Scene screening method, device, electronic device and storage medium for autonomous driving

Abstract

The present application provides a scene screening method, device, electronic device and storage medium for autonomous driving, the method comprising: obtaining multiple initial scene samples and the square deviation value of the scene parameter corresponding to each initial scene sample; determining the scene covariance between the initial scene sample and each other initial scene sample; determining the square deviation value of the scene parameter corresponding to the test scene sample according to the test scene sample, multiple initial scene samples and the square deviation value of the scene parameter corresponding to each initial scene sample, and all scene covariances; screening out autonomous driving scenes according to the minimum value of the square deviation value of the scene parameter corresponding to all initial scene samples, the test scene sample, and the square deviation value of the scene parameter corresponding to all test scene samples. Through the method of the present application, autonomous driving scenes that cannot be controlled by autonomous driving can be screened out.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a scene screening method, device, electronic device and storage medium for autonomous driving. Background Art

[0002] With the development of science and technology, autonomous driving technology is becoming more and more mature. However, autonomous driving technology still cannot fully control some sensitive scenarios. The so-called sensitive scenarios are when the sensitive scenarios change slightly, the scene parameters of autonomous driving will change from meeting the requirements to not meeting the requirements.

[0003] It is necessary to screen out sensitive scenarios that autonomous driving technology cannot fully handle from a large number of scenarios, and then use a large number of simulation tests to enable autonomous driving technology to handle these sensitive scenarios; however, there is currently no method to screen out sensitive scenarios that autonomous driving technology cannot fully handle. Summary of the invention

[0004] In view of this, the purpose of the present application is to provide a scene screening method, device, electronic device and storage medium for autonomous driving, which can screen out autonomous driving scenes that cannot be handled by autonomous driving.

[0005] In a first aspect, an embodiment of the present application provides a method for screening scenes for autonomous driving, the method comprising:

[0006] Acquire multiple initial scene samples and a square deviation value of a scene parameter corresponding to each initial scene sample; the square deviation value of the scene parameter is the square of the difference between the value of the scene parameter corresponding to the initial scene sample and a preset threshold value corresponding to the scene parameter;

[0007] For each initial scene sample, determining the scene covariance between the initial scene sample and each other initial scene sample;

[0008] For each test scene sample, determine the square deviation value of the scene parameter corresponding to the test scene sample according to the test scene sample, multiple initial scene samples, the square deviation value of the scene parameter corresponding to each initial scene sample, and all scene covariances;

[0009] The autonomous driving scenarios are screened out based on the minimum value of the square deviation of the scenario parameters corresponding to all initial scenario samples, the test scenario samples, and the square deviation of the scenario parameters corresponding to all test scenario samples.

[0010] In a possible implementation, determining the square deviation value of the scene parameter corresponding to the test scene sample according to the test scene sample, multiple initial scene samples, and the square deviation value of the scene parameter corresponding to each initial scene sample, and all scene covariances includes:

[0011] Determine the squared deviation value of the scene parameter corresponding to the test scene sample according to the following a posteriori formula;

[0012] The posterior formula is: y*|x*,X,Y~N(μ * ,∑*);

[0013] Wherein, y* is the square deviation value of the scene parameter corresponding to the test scene sample, x* is the test scene sample, X is a matrix of the number of initial scene samples × the number of sample parameters composed of the initial scene samples, Y is a matrix of 1 × the number of initial scene samples composed of the square deviation values ​​of the scene parameters corresponding to multiple initial scene samples, and N represents normal distribution;

[0014] μ * =k(x*,Z)K -1 Y;

[0015] ∑*=k(x*,x*)−k(x*,Z)K -1 k(X, x*);

[0016] k(a, b) is the scene covariance of a and b, a is x* or X, b is Z or x*, Z is the matrix of the number of sample parameters composed of the initial scene samples × the number of initial scene samples, and K is the matrix of the number of initial scene samples × the number of initial scene samples composed of all scene covariances.

[0017] In a possible implementation, the autonomous driving scene is screened out according to the minimum value of the square deviation values ​​of the scene parameters corresponding to all the initial scene samples, the test scene sample, and the square deviation values ​​of the scene parameters corresponding to all the test scene samples, including:

[0018] For each test scene sample, calculate the target value of the test scene sample according to the minimum value of the square deviation values ​​of the scene parameters corresponding to all initial scene samples and the square deviation value of the scene parameters corresponding to the test scene sample;

[0019] Filter out the autonomous driving scenarios based on the maximum value among all target values.

[0020] In a possible implementation, calculating the target value of the test scene sample according to the minimum value of the square deviation values ​​of the scene parameters corresponding to all the initial scene samples and the square deviation value of the scene parameters corresponding to the test scene sample includes:

[0021] Substitute the minimum value of the square deviation values ​​of the scene parameters corresponding to all the initial scene samples and the square deviation value of the scene parameters corresponding to the test scene samples into the scene screening formula to calculate the target value of the test scene samples;

[0022] The scenario screening formula is: α(x * ) = ∫max(0, y min -y * )p(y * |x * )dy * ;

[0023] Among them, α(x * ) is the target value of the test scene sample, y min It is the minimum value of the square deviation of the scene parameters corresponding to all initial scene samples, p represents probability, d represents derivative, and y* is the square deviation of the scene parameters corresponding to the test scene sample.

[0024] In a possible implementation, the autonomous driving scenarios are screened out according to the maximum value among all target values, including:

[0025] If the maximum value is 0, the initial scene sample corresponding to the minimum value of the squared deviation values ​​of the scene parameters corresponding to all the initial scene samples is determined as the autonomous driving scene;

[0026] If the maximum value is greater than 0, and the number of initial scene samples is greater than or equal to the preset threshold, the test scene sample corresponding to the maximum value is determined as the autonomous driving scene.

[0027] In a possible implementation, the method further includes:

[0028] If the maximum value is greater than 0 and the number of initial scene samples is less than a preset threshold, the squared deviation value corresponding to the scene parameter corresponding to the maximum value is obtained;

[0029] The test scene sample corresponding to the maximum value is updated as the initial scene sample, and the process jumps to determining the scene covariance between the initial scene sample and each other initial scene sample for each initial scene sample and continues to execute.

[0030] In a second aspect, an embodiment of the present application further provides a scene screening device for autonomous driving, the scene screening device for autonomous driving comprising:

[0031] An acquisition module is used to acquire a plurality of initial scene samples and a square deviation value of a scene parameter corresponding to each initial scene sample; the square deviation value of the scene parameter is the square of the difference between the value of the scene parameter corresponding to the initial scene sample and a preset threshold value corresponding to the scene parameter;

[0032] A determination module, used for determining, for each initial scene sample, a scene covariance between the initial scene sample and each other initial scene sample;

[0033] The determination module is further used to determine, for each test scene sample, the square deviation value of the scene parameter corresponding to the test scene sample according to the test scene sample, multiple initial scene samples, the square deviation value of the scene parameter corresponding to each initial scene sample, and all scene covariances;

[0034] The screening module is used to screen out the autonomous driving scenes according to the minimum value of the square deviation values ​​of the scene parameters corresponding to all initial scene samples, the test scene samples, and the square deviation values ​​of the scene parameters corresponding to all test scene samples.

[0035] In a possible implementation manner, the determination module is specifically used to: determine the square deviation value of the scene parameter corresponding to the test scene sample according to the following a posteriori formula;

[0036] The posterior formula is: y*|x*,X,Y~N(μ * ,∑*);

[0037] Wherein, y* is the square deviation value of the scene parameter corresponding to the test scene sample, x* is the test scene sample, X is a matrix of the number of initial scene samples × the number of sample parameters composed of the initial scene samples, Y is a matrix of 1 × the number of initial scene samples composed of the square deviation values ​​of the scene parameters corresponding to multiple initial scene samples, and N represents normal distribution;

[0038] μ * =k(x*,Z)K -1 Y;

[0039] ∑*=k(x*,x*)−k(x*,Z)K -1 k(X, x*);

[0040] k(a, b) is the scene covariance of a and b, a is x* or X, b is Z or x*, Z is the matrix of the number of sample parameters composed of the initial scene samples × the number of initial scene samples, and K is the matrix of the number of initial scene samples × the number of initial scene samples composed of all scene covariances.

[0041] In one possible implementation, the screening module is specifically used to calculate, for each test scene sample, a target value of the test scene sample based on the minimum value of the squared deviations of the scene parameters corresponding to all initial scene samples and the squared deviations of the scene parameters corresponding to the test scene sample; and screen out the autonomous driving scene based on the maximum value of all target values.

[0042] In a possible implementation, the screening module is further used to substitute the minimum value of the square deviation values ​​of the scene parameters corresponding to all the initial scene samples and the square deviation value of the scene parameters corresponding to the test scene samples into the scene screening formula to calculate the target value of the test scene samples;

[0043] The scenario screening formula is: α(x * ) = ∫max(0, y min -y * )p(y * |x * )dy * ;

[0044] Among them, α(x * ) is the target value of the test scene sample, y min It is the minimum value of the square deviation of the scene parameters corresponding to all initial scene samples, p represents probability, d represents derivative, and y* is the square deviation of the scene parameters corresponding to the test scene sample.

[0045] In a possible implementation, the screening module is further configured to, if the maximum value is 0, determine the initial scene sample corresponding to the minimum value among the squared deviation values ​​of the scene parameters corresponding to all the initial scene samples as the autonomous driving scene;

[0046] If the maximum value is greater than 0, and the number of initial scene samples is greater than or equal to the preset threshold, the test scene sample corresponding to the maximum value is determined as the autonomous driving scene.

[0047] In a possible implementation, the device further includes: an updating module, a jumping module;

[0048] The acquisition module is further used to acquire the square value of the deviation corresponding to the scene parameter corresponding to the maximum value if the maximum value is greater than 0 and the number of initial scene samples is less than a preset threshold;

[0049] An updating module, used for updating the test scene sample corresponding to the maximum value to the initial scene sample;

[0050] The jump module is used to jump to each initial scene sample and continue to execute the process of determining the scene covariance between the initial scene sample and each other initial scene sample.

[0051] In the third aspect, an embodiment of the present application also provides an electronic device, comprising: a processor, a storage medium and a bus, the storage medium storing machine-readable instructions executable by the processor, when the electronic device is running, the processor and the storage medium communicate through the bus, and the processor executes the machine-readable instructions to perform the steps of the scene screening method for autonomous driving as described in any one of the first aspects.

[0052] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the scene screening method for autonomous driving as described in any one of the first aspects are executed.

[0053] The embodiment of the present application provides a scene screening method, device, electronic device and storage medium for autonomous driving, the method comprising: obtaining a plurality of initial scene samples and the square deviation value of the scene parameter corresponding to each initial scene sample; determining the scene covariance between the initial scene sample and each other initial scene sample; determining the square deviation value of the scene parameter corresponding to the test scene sample according to the test scene sample, the plurality of initial scene samples and the square deviation value of the scene parameter corresponding to each initial scene sample, and all scene covariances; screening out the autonomous driving scene according to the minimum value of the square deviation value of the scene parameter corresponding to all initial scene samples, the test scene sample, and the square deviation value of the scene parameter corresponding to all test scene samples. The present application determines the square deviation value of the scene parameter corresponding to the test scene sample by testing the scene sample, the plurality of initial scene samples and the square deviation value of the scene parameter corresponding to each initial scene sample, and all scene covariances; and then screening out the autonomous driving scene according to the minimum value of the square deviation value of the scene parameter corresponding to all initial scene samples, the test scene sample, and the square deviation value of the scene parameter corresponding to all test scene samples, and can screen out the autonomous driving scene that the autonomous driving cannot control. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1 A flowchart of a scene screening method for autonomous driving provided in an embodiment of the present application is shown;

[0056] Figure 2 A flowchart of another method for scene screening of autonomous driving provided in an embodiment of the present application is shown;

[0057] Figure 3 A schematic diagram of the structure of a scene screening device for autonomous driving provided in an embodiment of the present application is shown;

[0058] Figure 4 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0059] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of explanation and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowchart used in this application shows the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowchart can be implemented out of sequence, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art can add one or more other operations to the flowchart under the guidance of the content of the present application, or remove one or more operations from the flowchart.

[0060] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0061] In order to enable those skilled in the art to use the contents of this application, the following implementation is provided in conjunction with a specific application scenario, the "field of autonomous driving technology". For those skilled in the art, the general principles defined herein may be applied to other embodiments and application scenarios without departing from the spirit and scope of this application. Although this application is mainly described around the "field of autonomous driving technology", it should be understood that this is only an exemplary embodiment.

[0062] It should be noted that the term "comprising" will be used in the embodiments of the present application to indicate the existence of the features declared thereafter, but does not exclude the addition of other features.

[0063] Reference Figure 1 As shown, it is a flow chart of a scene screening method for autonomous driving provided in an embodiment of the present application. The exemplary steps of the embodiment of the present application are described below:

[0064] S101, obtaining a plurality of initial scene samples and a squared deviation value of a scene parameter corresponding to each initial scene sample.

[0065] In an embodiment of the present application, multiple initial scene samples are obtained, and the scene parameters corresponding to the initial scene samples may be the time from the occurrence of a collision, etc. The square deviation value of the scene parameter is the square of the difference between the value of the scene parameter corresponding to the initial scene sample and the preset threshold value corresponding to the scene parameter.

[0066] Here, the closer the square deviation value of the scene parameter is to 0, the closer the initial scene sample is to the sensitive scene, that is, the closer it is to the screened autonomous driving scene.

[0067] For example, the value of the scene parameter of the initial scene sample is 5, the preset threshold of the scene parameter is 3, and the square value of the deviation of the scene parameter corresponding to the initial scene sample is (5-3) 2 =4.

[0068] S102: For each initial scene sample, determine the scene covariance between the initial scene sample and each other initial scene sample.

[0069] In an implementation manner of the present application, each initial scene sample corresponds to a matrix of 1× the number of sample parameters, where the sample parameters include the speed, distance, acceleration, weather, etc. of the autonomous driving vehicle; the scene covariance between the matrix corresponding to the initial scene sample and the matrix corresponding to each other initial scene sample is determined.

[0070] S103. For each test scene sample, determine the square deviation value of the scene parameter corresponding to the test scene sample according to the test scene sample, multiple initial scene samples, the square deviation value of the scene parameter corresponding to each initial scene sample, and all scene covariances.

[0071] Specifically, the squared deviation value of the scene parameter corresponding to the test scene sample is determined according to the following a posteriori formula.

[0072] Here, the posterior formula is: y*|x*,X,Y~N(μ * ,∑*);

[0073] Wherein, y* is the square deviation value of the scene parameter corresponding to the test scene sample, x* is the test scene sample, X is a matrix of the number of initial scene samples × the number of sample parameters composed of the initial scene samples, Y is a matrix of 1 × the number of initial scene samples composed of the square deviation values ​​of the scene parameters corresponding to multiple initial scene samples, and N represents normal distribution;

[0074] μ * =k(x*,Z)K -1 Y;

[0075] ∑*=k(x*,x*)−k(x*,Z)K -1 k(X, x*);

[0076] k(a, b) is the scene covariance of a and b, a is x* or X, b is Z or x*, Z is the matrix of the number of sample parameters composed of the initial scene samples × the number of initial scene samples, and K is the matrix of the number of initial scene samples × the number of initial scene samples composed of all scene covariances.

[0077] S104. Filter out the autonomous driving scenes according to the minimum value of the square deviations of the scene parameters corresponding to all the initial scene samples, the test scene samples, and the square deviations of the scene parameters corresponding to all the test scene samples.

[0078] The embodiment of the present application provides a scene screening method for autonomous driving, the method comprising: obtaining a plurality of initial scene samples and the squared deviation value of the scene parameter corresponding to each initial scene sample; determining the scene covariance between the initial scene sample and each other initial scene sample; determining the squared deviation value of the scene parameter corresponding to the test scene sample according to the test scene sample, the plurality of initial scene samples and the squared deviation value of the scene parameter corresponding to each initial scene sample, and the covariance of all scenes; screening out the autonomous driving scene according to the minimum value of the squared deviation value of the scene parameter corresponding to all initial scene samples, the test scene sample, and the squared deviation value of the scene parameter corresponding to all test scene samples. The present application determines the squared deviation value of the scene parameter corresponding to the test scene sample by testing the scene sample, the plurality of initial scene samples and the squared deviation value of the scene parameter corresponding to each initial scene sample, and the covariance of all scenes; and then screening out the autonomous driving scene according to the minimum value of the squared deviation value of the scene parameter corresponding to all initial scene samples, the test scene sample, and the squared deviation value of the scene parameter corresponding to all test scene samples, and can screen out the autonomous driving scene that the autonomous driving cannot control.

[0079] Reference Figure 2 As shown, it is a flow chart of another scene screening method for autonomous driving provided in an embodiment of the present application. The exemplary steps of the embodiment of the present application are described below:

[0080] S201. For each test scene sample, calculate a target value of the test scene sample according to the minimum value of the square deviations of the scene parameters corresponding to all initial scene samples and the square deviations of the scene parameters corresponding to the test scene sample.

[0081] Specifically, the minimum value of the square deviation values ​​of the scene parameters corresponding to all initial scene samples and the square deviation value of the scene parameters corresponding to the test scene samples are substituted into the scene screening formula to calculate the target value of the test scene samples.

[0082] Here, the scenario screening formula is: α(x * ) = ∫max(0, ymin -y * )p(y * |x * )dy * .

[0083] Among them, α(x * ) is the target value of the test scene sample, y min It is the minimum value of the square deviation of the scene parameters corresponding to all initial scene samples, p represents probability, d represents derivative, and y* is the square deviation of the scene parameters corresponding to the test scene sample.

[0084] Here, the output of this formula is α(x * ) is 0 and y min -y * If the squared deviations of the scene parameters corresponding to all test scene samples are less than y min , then y min -y * It must be a positive number, that is, y min -y * is greater than 0, output α(x * ) must be a positive number, that is, the target value must be a positive number, and the larger the target value, the greater the * The smaller.

[0085] In addition, if the squared deviations of the scene parameters corresponding to all test scene samples are greater than y min ,y min -y * It must be a negative number, that is, y min -y * Less than 0, output α(x * ) must be 0, that is, the target value must be 0, so α(x * ) has only two output conditions, greater than 0 or equal to 0.

[0086] S202: Filter out the autonomous driving scenarios according to the maximum value among all target values.

[0087] Optionally, if the maximum value is 0, the initial scene sample corresponding to the minimum value of the squared deviation values ​​of the scene parameters corresponding to all the initial scene samples is determined as the autonomous driving scene.

[0088] In the implementation mode of the present application, if the maximum value is 0, it means that the squared deviation values ​​of the scene parameters corresponding to all test scene samples are greater than y min, that is, the initial scene sample corresponding to the minimum value of the square deviation values ​​of the scene parameters corresponding to all the initial scene samples is closest to the sensitive scene. Therefore, the initial scene sample corresponding to the minimum value of the square deviation values ​​of the scene parameters corresponding to all the initial scene samples is used as the autonomous driving scene.

[0089] Optionally, if the maximum value is greater than 0, and the number of initial scene samples is greater than or equal to a preset threshold, the test scene sample corresponding to the maximum value is determined as an autonomous driving scene.

[0090] In the implementation mode of the present application, if the maximum value is greater than 0, it means that the squared deviation values ​​of the scene parameters corresponding to the test scene samples are all less than y min , and the test scene sample corresponding to the maximum value is closest to the sensitive scene.

[0091] In addition, if the number of initial scene samples is greater than or equal to the preset threshold, it means that the autonomous driving scene obtained at this time has met expectations, and the test scene sample corresponding to the maximum value is determined as the autonomous driving scene.

[0092] Optionally, if the maximum value is greater than 0 and the number of initial scene samples is less than a preset threshold, the squared deviation value corresponding to the scene parameter corresponding to the maximum value is obtained; the test scene sample corresponding to the maximum value is updated to the initial scene sample, and the process jumps to each initial scene sample, determines the scene covariance between the initial scene sample and each other initial scene sample, and continues to execute.

[0093] In an implementation manner of the present application, if the number of initial scene samples is less than or equal to a preset threshold, it means that the autonomous driving scene obtained at this time does not meet expectations, and the test scene sample corresponding to the maximum value is updated to the initial scene sample, and the process jumps to each initial scene sample, and determines the scene covariance between the initial scene sample and each other initial scene sample to continue execution.

[0094] Reference Figure 3 2 is a schematic diagram of a scene screening device for autonomous driving provided in an embodiment of the present application, and the scene screening device for autonomous driving includes:

[0095] An acquisition module 301 is used to acquire a plurality of initial scene samples and a square deviation value of a scene parameter corresponding to each initial scene sample; the square deviation value of the scene parameter is the square of the difference between the value of the scene parameter corresponding to the initial scene sample and a preset threshold value corresponding to the scene parameter;

[0096] A determination module 302 is used to determine, for each initial scene sample, a scene covariance between the initial scene sample and each other initial scene sample;

[0097] The determination module 302 is further used to determine, for each test scene sample, a square deviation value of a scene parameter corresponding to the test scene sample according to the test scene sample, multiple initial scene samples, and the square deviation value of the scene parameter corresponding to each initial scene sample, and all scene covariances;

[0098] The screening module 303 is used to screen out the autonomous driving scenes according to the minimum value of the square deviation values ​​of the scene parameters corresponding to all the initial scene samples, the test scene samples, and the square deviation values ​​of the scene parameters corresponding to all the test scene samples.

[0099] In a possible implementation, the determination module 302 is specifically configured to: determine the square deviation value of the scene parameter corresponding to the test scene sample according to the following a posteriori formula;

[0100] The posterior formula is: y*|x*,X,Y~N(μ * ,∑*);

[0101] Wherein, y* is the square deviation value of the scene parameter corresponding to the test scene sample, x* is the test scene sample, X is a matrix of the number of initial scene samples × the number of sample parameters composed of the initial scene samples, Y is a matrix of 1 × the number of initial scene samples composed of the square deviation values ​​of the scene parameters corresponding to multiple initial scene samples, and N represents normal distribution;

[0102] μ * =k(x*,Z)K -1 Y;

[0103] ∑*=k(x*,x*)−k(x*,Z)K -1 k(X, x*);

[0104] k(a, b) is the scene covariance of a and b, a is x* or X, b is Z or x*, Z is the matrix of the number of sample parameters composed of the initial scene samples × the number of initial scene samples, and K is the matrix of the number of initial scene samples × the number of initial scene samples composed of all scene covariances.

[0105] In one possible implementation, the screening module 303 is specifically used to calculate, for each test scene sample, a target value of the test scene sample based on the minimum value of the squared deviations of the scene parameters corresponding to all initial scene samples and the squared deviations of the scene parameters corresponding to the test scene sample; and screen out the autonomous driving scene based on the maximum value of all target values.

[0106] In a possible implementation, the screening module 303 is further used to substitute the minimum value of the square deviation values ​​of the scene parameters corresponding to all the initial scene samples and the square deviation value of the scene parameters corresponding to the test scene samples into the scene screening formula to calculate the target value of the test scene samples;

[0107] The scenario screening formula is: α(x * ) = ∫max(0, y min -y * )p(y * |x * )dy * ;

[0108] Among them, α(x * ) is the target value of the test scene sample, y min It is the minimum value of the square deviation of the scene parameters corresponding to all initial scene samples, p represents probability, d represents derivative, and y* is the square deviation of the scene parameters corresponding to the test scene sample.

[0109] In a possible implementation, the screening module 303 is further configured to, if the maximum value is 0, determine the initial scene sample corresponding to the minimum value of the squared deviation values ​​of the scene parameters corresponding to all the initial scene samples as the autonomous driving scene;

[0110] If the maximum value is greater than 0, and the number of initial scene samples is greater than or equal to the preset threshold, the test scene sample corresponding to the maximum value is determined as the autonomous driving scene.

[0111] In a possible implementation, the device further includes: an updating module 304, a jumping module 305;

[0112] The acquisition module 301 is further used to acquire the square value of the deviation corresponding to the scene parameter corresponding to the maximum value if the maximum value is greater than 0 and the number of initial scene samples is less than a preset threshold;

[0113] An updating module 304 is used to update the test scene sample corresponding to the maximum value to the initial scene sample;

[0114] The jump module 305 is used to jump to the step of determining the scene covariance between each initial scene sample and each other initial scene sample and continue to execute.

[0115] The embodiment of the present application provides a scene screening device for autonomous driving, which includes: an acquisition module 301, which is used to acquire a plurality of initial scene samples and the square deviation value of the scene parameter corresponding to each initial scene sample; a determination module 301, which is used to determine the scene covariance between the initial scene sample and each other initial scene sample; a determination module 302, which is also used to determine the square deviation value of the scene parameter corresponding to the test scene sample according to the test scene sample, the plurality of initial scene samples and the square deviation value of the scene parameter corresponding to each initial scene sample, and all scene covariances; a screening module 302, which is used to screen out the autonomous driving scene according to the minimum value of the square deviation value of the scene parameter corresponding to all initial scene samples, the test scene sample, and the square deviation value of the scene parameter corresponding to all test scene samples. The present application determines the square deviation value of the scene parameter corresponding to the test scene sample through the test scene sample, the plurality of initial scene samples and the square deviation value of the scene parameter corresponding to each initial scene sample, and all scene covariances; and then screens out the autonomous driving scene according to the minimum value of the square deviation value of the scene parameter corresponding to all initial scene samples, the test scene sample, and the square deviation value of the scene parameter corresponding to all test scene samples, so as to screen out the autonomous driving scene that the autonomous driving cannot control.

[0116] like Figure 4 As shown, an electronic device 400 provided in an embodiment of the present application includes: a processor 401, a memory 402 and a bus, the memory 402 stores machine-readable instructions executable by the processor 401, when the electronic device is running, the processor 401 and the memory 402 communicate through the bus, and the processor 401 executes the machine-readable instructions to perform the steps of the scene screening method for autonomous driving as described above.

[0117] Specifically, the above-mentioned memory 402 and processor 401 can be general-purpose memories and processors, which are not specifically limited here. When the processor 401 runs the computer program stored in the memory 402, it can execute the above-mentioned scene screening method for autonomous driving.

[0118] Corresponding to the above-mentioned scene screening method for autonomous driving, an 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 steps of the above-mentioned scene screening method for autonomous driving are executed.

[0119] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the method embodiment, and will not be repeated in this application. In the several embodiments provided in this application, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules 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 through some communication interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0120] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0121] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0122] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the information processing method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard drives, ROM, RAM, magnetic disks, or optical disks.

[0123] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A scene screening method for autonomous driving, characterized in that: The scene screening method comprises: Acquire multiple initial scene samples and a square deviation value of a scene parameter corresponding to each of the initial scene samples; the square deviation value of the scene parameter is the square of a difference between a value of the scene parameter corresponding to the initial scene sample and a preset threshold value corresponding to the scene parameter; For each initial scene sample, determining a scene covariance between the initial scene sample and each other initial scene sample; For each test scene sample, according to the test scene sample, the multiple initial scene samples and the square deviation value of the scene parameter corresponding to each of the initial scene samples, and all scene covariances, the square deviation value of the scene parameter corresponding to the test scene sample is determined, including: determining the square deviation value of the scene parameter corresponding to the test scene sample according to the following a posteriori formula; the a posteriori formula is: y*|x*,X,Y~N( , ); where y* is the squared deviation of the scene parameter corresponding to the test scene sample, x* is the test scene sample, and X is the number of initial scene samples composed of the initial scene samples. The matrix of the number of scene parameters, Y is composed of the squared deviations of the scene parameters corresponding to multiple initial scene samples. The matrix of the number of initial scene samples, N represents normal distribution; =k(x*,z)K -1 Y; =k(x*,x*)-k(x*,Z)K -1 k(X, x*); is the scene covariance of a and b, a is x* or X, b is Z or x*, and Z is the number of sample parameters composed of the initial scene samples The matrix of the number of initial scene samples, K is the number of initial scene samples composed of all scene covariances The matrix of the number of initial scene samples; The method further comprises: for each test scene sample, substituting the minimum value of the square deviation values ​​of the scene parameters corresponding to all the initial scene samples and the square deviation value of the scene parameters corresponding to the test scene samples into a scene screening formula to calculate the target value of the test scene sample; the scene screening formula is: ;in, is the target value of the test scenario sample, and the output of the scenario screening formula is 0 and The maximum value in is the minimum value of the squared deviations of the scene parameters corresponding to all the initial scene samples, represents the probability, d Represents seeking guidance, y* is the squared deviation value of the scene parameter corresponding to the test scene sample; if the maximum value of the target value is 0, the initial scene sample corresponding to the minimum value of the squared deviation values ​​of the scene parameters corresponding to all the initial scene samples is determined as the autonomous driving scene; if the maximum value of the target value is greater than 0, and the number of the initial scene samples is greater than or equal to a preset threshold; the test scene sample corresponding to the maximum value of the target value is determined as the autonomous driving scene.

2. The scene screening method for autonomous driving according to claim 1, characterized in that: The method further comprises: If the maximum value is greater than 0 and the number of the initial scene samples is less than a preset threshold, obtaining a square deviation value corresponding to the scene parameter corresponding to the maximum value; The test scene sample corresponding to the maximum value is updated as the initial scene sample, and the process jumps to, for each initial scene sample, determining the scene covariance between the initial scene sample and each other initial scene sample and continues to execute.

3. A scene screening device for autonomous driving, characterized in that: The scene screening device for autonomous driving comprises: An acquisition module, used to acquire a plurality of initial scene samples and a square deviation value of a scene parameter corresponding to each of the initial scene samples; the square deviation value of the scene parameter is the square of a difference between a value of the scene parameter corresponding to the initial scene sample and a preset threshold value corresponding to the scene parameter; A determination module, configured to determine, for each initial scene sample, a scene covariance between the initial scene sample and each other initial scene sample; The determination module is further used to determine, for each test scene sample, the square deviation value of the scene parameter corresponding to the test scene sample according to the test scene sample, the multiple initial scene samples, the square deviation value of the scene parameter corresponding to each of the initial scene samples, and all scene covariances, including: determining the square deviation value of the scene parameter corresponding to the test scene sample according to the following a posteriori formula; the a posteriori formula is: y*|x*,X,Y~N( , ); where y* is the squared deviation of the scene parameter corresponding to the test scene sample, x* is the test scene sample, and X is the number of initial scene samples composed of the initial scene samples. The matrix of the number of scene parameters, Y is composed of the squared deviations of the scene parameters corresponding to multiple initial scene samples. The matrix of the number of initial scene samples, N represents normal distribution; =k(x*,z)K -1 Y; =k(x*,x*)-k(x*,Z)K -1 k(X, x*); is the scene covariance of a and b, a is x* or X, b is Z or x*, and Z is the number of sample parameters composed of the initial scene samples The matrix of the number of initial scene samples, K is the number of initial scene samples composed of all scene covariances The matrix of the number of initial scene samples; The screening module is used to screen out the autonomous driving scenes according to the minimum value of the square deviation values ​​of the scene parameters corresponding to all the initial scene samples, the test scene sample, and the square deviation values ​​of the scene parameters corresponding to all the test scene samples, including: for each test scene sample, substituting the minimum value of the square deviation values ​​of the scene parameters corresponding to all the initial scene samples and the square deviation value of the scene parameters corresponding to the test scene sample into a scene screening formula to calculate the target value of the test scene sample; the scene screening formula is: ;in, is the target value of the test scenario sample, and the output of the scenario screening formula is 0 and The maximum value in is the minimum value of the squared deviations of the scene parameters corresponding to all the initial scene samples, represents the probability, d Represents seeking guidance, y* is the squared deviation value of the scene parameter corresponding to the test scene sample; if the maximum value of the target value is 0, the initial scene sample corresponding to the minimum value of the squared deviation values ​​of the scene parameters corresponding to all the initial scene samples is determined as the autonomous driving scene; if the maximum value of the target value is greater than 0, and the number of the initial scene samples is greater than or equal to a preset threshold; the test scene sample corresponding to the maximum value of the target value is determined as the autonomous driving scene.

4. An electronic device, characterized in that: include: A processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the scene screening method for autonomous driving as described in claim 1 or 2.

5. 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 scene screening method for autonomous driving as described in claim 1 or 2 are executed.

Images