An autonomous driving test method, device, equipment and medium

Abstract

The present application relates to an autonomous driving test method, device, equipment and medium. By obtaining a plurality of test scenarios and the correlation information of a plurality of target scenario elements corresponding to a target test function, correlation analysis processing is performed on any two of the plurality of target scenario elements to obtain a target correlation matrix; the weight of each target scenario element is obtained according to the target correlation matrix, and based on the weights of the plurality of target scenario elements and the correlation information of the plurality of target scenario elements, a target test scenario is determined from the plurality of test scenarios, which can accurately and efficiently screen out invalid scenarios. Testing the target test function of a vehicle based on the target test scenario can avoid the waste of test resources caused by testing the target test function of a vehicle using invalid scenarios and the problem of inaccurate test results caused by testing with invalid scenarios, thereby improving the accuracy and overall test efficiency of autonomous driving tests.

Description

Technical Field

[0001] The present invention relates to the technical field of autonomous driving, and particularly to a method, device, equipment and medium for autonomous driving testing. Background Art

[0002] In the field of autonomous driving technology, testing autonomous driving can ensure general autonomous driving functions. Among them, the construction of virtual test scenarios is an important part of autonomous driving testing. The test scenarios in autonomous driving can refer to the combination of driving sites and environments. Specifically, they can include factors such as roads, traffic elements, weather, and lighting. In the related art, real vehicle road sampling can be carried out, and algorithms such as Monte Carlo are used to extract generalization factors for scenario generalization, and simulation tools are used for scenario generation. During the process of scenario generation, although a large amount of scenario data can be generated based on the generalization factors, due to the random nature of the algorithms, some randomly generated scenarios do not exist in the actual environment, resulting in the generation of invalid test scenarios. Identifying valid scenario data from hundreds of thousands or even millions of generalized scenarios manually will lead to a waste of a large amount of human resources and time resources. If invalid scenarios are directly used for automated testing, on the one hand, it wastes testing resources, and on the other hand, the statistical analysis of the test results also interferes with the actual evaluation of autonomous driving. Summary of the Invention

[0003] To solve the above technical problems, the present invention provides a method, device, equipment and medium for autonomous driving testing, which can quickly and conveniently eliminate invalid test scenarios, save testing resources, and improve the efficiency and accuracy of autonomous driving testing.

[0004] According to the first aspect of the embodiments of the present disclosure, a method for autonomous driving testing is provided. The method includes:

[0005] Obtain a plurality of test scenarios and the correlation information of a plurality of target scenario elements corresponding to the target test function; the correlation information represents the degree of correlation between each target scenario element and the target test function;

[0006] Perform correlation analysis processing on any two of the plurality of target scenario elements to obtain a target correlation matrix; each correlation information in the target correlation matrix refers to the comparison result of the correlation degrees of the two target scenario elements corresponding to each correlation information with the target test function;

[0007] Obtain the weight of each target scenario element according to the target correlation matrix;

[0008] Determine the target test scenario from the plurality of test scenarios based on the weights of the plurality of target scenario elements and the correlation information of the plurality of target scenario elements;

[0009] Test the target test function of the vehicle based on the target test scenario.

[0010] In a possible implementation, each association degree information in the target association degree matrix is an association value within a target range. The association analysis and processing of any two target scenario elements among the multiple target scenario elements to obtain the association degree matrix includes:

[0011] Perform association analysis and processing on any two target scenario elements among the multiple target scenario elements to obtain a first association degree matrix;

[0012] Determine the consistency index value according to the eigenvalues of the first association degree matrix;

[0013] Generate a random consistency index value based on the number of the multiple target scenario elements and the target range;

[0014] When the consistency index value and the random consistency index value do not meet the preset conditions, re-perform the association analysis and processing on any two target scenario elements among the multiple target scenario elements to obtain a second association degree matrix, until the consistency index value and the random consistency index value meet the preset conditions, and use the association degree matrix that meets the preset conditions as the target association degree matrix.

[0015] In a possible implementation, the determining the target test scenario from the multiple test scenarios based on the weights of the multiple target scenario elements and the correlation information of the multiple target scenario elements includes:

[0016] Determine the effective values of the multiple test scenarios based on the weights of the multiple target scenario elements and the correlation information of the multiple target scenario elements;

[0017] Sort the multiple test scenarios according to the effective values of the multiple test scenarios;

[0018] Use the preset number of test scenarios with higher rankings as the target test scenarios.

[0019] In a possible implementation, the obtaining the weight of each target scenario element according to the target association degree matrix includes:

[0020] Normalize the target association degree matrix to obtain a target matrix;

[0021] Perform row addition and averaging processing on the target matrix to obtain the weight of each target scenario element.

[0022] Before obtaining the relevance information of multiple target scenario elements corresponding to multiple test scenarios and a target test function, the method further includes:

[0023] Pre-generate a relevance function for each test function, where the relevance function characterizes the relationship between each test function and the corresponding scenario elements;

[0024] Determine the scenario elements corresponding to each test function according to the relevance function of each test function.

[0025] In a possible implementation manner, the method further includes:

[0026] Determine the relevance information of multiple target scenario elements corresponding to each test function according to the number of occurrences and / or duration of the scenario elements corresponding to each test function in the multiple test scenarios.

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

[0028] Determine the number of the multiple target scenario elements;

[0029] Set the target range according to the number of the multiple target scenario elements.

[0030] According to a second aspect of the embodiments of the present disclosure, there is provided an autonomous driving test device, which may include:

[0031] An information acquisition module configured to acquire multiple test scenarios and the relevance information of multiple target scenario elements corresponding to a target test function; the relevance information characterizes the degree of correlation between each target scenario element and the target test function;

[0032] A matrix determination module configured to perform correlation analysis processing on any two of the multiple target scenario elements to obtain a target correlation matrix; each correlation information in the target correlation matrix refers to the comparison result of the correlation degrees of the two target scenario elements corresponding to each correlation information with the target test function;

[0033] A weight determination module configured to obtain the weight of each target scenario element according to the target correlation matrix;

[0034] A target test scenario determination module configured to determine a target test scenario from the multiple test scenarios based on the weights of the multiple target scenario elements and the relevance information of the multiple target scenario elements;

[0035] A test module configured to test the target test function of the vehicle based on the target test scenario.

[0036] In a possible implementation, the matrix determination module may include:

[0037] A first matrix determination unit, configured to perform correlation analysis processing on any two of the multiple target scenario elements to obtain a first correlation matrix;

[0038] A consistency index value determination unit, configured to determine a consistency index value according to the eigenvalues of the first correlation matrix;

[0039] A random value determination unit, configured to generate a random consistency index value based on the number of the multiple target scenario elements and the target range;

[0040] A target correlation matrix determination unit, configured to, when the consistency index value and the random consistency index value do not meet the preset conditions, re-perform correlation analysis processing on any two of the multiple target scenario elements to obtain a second correlation matrix, and until the consistency index value and the random consistency index value meet the preset conditions, use the correlation matrix that meets the preset conditions as the target correlation matrix.

[0041] In a possible implementation, the target test scenario determination module includes:

[0042] A valid value determination unit, configured to determine valid values of the multiple test scenarios based on the weights of the multiple target scenario elements and the correlation information of the multiple target scenario elements;

[0043] A sorting unit, configured to sort the multiple test scenarios according to the valid values of the multiple test scenarios;

[0044] A target test scenario determination unit, configured to use a preset number of test scenarios with higher rankings as the target test scenarios.

[0045] In a possible implementation, the weight determination module includes:

[0046] A target matrix determination unit, configured to perform normalization processing on the target correlation matrix to obtain a target matrix;

[0047] A weight determination unit, configured to perform row-by-row summation and averaging processing on the target matrix to obtain the weight of each target scenario element.

[0048] In a possible implementation, the device further includes:

[0049] A correlation function generation module, configured to pre-generate a correlation function for each test function, where the correlation function characterizes that each test function is related to the corresponding scenario element;

[0050] A scene element determination module, configured to determine scene elements corresponding to each test function according to a correlation function of each test function.

[0051] In a possible implementation manner, the apparatus further includes:

[0052] A correlation information determination module, configured to determine correlation information of a plurality of target scene elements corresponding to each test function according to the number of occurrences and / or duration of the scene elements corresponding to each test function in the plurality of test scenarios.

[0053] In a possible implementation manner, the apparatus further includes:

[0054] A quantity determination module, configured to determine the quantity of the plurality of target scene elements;

[0055] A target range setting module, configured to set the target range according to the quantity of the plurality of target scene elements.

[0056] According to a third aspect of the embodiments of the present disclosure, there is provided an electronic device, including:

[0057] A processor;

[0058] A memory for storing executable instructions of the processor;

[0059] Wherein, the processor is configured to execute the instructions to implement the method according to any one of the above first aspects.

[0060] According to a fourth aspect of the embodiments of the present disclosure, there is provided a computer-readable storage medium, when instructions in the computer-readable storage medium are executed by a processor of an electronic device, enabling the electronic device to execute the method according to any one of the first aspects of the embodiments of the present disclosure.

[0061] According to a fifth aspect of the embodiments of the present disclosure, there is provided a computer program product, enabling a computer to execute the method according to any one of the first aspects of the embodiments of the present disclosure.

[0062] Implementing the present application has the following beneficial effects:

[0063] In this application, by obtaining multiple test scenarios and the correlation information of multiple target scenario elements corresponding to the target test function, performing correlation analysis processing on any two of the multiple target scenario elements to obtain a target correlation matrix, the correlation between the target test function and the target scenario elements can be well reflected; based on the target correlation matrix, the weight of each target scenario element is obtained, and based on the weights of the multiple target scenario elements and the correlation information of the multiple target scenario elements, the target test scenario is determined from the multiple test scenarios, which can accurately and efficiently screen out invalid scenarios. Testing the target test function of the vehicle based on the target test scenario can avoid the waste of test resources caused by testing the target test function of the vehicle using invalid scenarios and the problem of inaccurate test results caused by testing with invalid scenarios, thereby improving the accuracy and overall test efficiency of autonomous driving testing. Description of the Drawings

[0064] To more clearly illustrate the technical solutions of this application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of this application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0065] Figure 1 It is a flowchart of an autonomous driving test method provided by an embodiment of this application;

[0066] Figure 2 It is a flowchart of obtaining a correlation matrix shown according to an exemplary embodiment;

[0067] Figure 3 It is a flowchart of obtaining the weight of each target scenario element according to the target correlation matrix provided by an embodiment of this application;

[0068] Figure 4 It is a flowchart of determining the target test scenario from multiple test scenarios provided by an embodiment of this application;

[0069] Figure 5 It is a flowchart of an autonomous driving test method provided by another embodiment of this application;

[0070] Figure 6 It is a diagram of an autonomous driving test device provided by an embodiment of this application;

[0071] Figure 7 It is a block diagram of an electronic device for autonomous driving testing shown by an embodiment of this application. Detailed Embodiments

[0072] To enable those skilled in the art to better understand the technical solutions in this application, the following will clearly and completely describe the technical solutions in the embodiments of this application in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of this application.

[0073] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0074] To implement the technical solutions of this application and make it easier for more engineering and technical workers to understand and apply this application, the working principle of this application will be further elaborated in conjunction with specific embodiments.

[0075] This application can be applied to the field of autonomous driving, and particularly relates to an autonomous driving test method, device, equipment, and medium.

[0076] It should be noted that the following shows a possible step sequence, and in fact, it is not necessary to strictly follow this sequence. Some steps can be executed in parallel without relying on each other. The user information involved in this disclosure (including but not limited to user device information, user personal information, user behavior information, etc.) is information that has been authorized by the user or fully authorized by all parties.

[0077] Figure 1 is a flowchart of an autonomous driving test method shown according to an exemplary embodiment. As Figure 1 shown, the autonomous driving test method includes the following steps:

[0078] In step S11, obtain multiple test scenarios and the correlation information of multiple target scenario elements corresponding to the target test function.

[0079] In the embodiments of this specification, the test scenario may refer to a simulation test scenario for testing the automatic driving of a vehicle, which may be a combination of a driving site and the environment. Specifically, the test scenario may include factors such as roads, traffic elements, weather, and lighting. In some examples, the test scenario may be a test scenario generated by a generalization algorithm. The target test function may be a function to be tested among the test functions related to automatic driving. For example, the target test function may be functions such as a lane keeping function, an AEB function (detecting a safe distance for alarm prompts and being able to automatically start, which is vehicle automatic braking), or an ACC function (adaptive cruise control function).

[0080] In the embodiments of this specification, multiple target scenario elements may refer to multiple of the multiple scenario elements, and the multiple target scenario elements may be scenario elements corresponding to the target test function. Scenario elements may be the factors constituting the test scenario, and scenario elements may be elements such as rainfall, vehicle speed, and other vehicles cutting in, etc.

[0081] In the embodiments of this specification, the relevance information may characterize the degree of relevance between the target scenario element and the target test function. Specifically, the relevance information may be a correlation value. When the correlation value is larger, it indicates that the degree of relevance between the target scenario element and the target test function is higher, and vice versa, it indicates that the degree of relevance between the target scenario element and the target test function is lower.

[0082] In step S12, correlation analysis processing is performed on any two of the multiple target scenario elements to obtain a target correlation matrix.

[0083] In the embodiments of this specification, each correlation information in the target correlation matrix may refer to the comparison result of the correlation degrees obtained by the two target scenario elements corresponding to each correlation information and the target test function.

[0084] In some embodiments, each correlation information in the target correlation matrix may be a correlation value within a target range. For example, each correlation information may take a natural number within the range of 1 - 5, or one of the values 1, 3, 5, 7, 9, and the larger the value, the higher the degree of correlation between the target scenario element and the target test function.

[0085] Correspondingly, as Figure 2 shown, step S12, performing correlation analysis processing on any two of the multiple target scenario elements to obtain a correlation matrix may include:

[0086] In step S121, correlation analysis processing is performed on any two of the multiple target scenario elements to obtain a first correlation matrix.

[0087] In the embodiments of this specification, the correlation analysis process may be to compare the correlation degree information of any two of the multiple target scenario elements, and obtain the correlation value within the target range according to the high or low correlation between the two compared target scenario elements and the target test function respectively.

[0088] In step S122, according to the eigenvalue of the first correlation matrix, the consistency index value is determined.

[0089] The method for solving the eigenvalue is relatively well-known, and the present disclosure will not describe it. In the embodiments of this specification, after obtaining the eigenvalue γ of the first correlation matrix max the consistency index value CI can be obtained according to the following formula:

[0090] CI = (γ max - n) / (n - 1), where n is the number of target scenario elements.

[0091] In step S123, based on the number of multiple target scenario elements and the target range, a random consistency index value is generated.

[0092] In the embodiments of this specification, according to the number of multiple target scenario elements and the target range, multiple numbers can be randomly generated. The number of these multiple numbers is the square of the number of multiple target scenario elements. A multi-order matrix is generated from these multiple numbers, the eigenvalue of the multi-order matrix is determined, and the consistency index value of the multi-order matrix is obtained. Repeat the steps of randomly generating multiple numbers according to the number of multiple target scenario elements and the target range multiple times, generating a multi-order matrix from these multiple numbers, and determining the eigenvalue of the multi-order matrix to obtain corresponding multiple consistency index values; calculate the average of the multiple consistency index values to obtain the random consistency index value.

[0093] In a specific embodiment, taking the number of target scenario elements as n and the target range as the 9 natural numbers 1, 2, 3... 9 as an example, the value range of the randomly generated multiple numbers can be 1, 2, 3... 9 (the target range), and 1 / 2, 1 / 3, 1 / 4... 1 / 9 (the correlation range of the target range, that is, taking the reciprocal), a total of 17 numbers. Randomly generate n 2 numbers from this value range, which can form an n-order matrix, calculate the eigenvalue of the n-order matrix, and use this eigenvalue as the consistency index value of the n-order matrix. Repeat the above process m times to obtain m consistency index values, and take the average of the m consistency index values to obtain the random consistency index value. This can make the random consistency index value more random. Optionally, the value of m can be made as large as possible to improve the randomness of the random consistency index value.

[0094] In step S124, when the consistency index value and the random consistency index value do not meet the preset conditions, re - conduct the correlation analysis process for any two of the multiple target scenario elements to obtain a second correlation matrix. Until the consistency index value and the random consistency index value meet the preset conditions, use the correlation matrix that meets the preset conditions as the target correlation matrix.

[0095] In the embodiments of this specification, if the consistency index value and the random consistency index value do not meet the preset conditions, it can indicate that the first correlation matrix is relatively unreasonable. Then, the correlation analysis process can be re - conducted to obtain a second correlation matrix. When the consistency index value and the random consistency index value of the latest correlation matrix meet the preset conditions, it can indicate that the correlation matrix that meets the preset conditions is set reasonably. Thus, use the correlation matrix that meets the preset conditions as the target correlation matrix.

[0096] In the embodiments of this specification, the preset condition can be that the ratio of the consistency index value to the random consistency index value is less than a preset threshold. Preferably, the preset condition can be that the ratio of the consistency index value to the random consistency index value is less than 0.1.

[0097] By conducting the correlation analysis process for any two of the multiple target scenario elements to obtain a first correlation matrix, determining the consistency index value based on the eigenvalues of the first correlation matrix, generating a random consistency index value based on the number and target range of the multiple target scenario elements. When the consistency index value and the random consistency index value do not meet the preset conditions, re - conduct the correlation analysis process for any two of the multiple target scenario elements until a correlation matrix whose consistency index value and random consistency index value meet the preset conditions is obtained. This correlation matrix that meets the preset conditions can be used as the target correlation matrix, thereby improving the rationality of setting the target correlation matrix, avoiding determining the target test scenario using an unreasonable correlation matrix, and thus improving the overall accuracy of the autonomous driving test method.

[0098] Optionally, first determine the number of multiple target scenario elements, and set the target range according to the number of the multiple target scenario elements. For a larger number of target scenario elements, a larger target range can be set to refine the value of the correlation information, so that the correlation between the target test function and the multiple target scenario elements is more specific, improving the accuracy of the correlation analysis process, thereby improving the accuracy of the target test scenario and the overall efficiency and accuracy of the target test.

[0099] In step S13, obtain the weight of each target scenario element according to the target correlation matrix.

[0100] In the embodiments of the present specification, the weight of each target scenario element can represent the relative importance degree of each target scenario element in the target test function.

[0101] In some embodiments, as Figure 3 shown, step S13 of obtaining the weight of each target scenario element according to the target correlation matrix may include:

[0102] In step S131, normalize the target correlation matrix to obtain a target matrix.

[0103] In step S132, perform row addition and averaging on the target matrix to obtain the weight of each target scenario element.

[0104] Taking the target scenario elements of the target test function as A1, A2, and A3 as an example, the target range is {1, 2, 3, 4, 5}, and the target correlation matrix may be the following matrix:

[0105] Target scene element <![CDATA[A1]]> <![CDATA[A2]]> <![CDATA[A3]]> <![CDATA[A1]]> <![CDATA[a 11 = 1]]> <![CDATA[a 12 = 3]]> <![CDATA[a 13 = 5]]> <![CDATA[A2]]> <![CDATA[a 21 = 1 / 3]]> <![CDATA[a 22 = 1]]> <![CDATA[a 23 = 1 / 4]]> <![CDATA[A3]]> <![CDATA[a 31 = 1 / 5]]> <![CDATA[a 32 = 4]]> <![CDATA[a 33 = 1]]>

[0106] Among them, a ij can represent the comparison result of the correlation degree between the corresponding two target scenario elements and the target test function. i represents the row number, and j represents the column number. For example, a 11 =1 can represent that the correlation degree of the target scenario element A1 to the target test function is the same as the correlation degree of the target scenario element A1 to the target test function. a 12 =3 can represent that the correlation degree of the target scenario element A1 to the target test function is higher than the correlation degree of the target scenario element A2 to the target test function by a third degree value. a 13 =5 can represent that the correlation degree of the target scenario element A1 to the target test function is higher than the correlation degree of the target scenario element A3 to the target test function by a fifth degree value. a 21 =1 / 3 can represent that the correlation degree of the target scenario element A1 to the target test function is higher than the correlation degree of the target scenario element A2 to the target test function by a third degree value. The same applies to the other comparison results a. Normalizing the target correlation matrix may be to calculate the proportion of each column a ij in the target correlation matrix after summing in that column to obtain the target matrix, and perform row addition and averaging on the target matrix to obtain the weights w1, w2, and w3 of each target scenario element.

[0107] By normalizing the target correlation matrix to obtain a target matrix, performing row addition and averaging on the target matrix to obtain the weight of each target scenario element, it is possible to quickly generate the weight of each target scenario element, improving the processing efficiency and accuracy of the weight of the target scenario element.

[0108] In step S14, based on the weights of multiple target scenario elements and the correlation information of multiple target scenario elements, a target test scenario is determined from multiple test scenarios.

[0109] In the embodiments of this specification, the screening conditions for the target test scenario can be determined based on the weights of multiple target scenario elements and the correlation information of multiple target scenario elements, and the target test scenario is determined from multiple test scenarios using the screening conditions.

[0110] In some embodiments, as Figure 4 shown, step S14, determining a target test scenario from multiple test scenarios based on the weights of multiple target scenario elements and the correlation information of multiple target scenario elements may include:

[0111] In step S141, based on the weights of multiple target scenario elements and the correlation information of multiple target scenario elements, the effective values of multiple test scenarios are determined.

[0112] In the embodiments of this specification, it may be to perform multiplication and summation processing on the weights and correlation information of each target scenario element in each test scenario to obtain the effective value of each test scenario. For example, the target scenario elements corresponding to the target test function are A1, A2, and A3, and the weights of the target scenario elements A1, A2, and A3 are w1, w2, and w3 respectively. The correlation value of the target scenario element A1 in test scenario 1 is c11, the correlation value of the target scenario element A2 is c12, and the correlation value of the target scenario element A3 is c13. Then, the effective value of test scenario 1 can be determined as C1 = w1*c11 + w2*c12 + w3*c13. The correlation value of the target scenario element A1 in test scenario 2 is c21, the correlation value of the target scenario element A2 is c22, and the correlation value of the target scenario element A3 is c23. Then, the effective value of test scenario 2 can be determined as C2 = w1*c21 + w2*c22 + w3*c23. The correlation value of the target scenario element A1 in test scenario 3 is c31, the correlation value of the target scenario element A2 is c32, and the correlation value of the target scenario element A3 is c33. Then, the effective value of test scenario 3 can be determined as C3 = w1*c31 + w2*c32 + w3*c33.

[0113] In step S142, the multiple test scenarios are sorted according to the effective values of the multiple test scenarios.

[0114] In the embodiments of this specification, the higher the effective value of the test scenario, the more reference value the test scenario has for testing the target test function. On the contrary, it can indicate that the test scenario is more ineffective for testing the target test function, resulting in a higher waste of test resources and a stronger interference with the test results.

[0115] In step S143, the preset number of test scenarios with higher rankings are used as target test scenarios.

[0116] In step S15, the target test function of the vehicle is tested based on the target test scenarios.

[0117] Taking the target scenario elements corresponding to the foregoing target test function as A1, A2, and A3, and the weights of the target scenario elements A1, A2, and A3 as w1, w2, and w3 as an example, C1, C2, and C3 can be sorted, and the two test scenarios with higher rankings are used as target test scenarios, and the target test function of the vehicle is tested based on the two target test scenarios.

[0118] Optionally, test scenarios with valid values higher than a preset threshold can also be used as target test scenarios.

[0119] In one example, for the AEB function, test scenarios of other vehicles cutting in or out are valid scenarios for testing the AEB function. According to the target correlation matrix, the weight of each cut-in or cut-out occurrence can be obtained. However, in different test scenarios, the number of cut-in or cut-out actions is different. Based on different numbers as correlation information, the weights can be combined to compare the effectiveness of different test scenarios. For example, test scenario a and test scenario b are two test scenario files. Test scenario a has 3 vehicle cut-ins, and test scenario b has 5 vehicle cut-ins. Then the valid value of test scenario a is lower than the valid value of test scenario b.

[0120] In practical applications, there are many target scenario elements corresponding to a target test function, and the number of test scenarios is also very large. By determining the target test scenarios through the valid values of the test scenarios, a large number of invalid test scenarios can be reasonably screened out, the effectiveness of the target test scenarios can be improved, and thus the overall test efficiency can be improved.

[0121] In some embodiments, as Figure 5 shown, before obtaining multiple test scenarios and the correlation information of multiple target scenario elements corresponding to the target test function, the method may further include:

[0122] In step S51, a correlation function for each test function is pre-generated.

[0123] In the embodiments of this specification, the correlation function can characterize that each test function is related to the corresponding scenario element. For example, the correlation function of a test function can be b1 + b2 + b3 + b4, where b1 can be the number of occurrences of scenario element B1, b2 can be the duration of occurrence of scenario element B2, b3 can be whether scenario element B3 appears (appearance is recorded as 1, non - appearance is recorded as 0), and b4 can be the number of kilometers of scenario element B4. It should be noted that the correlation function can be set according to the actual situation, and its structure can be simplified or complicated according to the scenario elements corresponding to the test function. This application does not make any limitations in this regard.

[0124] In step S52, determine the scenario element corresponding to each test function according to the correlation function of each test function.

[0125] In the embodiments of this specification, the scenario element corresponding to each test function can be determined through the scenario elements involved in the correlation function of each test function.

[0126] Optionally, the method may further include: determining the correlation information of multiple target scenario elements corresponding to each test function according to the number of occurrences and / or duration of the scenario elements corresponding to each test function in multiple test scenarios. An appropriate correlation information acquisition method can be formulated for different test functions, which is fully compatible with different test functions.

[0127] In actual use, after determining the target test function, the scenario element corresponding to the target test function can be quickly obtained according to the correspondence between the test function and the scenario element, thereby improving the acquisition efficiency of the target scenario element.

[0128] This application obtains multiple test scenarios and the correlation information of multiple target scenario elements corresponding to the target test function, performs correlation analysis processing on any two of the multiple target scenario elements, and obtains a target correlation matrix, which can well reflect the correlation between the target test function and the target scenario element; obtaining the weight of each target scenario element according to the target correlation matrix, and based on the weights of the multiple target scenario elements and the correlation information of the multiple target scenario elements, determining the target test scenario from the multiple test scenarios can accurately and efficiently screen out invalid scenarios. Testing the target test function of the vehicle based on the target test scenario can avoid the waste of test resources caused by testing the target test function of the vehicle using invalid scenarios and the problem of inaccurate test results caused by testing with invalid scenarios, thereby improving the accuracy and overall test efficiency of autonomous driving testing.

[0129] On the other hand, this application also provides an embodiment of an autonomous driving test device, as Figure 6 shown, the device may include:

[0130] An information acquisition module 61, configured to acquire a plurality of test scenarios and correlation information of a plurality of target scenario elements corresponding to a target test function; the correlation information characterizes the degree of correlation between each target scenario element and the target test function;

[0131] A matrix determination module 62, configured to perform correlation analysis processing on any two of the plurality of target scenario elements to obtain a target correlation matrix; each correlation information in the target correlation matrix refers to the comparison result of the correlation degrees of the two target scenario elements corresponding to each correlation information with the target test function;

[0132] A weight determination module 63, configured to obtain the weight of each target scenario element according to the target correlation matrix;

[0133] A target test scenario determination module 64, configured to determine a target test scenario from the plurality of test scenarios based on the weights of the plurality of target scenario elements and the correlation information of the plurality of target scenario elements;

[0134] A test module 65, configured to test the target test function of the vehicle based on the target test scenario.

[0135] In this application, by acquiring a plurality of test scenarios and correlation information of a plurality of target scenario elements corresponding to a target test function, performing correlation analysis processing on any two of the plurality of target scenario elements to obtain a target correlation matrix, the correlation degree between the target test function and the target scenario elements can be well reflected; obtaining the weight of each target scenario element according to the target correlation matrix, and determining a target test scenario from the plurality of test scenarios based on the weights of the plurality of target scenario elements and the correlation information of the plurality of target scenario elements can accurately and efficiently screen out invalid scenarios, and testing the target test function of the vehicle based on the target test scenario can avoid the waste of test resources caused by testing the target test function of the vehicle using invalid scenarios and the problem of inaccurate test results caused by testing with invalid scenarios, thereby improving the accuracy of autonomous driving testing and the overall test efficiency.

[0136] In a possible implementation manner, the matrix determination module may include:

[0137] A first matrix determination unit, configured to perform correlation analysis processing on any two of the plurality of target scenario elements to obtain a first correlation matrix;

[0138] A consistency index value determination unit, configured to determine a consistency index value according to the eigenvalues of the first correlation matrix;

[0139] A random value determination unit, configured to generate a random consistency index value based on the number of the multiple target scenario elements and the target range;

[0140] A target correlation matrix determination unit, configured to, when the consistency index value and the random consistency index value do not meet a preset condition, re - perform an association analysis process on any two of the multiple target scenario elements to obtain a second correlation matrix, and until the consistency index value and the random consistency index value meet the preset condition, use the correlation matrix that meets the preset condition as the target correlation matrix.

[0141] In a possible implementation manner, the target test scenario determination module includes:

[0142] A valid value determination unit, configured to determine valid values of the multiple test scenarios based on the weights of the multiple target scenario elements and the correlation information of the multiple target scenario elements;

[0143] A sorting unit, configured to sort the multiple test scenarios according to the valid values of the multiple test scenarios;

[0144] A target test scenario determination unit, configured to use a preset number of test scenarios with higher rankings as the target test scenarios.

[0145] In a possible implementation manner, the weight determination module includes:

[0146] A target matrix determination unit, configured to perform a normalization process on the target correlation matrix to obtain a target matrix;

[0147] A weight determination unit, configured to perform row - adding and averaging processes on the target matrix to obtain the weight of each target scenario element.

[0148] In a possible implementation manner, the device further includes:

[0149] A correlation function generation module, configured to pre - generate a correlation function for each test function, and the correlation function represents that each test function is related to the corresponding scenario element;

[0150] A scenario element determination module, configured to determine the scenario element corresponding to each test function according to the correlation function of each test function.

[0151] In a possible implementation manner, the device further includes:

[0152] A relevance information determination module, configured to determine the relevance information of multiple target scenario elements corresponding to each test function according to the number of occurrences and / or duration of the scenario elements corresponding to each test function in the multiple test scenarios.

[0153] In a possible implementation manner, the device further includes:

[0154] A quantity determination module, configured to determine the quantity of the multiple target scenario elements;

[0155] A target range setting module, configured to set the target range according to the quantity of the multiple target scenario elements.

[0156] Figure 7 It is a block diagram of an electronic device for autonomous driving testing shown according to an exemplary embodiment. The electronic device can be a server or an interrupt, and its internal structure diagram can be as Figure 7 shown. The electronic device includes a processor, a memory, a network interface, a display screen, and an input device connected through a system bus. Among them, the processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic 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 network interface of the electronic device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements an autonomous driving test method. The display screen of the electronic device can be a liquid crystal display screen or an electronic ink display screen. The input device of the electronic device can be a touch layer covered on the display screen, or a button, a trackball, or a touchpad provided on the housing of the electronic device, or an external keyboard, a touchpad, or a mouse, etc. The electronic device can be connected to a test bench, and this application does not make any limitations in this regard.

[0157] Those skilled in the art can understand that Figure 7 the structure shown in

[0158] is only a block diagram of a part of the structure related to the solution of the present disclosure, and does not constitute a limitation on the electronic device to which the solution of the present disclosure is applied. The specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0159] The present application further provides a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device can implement the testing method in any of the above embodiments.

[0160] The present application further provides a computer program product, including a computer program / instructions. When the computer program / instructions are executed by a processor, the testing method in any of the above embodiments is implemented.

[0161] In the specification provided herein, a large number of specific details are set forth. However, it can be understood that the embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures, and technologies have not been shown in detail so as not to obscure the understanding of this specification.

[0162] Similarly, it should be understood that, in order to streamline the present disclosure and assist in understanding one or more of the various inventive aspects, in the foregoing description of the exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together in a single embodiment, figure, or description thereof. However, the disclosed method should not be construed as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the claims of the present invention, the inventive aspects lie in less than all of the features of the single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into the detailed description, with each claim standing on its own as a separate embodiment of the present invention.

[0163] Those skilled in the art can understand that the modules in the devices in the embodiments can be adaptively changed and disposed in one or more devices different from the embodiments. The modules or units or components in the embodiments can be combined into one module or unit or component, and in addition, they can be divided into multiple sub-modules or sub-units or sub-components. Except that at least some of such features and / or processes or units are mutually exclusive, any combination can be adopted to combine all the features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all the processes or units of any method or device so disclosed. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) can be replaced by an alternative feature providing the same, equivalent, or similar purpose.

[0164] In addition, those skilled in the art can understand that although the embodiments described herein include some features included in other embodiments rather than other features, the combination of features of different embodiments means that it is within the scope of the present invention and forms different embodiments. For example, in the claims of the present invention, any one of the claimed embodiments can be used in any combination.

[0165] The present invention can also be implemented as a device or system program (such as a computer program and a computer program product) for performing part or all of the methods described herein. Such a program implementing the present invention can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, can be provided on a carrier signal, or can be provided in any other form.

[0166] It should be noted that the above embodiments are illustrative of the present invention rather than limiting the present invention, and those skilled in the art can design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in the claim, etc. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention can be implemented by means of hardware including several different elements and by means of a suitably programmed computer. In the unit claims listing several systems, several of these systems can be embodied by the same item of hardware. The use of the words first, second, and third, etc. does not denote any order and these words can be interpreted as names.

Claims

1. An autonomous driving test method, characterized in that, The method includes: Obtaining a plurality of test scenarios and correlation information of a plurality of target scenario elements corresponding to a target test function; the correlation information characterizes the degree of correlation between each target scenario element and the target test function; Performing correlation analysis processing on any two of the plurality of target scenario elements to obtain a first correlation degree matrix; Determining a consistency index value according to the eigenvalues of the first correlation degree matrix; Generating a random consistency index value based on the number and target range of the plurality of target scenario elements; When the consistency index value and the random consistency index value do not meet the preset conditions, re-performing correlation analysis processing on any two of the plurality of target scenario elements to obtain a second correlation degree matrix, until the consistency index value and the random consistency index value meet the preset conditions, and taking the correlation degree matrix that meets the preset conditions as the target correlation degree matrix; each correlation degree information in the target correlation degree matrix refers to the comparison result of the correlation degrees of the two target scenario elements corresponding to each correlation degree information with the target test function, and each correlation degree information in the target correlation degree matrix is an association value within the target range; Obtaining the weight of each target scenario element according to the target correlation degree matrix; Determining a target test scenario from the plurality of test scenarios based on the weights of the plurality of target scenario elements and the correlation information of the plurality of target scenario elements, the target test scenario is determined by an effective value, and the effective value is determined based on the weights of the plurality of target scenario elements and the correlation information of the plurality of target scenario elements; Testing the target test function of the vehicle based on the target test scenario.

2. The method according to claim 1, wherein The determining a target test scenario from the plurality of test scenarios based on the weights of the plurality of target scenario elements and the correlation information of the plurality of target scenario elements includes: Determining the effective values of the plurality of test scenarios based on the weights of the plurality of target scenario elements and the correlation information of the plurality of target scenario elements; Sorting the plurality of test scenarios according to the effective values of the plurality of test scenarios; taking a preset number of test scenarios with a higher ranking as the target test scenarios.

3. The method according to claim 1, wherein The obtaining the weight of each target scenario element according to the target correlation degree matrix includes: Normalizing the target correlation degree matrix to obtain a target matrix; Performing row addition and averaging processing on the target matrix to obtain the weight of each target scenario element.

4. The method according to claim 1, characterized in that Before obtaining the plurality of test scenarios and the correlation information of the plurality of target scenario elements corresponding to the target test function, the method further includes: Pre-generating a correlation function for each test function, the correlation function characterizing that each test function is related to the corresponding scenario element; Determining the scenario elements corresponding to each test function according to the correlation function of each test function.

5. The method according to claim 4, characterized in that After pre-generating the correlation function for each test function, the method further includes: Determining the correlation information of the plurality of target scenario elements corresponding to each test function according to the number of times and / or duration of occurrence of the scenario elements corresponding to each test function in the plurality of test scenarios.

6. The method according to claim 1, wherein The method further includes: determining the number of the multiple target scenario elements; setting the target range according to the number of the multiple target scenario elements.

7. An automatic driving test device, characterized in that, The apparatus includes: an information acquisition module, configured to acquire a plurality of test scenarios and correlation information of a plurality of target scenario elements corresponding to a target test function; the correlation information characterizes the degree of correlation between each target scenario element and the target test function; a first matrix determination module, configured to perform correlation analysis processing on any two of the multiple target scenario elements to obtain a first correlation degree matrix; a consistency index value determination module, configured to determine a consistency index value according to the eigenvalues of the first correlation degree matrix; a random value determination module, configured to generate a random consistency index value based on the number of the multiple target scenario elements and the target range; a target correlation degree matrix determination module, configured to, when the consistency index value and the random consistency index value do not meet a preset condition, re-perform correlation analysis processing on any two of the multiple target scenario elements to obtain a second correlation degree matrix, and until the consistency index value and the random consistency index value meet the preset condition, use the correlation degree matrix that meets the preset condition as the target correlation degree matrix; each correlation degree information in the target correlation degree matrix refers to a comparison result of the correlation degrees of the two target scenario elements corresponding to each correlation degree information with the target test function, and each correlation degree information in the target correlation degree matrix is a correlation value within the target range; a weight determination module, configured to obtain the weight of each target scenario element according to the target correlation degree matrix; a target test scenario determination module, configured to determine a target test scenario from the multiple test scenarios based on the weights of the multiple target scenario elements and the correlation information of the multiple target scenario elements, the target test scenario is determined by an effective value, and the effective value is determined based on the weights of the multiple target scenario elements and the correlation information of the multiple target scenario elements; a test module, configured to test the target test function of the vehicle based on the target test scenario.

8. An electronic device, characterized in that, including: a processor; a memory for storing executable instructions of the processor; wherein, the processor is configured to execute the instructions to implement the autonomous driving test method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device is enabled to execute the autonomous driving test method according to any one of claims 1 to 6.

Images