Real-time high-precision map generation method and device, electronic equipment and storage medium

Abstract

The invention provides a real-time high-precision map generation method and device, electronic equipment and a storage medium. The method relates to the technical field of maps. The method may include: acquiring map data of a target geographic area, the map data including real-time data acquired by a vehicle for the target geographic area, the real-time data including real-time road data and real-time vehicle data; the real-time high-precision map model is called, a real-time high-precision map corresponding to the map data is generated, and the target moment of the real-time high-precision map corresponds to the collection moment of the real-time data. According to the embodiment of the invention, the real-time high-precision map can be predicted through the real-time map data, so that the high-precision map with high real-time degree is accurately and efficiently generated, the updating difficulty of the high-precision map is reduced, the real-time performance of the high-precision map is ensured, the timeliness lag of the high-precision map can be avoided, and the accuracy of the high-precision map is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of map technology, and in particular to a method, device, electronic device and storage medium for generating a real-time high-precision map. Background Art

[0002] A high-precision map is a map used for autonomous driving technology that contains map elements such as road shapes, road markings, traffic signs, and obstacles. The accuracy of a high-precision map is high, at least down to the centimeter level.

[0003] The accuracy of high-precision maps is crucial for the realization of precise navigation applications, which leads to the high reliance of autonomous vehicles on high-precision maps. Therefore, the accuracy of high-precision maps is particularly important for the smooth driving of autonomous vehicles. In urban environments, the update speed of various geographical areas is very fast, so the real-time degree of high-precision maps will directly affect the accuracy of high-precision maps. Summary of the invention

[0004] In view of this, the purpose of the present disclosure is to propose a method, device, electronic device and storage medium for generating a real-time high-precision map, which can solve the existing problems in a targeted manner.

[0005] Based on the above-mentioned purpose, in a first aspect, the present disclosure proposes a method for generating a real-time high-precision map, comprising: acquiring map data of a target geographical area, wherein the map data includes real-time data collected by vehicles on the target geographical area, and the real-time data includes real-time road data and real-time vehicle data; calling a real-time high-precision map model to generate a real-time high-precision map corresponding to the map data, wherein the real-time high-precision map is used to represent high-precision map information of the target geographical area at a target time, and the target time corresponds to the collection time of the real-time data.

[0006] In the second aspect, there is also provided an acquisition unit, configured to acquire map data of a target geographical area, wherein the map data includes real-time data collected by vehicles on the target geographical area, and the real-time data includes real-time road data and real-time vehicle data; a generation unit, configured to call a real-time high-precision map model to generate a real-time high-precision map corresponding to the map data, wherein the real-time high-precision map is used to represent high-precision map information of the target geographical area at a target time, and the target time corresponds to the collection time of the real-time data.

[0007] In a third aspect, an electronic device is also provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in the first aspect.

[0008] According to a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, and the program is executed by a processor to implement any method described in the first aspect.

[0009] In general, the present disclosure has at least the following beneficial effects: real-time high-precision maps can be predicted through real-time map data, thereby accurately and efficiently generating high-precision maps with a high degree of real-timeness, reducing the difficulty of updating high-precision maps, and ensuring the real-time nature of high-precision maps. Therefore, the timeliness lag of high-precision maps can be avoided, and the accuracy of high-precision maps can be improved, which helps to ensure the smooth driving of autonomous driving vehicles that rely on high-precision maps. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the multiple drawings represent the same or similar parts or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings only depict some embodiments disclosed in accordance with the present disclosure and should not be regarded as limiting the scope of the present disclosure.

[0011] Figure 1 A flowchart showing a method for generating a real-time high-precision map according to an embodiment of the present disclosure;

[0012] Figure 2 A flowchart showing the training steps of a real-time high-precision map model according to an embodiment of the present disclosure;

[0013] Figure 3 A schematic diagram showing a real-time high-precision map model according to an embodiment of the present disclosure;

[0014] Figure 4 A schematic diagram showing a device for generating a real-time high-precision map according to an embodiment of the present disclosure;

[0015] Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present disclosure is shown;

[0016] Figure 6 A schematic diagram of a storage medium provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0017] The present disclosure is further described in detail below in conjunction with the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0018] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present disclosure may be combined with each other. The present disclosure will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0019] Figure 1 A method for generating a real-time high-precision map of the present disclosure is shown. In an embodiment of the present disclosure, the method includes the following steps:

[0020] Step S101, obtaining map data of a target geographical area, wherein the map data includes real-time data collected by vehicles on the target geographical area, and the real-time data includes real-time road data and real-time vehicle data.

[0021] In this embodiment, the above-mentioned execution subject can obtain map data of the target geographical area from this device or other electronic devices. Map data refers to data related to the map of the target geographical area. Specifically, the map data includes real-time data collected by the vehicle on the target geographical area. The real-time data includes real-time road data and real-time vehicle data. In practice, real-time road data is real-time data related to the road, for example, real-time road data may include multi-angle road images, and in addition, real-time road data may also include lane line information of the road. Real-time vehicle data is real-time data related to the vehicle, for example, it may include at least one of the following: inertial measurement data, wheel speed data, speed data.

[0022] Specifically, the multi-angle road image is a road image captured in real time by sensors at multiple shooting angles of the vehicle. When the vehicle is an autonomous driving vehicle, the multi-angle road image may be a surround view image.

[0023] The inertial measurement data can be detected in real time by the vehicle's inertial measurement unit (IMU) or inertial navigation system. Wheel speed data, also known as wheel speed meter data, is data measured in real time by a wheel speed sensor.

[0024] Alternatively, both the inertial measurement data and the wheel speed data may be obtained by combining wheel speed and inertial navigation measurement of the vehicle.

[0025] Step S102, calling a real-time high-precision map model to generate a real-time high-precision map corresponding to the map data, wherein the real-time high-precision map is used to characterize the high-precision map information of the target geographic area at a target time, and the target time corresponds to the collection time of the real-time data.

[0026] In this embodiment, the execution subject may call the real-time high-precision map model to generate a real-time high-precision map corresponding to the map data. Specifically, the execution subject may input the map data into the high-precision map model and obtain a real-time high-precision map output from the model, and the real-time high-precision map corresponds to the map data.

[0027] A real-time high-precision map indicates the high-precision map information at the target time, and the target time corresponds to the collection time of the real-time data. Specifically, the update cycle of the high-precision map is longer than the instantaneous collection time, so a real-time high-precision map reflects the high-precision map information for a longer period of time. Accordingly, the target time corresponds to the collection time, which means that the target time includes the collection time, and the duration of the target time is greater than the duration of the collection time.

[0028] Real-time high-precision maps are usually vector maps. For a target geographic area, the multi-angle road images collected by a vehicle can fully cover it. Therefore, the real-time high-precision map corresponding to the target geographic area is a local real-time high-precision map. The real-time high-precision map model can be a trained deep learning model, such as the Transformer model, MapTR model, or BEV Former model.

[0029] This embodiment can predict real-time high-precision maps through real-time map data, thereby accurately and efficiently generating high-precision maps with a high degree of real-timeness, reducing the difficulty of updating high-precision maps, and ensuring the real-time nature of high-precision maps. Therefore, it is possible to avoid timeliness lags in high-precision maps and improve the accuracy of high-precision maps, which helps to ensure the smooth driving of autonomous driving vehicles that rely on high-precision maps.

[0030] In some optional implementations of any embodiment of the present disclosure, the map data also includes a historical precision map of the target geographic area, the historical precision map includes a historical high-precision map or a historical standard precision map, and the historical moment indicated by the historical precision map precedes the target moment.

[0031] In these implementations, the above-mentioned execution entity can also input historical precision maps with low real-time accuracy into the model, so that the model can output accurate real-time high-precision maps end-to-end based on the map with higher accuracy.

[0032] The historical high-precision map is used to represent the high-precision map information of the target geographic area at a historical moment, and the historical standard map is used to represent the standard map information of the target geographic area at a historical moment. The historical moment indicated by the historical high-precision map and the historical standard map is prior to the target moment indicated by the real-time data.

[0033] These implementation methods can use real-time data to make real-time adjustments to historical precision maps to obtain accurate real-time high-precision maps.

[0034] In some optional implementations of any one of the embodiments of the present disclosure, the real-time high-precision map model includes an encoder, an embedding network, a deep neural network and a decoder; calling the real-time high-precision map model to generate a real-time high-precision map corresponding to the map data includes: inputting the map data into the encoder to obtain encoding information; inputting the encoding information into the embedding network to obtain intermediate information; inputting the intermediate information into the deep neural network, and inputting the output of the deep neural network into the decoder to obtain a real-time high-precision map.

[0035] In these implementations, the encoder, embedding network, deep neural network and decoder in the real-time high-precision map model can be in a cascade relationship. The above-mentioned execution subject can input the map data into the encoder, and input the encoded information output by the encoder into the embedding network (embedding) for dimension change, and input the intermediate information output by the embedding network into the deep neural network (DNN). Finally, the above-mentioned execution subject can input the output of the deep neural network into the decoder to obtain a real-time high-precision map.

[0036] These implementation methods can cascade the map data through multiple network structures with different functions in the model, which helps to obtain more accurate real-time high-precision maps.

[0037] Optionally, the encoder includes a first encoder, a second encoder and a third encoder; the step of inputting the map data into the encoder to obtain encoding information includes: inputting the inertial measurement data and wheel speed data into the first encoder to obtain timing signal encoding information, inputting the road image into the second encoder to obtain road image encoding information, and inputting the historical precision map into the third encoder to obtain precision map encoding information.

[0038] The encoders in these implementations can be different encoders for different types of data, which can effectively improve the model's encoding capabilities for different types of data and help output more accurate encoding information.

[0039] Optionally, the embedding network includes a first embedding network, a second embedding network and a third embedding network; the step of inputting the encoded information into the embedding network to obtain intermediate information comprises: inputting the timing signal encoded information into the first embedding network to obtain first intermediate information, inputting the road image encoded information into the second embedding network to obtain second intermediate information, and inputting the precision map encoded information into the third embedding network to obtain third intermediate information; and the step of inputting the intermediate information into the deep neural network comprises: inputting the first intermediate information, the second intermediate information and the third intermediate information into the deep neural network.

[0040] Specifically, the above-mentioned execution entity can adopt different embedding networks for different types of data, thereby effectively improving the model's embedding processing capabilities for different types of data, and helping the embedding network to perform more accurate dimensionality increase or dimensionality reduction processing.

[0041] Figure 2 The training steps of the real-time high-precision map model according to the embodiment of the present disclosure are shown. Figure 2 As shown, the step may include: step S201, inputting the map data sample into the real-time high-precision map model to be trained to obtain a prediction result of the real-time high-precision map, wherein the map data sample has a corresponding annotated real-time high-precision map; step S202, training the real-time high-precision map model to be trained based on the prediction result and the annotated real-time high-precision map to obtain the real-time high-precision map model.

[0042] In this embodiment, the training step can be performed by the above-mentioned execution subject or other electronic devices. Taking the above-mentioned execution subject as an example: the above-mentioned execution subject can input the samples used for training the model into the real-time high-precision map model to be trained, and let the samples be forward propagated in the real-time high-precision map model to be trained, so as to obtain the result of predicting the real-time high-precision map, that is, the prediction result. The above-mentioned execution subject can use the prediction result and the annotated real-time high-precision map to determine the loss value of the real-time advanced map model to be trained, and use the loss value to perform back propagation in the model, so as to train the model and obtain the trained real-time high-precision map model.

[0043] In practice, the sample may be a map data sample, and each map data sample has a corresponding real-time high-precision map as label data (label), that is, a labeled real-time high-precision map.

[0044] In this embodiment, the real-time high-precision map can be used as annotation data to train a model that can predict the real-time high-precision map based on the real-time map data.

[0045] Optionally, the map data sample exists in a map data sample set, the map data sample exists in a map data sample set, the map data sample set contains multiple versions of historical precision maps, the multiple versions of historical precision maps are at least two different historical precision maps corresponding to the same geographical area; the map data sample set contains a historical precision map that is missing at least one road element.

[0046] Specifically, the execution subject can add low-precision maps to the training, so that the historical precision maps can be adjusted using real-time data to obtain real-time high-precision maps. In order to enrich the samples, the execution subject can use multiple versions of historical precision maps for training. In practice, multiple versions of historical precision maps can correspond to the same annotated real-time high-precision map.

[0047] In addition, the execution subject may also use a historical precision map with missing road elements, i.e., a mask image. Road elements refer to object elements in a road image, such as traffic lights, vehicles, etc. The execution subject or other electronic device may perform local destruction on the historical precision map, each of which includes at least one road element. For example, at least one target, i.e., a road element, in the historical precision map may be randomly destroyed.

[0048] These implementations can train high-precision map models through diverse samples, thereby helping to train models with stronger predictive capabilities. Specifically, multiple versions of historical precision maps can help the model learn to adjust to different historical precision maps. In addition, by training historical precision maps with missing road elements and outputting complete images, the model can learn to generate stronger high-precision maps.

[0049] like Figure 3 As shown in the figure, a schematic diagram of a high-precision map model is shown. The dotted lines in the figure represent non-essential content.

[0050] The embodiment of the present disclosure provides a device for generating a real-time high-precision map, and the device for generating a real-time high-precision map is used to execute the method for generating a real-time high-precision map described in the above embodiment, such as Figure 4 As shown, the device includes: an acquisition unit 401, configured to acquire map data of a target geographical area, wherein the map data includes real-time data collected by vehicles for the target geographical area, and the real-time data includes real-time road data and real-time vehicle data; a generation unit 402, configured to call a real-time high-precision map model to generate a real-time high-precision map corresponding to the map data, wherein the real-time high-precision map is used to characterize high-precision map information of the target geographical area at a target time, and the target time corresponds to the collection time of the real-time data.

[0051] Optionally, the map data also includes a historical precision map of the target geographic area, the historical precision map includes a historical high-precision map or a historical standard-precision map, and the historical moment indicated by the historical precision map is prior to the target moment.

[0052] Optionally, the real-time high-precision map model includes an encoder, an embedding network, a deep neural network and a decoder; the generation unit is further configured to execute the call of the real-time high-precision map model in the following manner to generate a real-time high-precision map corresponding to the map data: input the map data into the encoder to obtain encoding information; input the encoding information into the embedding network to obtain intermediate information; input the intermediate information into the deep neural network, and input the output of the deep neural network into the decoder to obtain a real-time high-precision map.

[0053] Optionally, the encoder includes a first encoder, a second encoder and a third encoder; the generating unit is further configured to execute the step of inputting the map data into the encoder to obtain encoding information in the following manner: inputting the inertial measurement data and the wheel speed data into the first encoder to obtain timing signal encoding information, inputting the road image into the second encoder to obtain road image encoding information, and inputting the historical precision map into the third encoder to obtain precision map encoding information.

[0054] Optionally, the embedding network includes a first embedding network, a second embedding network and a third embedding network; the generating unit is further configured to execute the step of inputting the encoded information into the embedding network to obtain intermediate information in the following manner: inputting the timing signal encoded information into the first embedding network to obtain first intermediate information, inputting the road image encoded information into the second embedding network to obtain second intermediate information, and inputting the precision map encoded information into the third embedding network to obtain third intermediate information; and the generating unit is further configured to execute the step of inputting the intermediate information into the deep neural network in the following manner: inputting the first intermediate information, the second intermediate information and the third intermediate information into the deep neural network.

[0055] Optionally, the training step of the real-time high-precision map model includes: inputting a map data sample into the real-time high-precision map model to be trained to obtain a prediction result of the real-time high-precision map, wherein the map data sample has a corresponding annotated real-time high-precision map; based on the prediction result and the annotated real-time high-precision map, training the real-time high-precision map model to be trained to obtain the real-time high-precision map model.

[0056] Optionally, the map data sample exists in a map data sample set, and the map data sample set contains multiple versions of historical precision maps, and the multiple versions of historical precision maps are at least two different historical precision maps corresponding to the same geographical area; the map data sample set contains a historical precision map that is missing at least one road element.

[0057] The real-time high-precision map generation device provided by the above-mentioned embodiment of the present disclosure and the real-time high-precision map generation method provided by the embodiment of the present disclosure are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0058] The disclosed embodiment also provides an electronic device corresponding to the method for generating a real-time high-precision map provided in the aforementioned embodiment, so as to execute the aforementioned method for generating a real-time high-precision map. The disclosed embodiment is not limited thereto.

[0059] Please refer to Figure 5 , which shows a schematic diagram of an electronic device provided by some embodiments of the present disclosure. Figure 5 As shown, the electronic device 50 includes: a processor 500, a memory 501, a bus 502 and a communication interface 503, and the processor 500, the communication interface 503 and the memory 501 are connected via the bus 502; the memory 501 stores a computer program that can be run on the processor 500, and the processor 500 executes the method provided in any of the aforementioned embodiments of the present disclosure when running the computer program.

[0060] The memory 501 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 503 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0061] The bus 502 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. Among them, the memory 501 is used to store programs, and the processor 500 executes the program after receiving the execution instruction. The method for generating a real-time high-precision map disclosed in any implementation of the aforementioned embodiment of the present disclosure may be applied to the processor 500, or implemented by the processor 500.

[0062] The processor 500 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit in the processor 500 or the instruction in the form of software. The above processor 500 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present disclosure can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in conjunction with the embodiments of the present disclosure can be directly embodied as a hardware decoding processor to be executed, or the hardware and software modules in the decoding processor can be executed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 501, and the processor 500 reads the information in the memory 501 and completes the steps of the above method in combination with its hardware.

[0063] The electronic device provided by the embodiment of the present disclosure and the method for generating a real-time high-precision map provided by the embodiment of the present disclosure are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by them.

[0064] The present disclosure also provides a computer-readable storage medium corresponding to the method for generating a real-time high-precision map provided in the above-mentioned embodiment. Figure 6 The computer-readable storage medium shown is a CD 60 on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it will execute the method for generating a real-time high-precision map provided by any of the aforementioned embodiments.

[0065] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical or magnetic storage media, which are not listed here one by one.

[0066] The computer-readable storage medium provided by the above-mentioned embodiments of the present disclosure and the method for generating a real-time high-precision map provided by the embodiments of the present disclosure are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0067] It should be noted that:

[0068] In the above text, the term "comprises", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be noted that the scope of the method and device in the embodiments of the present disclosure is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0069] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present disclosure, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present disclosure.

[0070] The embodiments of the present disclosure are described above in conjunction with the accompanying drawings, which are only specific implementation modes of the present disclosure. However, the present disclosure is not limited to the above-mentioned specific implementation modes. The above-mentioned specific implementation modes are merely illustrative and not restrictive. Under the guidance of the present disclosure, ordinary technicians in this field can also make many forms without departing from the purpose of the present disclosure and the scope of protection of the claims, all of which are within the protection of the present disclosure.

Claims

1. A method for generating a real-time high-precision map, characterized in that: include: Acquire map data of a target geographical area, wherein the map data includes real-time data collected by a vehicle on the target geographical area, and the real-time data includes real-time road data and real-time vehicle data; A real-time high-precision map model is called to generate a real-time high-precision map corresponding to the map data, wherein the real-time high-precision map is used to characterize the high-precision map information of the target geographic area at a target time, and the target time corresponds to the collection time of the real-time data.

2. The method according to claim 1, characterized in that The map data also includes a historical precision map of the target geographic area, wherein the historical precision map includes a historical high-precision map or a historical standard-precision map, and the historical moment indicated by the historical precision map precedes the target moment.

3. The method according to claim 1, characterized in that The real-time high-precision map model includes an encoder, an embedding network, a deep neural network and a decoder; The calling of the real-time high-precision map model to generate a real-time high-precision map corresponding to the map data includes: Inputting the map data into the encoder to obtain encoding information; Inputting the encoded information into the embedding network to obtain intermediate information; The intermediate information is input into the deep neural network, and the output of the deep neural network is input into the decoder to obtain a real-time high-precision map.

4. The method according to claim 3, characterized in that The encoder includes a first encoder, a second encoder and a third encoder; The step of inputting the map data into the encoder to obtain encoding information comprises: The inertial measurement data and wheel speed data are input into the first encoder to obtain timing signal encoding information, the road image is input into the second encoder to obtain road image encoding information, and the historical precision map is input into the third encoder to obtain precision map encoding information.

5. The method according to claim 4, characterized in that The embedded network includes a first embedded network, a second embedded network and a third embedded network; The step of inputting the encoded information into the embedding network to obtain intermediate information comprises: Inputting the time series signal encoding information into the first embedding network to obtain first intermediate information, inputting the road image encoding information into the second embedding network to obtain second intermediate information, and inputting the accuracy map encoding information into the third embedding network to obtain third intermediate information; and The step of inputting the intermediate information into the deep neural network comprises: The first intermediate information, the second intermediate information, and the third intermediate information are input into the deep neural network.

6. The method according to claim 1 or 2, characterized in that: The training steps of the real-time high-precision map model include: Inputting a map data sample into a real-time high-precision map model to be trained to obtain a prediction result of a real-time high-precision map, wherein the map data sample has a corresponding annotated real-time high-precision map; Based on the prediction result and the annotated real-time high-precision map, the real-time high-precision map model to be trained is trained to obtain the real-time high-precision map model.

7. The method according to claim 6, characterized in that The map data sample exists in a map data sample set, and the map data sample set contains multiple versions of historical accuracy maps, and the multiple versions of historical accuracy maps are at least two different historical accuracy maps corresponding to the same geographical area; The map data sample set contains a historical precision map that is missing at least one road element.

8. A device for generating a real-time high-precision map, characterized in that: include: An acquisition unit is configured to acquire map data of a target geographical area, wherein the map data includes real-time data collected by a vehicle on the target geographical area, and the real-time data includes real-time road data and real-time vehicle data; The generation unit is configured to call a real-time high-precision map model to generate a real-time high-precision map corresponding to the map data, wherein the real-time high-precision map is used to represent the high-precision map information of the target geographic area at a target time, and the target time corresponds to the collection time of the real-time data.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: The processor runs the computer program to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the method according to any one of claims 1 to 7.

Images