Vehicle map update determination method and device, electronic equipment and storage medium
By obtaining static and dynamic map data of the vehicle on the target driving route, and using pre-trained visual models and information comparison models to determine the target confidence map data and its difference data, the problem of untimely and inefficient vehicle map updates is solved, and efficient and accurate updates are achieved.
Patent Information
- Application Number
- CN202510551674.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-29
- Publication Date
- 2025-09-12
AI Technical Summary
In the existing technology, the vehicle map update cannot determine the actual difference in real time, resulting in untimely updates, long cycles, low efficiency and poor accuracy.
By obtaining static and dynamic map data of the vehicle on the target driving route, using pre-trained visual models and information comparison models, the target confidence map data and its difference data are determined, and the trained information comparison model or external map database is optimized and updated based on the confidence and pre-set confidence threshold.
It realizes the timely update of vehicle maps, improves the update efficiency and accuracy, shortens the update cycle, and enhances the efficiency of solving map difference problems.
Smart Images
Figure CN120631998A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of vehicle intelligent driving technology, and in particular to a method, device, electronic device and storage medium for determining vehicle map updates. Background Art
[0002] At present, the on-board map software of more and more vehicles is equipped with a feedback portal for the on-board map. The feedback portal can support users to feedback the differences in the actual use of the on-board map through voice or manual text input, thereby updating the on-board map. However, the manual input method requires users to manually input a large amount of information, and the input information is highly subjective. There is no other information to verify the correctness and accuracy of the user's input information, resulting in the inability of the on-board map to determine the actual differences in real time, which in turn causes untimely map updates, long update cycles, low update efficiency and poor update accuracy. Summary of the Invention
[0003] The embodiments of the present application provide a method, device, electronic device and storage medium for determining vehicle map updates. The embodiments provided in the present application solve the technical problem in the prior art that vehicle map updates cannot determine the actual differences in real time, which in turn causes the map updates to be untimely, the update cycle to be long, the update efficiency to be low and the update accuracy to be poor.
[0004] In a first aspect of an embodiment of the present application, an embodiment of the present application provides a method for determining a vehicle map update, comprising:
[0005] Obtaining static map data and dynamic map data of a vehicle on a target driving route, wherein the dynamic map data is determined based on a pre-trained visual model and image data of the target driving route collected by the vehicle during driving, and the static map data is used to represent map data of the target driving route determined from an external map database;
[0006] Inputting the static map data and the dynamic map data into a trained information comparison model to determine target confidence map data and difference data in the target confidence map data, wherein the target confidence map data is used to represent the map data with the highest confidence between the static map data and the dynamic map data, and the difference data is used to represent data that differs between the static map data and the dynamic map data;
[0007] The trained information comparison model or the external map database is optimized and updated based on the target confidence of the target confidence map data, a preset confidence threshold, and the difference data in the target confidence map data.
[0008] In a feasible implementation manner, inputting the static map data and the dynamic map data into a trained information comparison model to determine target confidence map data and difference data in the target confidence map data includes:
[0009] Inputting the static map data and the dynamic map data into a trained information comparison model to determine whether the static map data and the dynamic map data match;
[0010] If there is no match, when the first confidence level of the static map data is greater than the second confidence level of the dynamic map data, determining the static map data as target confidence map data, and determining data in the static map data that differs from the dynamic map data as difference data in the target confidence map data;
[0011] When the second confidence of the dynamic map data is greater than the first confidence of the static map data, the dynamic map data is determined to be target confidence map data, and data in the dynamic map data that differs from the static map data is determined to be difference data in the target confidence map data.
[0012] In a feasible implementation, the preset confidence threshold includes a first preset confidence threshold, and the optimizing and updating of the trained information comparison model or the external map database based on the target confidence of the target confidence map data, the preset confidence threshold, and the difference data in the target confidence map data includes:
[0013] If the target confidence of the target confidence map data is greater than or equal to the first preset confidence threshold, determining that the target confidence map data is high confidence data;
[0014] If the high-confidence data is static map data, optimizing the trained information comparison model parameters based on the difference data in the high-confidence data;
[0015] If the high-confidence data is dynamic map data, the difference data in the high-confidence data is fed back to the external map database, so as to update the static map data in the external map database.
[0016] In a feasible implementation, the preset confidence threshold includes a second preset confidence threshold, and the optimizing and updating of the trained information comparison model or the external map database based on the target confidence of the target confidence map data, the preset confidence threshold, and the difference data in the target confidence map data includes:
[0017] If the target confidence of the target confidence map data is greater than or equal to the second preset confidence threshold and less than the first preset confidence threshold, determining that the target confidence map data is medium confidence data;
[0018] If the medium confidence data is static map data, optimizing the trained information comparison model parameters based on user feedback on authorization information of the static map data and difference data in the medium confidence data;
[0019] If the medium confidence data is dynamic map data, based on the authorization information of the static map data fed back by the user, the difference data in the medium confidence data is fed back to the external map database, so as to update the static map data in the external map database.
[0020] In a feasible implementation manner, after optimizing and updating the trained information comparison model or the external map database based on the target confidence of the target confidence map data, the preset confidence threshold, and the difference data in the target confidence map data, the determination method further includes:
[0021] If the target confidence of the target confidence map data is less than the second preset confidence threshold, determining that the target confidence map data is low confidence data;
[0022] The low-confidence data and the difference data in the low-confidence data are deleted.
[0023] In a feasible implementation, the trained information comparison model is determined by:
[0024] Acquiring historical dynamic map data and historical static map data of the vehicle on a historical driving route, wherein the historical dynamic map data includes historical dynamic map confidence, and the historical static map data includes historical static lane line data, historical static obstacle features, and historical static speed limit signs on the historical driving route;
[0025] Based on the historical dynamic map data and the historical static map data, an initial information comparison model is trained to determine a trained information comparison model.
[0026] In a feasible implementation, the training of the initial information comparison model based on the historical dynamic map data and the historical static map data to determine the trained information comparison model includes:
[0027] Inputting the historical static lane line data, historical static obstacle features, and historical static speed limit signs into an initial information comparison model to determine a historical dynamic map confidence corresponding to the historical static map data;
[0028] Based on the historical dynamic map confidence, the historical static map confidence and a preset loss algorithm, the initial information comparison model is trained to determine a trained information comparison model.
[0029] In a second aspect of the embodiments of the present application, an apparatus for determining a vehicle map update is provided. The apparatus for determining a vehicle map update includes:
[0030] an acquisition module, configured to acquire static map data and dynamic map data of a vehicle on a target driving route, wherein the dynamic map data is determined based on a pre-trained visual model and image data of the target driving route collected by the vehicle during driving, and the static map data is used to represent map data of the target driving route determined from an external map database;
[0031] a first determination module, configured to input the static map data and the dynamic map data into a trained information comparison model, and determine target confidence map data and difference data in the target confidence map data, wherein the target confidence map data is used to represent the map data with the highest confidence between the static map data and the dynamic map data, and the difference data is used to represent data that differs between the static map data and the dynamic map data;
[0032] The second determining module is configured to optimize and update the trained information comparison model or the external map database based on the target confidence of the target confidence map data, a preset confidence threshold, and the difference data in the target confidence map data.
[0033] In a third aspect of an embodiment of the present application, an embodiment of the present application provides an electronic device comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor and the memory communicate through the bus, and the machine-readable instructions are executed by the processor to perform the steps of the above-mentioned method for determining the vehicle map update.
[0034] In a fourth aspect of the embodiments of the present application, the embodiments of the present application provide a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method for determining the vehicle map update as described above are executed.
[0035] The embodiments of the present application provide a method, device, electronic device and storage medium for determining vehicle map updates. Compared with the prior art, the embodiments provided by the present application obtain static map data and dynamic map data of the vehicle on the target driving route, and input the static map data and dynamic map data into a trained information comparison model to determine the target confidence map data and the difference data in the target confidence map data. Then, based on the target confidence of the target confidence map data, the preset confidence threshold and the difference data in the target confidence map data, the trained information comparison model or the external map database is optimized and updated. The present application uses the trained information comparison model to compare the static map data and the dynamic map data to determine the target confidence map data and the difference data in the target confidence map data, thereby realizing the problem of determining the real difference in the vehicle map. The addition of the trained information comparison model improves the update efficiency and update accuracy of the trained information comparison model, the external map database and the vehicle map, greatly shortens the update cycle, and enhances the timeliness of the vehicle map update. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] Figure 1 A flowchart of a method for determining a vehicle map update provided by an embodiment of the present application is shown;
[0037] Figure 2 A structural block diagram of a vehicle map update determination device provided by an embodiment of the present application is shown;
[0038] Figure 3 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown.
[0039] Figure 2 and Figure 3 The corresponding relationship between the reference numerals and the names of the drawings is as follows:
[0040] 200 : Device for determining vehicle map update; 210 : Acquisition module; 220 : First determination module; 230 : Second determination module; 300 : Electronic device; 310 : Processor; 320 : Memory; 320 : Bus. DETAILED DESCRIPTION
[0041] In order to better understand the technical solutions provided by the embodiments of this specification, the technical solutions of the embodiments of this specification are described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of this specification and the specific features in the embodiments are detailed descriptions of the technical solutions of the embodiments of this specification, rather than limitations on the technical solutions of this specification. In the absence of conflict, the embodiments of this specification and the technical features in the embodiments can be combined with each other.
[0042] In this article, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also include elements inherent to such process, method, article or equipment. In the absence of further restrictions, the elements defined by the statement "comprising a ..." do not exclude the presence of other identical elements in the process, method, article or equipment comprising the elements. The term "two or more" includes two or more than two cases.
[0043] First, the application scenarios to which this application is applicable are introduced. The embodiments provided in this application are applicable to the field of vehicle intelligent driving technology, and in particular, relate to a method, device, electronic device and storage medium for determining vehicle map updates.
[0044] At present, the on-board map software of more and more vehicles is equipped with a feedback portal for the on-board map. The feedback portal can support users to feedback the differences in the actual use of the on-board map through voice or manual text input, thereby updating the on-board map. However, the manual input method requires users to manually input a large amount of information, and the input information is highly subjective. There is no other information to verify the correctness and accuracy of the user's input information, resulting in the inability of the on-board map to determine the actual differences in real time, which in turn causes untimely map updates, long update cycles, low update efficiency and poor update accuracy.
[0045] Based on this, the embodiments of the present application provide a method, device, electronic device and storage medium for determining vehicle map updates. The embodiments provided by the present application solve the problem in the prior art that the vehicle map cannot determine the actual differences in real time, which in turn causes the technical problems of untimely map updates, long update cycles, low update efficiency and poor update accuracy.
[0046] Figure 1 FIG1 shows a flow chart of a method for determining a vehicle map update provided by an embodiment of the present application. Figure 1 As shown, the method for determining the vehicle map update includes the following steps:
[0047] S101. Obtain static map data and dynamic map data of a vehicle on a target driving route, wherein the dynamic map data is determined based on a pre-trained visual model and image data of the target driving route collected by the vehicle during driving, and the static map data is used to represent map data of the target driving route determined from an external map database.
[0048] In this step, when the vehicle in the embodiment provided by the present application is traveling on the target driving route, the vehicle can determine in real time the static map data that matches the target driving route through the external map database connected to the vehicle computer. At the same time, the embodiment provided by the present application collects image data on the target driving route through the image acquisition device of the vehicle computer, and inputs the image data into the pre-trained visual model to determine the dynamic map data related to the above-mentioned target driving route, wherein the target driving route is used to represent the driving route selected by the user when using the vehicle.
[0049] It can be understood that the static map data in the embodiments provided in the present application can be: target prohibition information on the target driving route, target speed limit sign information on the target driving route, target speed limit value, target lane line information on the target driving route, latitude and longitude information of target obstacles on the target driving route (i.e., location information), and historical update information of the target driving route, etc.
[0050] The dynamic map data in the embodiments provided in the present application may be: predicted obstacle information on the target driving route, predicted lane line information on the target driving route, predicted prohibition information on the target driving route, and the prediction confidence corresponding to the above prediction information, that is, the dynamic map data output by the pre-trained visual model in the embodiments provided in the present application is dynamic map data with prediction confidence label information.
[0051] Among them, the external map database in the embodiment provided by this application can be a map software development kit (SDK), such as: a third-party database or APP that can collect and store static map data. The external map database in the embodiment provided by this application can be customized and selected according to different usage scenarios and application conditions. The external map database of this application includes but is not limited to: cloud GPS system, Baidu, AutoNavi, and Hua Xiaozhu and other APPs.
[0052] In the above, the image acquisition devices in the embodiments provided in this application can be customized and selected according to different usage scenarios and application conditions, such as cameras, image collectors, radars, and lidars.
[0053] The process of inputting image data into a pre-trained visual model for training and outputting result features in the embodiments provided in the present application can be a visual true lane-level function or a city assisted navigation function (Navigate on Autopilot, NOA) on a vehicle computer.
[0054] The true lane-level visual function refers to obtaining road images through visual sensors such as cameras and processing these images using computer vision algorithms to accurately identify information such as lane lines, road signs, and traffic signs, ultimately achieving high-precision positioning and navigation at the lane level.
[0055] The city assisted navigation function (Navigate on Autopilot, NOA) is part of the vehicle's driving assistance system, used to provide automatic navigation and driving assistance for vehicles in complex urban environments. This function uses multiple sensors (such as cameras, radars, and lidars, etc.) and high-precision map data, combined with advanced algorithms and technologies, to help vehicles achieve safer and more efficient autonomous driving or assisted driving.
[0056] S102. Input the static map data and the dynamic map data into a trained information comparison model to determine target confidence map data and difference data in the target confidence map data, wherein the target confidence map data is used to represent the map data with the highest confidence between the static map data and the dynamic map data, and the difference data is used to represent data that differs between the static map data and the dynamic map data.
[0057] In this step, the embodiment provided by the present application first performs data alignment on the dynamic map data after inputting the static map data and the dynamic map data into the trained information comparison model, and then constructs a joint feature space based on the aligned static map data and the dynamic map data, and determines a first confidence level of the static map data and a second confidence level of the dynamic map data, respectively. Then, the first confidence level and the second confidence level are respectively compared with a preset confidence threshold, and the map data with the highest confidence level is determined as the target confidence map data. At the same time, the data with differences between the static map data and the dynamic map data is determined as difference data.
[0058] It can be understood that data alignment of dynamic map data is specifically to align the position points identified in the image data with the position information of each point in the static map data to ensure the temporal and spatial consistency of the comparison. Among them, the embodiment provided by the present application allows the position point information identified in the image data to deviate from the position information of each point in the static map data within a certain range (the deviation can be within a reasonable range).
[0059] Exemplarily, static map data and dynamic map data are input into a trained information comparison model to determine whether the static map data and dynamic map data match; if they do not match, when the first confidence of the static map data is greater than the second confidence of the dynamic map data, the static map data is determined to be the target confidence map data, and the data in the static map data that differs from the dynamic map data is determined to be the difference data in the target confidence map data; when the second confidence of the dynamic map data is greater than the first confidence of the static map data, the dynamic map data is determined to be the target confidence map data, and the data in the dynamic map data that differs from the static map data is determined to be the difference data in the target confidence map data.
[0060] In the above, the embodiment provided by the present application needs to determine whether the static map data and the dynamic map data match, and when the static map data and the dynamic map data match, the static map data / dynamic map data is determined as the target confidence map data, and the target confidence map data is determined to have no difference data.
[0061] In the embodiment provided by the present application, when static map data and dynamic map data do not match, map data with high confidence is determined as target confidence map data, and data in the static map data that differs from the dynamic map data is determined as difference data in the target confidence map data.
[0062] Exemplarily, the present application determines the trained information comparison model in the following manner:
[0063] Obtain historical dynamic map data and historical static map data of the vehicle on its historical driving route, wherein the historical dynamic map data includes historical dynamic map confidence, and the historical static map data includes historical static lane line data, historical static obstacle features, and historical static speed limit signs on the historical driving route; train an initial information comparison model based on the historical dynamic map data and historical static map data to determine the trained information comparison model.
[0064] It should be noted that, in the embodiment provided by the present application, after the historical dynamic map data and the historical static map data are input into the initial information comparison model, the historical dynamic map data and the historical static map data are first aligned. Specifically, the position points identified in the historical dynamic map data are aligned with the position information of each point in the historical static map data to ensure the temporal and spatial consistency of the comparison. Among them, the embodiment provided by the present application allows a certain range of deviation between the position point information identified in the historical dynamic map data and the position information of each point in the static map data.
[0065] It can be understood that the embodiment provided in this application constructs a joint feature space of historical information after determining the historical dynamic map data and the historical static map data, and then uses the initial information comparison model to perform comparative prediction, outputs the historical prediction confidence map data, and uses the historical prediction confidence map data to determine the historical difference data to realize the training of the initial information comparison model.
[0066] Among them, the initial information comparison model in the embodiment provided in this application can be customized and used according to different application scenarios and usage conditions. The initial information comparison model provided in this application can be specific but not limited to: residual network (Residual Network, ResNet) or convolutional neural network (EfficientNet), etc.
[0067] Exemplarily, training an initial information comparison model based on historical dynamic map data and historical static map data to determine a trained information comparison model includes:
[0068] The historical static lane line data, historical static obstacle features, and historical static speed limit signs are input into the initial information comparison model to determine the historical dynamic map confidence corresponding to the historical static map data; based on the historical dynamic map confidence, the historical static map confidence, and the preset loss algorithm, the initial information comparison model is trained to determine the trained information comparison model.
[0069] It should be noted that this application continuously learns the distribution differences between historical dynamic map confidence and historical static map confidence based on a preset loss algorithm, and trains the initial information comparison model based on the above distribution differences, and then constructs a trained information comparison model, wherein the historical dynamic map confidence is calibrated by Bayesian calibration for confidence output.
[0070] It can be understood that the preset loss algorithm in the embodiments provided in this application can be customized and selected according to different application scenarios. The preset loss algorithm of this application can be a KL divergence loss algorithm or a cross entropy quantization loss algorithm.
[0071] S103 : Optimizing and updating the trained information comparison model or the external map database based on the target confidence of the target confidence map data, a preset confidence threshold, and difference data in the target confidence map data.
[0072] In this step, the embodiment provided in the present application divides the target confidence of the target confidence map data into groups according to the numerical values based on a preset confidence threshold, and then determines the optimized update method of the target confidence map data corresponding to the target confidence within different numerical ranges.
[0073] It can be understood that if the target confidence map data in the embodiment provided in this application is static map data, it means that the static map data is more accurate than the dynamic map data. The trained information comparison model for outputting dynamic map data will be optimized and updated based on the static map data.
[0074] If the target confidence map data in the embodiment provided in this application is dynamic map data, it means that the dynamic map data is more accurate than the static map data. The external map database that outputs the static map data will be optimized and updated based on the dynamic map data.
[0075] Exemplarily, the preset confidence threshold includes a first preset confidence threshold; if the target confidence of the target confidence map data is greater than or equal to the first preset confidence threshold, the target confidence map data is determined to be high confidence data; if the high confidence data is static map data, the trained information comparison model parameters are optimized based on difference data in the high confidence data;
[0076] If the high-confidence data is dynamic map data, the difference data in the high-confidence data is fed back to the external map database so as to update the static map data in the external map database.
[0077] It should be noted that the high-confidence data determined in this application, whether it is static map data or dynamic map data, indicates that there are potential differences in the map output of the vehicle computer. Therefore, the difference data needs to be sorted and recorded. This application will record the problem points in the difference data and the corresponding location information, identification information, image information, and continuous video stream information within a preset time before and after the problem points, and then send the difference data and the problem points corresponding to the difference data to the data source for updating and optimization. After updating and optimization, the processed map data will be fed back to the user so that the user can view the processing flow and processing status of the difference data in real time.
[0078] It can be understood that if the high-confidence data is static map data, it means that there are problems with the dynamic map data on the target driving route, and the parameters of the trained information comparison model that outputs the dynamic map data need to be updated and optimized; if the high-confidence data is dynamic map data, it means that there are problems with the static map data on the target driving route, and the external map database that outputs the static map data needs to be updated and optimized.
[0079] Here, the method of providing feedback to the user for viewing the processing flow in the embodiment provided by this application includes but is not limited to: displaying the progress of the processing flow on the instrument display interface of the vehicle computer and viewing it on the APP (or remote host computer).
[0080] Exemplarily, the preset confidence threshold includes a second preset confidence threshold. If the target confidence of the target confidence map data is greater than or equal to the second preset confidence threshold and less than the first preset confidence threshold, the target confidence map data is determined to be medium confidence data; if the medium confidence data is static map data, the trained information comparison model parameters are optimized based on the authorization information of the static map data and the difference data in the medium confidence data according to user feedback; if the medium confidence data is dynamic map data, the difference data in the medium confidence data is fed back to the external map database based on the authorization information of the static map data according to user feedback, so as to update the static map data in the external map database.
[0081] It should be noted that, in the embodiment provided by the present application, after determining that the target confidence map data is medium confidence data, it is necessary to feed back the determined medium confidence data to the user, and the user confirms whether to send the above confidence data to an external map database or a trained information comparison model for optimization and updating.
[0082] If the user does not confirm or authorize the instruction to feedback the confidence data, the above-mentioned confidence data will be directly deleted; if the user confirms or authorizes the instruction to feedback the confidence data, the user's feedback authorization information for the static map data will be received, and the difference data in the confidence data will be fed back to the external map database or the trained information comparison model for optimization and update.
[0083] It can be understood that if the medium confidence data is static map data, it means that there are problems with the dynamic map data on the target driving route, and the trained information comparison model parameters for outputting the dynamic map data need to be updated and optimized; if the medium confidence data is dynamic map data, it means that there are problems with the static map data on the target driving route, and the external map database for outputting the static map data needs to be updated and optimized. Specifically, the map manufacturer can repair, update and optimize the corresponding problem points.
[0084] Exemplarily, if the target confidence of the target confidence map data is less than a second preset confidence threshold, the target confidence map data is determined to be low confidence data; and the low confidence data and the difference data in the low confidence data are deleted.
[0085] It should be noted that the embodiment provided in this application will delete the target confidence map data whose target confidence is less than the second preset confidence threshold, so as to prevent the background from affecting the memory and performance of the entire vehicle computer due to storage excess.
[0086] The method for determining the vehicle map update provided by the embodiment of the present application, compared with the prior art, obtains static map data and dynamic map data of the vehicle on the target driving route, and inputs the static map data and dynamic map data into a trained information comparison model to determine the target confidence map data and the difference data in the target confidence map data. Then, based on the target confidence of the target confidence map data, the preset confidence threshold and the difference data in the target confidence map data, the trained information comparison model or the external map database is optimized and updated. The present application utilizes the trained information comparison model, By comparing static map data and dynamic map data, the target confidence map data and the difference data in the target confidence map data are determined, so as to determine the real difference problem of the vehicle map, and save the location and video image information corresponding to the difference problem. The addition of the trained information comparison model improves the update efficiency and update accuracy of the trained information comparison model, the external map database and the vehicle map. The present application can automatically provide feedback on the determined target confidence map data, improve the efficiency of solving the map difference problem, greatly shorten the update cycle, and enhance the timeliness of the vehicle map update.
[0087] See also Figure 2 , Figure 2 A structural block diagram of a vehicle map update determination device provided by an embodiment of the present application is shown. Figure 2 As shown, the vehicle map update determination device 200 includes:
[0088] The acquisition module 210 is used to obtain static map data and dynamic map data of the vehicle on the target driving route, wherein the dynamic map data is determined based on a pre-trained visual model and image data of the target driving route collected by the vehicle during driving, and the static map data is used to represent the map data of the target driving route determined from an external map database.
[0089] The first determination module 220 is used to input the static map data and the dynamic map data into the trained information comparison model to determine the target confidence map data and the difference data in the target confidence map data, wherein the target confidence map data is used to represent the map data with the highest confidence between the static map data and the dynamic map data, and the difference data is used to represent the data that differs between the static map data and the dynamic map data.
[0090] The second determining module 230 is configured to optimize and update the trained information comparison model or the external map database based on the target confidence of the target confidence map data, a preset confidence threshold, and difference data in the target confidence map data.
[0091] Exemplarily, the first determining module 220 is specifically configured to:
[0092] The static map data and the dynamic map data are input into the trained information comparison model to determine whether the static map data and the dynamic map data match.
[0093] If there is no match, when the first confidence of the static map data is greater than the second confidence of the dynamic map data, the static map data is determined to be the target confidence map data, and the data in the static map data that differs from the dynamic map data is determined to be the difference data in the target confidence map data.
[0094] When the second confidence of the dynamic map data is greater than the first confidence of the static map data, the dynamic map data is determined as the target confidence map data, and data in the dynamic map data that differs from the static map data is determined as difference data in the target confidence map data.
[0095] Exemplarily, the preset confidence threshold includes a first preset confidence threshold, and the second determination module 230 is specifically configured to:
[0096] If the target confidence of the target confidence map data is greater than or equal to a first preset confidence threshold, the target confidence map data is determined to be high-confidence data.
[0097] If the high-confidence data is static map data, the trained information comparison model parameters are optimized based on the difference data in the high-confidence data.
[0098] If the high-confidence data is dynamic map data, the difference data in the high-confidence data is fed back to the external map database so as to update the static map data in the external map database.
[0099] Exemplarily, the preset confidence threshold includes a second preset confidence threshold, and the second determination module 230 is specifically configured to:
[0100] If the target confidence of the target confidence map data is greater than or equal to the second preset confidence threshold and less than the first preset confidence threshold, it is determined that the target confidence map data is medium confidence data.
[0101] If the medium confidence data is static map data, the trained information comparison model parameters are optimized based on the user feedback on the authorization information of the static map data and the difference data in the medium confidence data.
[0102] If the medium confidence data is dynamic map data, then based on the authorization information of the static map data fed back by the user, the difference data in the medium confidence data is fed back to the external map database so as to update the static map data in the external map database.
[0103] Exemplarily, the second determining module 230 is specifically configured to:
[0104] If the target confidence of the target confidence map data is less than a second preset confidence threshold, it is determined that the target confidence map data is low confidence data.
[0105] Delete low-confidence data and discrepancy data in low-confidence data.
[0106] Exemplarily, the trained information comparison model is determined in the following manner:
[0107] Acquire historical dynamic map data and historical static map data of the vehicle on the historical driving route, wherein the historical dynamic map data includes historical dynamic map confidence, and the historical static map data includes historical static lane line data, historical static obstacle features, and historical static speed limit signs on the historical driving route.
[0108] Based on the historical dynamic map data and the historical static map data, the initial information comparison model is trained to determine the trained information comparison model.
[0109] Exemplarily, training an initial information comparison model based on historical dynamic map data and historical static map data to determine a trained information comparison model includes:
[0110] The historical static lane line data, historical static obstacle features and historical static speed limit signs are input into the initial information comparison model to determine the historical dynamic map confidence corresponding to the historical static map data.
[0111] Based on the historical dynamic map confidence, the historical static map confidence and the preset loss algorithm, the initial information comparison model is trained to determine the trained information comparison model.
[0112] The vehicle map update determination device 200 provided in the embodiment of the present application, compared with the prior art, obtains static map data and dynamic map data of the vehicle on the target driving route, and inputs the static map data and dynamic map data into the trained information comparison model to determine the target confidence map data and the difference data in the target confidence map data. Then, based on the target confidence of the target confidence map data, the preset confidence threshold and the difference data in the target confidence map data, the trained information comparison model or the external map database is optimized and updated. The present application utilizes the trained information comparison model to determine the target confidence map data and the difference data in the target confidence map data. , compare the static map data and the dynamic map data, determine the target confidence map data and the difference data in the target confidence map data, realize the determination of the real difference problem of the vehicle map, and save the location and video image information corresponding to the difference problem. The addition of the trained information comparison model improves the update efficiency and update accuracy of the trained information comparison model, the external map database and the vehicle map. This application can automatically feedback the determined target confidence map data, improve the efficiency of solving the map difference problem, greatly shorten the update cycle, and enhance the timeliness of the vehicle map update.
[0113] Figure 3 1 is a schematic diagram showing the structure of an electronic device provided in an embodiment of the present application. Figure 3 As shown, the electronic device 300 includes a processor 310 , a memory 320 , and a bus 330 .
[0114] The memory 320 stores machine-readable instructions executable by the processor 310. When the electronic device 300 is running, the processor 310 communicates with the memory 320 via the bus 330. When the machine-readable instructions are executed by the processor 310, the above-mentioned Figure 1 The steps of the method for determining the vehicle map update in the method embodiment shown are specifically implemented in accordance with the method embodiment and will not be described in detail here.
[0115] The embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the computer program can execute the above-mentioned Figure 1 The steps of the method for determining the vehicle map update in the method embodiment shown are specifically implemented in accordance with the method embodiment and will not be described in detail here.
[0116] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0117] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.
[0118] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-readable program code.
[0119] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0120] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.
[0121] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.
[0122] An embodiment of the present application also provides a computer program product, which includes computer software instructions. When the computer software instructions are executed on a processing device, the processing device executes the process of the method for determining the vehicle map update.
[0123] A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)).
[0124] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0125] In the several embodiments provided in this application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.
[0126] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.
[0127] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.
[0128] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.
[0129] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.
[0130] Although the preferred embodiments of this specification have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of this specification.
[0131] Obviously, those skilled in the art may make various changes and modifications to this specification without departing from the spirit and scope of this specification. Thus, if such changes and modifications fall within the scope of the claims of this specification and their equivalents, this specification is intended to include such changes and modifications.
Claims
1. A method for determining a vehicle map update, characterized in that: include: Obtaining static map data and dynamic map data of a vehicle on a target driving route, wherein the dynamic map data is determined based on a pre-trained visual model and image data of the target driving route collected by the vehicle during driving, and the static map data is used to represent map data of the target driving route determined from an external map database; Inputting the static map data and the dynamic map data into a trained information comparison model to determine target confidence map data and difference data in the target confidence map data, wherein the target confidence map data is used to represent the map data with the highest confidence between the static map data and the dynamic map data, and the difference data is used to represent data that differs between the static map data and the dynamic map data; The trained information comparison model or the external map database is optimized and updated based on the target confidence of the target confidence map data, a preset confidence threshold, and the difference data in the target confidence map data.
2. The method for determining the vehicle map update according to claim 1, characterized in that: Inputting the static map data and the dynamic map data into a trained information comparison model to determine target confidence map data and difference data in the target confidence map data includes: Inputting the static map data and the dynamic map data into a trained information comparison model to determine whether the static map data and the dynamic map data match; If there is no match, when the first confidence level of the static map data is greater than the second confidence level of the dynamic map data, determining the static map data as target confidence map data, and determining data in the static map data that differs from the dynamic map data as difference data in the target confidence map data; When the second confidence of the dynamic map data is greater than the first confidence of the static map data, the dynamic map data is determined to be target confidence map data, and data in the dynamic map data that differs from the static map data is determined to be difference data in the target confidence map data.
3. The method for determining the vehicle map update according to claim 2, characterized in that: The preset confidence threshold includes a first preset confidence threshold, and the optimizing and updating of the trained information comparison model or the external map database based on the target confidence of the target confidence map data, the preset confidence threshold, and the difference data in the target confidence map data includes: If the target confidence of the target confidence map data is greater than or equal to the first preset confidence threshold, determining that the target confidence map data is high confidence data; If the high-confidence data is static map data, optimizing the trained information comparison model parameters based on the difference data in the high-confidence data; If the high-confidence data is dynamic map data, the difference data in the high-confidence data is fed back to the external map database, so as to update the static map data in the external map database.
4. The method for determining the vehicle map update according to claim 3, characterized in that: The preset confidence threshold includes a second preset confidence threshold, and the optimizing and updating of the trained information comparison model or the external map database based on the target confidence of the target confidence map data, the preset confidence threshold, and the difference data in the target confidence map data includes: If the target confidence of the target confidence map data is greater than or equal to the second preset confidence threshold and less than the first preset confidence threshold, determining that the target confidence map data is medium confidence data; If the medium confidence data is static map data, optimizing the trained information comparison model parameters based on user feedback on authorization information of the static map data and difference data in the medium confidence data; If the medium confidence data is dynamic map data, based on the authorization information of the static map data fed back by the user, the difference data in the medium confidence data is fed back to the external map database, so as to update the static map data in the external map database.
5. The method for determining the vehicle map update according to claim 4, characterized in that: After optimizing and updating the trained information comparison model or the external map database based on the target confidence of the target confidence map data, the preset confidence threshold, and the difference data in the target confidence map data, the determining method further includes: If the target confidence of the target confidence map data is less than the second preset confidence threshold, determining that the target confidence map data is low confidence data; The low-confidence data and the difference data in the low-confidence data are deleted.
6. The method for determining the vehicle map update according to claim 1, characterized in that: The trained information contrast model is determined by: Acquiring historical dynamic map data and historical static map data of the vehicle on a historical driving route, wherein the historical dynamic map data includes historical dynamic map confidence, and the historical static map data includes historical static lane line data, historical static obstacle features, and historical static speed limit signs on the historical driving route; Based on the historical dynamic map data and the historical static map data, an initial information comparison model is trained to determine a trained information comparison model.
7. The method for determining the vehicle map update according to claim 6, characterized in that: The training of the initial information comparison model based on the historical dynamic map data and the historical static map data to determine the trained information comparison model includes: Inputting the historical static lane line data, historical static obstacle features, and historical static speed limit signs into an initial information comparison model to determine a historical dynamic map confidence corresponding to the historical static map data; Based on the historical dynamic map confidence, the historical static map confidence and a preset loss algorithm, the initial information comparison model is trained to determine a trained information comparison model.
8. A device for determining a vehicle map update, characterized in that: The vehicle map update determining device includes: an acquisition module, configured to acquire static map data and dynamic map data of a vehicle on a target driving route, wherein the dynamic map data is determined based on a pre-trained visual model and image data of the target driving route collected by the vehicle during driving, and the static map data is used to represent map data of the target driving route determined from an external map database; a first determination module, configured to input the static map data and the dynamic map data into a trained information comparison model, and determine target confidence map data and difference data in the target confidence map data, wherein the target confidence map data is used to represent the map data with the highest confidence between the static map data and the dynamic map data, and the difference data is used to represent data that differs between the static map data and the dynamic map data; The second determining module is configured to optimize and update the trained information comparison model or the external map database based on the target confidence of the target confidence map data, a preset confidence threshold, and the difference data in the target confidence map data.
9. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor and the memory communicate via the bus, and the machine-readable instructions are executed by the processor to execute the steps of the method for determining the vehicle map update as described in any one of claims 1 to 7 above.
10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for determining the vehicle map update as described in any one of claims 1 to 7 are executed.