Positioning model optimization method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a positioning model optimization method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring first pose information of a vehicle based on satellite navigation data, inputting environmental data into a positioning model, and outputting second pose information of the vehicle; obtaining a pose information difference value between the second pose information and the first pose information, and determining an acquisition state corresponding to the data acquisition moment based on a size relationship between the pose information difference value and a first preset threshold value; obtaining a preset statistical value between the pose information difference values corresponding to all the data acquisition moments in the real vehicle road test process, and obtaining a first optimization index value based on the size relationship between the pose information difference values corresponding to the data acquisition moments and the preset statistical value; and optimizing the positioning model based on the acquisition state and the first optimization index value. According to the method provided by the invention, the pose information determined based on the positioning model can be more accurate.

Description

Technical Field

[0001] The present application relates to the field of vehicle testing technology, and in particular to a positioning model optimization method, device, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] During the actual vehicle testing, the vehicle's position information can be obtained by inputting the road test data into the localization algorithm. However, this method may result in inaccurate position information due to the poor performance of the localization algorithm. Summary of the invention

[0003] Based on this, it is necessary to provide a positioning model optimization method, device, computer equipment, computer-readable storage medium and computer program product that can improve the accuracy of vehicle posture information in response to the above technical problems.

[0004] In a first aspect, the present application provides a positioning model optimization method, the method comprising:

[0005] During the actual vehicle road test, real-time acquisition of satellite navigation data of the vehicle and environmental data collected by the vehicle's environmental perception sensor;

[0006] Based on the satellite navigation data, obtaining the first posture information of the vehicle, and inputting the environmental data into a positioning model, and outputting the second posture information of the vehicle;

[0007] For each data collection moment during the actual vehicle road test, a posture information difference between the second posture information and the first posture information is obtained, and based on a magnitude relationship between the posture information difference and a first preset threshold, a collection state corresponding to the data collection moment is determined;

[0008] Obtaining a preset statistical value between the posture information difference values ​​corresponding to all data collection moments during the actual vehicle road test, and obtaining a first optimization index value for each data collection moment based on a magnitude relationship between the posture information difference value corresponding to the data collection moment and the preset statistical value;

[0009] The positioning model is optimized based on the acquisition status and the first optimization index value.

[0010] In one embodiment, the posture information includes the lateral coordinate, longitudinal coordinate and heading angle of the vehicle, the longitudinal direction is consistent with the driving direction of the vehicle, and the lateral direction is perpendicular to the longitudinal direction and parallel to the ground; based on the size relationship between the posture information difference and the first preset threshold, determining the corresponding collection state at the data collection time, including:

[0011] When at least one of the lateral coordinate difference, the longitudinal coordinate difference or the heading angle difference is greater than a first preset threshold corresponding to the difference, determining the collection state corresponding to the data collection moment as an abnormal state;

[0012] When the lateral coordinate difference, the longitudinal coordinate difference and the heading angle difference are all not greater than the first preset threshold value corresponding to each difference, the collection state corresponding to the data collection moment is determined as a normal state.

[0013] In one embodiment, obtaining the first optimization index value based on the size relationship between the posture information difference value corresponding to the data collection moment and the preset statistical value includes:

[0014] For each type of posture information, obtain a first number of posture information difference values ​​that are less than a preset statistical value, and a ratio of the first number to the total number of the posture information difference values;

[0015] The first optimization index value is determined based on the size relationship between the ratio and a preset ratio.

[0016] In one embodiment, the optimizing the positioning model based on the acquisition status and the first optimization index value includes:

[0017] For each two adjacent data collection moments, based on the satellite navigation data corresponding to the two data collection moments, the longitude and latitude coordinates of the vehicle at the two data collection moments are obtained, and the straight-line distance between the longitude and latitude coordinates at the two data collection moments is obtained;

[0018] Acquire a second number of the abnormal states, and a preset function of the optimization index value with respect to the number and the distance;

[0019] Inputting the second number and the straight-line distance into the preset function, and outputting a second optimization index value;

[0020] The positioning model is optimized based on the first optimization index value and the second optimization index value.

[0021] In one embodiment, the optimizing the positioning model based on the first optimization index value and the second optimization index value includes:

[0022] The positioning model is optimized based on the first optimization index value corresponding to each posture information and the size relationship between the second optimization index value and the second preset threshold.

[0023] In one embodiment, the optimizing the positioning model based on the first optimization index value corresponding to each posture information and the magnitude relationship between the second optimization index value and the second preset threshold value includes:

[0024] When the first optimization index values ​​corresponding to all types of posture information indicate that the ratio is greater than the preset ratio, and the second optimization index value is greater than the second preset threshold, the positioning model is not optimized;

[0025] When there is a first optimization index value corresponding to at least one type of posture information indicating that the ratio is greater than the preset ratio, or the second optimization index value is not greater than the second preset threshold, optimizing the positioning model using a first optimization scheme;

[0026] In the case where there is a first optimization index value corresponding to at least one posture information indicating that the ratio is greater than the preset ratio, and the second optimization index value is not greater than the second preset threshold, a second optimization scheme is used to optimize the positioning model.

[0027] In a second aspect, the present application also provides a positioning model optimization device, the device comprising:

[0028] A first acquisition module is used to acquire satellite navigation data of the vehicle and environmental data collected by an environmental perception sensor of the vehicle in real time during a real vehicle road test;

[0029] A second acquisition module is used to acquire the first posture information of the vehicle based on the satellite navigation data, input the environmental data into a positioning model, and output the second posture information of the vehicle;

[0030] A third acquisition module is used to obtain a posture information difference between the second posture information and the first posture information at each data collection moment during the actual vehicle road test, and determine a collection state corresponding to the data collection moment based on a magnitude relationship between the posture information difference and a first preset threshold;

[0031] A fourth acquisition module is used to obtain a preset statistical value between the posture information difference values ​​corresponding to all data collection moments during the real vehicle road test, and for each data collection moment, based on the size relationship between the posture information difference value corresponding to the data collection moment and the preset statistical value, obtain a first optimization index value;

[0032] An optimization module is used to optimize the positioning model based on the acquisition status and the first optimization index value.

[0033] In a third aspect, the present application further provides a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method in any one of the above embodiments when executing the computer program.

[0034] In a fourth aspect, the present application further provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method in any one of the above embodiments are implemented.

[0035] In a fifth aspect, the present application further provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of the method in any one of the above embodiments are implemented.

[0036] The above positioning model optimization method, device, computer equipment, computer-readable storage medium and computer program product, in the process of the vehicle conducting a real-vehicle road test, obtains the vehicle's satellite navigation data and the environmental data collected by the vehicle's environmental perception sensor in real time; based on the satellite navigation data, obtains the vehicle's first posture information, and inputs the environmental data into the positioning model, and outputs the vehicle's second posture information; for each data collection moment in the real-vehicle road test, obtains the posture information difference between the second posture information and the first posture information, and determines the corresponding collection state at the data collection moment based on the size relationship between the posture information difference and the first preset threshold; obtains the preset statistical value between the posture information difference corresponding to all data collection moments in the real-vehicle road test, and for each data collection moment, based on the size relationship between the posture information difference corresponding to the data collection moment and the preset statistical value, obtains the first optimization index value; based on the collection state and the first optimization index value, optimizes the positioning model. The method provided in the present application optimizes the positioning model so that the posture information determined based on the positioning model is more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 A schematic diagram of a flow chart of a positioning model optimization method in one embodiment;

[0039] Figure 2 A schematic diagram of a flow chart of a method for determining an acquisition state in an embodiment;

[0040] Figure 3 A coordinate diagram of a posture information difference in one embodiment;

[0041] Figure 4 is a flowchart of a method for determining an acquisition state in one embodiment;

[0042] Figure 5 It is a structural block diagram of a positioning model optimization device in one embodiment;

[0043] Figure 6 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0045] In one embodiment, Figure 1 As shown, a positioning model optimization method is provided. This embodiment uses the method applied to a terminal as an example for illustration. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0046] S102. During the actual vehicle road test, satellite navigation data of the vehicle and environmental data collected by the vehicle's environmental perception sensor are obtained in real time.

[0047] Among them, real vehicle road testing refers to testing the vehicle in a real road environment; satellite navigation data refers to data related to the vehicle's positioning and navigation collected by the satellite navigation system; optionally, environmental perception sensors may include but are not limited to cameras, lidars, millimeter-wave radars and infrared sensors.

[0048] S104, based on the satellite navigation data, obtaining the first posture information of the vehicle, and inputting the environmental data into the positioning model, and outputting the second posture information of the vehicle.

[0049] Optionally, the positioning model may be, but is not limited to, a positioning algorithm.

[0050] Optionally, the first position information of the vehicle can be obtained by post-processing and solving the satellite navigation data, wherein the post-processing and solving process includes: first checking and correcting the abnormal values ​​in the satellite navigation data, and converting the corrected data from the receiver coordinate system to the earth coordinate system, then determining which satellites sent the satellite navigation data received by the receiver, and determining the distance between the receiver and these satellites, and finally inputting the satellite navigation data after coordinate conversion and the distance between the receiver and the satellite into the navigation solution model to obtain the first position information. Since the accuracy of the satellite navigation data is high, the first position information can be considered as the precise position information of the vehicle.

[0051] S106. For each data collection moment during the actual vehicle road test, obtain the posture information difference between the second posture information and the first posture information, and determine the collection state corresponding to the data collection moment based on the size relationship between the posture information difference and the first preset threshold.

[0052] Among them, there is at least one type of posture information, and the posture information difference refers to the difference between the second posture information corresponding to each posture information and the corresponding first posture information; for each posture information, a corresponding first preset threshold is preset, and the posture information difference is not greater than the first preset threshold, indicating that the difference between the first posture information and the second posture information is within a reasonable range, and the performance of the positioning model is good; the acquisition status characterizes whether the corresponding second posture information at the time of data acquisition is abnormal.

[0053] Optionally, the data collection time is a preset time included in a preset time series. Since the collection time between satellite navigation data and environmental data is different, and the collection time between environmental data collected by different environmental perception sensors is also different, a preset time series corresponding to the actual vehicle road test process can be preset, and the satellite navigation data and environmental data are aligned with the data collection time in the preset time series by using an interpolation method.

[0054] S108. Obtain preset statistical values ​​between the posture information differences corresponding to all data collection moments during the actual vehicle road test, and for each data collection moment, obtain a first optimization index value based on the size relationship between the posture information difference value corresponding to the data collection moment and the preset statistical value.

[0055] Optionally, the preset statistical value may be but is not limited to the mean value, standard deviation or 3sigma value between the posture information difference values, wherein the 3sigma value refers to the value that is 3 standard deviations away from the mean value in a normal distribution, and each type of posture information corresponds to a preset statistical value; the first optimization index value represents whether there is an abnormality in the corresponding second posture information at the time of data collection.

[0056] S110: Optimize the positioning model based on the acquisition status and the first optimization index value.

[0057] Optionally, the positioning model is optimized, mainly to optimize various performances of the positioning model, for example, the accuracy, robustness, computational efficiency and power consumption of the positioning model may be optimized.

[0058] In the above positioning model optimization method, during the actual vehicle road test, the satellite navigation data of the vehicle and the environmental data collected by the vehicle's environmental perception sensor are obtained in real time; based on the satellite navigation data, the first posture information of the vehicle is obtained, and the environmental data is input into the positioning model to output the second posture information of the vehicle; for each data collection moment in the actual vehicle road test, the posture information difference between the second posture information and the first posture information is obtained, and based on the size relationship between the posture information difference and the first preset threshold, the corresponding collection state at the data collection moment is determined; the preset statistical value between the posture information difference corresponding to all data collection moments in the actual vehicle road test is obtained, and for each data collection moment, based on the size relationship between the posture information difference corresponding to the data collection moment and the preset statistical value, the first optimization index value is obtained; based on the collection state and the first optimization index value, the positioning model is optimized. The method provided in the present application optimizes the positioning model so that the posture information determined based on the positioning model is more accurate.

[0059] In some embodiments, Figure 2 As shown, the posture information includes the lateral coordinate, longitudinal coordinate and heading angle of the vehicle, the longitudinal direction is consistent with the driving direction of the vehicle, the lateral direction is perpendicular to the longitudinal direction and parallel to the ground; based on the size relationship between the posture information difference and the first preset threshold, the corresponding collection state at the data collection time is determined, including:

[0060] S202: When at least one of the lateral coordinate difference, the longitudinal coordinate difference or the heading angle difference is greater than a first preset threshold corresponding to the difference, determine the collection state corresponding to the data collection moment as an abnormal state.

[0061] S204: When the lateral coordinate difference, the longitudinal coordinate difference and the heading angle difference are not greater than the first preset threshold corresponding to each difference, determine the collection state corresponding to the data collection moment as a normal state.

[0062] Among them, the abnormal state represents that the corresponding second posture information at the time of data collection exceeds a reasonable range, and the normal state represents that the corresponding second posture information at the time of data collection is within a reasonable range; the heading angle refers to the angle between the vehicle's driving direction and the reference direction. For example, the reference direction can be the magnetic North Pole or the geographic North Pole of the earth.

[0063] Alternatively, if Figure 3In the coordinate system, the origin of the coordinate represents the position coordinate of the vehicle in the first pose information, the end point of arrow B represents the position coordinate of the vehicle in the second pose information, arrow B represents the pose information difference, arrow A represents the driving direction of the vehicle, dx is the projection of arrow B on arrow A, that is, the longitudinal coordinate difference, and dy is the projection of arrow B on arrow C, that is, the lateral coordinate difference. Figure 4 It is a flowchart of the acquisition state determination process, where threshold 1, threshold 2 and threshold 3 are first preset thresholds corresponding to the lateral coordinate difference, the longitudinal coordinate difference and the heading angle difference respectively.

[0064] In this embodiment, the acquisition status is determined based on the magnitude relationship between the lateral coordinate difference, the longitudinal coordinate difference and the heading angle difference and their respective corresponding first preset thresholds, so that the determined acquisition status is more accurate.

[0065] In some embodiments, based on the size relationship between the corresponding posture information difference value at the data collection moment and the preset statistical value, a first optimization index value is obtained, including: for each posture information, obtaining a first number of posture information difference values ​​that are less than the preset statistical value, and the ratio of the first number to the total number of posture information difference values; based on the size relationship between the ratio and the preset ratio, determining the first optimization index value.

[0066] Optionally, if the preset statistical value is a 3 sigma value, the preset ratio is 99.74%.

[0067] Optionally, multiple numerical ranges may be preset, each numerical range corresponds to an index value, and a preset function between the ratio and the preset ratio is set. The ratio and the preset ratio are input into the preset function to obtain the function value, the numerical range in which the function value is located is determined, and the index value corresponding to the numerical range is determined as the first optimization index value.

[0068] In this embodiment, a first number of posture information difference values ​​that are less than a preset statistical value and the ratio of the first number to the total number of posture information difference values ​​are obtained, and a first optimization index value is determined based on the relationship between the ratio and the preset ratio. The first optimization index value determined in this way is more accurate, thereby making the subsequent optimization of the positioning model more accurate.

[0069] In some embodiments, the positioning model is optimized based on the collection status and the first optimization index value, including: for each two adjacent data collection moments, based on the corresponding satellite navigation data of the two data collection moments, obtaining the longitude and latitude coordinates of the vehicle at the two data collection moments, and obtaining the straight-line distance between the longitude and latitude coordinates of the two data collection moments; obtaining a second number of abnormal states, and a preset function of the optimization index value with respect to the number and distance; inputting the second number and the straight-line distance into the preset function, and outputting the second optimization index value; optimizing the positioning model based on the first optimization index value and the second optimization index value.

[0070] The second optimization index value indicates whether there is any abnormality in the second posture information.

[0071] Optionally, after calculating the corresponding straight-line distances between every two adjacent data collection moments, the sum of all straight-line distances is calculated, and the second number and the sum of the straight-line distances are input into a preset function to output a second optimization index value. For example, the preset function may be: second optimization index value = sum of straight-line distances / (second number + 1).

[0072] In this embodiment, the straight-line distance between the longitude and latitude coordinates of every two data collection moments is obtained, and the straight-line distance and the second number of abnormal states are input into a preset function, and a second optimization index value is output, so that the determined second optimization index value is more accurate, thereby making the subsequent optimization of the positioning model more accurate.

[0073] In some embodiments, the positioning model is optimized based on the first optimization index value and the second optimization index value, including: optimizing the positioning model based on the first optimization index value corresponding to each posture information, and the size relationship between the second optimization index value and the second preset threshold.

[0074] Optionally, different optimization schemes may be used to optimize the positioning model based on the numerical value of the first optimization index value and the magnitude relationship between the second optimization index value and the second preset threshold value.

[0075] In this embodiment, the positioning model is comprehensively optimized based on the first optimization index value and the second optimization index value, so that the optimization of the positioning model is more accurate.

[0076] In some embodiments, the positioning model is optimized based on the first optimization index value corresponding to each type of posture information, and the size relationship between the second optimization index value and the second preset threshold, including: when the first optimization index values ​​corresponding to all types of posture information represent a ratio greater than a preset ratio, and the second optimization index value is greater than the second preset threshold, the positioning model is not optimized; when there is at least one type of posture information whose first optimization index value represents a ratio greater than the preset ratio, or the second optimization index value is not greater than the second preset threshold, the positioning model is optimized using the first optimization scheme; when there is at least one type of posture information whose first optimization index value represents a ratio greater than the preset ratio, and the second optimization index value is not greater than the second preset threshold, the positioning model is optimized using the second optimization scheme.

[0077] Optionally, the first optimization scheme and the second optimization scheme can simultaneously optimize various performances of the positioning model to different degrees, or the first optimization scheme can optimize some performances and the second optimization scheme can optimize other performances. The embodiments of the present application do not specifically limit this.

[0078] In this embodiment, based on the size relationship between the ratio represented by the first optimization index value corresponding to each posture information and the preset ratio, and the size relationship between the second optimization index value and the second preset threshold, different optimization schemes are used to optimize the positioning model, so that the result of optimizing the positioning model is more accurate.

[0079] In one embodiment, another positioning model optimization method is provided, the method comprising the following contents:

[0080] During the actual vehicle road test, the satellite navigation data of the vehicle and the environmental data collected by the vehicle's environmental perception sensor are obtained in real time; based on the satellite navigation data, the first posture information of the vehicle is obtained, and the environmental data is input into the positioning model to output the second posture information of the vehicle, the posture information includes the lateral coordinate, longitudinal coordinate and heading angle of the vehicle; the posture information difference between the first posture information and the second posture information is obtained, and the posture information difference includes the lateral coordinate difference, the longitudinal coordinate difference and the heading angle difference; when at least one of the lateral coordinate difference, the longitudinal coordinate difference or the heading angle difference is greater than the corresponding difference threshold, the state of the posture information at the corresponding collection time is determined as an abnormal state; for each two adjacent data At the data collection moment, based on the corresponding satellite navigation data at the two data collection moments, the longitude and latitude coordinates of the vehicle at the two data collection moments are obtained, and the straight-line distance between the longitude and latitude coordinates at the two data collection moments is obtained; a second number of abnormal states and a preset function of the optimization index value with respect to the number and the distance are obtained; the second number and the straight-line distance are input into the preset function, and a second optimization index value is output; a preset statistical value between the posture information difference values ​​corresponding to all data collection moments during the actual vehicle road test is obtained, and for each data collection moment, a first optimization index value is obtained based on the size relationship between the posture information difference value corresponding to the data collection moment and the preset statistical value; based on the first optimization index value and the second optimization index value, the positioning model is optimized.

[0081] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0082] Based on the same inventive concept, the embodiment of the present application also provides a positioning model optimization device for implementing the positioning model optimization method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more positioning model optimization device embodiments provided below can refer to the limitations of the positioning model optimization method above, and will not be repeated here.

[0083] In an exemplary embodiment, Figure 5As shown, a positioning model optimization device 500 is provided, comprising: a first acquisition module 501, a second acquisition module 502, a third acquisition module 503, a fourth acquisition module 504 and an optimization module 505, wherein:

[0084] The first acquisition module 501 is used to acquire the satellite navigation data of the vehicle and the environmental data collected by the vehicle's environmental perception sensor in real time during the vehicle road test.

[0085] The second acquisition module 502 is used to acquire the first posture information of the vehicle based on the satellite navigation data, input the environmental data into a positioning model, and output the second posture information of the vehicle.

[0086] The third acquisition module 503 is used to obtain the posture information difference between the second posture information and the first posture information at each data collection moment during the actual vehicle road test, and determine the collection state corresponding to the data collection moment based on the size relationship between the posture information difference and the first preset threshold.

[0087] The fourth acquisition module 504 is used to obtain the preset statistical value between the posture information difference values ​​corresponding to all data collection moments during the actual vehicle road test, and for each data collection moment, obtain the first optimization index value based on the size relationship between the posture information difference value corresponding to the data collection moment and the preset statistical value.

[0088] The optimization module 505 is used to optimize the positioning model based on the acquisition status and the first optimization index value.

[0089] In some embodiments, the posture information includes the lateral coordinate, longitudinal coordinate and heading angle of the vehicle, the longitudinal direction is consistent with the driving direction of the vehicle, and the lateral direction is perpendicular to the longitudinal direction and parallel to the ground; the third acquisition module 503 is also used to determine the corresponding collection state at the data collection moment as an abnormal state when at least one of the lateral coordinate difference, the longitudinal coordinate difference or the heading angle difference is greater than the first preset threshold corresponding to the difference; when the lateral coordinate difference, the longitudinal coordinate difference and the heading angle difference are not greater than the first preset threshold corresponding to each difference, determine the corresponding collection state at the data collection moment as a normal state.

[0090] In some embodiments, the fourth acquisition module 504 is also used to obtain, for each type of posture information, a first number of posture information difference values ​​that are less than a preset statistical value, and the ratio of the first number to the total number of posture information difference values; and determine the first optimization index value based on the relationship between the ratio and the preset ratio.

[0091] In some embodiments, the optimization module 505 includes:

[0092] The first acquisition submodule is used to obtain the longitude and latitude coordinates of the vehicle at each two adjacent data collection moments based on the satellite navigation data corresponding to the two data collection moments, and obtain the straight-line distance between the longitude and latitude coordinates at the two data collection moments.

[0093] The second acquisition submodule is used to acquire a second number of the abnormal state and a preset function of the optimization index value with respect to the number and the distance.

[0094] An input submodule is used to input the second number and the straight-line distance into the preset function and output a second optimization index value.

[0095] An optimization submodule is used to optimize the positioning model based on the first optimization index value and the second optimization index value.

[0096] In some embodiments, the optimization submodule includes:

[0097] The optimization unit is used to optimize the positioning model based on the first optimization index value corresponding to each posture information and the size relationship between the second optimization index value and the second preset threshold.

[0098] In some embodiments, the optimization unit is also used to not optimize the positioning model when the first optimization index values ​​corresponding to all types of posture information indicate that the ratio is greater than the preset ratio and the second optimization index value is greater than the second preset threshold; to optimize the positioning model using the first optimization scheme when there is at least one type of posture information whose first optimization index value corresponds to the ratio is greater than the preset ratio, or the second optimization index value is not greater than the second preset threshold; to optimize the positioning model using the second optimization scheme when there is at least one type of posture information whose first optimization index value corresponds to the ratio is greater than the preset ratio and the second optimization index value is not greater than the second preset threshold.

[0099] Each module in the above positioning model optimization device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0100] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 6As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (Near Field Communication, NFC) or other technologies. When the computer program is executed by the processor, a positioning model optimization method is implemented.

[0101] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0102] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the following steps when executing the computer program: in a process of a vehicle being subjected to a real-vehicle road test, obtaining in real time satellite navigation data of the vehicle and environmental data collected by an environmental perception sensor of the vehicle; obtaining first posture information of the vehicle based on the satellite navigation data, inputting the environmental data into a positioning model, and outputting second posture information of the vehicle; for each data collection moment during the real-vehicle road test, obtaining a posture information difference between the second posture information and the first posture information, and determining a collection state corresponding to the data collection moment based on a size relationship between the posture information difference and a first preset threshold; obtaining a preset statistical value between the posture information differences corresponding to all data collection moments during the real-vehicle road test, and for each data collection moment, obtaining a first optimization index value based on a size relationship between the posture information difference corresponding to the data collection moment and the preset statistical value; optimizing the positioning model based on the collection state and the first optimization index value.

[0103] In one embodiment, the posture information implemented when the processor executes the computer program includes the lateral coordinate, longitudinal coordinate and heading angle of the vehicle, the longitudinal direction is consistent with the driving direction of the vehicle, and the lateral direction is perpendicular to the longitudinal direction and parallel to the ground; based on the size relationship between the posture information difference and the first preset threshold, the corresponding collection state at the data collection moment is determined, including: when at least one of the lateral coordinate difference, the longitudinal coordinate difference or the heading angle difference is greater than the first preset threshold corresponding to the difference, the corresponding collection state at the data collection moment is determined as an abnormal state; when the lateral coordinate difference, the longitudinal coordinate difference and the heading angle difference are not greater than the first preset threshold corresponding to each difference, the corresponding collection state at the data collection moment is determined as a normal state.

[0104] In one embodiment, when a processor executes a computer program, the first optimization index value is obtained based on the size relationship between the posture information difference value corresponding to the data collection moment and the preset statistical value, including: for each posture information, obtaining a first number of posture information difference values ​​that are less than the preset statistical value, and the ratio of the first number to the total number of posture information difference values; based on the size relationship between the ratio and the preset ratio, determining the first optimization index value.

[0105] In one embodiment, the processor optimizes the positioning model based on the collection status and the first optimization index value when executing a computer program, including: for each two adjacent data collection moments, based on the corresponding satellite navigation data of the two data collection moments, obtaining the longitude and latitude coordinates of the vehicle at the two data collection moments, and obtaining the straight-line distance between the longitude and latitude coordinates at the two data collection moments; obtaining a second number of abnormal states, and a preset function of the optimization index value with respect to the number and distance; inputting the second number and the straight-line distance into the preset function, and outputting a second optimization index value; optimizing the positioning model based on the first optimization index value and the second optimization index value.

[0106] In one embodiment, the processor optimizes the positioning model based on the first optimization index value and the second optimization index value when executing a computer program, including: optimizing the positioning model based on the first optimization index value corresponding to each posture information, and the relationship between the second optimization index value and the second preset threshold.

[0107] In one embodiment, when the processor executes a computer program, the positioning model is optimized based on the first optimization index value corresponding to each type of posture information, and the size relationship between the second optimization index value and the second preset threshold, including: when the first optimization index values ​​corresponding to all types of posture information represent that the ratio is greater than the preset ratio, and the second optimization index value is greater than the second preset threshold, the positioning model is not optimized; when there is at least one type of posture information corresponding to the first optimization index value representing that the ratio is greater than the preset ratio, or the second optimization index value is not greater than the second preset threshold, the positioning model is optimized using the first optimization scheme; when there is at least one type of posture information corresponding to the first optimization index value representing that the ratio is greater than the preset ratio, and the second optimization index value is not greater than the second preset threshold, the positioning model is optimized using the second optimization scheme.

[0108] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented: during a vehicle road test, satellite navigation data of the vehicle and environmental data collected by an environmental perception sensor of the vehicle are obtained in real time; based on the satellite navigation data, first posture information of the vehicle is obtained, and the environmental data is input into a positioning model to output second posture information of the vehicle; for each data collection moment during the real vehicle road test, a posture information difference between the second posture information and the first posture information is obtained, and based on the size relationship between the posture information difference and a first preset threshold, a collection state corresponding to the data collection moment is determined; a preset statistical value between the posture information difference values ​​corresponding to all data collection moments during the real vehicle road test is obtained, and for each data collection moment, a first optimization index value is obtained based on the size relationship between the posture information difference value corresponding to the data collection moment and the preset statistical value; and the positioning model is optimized based on the collection state and the first optimization index value.

[0109] In one embodiment, the posture information realized when the computer program is executed by the processor includes the lateral coordinate, longitudinal coordinate and heading angle of the vehicle, the longitudinal direction is consistent with the driving direction of the vehicle, and the lateral direction is perpendicular to the longitudinal direction and parallel to the ground; based on the size relationship between the posture information difference and the first preset threshold, the collection state corresponding to the data collection moment is determined, including: when at least one of the lateral coordinate difference, the longitudinal coordinate difference or the heading angle difference is greater than the first preset threshold corresponding to the difference, the collection state corresponding to the data collection moment is determined as an abnormal state; when the lateral coordinate difference, the longitudinal coordinate difference and the heading angle difference are not greater than the first preset threshold corresponding to each difference, the collection state corresponding to the data collection moment is determined as a normal state.

[0110] In one embodiment, when a computer program is executed by a processor, the first optimization index value is obtained based on the size relationship between the posture information difference value corresponding to the data collection moment and the preset statistical value, including: for each posture information, obtaining a first number of posture information difference values ​​that are less than the preset statistical value, and the ratio of the first number to the total number of posture information difference values; based on the size relationship between the ratio and the preset ratio, determining the first optimization index value.

[0111] In one embodiment, the computer program implemented when executed by the processor optimizes the positioning model based on the collection status and the first optimization index value, including: for each two adjacent data collection moments, based on the corresponding satellite navigation data of the two data collection moments, obtaining the longitude and latitude coordinates of the vehicle at the two data collection moments, and obtaining the straight-line distance between the longitude and latitude coordinates of the two data collection moments; obtaining a second number of the abnormal state, and a preset function of the optimization index value with respect to the number and distance; inputting the second number and the straight-line distance into the preset function, and outputting a second optimization index value; optimizing the positioning model based on the first optimization index value and the second optimization index value.

[0112] In one embodiment, the computer program implemented when executed by the processor optimizes the positioning model based on the first optimization index value and the second optimization index value, including: optimizing the positioning model based on the first optimization index value corresponding to each posture information, and the relationship between the second optimization index value and the second preset threshold.

[0113] In one embodiment, the computer program implemented when executed by the processor optimizes the positioning model based on the first optimization index value corresponding to each type of posture information and the size relationship between the second optimization index value and the second preset threshold, including: when the first optimization index values ​​corresponding to all types of posture information represent that the ratio is greater than the preset ratio, and the second optimization index value is greater than the second preset threshold, the positioning model is not optimized; when there is at least one type of posture information corresponding to the first optimization index value representing that the ratio is greater than the preset ratio, or the second optimization index value is not greater than the second preset threshold, the positioning model is optimized using the first optimization scheme; when there is at least one type of posture information corresponding to the first optimization index value representing that the ratio is greater than the preset ratio, and the second optimization index value is not greater than the second preset threshold, the positioning model is optimized using the second optimization scheme.

[0114] In one embodiment, a computer program product is provided, comprising a computer program, which implements the following steps when executed by a processor: in a process of a vehicle conducting a real-vehicle road test, obtaining satellite navigation data of the vehicle and environmental data collected by an environmental perception sensor of the vehicle in real time; obtaining first posture information of the vehicle based on the satellite navigation data, inputting the environmental data into a positioning model, and outputting second posture information of the vehicle; for each data collection moment in the real-vehicle road test, obtaining a posture information difference between the second posture information and the first posture information, and determining a collection state corresponding to the data collection moment based on a size relationship between the posture information difference and a first preset threshold; obtaining a preset statistical value between the posture information differences corresponding to all data collection moments in the real-vehicle road test, and for each data collection moment, obtaining a first optimization index value based on a size relationship between the posture information difference corresponding to the data collection moment and the preset statistical value; optimizing the positioning model based on the collection state and the first optimization index value.

[0115] In one embodiment, the posture information realized when the computer program is executed by the processor includes the lateral coordinate, longitudinal coordinate and heading angle of the vehicle, the longitudinal direction is consistent with the driving direction of the vehicle, and the lateral direction is perpendicular to the longitudinal direction and parallel to the ground; based on the size relationship between the posture information difference and the first preset threshold, the collection state corresponding to the data collection moment is determined, including: when at least one of the lateral coordinate difference, the longitudinal coordinate difference or the heading angle difference is greater than the first preset threshold corresponding to the difference, the collection state corresponding to the data collection moment is determined as an abnormal state; when the lateral coordinate difference, the longitudinal coordinate difference and the heading angle difference are not greater than the first preset threshold corresponding to each difference, the collection state corresponding to the data collection moment is determined as a normal state.

[0116] In one embodiment, when a computer program is executed by a processor, the first optimization index value is obtained based on the size relationship between the posture information difference value corresponding to the data collection moment and the preset statistical value, including: for each posture information, obtaining a first number of posture information difference values ​​that are less than the preset statistical value, and the ratio of the first number to the total number of posture information difference values; based on the size relationship between the ratio and the preset ratio, determining the first optimization index value.

[0117] In one embodiment, the computer program implemented when executed by the processor optimizes the positioning model based on the collection status and the first optimization index value, including: for each two adjacent data collection moments, based on the corresponding satellite navigation data of the two data collection moments, obtaining the longitude and latitude coordinates of the vehicle at the two data collection moments, and obtaining the straight-line distance between the longitude and latitude coordinates of the two data collection moments; obtaining a second number of the abnormal state, and a preset function of the optimization index value with respect to the number and distance; inputting the second number and the straight-line distance into the preset function, and outputting a second optimization index value; optimizing the positioning model based on the first optimization index value and the second optimization index value.

[0118] In one embodiment, the computer program implemented when executed by the processor optimizes the positioning model based on the first optimization index value and the second optimization index value, including: optimizing the positioning model based on the first optimization index value corresponding to each posture information, and the relationship between the second optimization index value and the second preset threshold.

[0119] In one embodiment, the computer program implemented when executed by the processor optimizes the positioning model based on the first optimization index value corresponding to each type of posture information and the size relationship between the second optimization index value and the second preset threshold, including: when the first optimization index values ​​corresponding to all types of posture information represent that the ratio is greater than the preset ratio, and the second optimization index value is greater than the second preset threshold, the positioning model is not optimized; when there is at least one type of posture information corresponding to the first optimization index value representing that the ratio is greater than the preset ratio, or the second optimization index value is not greater than the second preset threshold, the positioning model is optimized using the first optimization scheme; when there is at least one type of posture information corresponding to the first optimization index value representing that the ratio is greater than the preset ratio, and the second optimization index value is not greater than the second preset threshold, the positioning model is optimized using the second optimization scheme.

[0120] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0121] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0122] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0123] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the present application. It should be noted that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A positioning model optimization method, characterized in that: The method comprises: During the actual vehicle road test, real-time acquisition of satellite navigation data of the vehicle and environmental data collected by the vehicle's environmental perception sensor; Based on the satellite navigation data, obtaining the first posture information of the vehicle, inputting the environmental data into a positioning model, and outputting the second posture information of the vehicle; For each data collection moment during the actual vehicle road test, a posture information difference between the second posture information and the first posture information is obtained, and based on a magnitude relationship between the posture information difference and a first preset threshold, a collection state corresponding to the data collection moment is determined; Obtaining a preset statistical value between the posture information difference values ​​corresponding to all data collection moments during the actual vehicle road test, and obtaining a first optimization index value for each data collection moment based on a magnitude relationship between the posture information difference value corresponding to the data collection moment and the preset statistical value; The positioning model is optimized based on the acquisition status and the first optimization index value.

2. The method according to claim 1, characterized in that The position information includes the lateral coordinate, longitudinal coordinate and heading angle of the vehicle, the longitudinal direction is consistent with the driving direction of the vehicle, and the lateral direction is perpendicular to the longitudinal direction and parallel to the ground; Determining a corresponding acquisition state at the data acquisition moment based on a magnitude relationship between the posture information difference and a first preset threshold value includes: When at least one of the lateral coordinate difference, the longitudinal coordinate difference or the heading angle difference is greater than a first preset threshold corresponding to the difference, determining the collection state corresponding to the data collection moment as an abnormal state; When the lateral coordinate difference, the longitudinal coordinate difference and the heading angle difference are all not greater than the first preset threshold value corresponding to each difference, the collection state corresponding to the data collection moment is determined as a normal state.

3. The method according to claim 2, characterized in that The obtaining of the first optimization index value based on the size relationship between the posture information difference value corresponding to the data collection moment and the preset statistical value comprises: For each type of posture information, obtain a first number of posture information difference values ​​that are less than a preset statistical value, and a ratio of the first number to the total number of the posture information difference values; The first optimization index value is determined based on the size relationship between the ratio and a preset ratio.

4. The method according to claim 3, characterized in that The optimizing the positioning model based on the acquisition status and the first optimization index value includes: For each two adjacent data collection moments, based on the satellite navigation data corresponding to the two data collection moments, the longitude and latitude coordinates of the vehicle at the two data collection moments are obtained, and the straight-line distance between the longitude and latitude coordinates at the two data collection moments is obtained; Acquire a second number of the abnormal states, and a preset function of the optimization index value with respect to the number and the distance; Inputting the second number and the straight-line distance into the preset function, and outputting a second optimization index value; The positioning model is optimized based on the first optimization index value and the second optimization index value.

5. The method according to claim 4, characterized in that The optimizing the positioning model based on the first optimization index value and the second optimization index value includes: The positioning model is optimized based on the first optimization index value corresponding to each posture information and the size relationship between the second optimization index value and the second preset threshold.

6. The method according to claim 5, characterized in that The optimizing the positioning model based on the first optimization index value corresponding to each type of posture information and the magnitude relationship between the second optimization index value and a second preset threshold value includes: When the first optimization index values ​​corresponding to all types of posture information indicate that the ratio is greater than the preset ratio, and the second optimization index value is greater than the second preset threshold, the positioning model is not optimized; When there is a first optimization index value corresponding to at least one type of posture information indicating that the ratio is greater than the preset ratio, or the second optimization index value is not greater than the second preset threshold, optimizing the positioning model using a first optimization scheme; In the case where there is a first optimization index value corresponding to at least one posture information indicating that the ratio is greater than the preset ratio, and the second optimization index value is not greater than the second preset threshold, a second optimization scheme is used to optimize the positioning model.

7. A positioning model optimization device, characterized in that: The device comprises: A first acquisition module is used to acquire satellite navigation data of the vehicle and environmental data collected by an environmental perception sensor of the vehicle in real time during a real vehicle road test; A second acquisition module is used to acquire the first posture information of the vehicle based on the satellite navigation data, input the environmental data into a positioning model, and output the second posture information of the vehicle; A third acquisition module is used to obtain a posture information difference between the second posture information and the first posture information at each data collection moment during the actual vehicle road test, and determine a collection state corresponding to the data collection moment based on a magnitude relationship between the posture information difference and a first preset threshold; A fourth acquisition module is used to obtain a preset statistical value between the posture information difference values ​​corresponding to all data collection moments during the real vehicle road test, and for each data collection moment, based on the size relationship between the posture information difference value corresponding to the data collection moment and the preset statistical value, obtain a first optimization index value; An optimization module is used to optimize the positioning model based on the acquisition status and the first optimization index value.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images