种方式确定出的地图更新的增量数据更为精准，从而使得根据该地图更新的增量数据更新后的高精地图较为精准，提高自动驾驶的安全性。
地图更新的增量数据包括：更新后的数据、更新后的数据对应的位置以及更新后的数据对应的元素。
地图更新的增量数据中，更新后的数据对应的位置指的是该更新后的数据在高精地图中的位置信息，更新后的数据对应的元素指的是发生变化的高精地图中的元素的标识。举例来说，假设车道线发生变化，该场景中，地图更新的增量数据中的更新后的数据指的是变化后的车道线的数据，更新后的数据对应的位置指的是该变化后的车道线在高精地图中的位置，更新后的数据对应的元素指的是车道线。假设标识牌被移除，该场景中，地图更新的增量数据中的更新后的数据指的是移除标识牌后的环境数据，更新后的数据对应的位置指的是该移除标识牌的位置在高精地图中的位置，更新后的数据对应的元素指的是标识牌。
步骤103：接收服务器发送的地图更新的增量数据。
地图更新的增量数据为服务器根据错误信息确定需要更新高精地图时发送的数据。
步骤104：根据地图更新的增量数据更新高精地图。
自动驾驶控制器23接收服务器25发送的地图更新的增量数据后，根据地图更新的增量数据更新高精地图。自动驾驶控制器23可以通过通信模块24接收地图更新的增量数据。
可选地，自动驾驶控制器23根据更新后的数据对应的位置以及更新后的数据对应的元素，确定预先存储的高精地图中待更新的元素；将预先存储的高精地图中，待更新的元素对应的数据更新为更新后的数据，形成更新后的高精地图，将更新后的高精地图作为预先存储的高精地图。
举例来说，假设地图更新的增量数据包括：变化后的车道线的数据、该变 化后的车道线在高精地图中的目标位置以及车道线。自动驾驶控制器在接收到该地图更新的增量数据之后，确定出待更新的元素为目标位置的车道线，之后，将该目标位置的车道线的数据更新为地图更新的增量数据包括的更新后的数据。
本实施例中，一方面，由于只需要传输服务器25确定的地图更新的增量数据，减小了车辆与服务器之间的传输的数据量，降低了成本，另一方面，在更新时，只需要更新该地图更新的增量数据，提高了更新效率。
本实施例提供的高精地图更新方法，包括：根据从采集到的图像中提取出的目标语义信息以及预先存储的高精地图，确定目标语义信息是否能成功与高精地图中的数据匹配，当确定目标语义信息与高精地图中的数据匹配失败时，向服务器发送错误信息，以使服务器根据错误信息确定高精地图是否需要更新，接收服务器发送的地图更新的增量数据，其中，地图更新的增量数据为服务器根据错误信息确定需要更新高精地图时发送的数据，根据地图更新的增量数据更新高精地图。该高精地图更新方法中，一方面，在车辆确定目标语义信息与高精地图中的数据匹配失败时，才向服务器发送错误信息，相较于向服务器发送采集到的图像的方式，向服务器传输的数据量较小，降低了成本，另一方面，由于只需要传输服务器确定的地图更新的增量数据，进一步减小了车辆与服务器之间的传输的数据量，降低了成本，另一方面，在更新时，只需要更新该地图更新的增量数据，提高了更新效率。因此，该高精地图更新方法成本较低并且更新效率较高。
图3为本申请另一实施例提供的高精地图更新方法的流程示意图。如图3所示，本实施例的高精地图更新方法包括如下步骤。
步骤301：接收车辆发送的错误信息。
错误信息为车辆根据从采集到的图像中提取出的目标语义信息以及预先存储的高精地图，确定目标语义信息与高精地图中的数据匹配失败时发送的信息。
步骤302：根据错误信息，确定车辆中预先存储的高精地图是否需要更新。
该错误信息包括：位置信息、错误类型、图像以及根据图像确定出的目标语义信息。错误类型用于指示高精地图中与目标语义信息匹配失败的数据对应的元素。
服务器在接收到错误信息之后，可以通过实际勘察、多个车辆的错误信息比对、人工对照等方式确定实际环境是否发生变化，即确定高精地图是否需要更新。以下对确定高精地图是否需要更新的多种实现方式进行描述。
第一种实现方式中，服务器通过实际勘察方式，确定高精地图是否需要更 新。该实现方式可以为：到达错误信息中的位置信息对应的位置处，找到采集到的图像对应的实际环境；根据实际环境中的元素，确定目标语义信息提取是否正确，错误类型指示的元素是否发生变化；如果确定目标语义信息提取正确，且错误类型指示的元素发送变化，则确定高精地图需要更新。
第二种实现方式中，服务器通过多个车辆的错误信息比对的方式，确定高精地图是否需要更新。该实现方式可以为：确定错误信息中的位置信息对应的位置处的车辆的总数量；当该位置信息对应的位置处上传的错误信息的车辆的数量与车辆的总数量的比值大于预设阈值，则确定高精地图需要更新。
第三种实现方式中，服务器通过人工对照的方式，确定高精地图是否需要更新。该实现方式可以为：人工从图像中确定出多个元素，再确定高精地图中错误信息中的位置信息对应的位置处的元素是否与从图像中确定出的元素匹配，如果高精地图中该位置处的元素与从图像中确定出的元素不匹配，确定高精地图需要更新。
可以看出，基于错误信息包括位置信息、错误类型、图像以及根据图像确定出的目标语义信息，服务器在接收错误信息之后，可以有针对性地检查环境元素是否有更新，提高了高精地图更新的效率。本实施例中的环境元素可以指道路元素。
步骤303：当确定车辆中预先存储的高精地图需要更新时，确定地图更新的增量数据。
服务器获取地图采集车在位置信息对应的位置处采集到的错误类型对应的元素的更新后的数据；根据错误类型对应的元素的更新后的数据、更新后的数据对应的位置以及元素，生成地图更新的增量数据。
步骤304：向车辆发送地图更新的增量数据。
服务器在确定出地图更新的增量数据后，向车辆发送地图更新的增量数据，以使车辆根据该地图更新的增量数据，更新高精地图。
一种实现方式中，本实施例提供的高精地图的更新方法还包括如下步骤：当确定车辆中预先存储的高精地图不需要更新时，确定目标语义信息与图像对应的实际环境中的信息是否一致；当确定目标语义信息与图像对应的实际环境中的信息不一致时，确定车辆中的图像处理算法出现错误；当确定目标语义信息与图像对应的实际环境中的信息一致时，确定车辆中的用于匹配目标语义信息与高精地图中的数据的匹配算法出现错误。
服务器可以通过实际勘察、人工识别以及多个车辆上传的错误信息比对等方式，确定目标语义信息与图像对应的实际环境中的信息是否一致。
图像处理算法指的是从图像中提取目标语义信息的算法。
在上述实现方式中，可以排查图像处理算法是否出现错误，还可以排查匹配算法是否出现错误。
可选地，本实施例中的服务器可以指的是高精地图提供商的服务器。当确定图像处理算法出现错误或者匹配算法出现错误时，高精地图提供商的服务器可以向车辆制造商的服务器反馈确定出的错误信息。
本实施例中的服务器可以通过对图像处理算法的错误及匹配算法的错误进行统计，对算法准确性和可靠性进行评价。
本实施例提供的高精地图更新方法，包括：接收车辆发送的错误信息，其中，错误信息为车辆根据从采集到的图像中提取出的目标语义信息以及预先存储的高精地图，确定目标语义信息与高精地图中的数据匹配失败时发送的信息，根据错误信息，确定车辆中预先存储的高精地图是否需要更新，当确定车辆中预先存储的高精地图需要更新时，确定地图更新的增量数据，向车辆发送地图更新的增量数据。该高精地图更新方法，一方面，在车辆确定目标语义信息与高精地图中的数据匹配失败时，才向服务器发送错误信息，相较于向服务器发送采集到的图像的方式，向服务器传输的数据量较小，降低了成本，另一方面，由于只需要传输服务器确定的地图更新的增量数据，进一步减小了车辆与服务器之间的传输的数据量，降低了成本，另一方面，在更新时，只需要更新该地图更新的增量数据，提高了更新效率。因此，该高精地图更新方法成本较低并且更新效率较高。
图4为本申请一实施例提供的高精地图更新装置的结构示意图。如图4所示，本实施例提供的高精地图更新装置包括如下模块：确定模块41、发送模块42、接收模块43以及更新模块44。
确定模块41设置为根据从采集到的图像中提取出的目标语义信息以及预先存储的高精地图，确定目标语义信息是否能成功与高精地图中的数据匹配。
发送模块42设置为当确定目标语义信息与高精地图中的数据匹配失败时，向服务器发送错误信息，以使服务器根据错误信息确定高精地图是否需要更新。
接收模块43设置为接收服务器发送的地图更新的增量数据。
地图更新的增量数据为服务器根据错误信息确定需要更新高精地图时发送的数据。
更新模块44设置为根据地图更新的增量数据更新高精地图。
可选地，错误信息包括：位置信息、错误类型、图像以及根据图像确定出 的目标语义信息。错误类型用于指示高精地图中与目标语义信息匹配失败的数据对应的元素。
该装置还包括获取模块，设置为获取车辆的位置信息。
可选地，地图更新的增量数据包括：更新后的数据、更新后的数据对应的位置以及更新后的数据对应的元素。
一种实现方式中，更新模块44设置为：根据更新后的数据对应的位置以及更新后的数据对应的元素，确定预先存储的高精地图中待更新的元素；将预先存储的高精地图中，待更新的元素对应的数据更新为更新后的数据，形成更新后的高精地图，将更新后的高精地图作为预先存储的高精地图。
本申请实施例所提供的高精地图更新装置可执行本申请图1所示实施例及多种可选的实现方式所提供的高精地图更新方法，具备执行方法相应的功能模块。
图5为本申请另一实施例提供的高精地图更新装置的结构示意图。如图5所示，本实施例提供的高精地图更新装置包括如下模块：接收模块51、第一确定模块52、第二确定模块53以及发送模块54。
接收模块51设置为接收车辆发送的错误信息。
错误信息为车辆根据从采集到的图像中提取出的目标语义信息以及预先存储的高精地图，确定目标语义信息与高精地图中的数据匹配失败时发送的信息。
第一确定模块52设置为根据错误信息，确定车辆中预先存储的高精地图是否需要更新。
第二确定模块53设置为当确定车辆中预先存储的高精地图需要更新时，确定地图更新的增量数据。
发送模块54设置为向车辆发送地图更新的增量数据。
可选地，错误信息包括：位置信息、错误类型、图像以及根据图像确定出的目标语义信息。错误类型用于指示高精地图中与目标语义信息匹配失败的数据对应的元素。
第二确定模块53设置为通过如下方式确定地图更新的增量数据：获取地图采集车在位置信息对应的位置处采集到的错误类型对应的元素的更新后的数据；根据错误类型对应的元素的更新后的数据、更新后的数据对应的位置以及元素，生成地图更新的增量数据。
该装置还包括：第三确定模块、第四确定模块以及第五确定模块。
第三确定模块设置为当确定车辆中预先存储的高精地图不需要更新时，确 定目标语义信息与图像对应的实际环境中的信息是否一致。
第四确定模块设置为当确定目标语义信息与图像对应的实际环境中的信息不一致时，确定车辆中的图像处理算法出现错误。
第五确定模块设置为当确定目标语义信息与图像对应的实际环境中的信息一致时，确定车辆中的用于匹配目标语义信息与高精地图中的数据的匹配算法出现错误。
本申请实施例所提供的高精地图更新装置可执行本申请图3所示实施例及多种可选的实现方式中所提供的高精地图更新方法，具备执行方法相应的功能模块。
图6为本申请提供的车辆的结构示意图。如图6所示，该车辆包括处理器60以及存储器61。该车辆中处理器60的数量可以是一个或多个，图6中以一个处理器60为例；该车辆的处理器60和存储器61可以通过总线或其他方式连接，图6中以通过总线连接为例。
存储器61作为一种计算机可读存储介质，可设置为存储软件程序、计算机可执行程序以及模块，如本申请实施例中的高精地图更新方法对应的程序指令以及模块(例如，高精地图更新装置中的确定模块41、发送模块42、接收模块43以及更新模块44)。处理器60通过运行存储在存储器61中的软件程序、指令以及模块，从而执行车辆的多种功能应用以及数据处理，即实现上述的高精地图更新方法。
存储器61可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序；存储数据区可存储根据车辆的使用所创建的数据等。存储器61可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他非易失性固态存储器件。在一些实施例中，存储器61可包括相对于处理器60远程设置的存储器，这些远程存储器可以通过网络连接至车辆。上述网络的实施例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
图7为本申请提供的服务器的结构示意图。如图7所示，该服务器包括处理器70以及存储器71。该服务器中处理器70的数量可以是一个或多个，图7中以一个处理器70为例；该服务器的处理器70和存储器71可以通过总线或其他方式连接，图7中以通过总线连接为例。
存储器71作为一种计算机可读存储介质，可设置为存储软件程序、计算机可执行程序以及模块，如本申请实施例中的高精地图更新方法对应的程序指令以及模块(例如，高精地图更新装置中的接收模块51、第一确定模块52、第二 确定模块53以及发送模块54)。处理器70通过运行存储在存储器71中的软件程序、指令以及模块，从而执行服务器的多种功能应用以及数据处理，即实现上述的高精地图更新方法。
存储器71可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序；存储数据区可存储根据服务器的使用所创建的数据等。存储器71可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他非易失性固态存储器件。在一些实施例中，存储器71可包括相对于处理器70远程设置的存储器，这些远程存储器可以通过网络连接至服务器。上述网络的实施例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
本申请还提供一种包含计算机可执行指令的存储介质，所述计算机可执行指令在由计算机处理器执行时用于执行一种高精地图更新方法，该方法包括：根据从采集到的图像中提取出的目标语义信息以及预先存储的高精地图，确定所述目标语义信息是否能成功与所述高精地图中的数据匹配；当确定所述目标语义信息与所述高精地图中的数据匹配失败时，向服务器发送错误信息，以使所述服务器根据所述错误信息确定所述高精地图是否需要更新；接收所述服务器发送的地图更新的增量数据，其中，所述地图更新的增量数据为所述服务器根据所述错误信息确定需要更新所述高精地图时发送的数据；根据所述地图更新的增量数据更新所述高精地图。
本申请还提供一种包含计算机可执行指令的存储介质，所述计算机可执行指令在由计算机处理器执行时用于执行一种高精地图更新方法，该方法包括：接收车辆发送的错误信息，其中，所述错误信息为所述车辆根据从采集到的图像中提取出的目标语义信息以及预先存储的高精地图，确定所述目标语义信息与所述高精地图中的数据匹配失败时发送的信息；根据所述错误信息，确定所述车辆中预先存储的高精地图是否需要更新；当确定所述车辆中预先存储的高精地图需要更新时，确定地图更新的增量数据；向所述车辆发送所述地图更新的增量数据。
本申请实施例所提供的一种包含计算机可执行指令的存储介质，其计算机可执行指令不限于如上所述的方法操作，还可以执行本申请任意实施例所提供的高精地图更新方法中的相关操作。
通过以上关于实施方式的描述，所属领域的技术人员可以了解到，本申请可借助软件及通用硬件来实现，也可以通过硬件实现。基于这样的理解，本申请的技术方案可以以软件产品的形式体现出来，该计算机软件产品可以存储在计算机可读存储介质中，如计算机的软盘、只读存储器(Read-Only Memory, ROM)、随机存取存储器(Random Access Memory,RAM)、闪存(FLASH)、硬盘或光盘等，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请多个实施例所述的方法。
上述高精地图更新装置的实施例中，所包括的多个单元和模块只是按照功能逻辑进行划分的，但并不局限于上述的划分，只要能够实现相应的功能即可；多个功能单元的名称也只是为了便于相互区分，并不用于限制本申请的保护范围。

Claims (10)

  1. 一种高精地图更新方法，包括：
    根据从采集到的图像中提取出的目标语义信息以及预先存储的高精地图，确定所述目标语义信息是否能成功与所述高精地图中的数据匹配；
    在确定所述目标语义信息与所述高精地图中的数据匹配失败的情况下，向服务器发送错误信息，以使所述服务器根据所述错误信息确定所述高精地图是否需要更新；
    接收所述服务器发送的地图更新的增量数据，其中，所述地图更新的增量数据为所述服务器在根据所述错误信息确定需要更新所述高精地图的情况下发送的数据；
    根据所述地图更新的增量数据更新所述高精地图。
  2. 根据权利要求1所述的方法，其中，所述错误信息包括：位置信息、错误类型、所述图像以及根据所述图像确定出的目标语义信息；其中，所述错误类型用于指示所述高精地图中与所述目标语义信息匹配失败的数据对应的元素；
    所述向服务器发送错误信息之前，所述方法还包括：获取车辆的位置信息。
  3. 根据权利要求2所述的方法，其中，所述地图更新的增量数据包括：更新后的数据、所述更新后的数据对应的位置、以及所述更新后的数据对应的元素。
  4. 根据权利要求3所述的方法，其中，所述根据所述地图更新的增量数据更新所述高精地图，包括：
    根据所述更新后的数据对应的位置以及所述更新后的数据对应的元素，确定所述预先存储的高精地图中待更新的元素；
    将所述预先存储的高精地图中，所述待更新的元素对应的数据更新为所述更新后的数据，形成更新后的高精地图，将所述更新后的高精地图作为所述预先存储的高精地图。
  5. 一种高精地图更新方法，包括：
    接收车辆发送的错误信息，其中，所述错误信息为所述车辆在根据从采集到的图像中提取出的目标语义信息以及预先存储的高精地图，确定所述目标语义信息与所述高精地图中的数据匹配失败的情况下发送的信息；
    根据所述错误信息，确定所述车辆中预先存储的所述高精地图是否需要更新；
    在确定所述车辆中预先存储的所述高精地图需要更新的情况下，确定地图更新的增量数据；
    向所述车辆发送所述地图更新的增量数据。
  6. 根据权利要求5所述的方法，其中，所述错误信息包括：位置信息、错误类型、所述图像以及根据所述图像确定出的目标语义信息；其中，所述错误类型用于指示所述高精地图中与所述目标语义信息匹配失败的数据对应的元素；
    所述在确定所述车辆中预先存储的所述高精地图需要更新的情况下，确定地图更新的增量数据，包括：
    获取地图采集车在所述位置信息对应的位置处采集到的所述错误类型对应的元素的更新后的数据；
    根据所述错误类型对应的元素的更新后的数据、更新后的数据对应的位置以及所述元素，生成所述地图更新的增量数据。
  7. 根据权利要求6所述的方法，还包括：
    在确定所述车辆中预先存储的所述高精地图不需要更新的情况下，确定所述目标语义信息与所述图像对应的实际环境中的信息是否一致；
    在确定所述目标语义信息与所述图像对应的实际环境中的信息不一致的情况下，确定所述车辆中的用于从所述图像中提取所述目标语义信息的图像处理算法出现错误；
    在确定所述目标语义信息与所述图像对应的实际环境中的信息一致的情况下，确定所述车辆中的用于匹配所述目标语义信息与所述高精地图中的数据的匹配算法出现错误。
  8. 一种车辆，包括：
    至少一个处理器；
    存储器，设置为存储至少一个程序；
    当所述至少一个程序被所述至少一个处理器执行时，使得所述至少一个处理器实现如权利要求1-4中任一所述的高精地图更新方法。
  9. 一种服务器，包括：
    至少一个处理器；
    存储器，设置为存储至少一个程序；
    当所述至少一个程序被所述至少一个处理器执行时，使得所述至少一个处 理器实现如权利要求5-7中任一所述的高精地图更新方法。
  10. 一种计算机可读存储介质，存储有计算机程序，该程序被处理器执行时实现如权利要求1-4中任一所述的高精地图更新方法，或者，实现如权利要求5-7中任一所述的高精地图更新方法。
PCT/CN2021/104861 2020-07-08 2021-07-07 高精地图更新方法、车辆、服务器及存储介质 WO2022007818A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010659658.9 2020-07-08
CN202010659658.9A CN111858805A (zh) 2020-07-08 2020-07-08 高精地图更新方法、车辆、服务器及存储介质

Publications (1)

Publication Number Publication Date
WO2022007818A1 true WO2022007818A1 (zh) 2022-01-13

Family

ID=73152216

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/104861 WO2022007818A1 (zh) 2020-07-08 2021-07-07 高精地图更新方法、车辆、服务器及存储介质

Country Status (2)

Country Link
CN (1) CN111858805A (zh)
WO (1) WO2022007818A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114661842A (zh) * 2022-02-25 2022-06-24 阿波罗智联(北京)科技有限公司 地图匹配方法、装置和电子设备
CN115164912A (zh) * 2022-06-24 2022-10-11 宁波均胜智能汽车技术研究院有限公司 一种车辆位置的定位方法、装置和可读存储介质
CN115962787A (zh) * 2023-03-16 2023-04-14 安徽蔚来智驾科技有限公司 地图更新、自动驾驶控制方法、设备、介质及车辆
CN118053052A (zh) * 2024-04-16 2024-05-17 之江实验室 一种无监督的高精矢量地图元素异常检测方法

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111858805A (zh) * 2020-07-08 2020-10-30 中国第一汽车股份有限公司 高精地图更新方法、车辆、服务器及存储介质
EP4242590A1 (en) * 2020-11-30 2023-09-13 Huawei Technologies Co., Ltd. Map verification method and related apparatus
CN112419542B (zh) * 2020-12-07 2022-01-14 安徽江淮汽车集团股份有限公司 道路纠错上报方法、装置、设备及存储介质
CN112487008B (zh) * 2020-12-12 2023-02-07 武汉中海庭数据技术有限公司 一种高精度地图速度限制元素局部更新方法及装置
CN112960000A (zh) * 2021-03-15 2021-06-15 新石器慧义知行智驰(北京)科技有限公司 高精地图更新方法、装置、电子设备和存储介质
CN114185613A (zh) * 2021-11-30 2022-03-15 广州景骐科技有限公司 一种语义地图分块方法、装置、交通工具及存储介质
CN114252087B (zh) * 2021-12-22 2022-07-01 广州小鹏自动驾驶科技有限公司 一种地图数据的处理方法及装置、车辆和存储介质
CN114353781A (zh) * 2021-12-31 2022-04-15 广州小鹏自动驾驶科技有限公司 地图更新方法、地图更新装置、电子设备及存储介质
CN114581667A (zh) * 2022-03-04 2022-06-03 北京百度网讯科技有限公司 地图数据处理方法、装置、电子设备和存储介质
CN115218909B (zh) * 2022-06-09 2024-03-01 广东省智能网联汽车创新中心有限公司 一种高精度地图的更新方法及系统

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105973245A (zh) * 2016-04-28 2016-09-28 百度在线网络技术（北京）有限公司 利用无人驾驶车辆更新在线地图的方法和装置
CN108121764A (zh) * 2016-11-26 2018-06-05 星克跃尔株式会社 图像处理装置、图像处理方法、电脑程序及电脑可读取记录介质
CN109783593A (zh) * 2018-12-27 2019-05-21 驭势科技（北京）有限公司 一种可用于自动驾驶的地图更新系统与方法
CN109781122A (zh) * 2019-01-31 2019-05-21 北京经纬恒润科技有限公司 高精度地图更新方法及装置
CN110160544A (zh) * 2019-06-12 2019-08-23 北京深思敏行科技有限责任公司 一种基于边缘计算的高精度地图众包更新系统
CN110888434A (zh) * 2019-11-14 2020-03-17 腾讯科技（深圳）有限公司 自动驾驶方法、装置、计算机设备和计算机可读存储介质
CN111858805A (zh) * 2020-07-08 2020-10-30 中国第一汽车股份有限公司 高精地图更新方法、车辆、服务器及存储介质

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103077215B (zh) * 2012-12-28 2016-01-20 深圳先进技术研究院 事件驱动的三维城市模型数据库动态更新方法
CN103150398B (zh) * 2013-03-27 2016-08-31 武汉大学 一种基于国家基础地理信息数据的增量式地图更新方法
CN110388929B (zh) * 2018-04-20 2022-04-15 比亚迪股份有限公司 导航地图更新方法、装置及系统
CN110287276B (zh) * 2019-05-27 2021-08-31 百度在线网络技术（北京）有限公司 高精地图更新方法、装置及存储介质

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105973245A (zh) * 2016-04-28 2016-09-28 百度在线网络技术（北京）有限公司 利用无人驾驶车辆更新在线地图的方法和装置
CN108121764A (zh) * 2016-11-26 2018-06-05 星克跃尔株式会社 图像处理装置、图像处理方法、电脑程序及电脑可读取记录介质
CN109783593A (zh) * 2018-12-27 2019-05-21 驭势科技（北京）有限公司 一种可用于自动驾驶的地图更新系统与方法
CN109781122A (zh) * 2019-01-31 2019-05-21 北京经纬恒润科技有限公司 高精度地图更新方法及装置
CN110160544A (zh) * 2019-06-12 2019-08-23 北京深思敏行科技有限责任公司 一种基于边缘计算的高精度地图众包更新系统
CN110888434A (zh) * 2019-11-14 2020-03-17 腾讯科技（深圳）有限公司 自动驾驶方法、装置、计算机设备和计算机可读存储介质
CN111858805A (zh) * 2020-07-08 2020-10-30 中国第一汽车股份有限公司 高精地图更新方法、车辆、服务器及存储介质

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114661842A (zh) * 2022-02-25 2022-06-24 阿波罗智联(北京)科技有限公司 地图匹配方法、装置和电子设备
CN115164912A (zh) * 2022-06-24 2022-10-11 宁波均胜智能汽车技术研究院有限公司 一种车辆位置的定位方法、装置和可读存储介质
CN115962787A (zh) * 2023-03-16 2023-04-14 安徽蔚来智驾科技有限公司 地图更新、自动驾驶控制方法、设备、介质及车辆
CN118053052A (zh) * 2024-04-16 2024-05-17 之江实验室 一种无监督的高精矢量地图元素异常检测方法

Also Published As

Publication number Publication date
CN111858805A (zh) 2020-10-30

Similar Documents

Publication Publication Date Title
WO2022007818A1 (zh) 高精地图更新方法、车辆、服务器及存储介质
WO2021217859A1 (zh) 目标异常识别方法、装置、电子设备及存储介质
JP7485749B2 (ja) ビデオベースの位置決め及びマッピングの方法及びシステム
US11105638B2 (en) Method, apparatus, and computer readable storage medium for updating electronic map
WO2020043081A1 (zh) 定位技术
US9983022B2 (en) Vehicle position estimation system, device, method, and camera device
CN108388641B (zh) 一种基于深度学习的交通设施地图生成方法与系统
US20160102987A1 (en) Method for inferring type of road segment
CN110763246A (zh) 自动驾驶车辆路径规划方法、装置、车辆及存储介质
CN112667837A (zh) 图像数据自动标注方法及装置
JP2015148601A (ja) マッピング、位置特定、及び姿勢補正のためのシステム及び方法
CN104573733A (zh) 一种基于高清正射影像图的高精细地图生成系统及方法
WO2021051344A1 (zh) 高精度地图中车道线的确定方法和装置
US20200240790A1 (en) Localization with Neural Network Based Image Registration of Sensor Data and Map Data
US10963708B2 (en) Method, device and computer-readable storage medium with instructions for determining the lateral position of a vehicle relative to the lanes of a road
WO2017161588A1 (zh) 一种定位方法及装置
CN111930872A (zh) 一种高精度地图更新方法、服务器及可读存储介质
JP2023508705A (ja) データ伝送方法および装置
AU2020375559B2 (en) Systems and methods for generating annotations of structured, static objects in aerial imagery using geometric transfer learning and probabilistic localization
CN115344655A (zh) 地物要素的变化发现方法、装置及存储介质
CN115564865A (zh) 一种众包高精地图的构建方法、系统、电子设备及车辆
CN113139031A (zh) 用于自动驾驶的交通标识的生成方法及相关装置
CN114419922B (zh) 车位识别方法及装置
WO2022055382A1 (en) A method and device for determining a vehicle position
US20220404167A1 (en) Roadway occlusion detection and reasoning

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21837178

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21837178

Country of ref document: EP

Kind code of ref document: A1