Method, apparatus, device and medium for reducing the number of simulation test scenarios

Abstract

The application discloses a method, device, equipment and medium for reducing the number of simulation test scenes, which comprises the following steps: obtaining a pre-constructed vehicle model, a sensor model, a control model, and sampling values of each scene variable in an automatic driving system; obtaining a simulation result corresponding to a scene analysis index of the automatic driving system; constructing a functional relationship between the scene variable and the scene analysis index according to the simulation result and the sampling values of each scene variable; determining the sensitivity of each scene variable through the functional relationship, so as to analyze the influence of each scene variable on the automatic driving system; and determining a specific scene generation strategy of each scene variable based on the sensitivity, so as to determine the strategy of each scene variable when generating a specific scene according to the influence of each scene variable on the automatic driving system. The number of specific scenes that need to be tested by the automatic driving system is reduced, the risk of scene explosion is greatly reduced, the test process is accelerated, and the simulation test efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving testing technology, and in particular to a method, apparatus, equipment and medium for reducing the number of simulation test scenarios. Background Technology

[0002] Autonomous driving simulation testing uses mathematical modeling to digitally recreate autonomous driving application scenarios, establishing system models that closely resemble the real world. This allows for testing and verification of autonomous driving systems and algorithms directly through software simulation without the need for actual vehicles. With the development of autonomous driving technology, more and more vehicles equipped with autonomous driving assistance systems are beginning to operate on the road. Before autonomous vehicles can achieve commercial operation, rigorous testing of the autonomous driving system is essential. Current autonomous driving testing methods mainly include simulation testing, real-vehicle track experiments, and real-vehicle road testing. Compared to traditional real-vehicle testing, simulation testing has advantages such as high scenario coverage, safe testing process, and high testing efficiency. Therefore, it plays an increasingly important role in autonomous driving testing, and simulation scenarios are a core component of simulation testing.

[0003] In the PEGASUS analysis framework, simulation test scenarios are hierarchically categorized into functional scenarios, logical scenarios, and specific scenarios. Functional scenarios integrate information about the road, the vehicle, traffic participants, and itself, describing the scenario in textual form. Logical scenarios variate the information contained in the functional scenarios and assign corresponding parameter space ranges. Specific scenarios select specific parameters based on the parameter ranges of the logical scenarios to construct test cases. For a particular autonomous driving function, a large number of simulation tests (specific scenarios) are typically conducted to verify its safety. Assuming each logical scenario has 8 scenario parameters / variables, and each parameter / variable has 10 discrete dimensions, then the number of specific scenarios to be tested is 10. 8 As the number of parameters in a logical scenario increases, the number of specific scenarios that need to be tested grows exponentially, which can easily lead to a scenario explosion problem. Summary of the Invention

[0004] This invention provides a method, apparatus, equipment, and medium for reducing the number of simulation test scenarios, thereby reducing the number of specific scenarios for testing autonomous driving systems and improving testing efficiency.

[0005] According to one aspect of the present invention, a method for reducing the number of simulation test scenarios is provided, the method comprising:

[0006] Based on the pre-built vehicle model, sensor model, control model corresponding to the autonomous driving system, and sampled values ​​of various scenario variables corresponding to the autonomous driving system, the simulation results corresponding to the scenario analysis indicators of the autonomous driving system are obtained.

[0007] Based on the simulation results and the sampled values ​​of each scenario variable, a functional relationship between each scenario variable and the scenario analysis index is constructed.

[0008] Based on the functional relationship, the sensitivity corresponding to each scene variable is determined, and based on the sensitivity corresponding to each scene variable, the specific scene generation strategy corresponding to each scene variable is determined.

[0009] Optionally, determining the sensitivity corresponding to each of the scenario variables based on the functional relationship includes:

[0010] For each scenario variable, the first-order sensitivity corresponding to the scenario variable is determined based on the contribution of the scenario variable to the variance of the functional relationship.

[0011] Optionally, determining the sensitivity corresponding to each of the scenario variables based on the functional relationship further includes:

[0012] For each scenario variable, the total sensitivity corresponding to the scenario variable is determined based on the contribution of the scenario variable to the variance of the functional relationship, and the contribution of the interaction between the scenario variable and other scenario variables other than the scenario variable to the variance.

[0013] Optionally, determining the specific scene generation strategy corresponding to each scene variable based on the sensitivity of each scene variable includes:

[0014] For each of the aforementioned scenario variables, a sensitivity level corresponding to the scenario variable is determined based on the first sensitivity and / or the total sensitivity corresponding to the scenario variable.

[0015] Based on the sensitivity level and the value range of the scene variable, the specific scene generation strategy corresponding to the scene variable is determined.

[0016] Optionally, determining the specific scene generation strategy corresponding to the scene variable based on the sensitivity level and the value range corresponding to the scene variable includes:

[0017] Based on the sensitivity level and the value range corresponding to the scene variable, determine the value step size and the number of values ​​corresponding to the scene variable; or,

[0018] Based on the sensitivity level and the value range corresponding to the scene variable, a fixed value corresponding to the scene variable is determined.

[0019] Optionally, the step of obtaining simulation results corresponding to the scene analysis indicators of the autonomous driving system based on the pre-built vehicle model, sensor model, control model corresponding to the autonomous driving system, and sampled values ​​of various scene variables corresponding to the autonomous driving system includes:

[0020] Obtain pre-built vehicle models, sensor models, and control models corresponding to the autonomous driving system;

[0021] Obtain the scene variables, scene analysis indicators, and simulation scene templates corresponding to the autonomous driving system;

[0022] The scenario variables are sampled based on a preset sampling strategy. Based on the sampling results, the simulation scenario template, the vehicle model, the sensor model, and the control model, simulation results corresponding to the scenario analysis indicators are generated.

[0023] Optionally, generating simulation results corresponding to the scene analysis indicators based on the sampling results, the simulation scene template, the vehicle model, the sensor model, and the control model includes:

[0024] A scene model is generated based on the sampling results and the simulation scene template.

[0025] Based on the target environment information transmitted from the scene model to the sensor model, the target information transmitted from the sensor model to the control model, the control information transmitted from the control model to the vehicle model, the motion state information transmitted from the vehicle model to the control model, and the position information transmitted from the vehicle model to the scene model, the simulation results corresponding to the scene analysis indicators are determined.

[0026] According to another aspect of the present invention, an apparatus for reducing the number of simulation test scenarios is provided, the apparatus comprising:

[0027] The simulation module is used to obtain the simulation results corresponding to the scene analysis indicators of the autonomous driving system based on the pre-built vehicle model, sensor model, control model corresponding to the autonomous driving system, and sampled values ​​of various scene variables corresponding to the autonomous driving system.

[0028] The function construction module is used to construct the functional relationship between each of the scenario variables and the scenario analysis index based on the simulation results and the sampled values ​​of each of the scenario variables;

[0029] The sensitivity analysis module is used to determine the sensitivity of each scene variable based on the functional relationship, and to determine the specific scene generation strategy corresponding to each scene variable based on the sensitivity of each scene variable.

[0030] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0031] At least one processor; and

[0032] A memory communicatively connected to the at least one processor; wherein,

[0033] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the method for reducing the number of simulation test scenarios as described in any embodiment of the present invention.

[0034] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the method for reducing the number of simulation test scenarios as described in any embodiment of the present invention.

[0035] The technical solution of this invention obtains a pre-built vehicle model, sensor model, control model corresponding to the autonomous driving system, and sampled values ​​of various scene variables corresponding to the autonomous driving system. It then obtains simulation results corresponding to the scene analysis indicators of the autonomous driving system. Based on the simulation results and the sampled values ​​of each scene variable, it constructs a functional relationship between the scene variables and the scene analysis indicators. The sensitivity of each scene variable is determined through this functional relationship to analyze the impact of each scene variable on the autonomous driving system. Based on the sensitivity, a specific scene generation strategy for each scene variable is determined. This reduces the number of specific scenes that need to be tested for the autonomous driving system, thereby significantly reducing the risk of scene explosion, accelerating the testing process, and improving simulation testing efficiency.

[0036] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a flowchart illustrating a method for reducing the number of simulation test scenarios provided in Embodiment 1 of the present invention;

[0039] Figure 2 This is a flowchart illustrating a method for reducing the number of simulation test scenarios provided in Embodiment 2 of the present invention;

[0040] Figure 3A This is a flowchart illustrating a method for reducing the number of simulation test scenarios provided in Embodiment 3 of the present invention;

[0041] Figure 3B This is a schematic diagram of a method for reducing the number of simulation test scenarios provided in Embodiment 3 of the present invention;

[0042] Figure 4 This is a schematic diagram of the structure of a device for reducing the number of simulation test scenarios provided in Embodiment 4 of the present invention;

[0043] Figure 5 This is a schematic diagram of the structure of an electronic device provided in Embodiment 5 of the present invention. Detailed Implementation

[0044] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0045] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0046] Example 1

[0047] Figure 1This is a flowchart illustrating a method for reducing the number of simulation test scenarios provided in Embodiment 1 of the present invention. This embodiment is applicable to autonomous driving systems such as automatic emergency braking systems, hill start assist systems, anti-lock braking systems, traffic jam following systems, highway valet driving systems, and automatic parking systems. It analyzes the specific scenario generation strategies corresponding to each scenario variable in the autonomous driving system, and generates specific scenarios for simulation testing of the autonomous driving system based on these specific scenario generation strategies. This method can be executed by a device for reducing the number of simulation test scenarios. This device can be implemented in hardware and / or software, and can be configured in electronic devices such as mobile phones, computers, and tablets. Figure 1 As shown, the method includes:

[0048] S110. Based on the pre-built vehicle model, sensor model, control model corresponding to the autonomous driving system, and sampled values ​​of various scenario variables corresponding to the autonomous driving system, obtain the simulation results corresponding to the scenario analysis indicators of the autonomous driving system.

[0049] The vehicle model can be a dynamic model simulating the motion characteristics of a vehicle in actual operation. Specifically, it can be constructed based on preset vehicle parameters (such as vehicle dimensions, wheelbase, front and rear track width, minimum ground clearance, minimum turning diameter, vehicle mass, sprung mass, drag coefficient, etc.), aerodynamic parameters (such as air density, frontal area, etc.), and transmission parameters. Specifically, the vehicle model can be built in autonomous driving simulation software.

[0050] The sensor model can be a model simulating sensors installed in a vehicle during actual driving, including but not limited to lidar models, camera models, millimeter-wave radar models, etc. The sensor model constructed in this embodiment can include not only the types of sensors simulated for installation, but also the sensor type (physical level, signal level, truth level), sensor installation location, and basic parameter configuration information. Specifically, the sensor model can be built in autonomous driving simulation software.

[0051] Autonomous driving systems can be control systems used to assist vehicle driving, such as Advanced / Automatic Emergency Braking (AEB), hill start assist, anti-lock braking system, traffic jam assist, highway chauffeur service, and automatic parking systems. In this embodiment, the control model corresponding to the autonomous driving system can control the vehicle according to the corresponding autonomous driving control algorithm. The control model can be integrated with other models (vehicle model, sensor model, etc.) in the Simulink environment.

[0052] In this embodiment, the scenario variables corresponding to the autonomous driving system can be the input variables of the test scenarios that need to be tested in the autonomous driving system; the scenario variables can be different under different test scenarios. For example, for the AEB autonomous driving system, it can include test scenarios such as the Car-to-Car Rear Moving (CCRm) test and the Car-to-Car Rear Braking (CCRb) test. In the CCRm test scenario, the scenario variables can include variables such as the speed of the vehicle itself, the speed of the target vehicle, and the bias rate.

[0053] Specifically, in this embodiment, we can first define the scene variables corresponding to the autonomous driving system. The definition of scene variables includes determining their types and value ranges. Further, for each scene variable's type and value range, we can randomly sample each scene variable to obtain sampled values. The sampling strategy and quantity can be determined based on the test scenarios of the autonomous driving system; typically, a simple random sampling strategy with a sampling quantity of 10 is used. 2 ~10 3 The quantity is sufficient to meet the needs.

[0054] Furthermore, after obtaining the sampled values ​​of each scenario variable, the sampled values ​​can be transmitted to the vehicle model, sensor model, or control model. Then, based on the data interaction between the vehicle model, sensor model, and control model, a simulation test of the autonomous driving system can be formed, and the simulation results corresponding to the scenario analysis indicators can be obtained.

[0055] The scenario analysis metrics can be the output metrics of the test scenarios that need to be tested in the test scenarios of the autonomous driving system. In this embodiment, scenario analysis metrics can be defined at the same time as scenario variables.

[0056] For example, in the CCRm test scenario, the scenario analysis metric could be the minimum relative distance between the two vehicles. A positive value for the minimum relative distance indicates that no collision has occurred, while a negative value indicates that a collision has occurred.

[0057] It should be noted that the purpose of using the minimum relative distance between the two vehicles as the scene analysis indicator is that the CCRm test scenario is concerned with whether a collision occurs, and whether a collision occurs can be represented by 0 or 1. However, this single value method of either 0 or 1 is not conducive to building a functional relationship between scene variables and scene analysis indicators. Therefore, this embodiment can use the minimum relative distance between the two vehicles as the scene analysis indicator for the CCRm test scenario.

[0058] It is understood that the simulation results corresponding to the scenario analysis indicators in this embodiment can be directly output from the simulation test, or they can be obtained by analyzing the results output from the simulation test.

[0059] S120. Based on the simulation results and the sampled values ​​of each scenario variable, construct the functional relationship between each scenario variable and the scenario analysis index.

[0060] In this embodiment, after obtaining the simulation results of the scene analysis indicators, the functional relationship between each scene variable and the scene analysis indicators can be constructed based on the sampled values ​​of the scene variables and the simulation results of the scene analysis indicators. That is, the functional relationship between the input variables and the output variables can be constructed through the input values ​​and the output values.

[0061] Specifically, the functional relationship between scenario variables and scenario analysis indicators can be established using multinomial regression, Kriging function, radial basis function, or neural network methods.

[0062] S130. Determine the sensitivity corresponding to each scene variable based on the functional relationship, and determine the specific scene generation strategy corresponding to each scene variable based on the sensitivity corresponding to each scene variable.

[0063] Specifically, after constructing the functional relationships between each scenario variable and the scenario analysis indicators, the sensitivity of each scenario variable can be further determined based on these relationships. The sensitivity of a scenario variable can be information describing the magnitude of the impact of the uncertainty of the scenario variable on the specific output value of the scenario analysis indicator. Higher sensitivity indicates a greater impact of the uncertainty of the scenario variable on the specific output value of the scenario analysis indicator.

[0064] Optionally, for each scenario variable, different values ​​can be assigned to the scenario variable in the functional relationship, and the impact of the scenario variable on the output of the functional relationship, i.e., the sensitivity, can be analyzed based on the calculated value of the output of the functional relationship and the assigned value. In this embodiment, variance decomposition can also be used to analyze the sensitivity corresponding to each scenario variable in the functional relationship.

[0065] Furthermore, based on the sensitivity of each scenario variable, a specific scenario generation strategy can be determined for each scenario variable. This specific scenario generation strategy can be a strategy for generalizing scenario variables to obtain specific scenarios. For example, for the scenario variable of environmental information, if analysis shows its sensitivity is low, the specific scenario generation strategy for this scenario variable could be: when generalizing specific scenarios, use a larger dispersion to obtain fewer values ​​(e.g., 5). For the scenario variable of road length, if analysis shows its sensitivity is extremely low, the specific scenario generation strategy for this scenario variable could be: when generalizing specific scenarios, use a fixed value.

[0066] For example, in the prior art, if the number of scenario variables in a certain test scenario is 8, and each scenario variable needs 10 values ​​(i.e., 10 discrete dimensions) when generalizing to specific scenarios, then the number of generated specific scenarios is 10. 8 The method provided in this application shows that among the eight scene variables, five have relatively low sensitivity. Therefore, the number of values ​​for these less sensitive scene variables can be reduced, for example, to three. This results in three specific scenes being generated. 5 ×10 3 This greatly reduces the number of specific scenarios obtained through generalization.

[0067] In one specific implementation, sensitivity may include primary sensitivity and / or total sensitivity; wherein, primary sensitivity may be the contribution of the uncertainty of the scenario variable to the variance of the output of the functional relationship; total sensitivity may include the contribution of the uncertainty of the scenario variable to the variance of the output of the functional relationship, and the contribution of the interaction between the scenario variable and other scenario variables to the variance of the output of the functional relationship.

[0068] If sensitivity includes at least one of primary sensitivity and total sensitivity, the specific scene generation strategy corresponding to the scene variable can be determined based on the primary sensitivity and / or total sensitivity of the scene variable.

[0069] Optionally, determining the specific scene generation strategy corresponding to each scene variable based on the sensitivity of each scene variable includes: for each scene variable, determining the sensitivity level corresponding to the scene variable based on the primary sensitivity and / or the total sensitivity corresponding to the scene variable; and determining the specific scene generation strategy corresponding to the scene variable based on the sensitivity level and the value range corresponding to the scene variable.

[0070] If the sensitivity level is determined based on primary sensitivity and total sensitivity, then corresponding weights can be set for primary sensitivity and total sensitivity respectively, and the sensitivity level can be determined based on primary sensitivity, the weight corresponding to primary sensitivity, total sensitivity, and the weight corresponding to total sensitivity.

[0071] Furthermore, based on the sensitivity level and the value range of the scene variables, the specific scene generation strategy corresponding to the scene variables is determined; wherein, the specific scene generation strategy may include the value step size, the number of values, or fixed values.

[0072] For example, determining the specific scene generation strategy corresponding to the scene variable based on the sensitivity level and the value range corresponding to the scene variable includes: determining the value step size and number of values ​​corresponding to the scene variable based on the sensitivity level and the value range corresponding to the scene variable; or, determining the fixed value corresponding to the scene variable based on the sensitivity level and the value range corresponding to the scene variable.

[0073] Specifically, a higher sensitivity level can result in a smaller step size and a larger number of values. A lower sensitivity level can result in a larger step size and a smaller number of values. Alternatively, when the sensitivity level is below a set threshold, a fixed value method can be used to set the scene variable to a fixed value, and the number of fixed values ​​can be one or more.

[0074] Of course, you can also determine only the step size of the values, or only the number of values. By determining the step size or the number of values, you can use an equal-interval sampling method within the range of values ​​to achieve generalization for specific scenarios.

[0075] By using the above methods, the sensitivity of scene variables can be accurately determined, thereby enabling the accurate determination of specific scene generation strategies for each scene variable. This effectively reduces the number of specific scenes during autonomous driving system simulation testing, lowers the risk of scene explosion, and improves testing efficiency.

[0076] The technical solution of this embodiment obtains a pre-built vehicle model, sensor model, control model corresponding to the autonomous driving system, and sampled values ​​of various scene variables corresponding to the autonomous driving system. It then obtains the simulation results corresponding to the scene analysis indicators of the autonomous driving system. Based on the simulation results and the sampled values ​​of each scene variable, it constructs a functional relationship between the scene variables and the scene analysis indicators. The sensitivity of each scene variable is determined through the functional relationship to analyze the magnitude of the impact of each scene variable on the autonomous driving system. Based on the sensitivity, it determines the specific scene generation strategy for each scene variable. This reduces the number of specific scenes that the autonomous driving system needs to test, thereby greatly reducing the risk of scene explosion, accelerating the testing process, and improving the efficiency of simulation testing.

[0077] It should be noted that the method for reducing the number of simulation test scenarios provided in this embodiment is not limited to reducing the number of simulation test scenarios (i.e., the parameter dimensions of scenario variables) for autonomous driving systems, but can also be applied to reducing the number of simulation test scenarios for other control systems. Specifically, after constructing and simulating various models of the control system, a functional relationship between input and output can be constructed using sampled values ​​of input scenario variables and output scenario analysis indicators. This allows for the analysis of the sensitivity of each input scenario variable, and the determination of the specific scenario generation strategy corresponding to each variable based on the sensitivity.

[0078] Example 2

[0079] Figure 2 This is a flowchart illustrating a method for reducing the number of simulation test scenarios according to Embodiment 2 of the present invention. Based on the above embodiments, this embodiment further explains how to determine the sensitivity corresponding to each scenario variable according to the functional relationship. For example... Figure 2 As shown, the method includes:

[0080] S210. Based on the pre-built vehicle model, sensor model, control model corresponding to the autonomous driving system, and sampled values ​​of various scenario variables corresponding to the autonomous driving system, obtain the simulation results corresponding to the scenario analysis indicators of the autonomous driving system.

[0081] S220. Based on the simulation results and the sampled values ​​of each scenario variable, construct the functional relationship between each scenario variable and the scenario analysis index.

[0082] In this embodiment, the impact of each input on the output in the functional relationship can be analyzed using variance decomposition, i.e., the sensitivity of each scene variable. Assume that the functional relationship between each scene variable and the scene analysis index is constructed as: y = f(x), x = (x1, x2, ... x...). n ), where x i It follows a uniform distribution in [0,1], and f 2 Since (x) is integrable, the functional relationship between each scenario variable and the scenario analysis index can be decomposed into functions of individual attributes and combinations of attributes, as shown in the following equation:

[0083]

[0084] Where f0 is a constant, and each decomposition term satisfy:

[0085]

[0086] Therefore, the total variance of a functional relationship can also be decomposed into the effects of individual parameters and the combined effects of each parameter:

[0087]

[0088] Where D represents the total variance of the functional relationship, D i D represents the variance produced by a single scene variable. i…s This represents the variance resulting from the interaction between the scene variables from the i-th scene variable to the s-th scene variable.

[0089] S230. For each of the scenario variables, determine the first-order sensitivity corresponding to the scenario variable based on the contribution of the scenario variable to the variance of the functional relationship.

[0090] In this context, first-order sensitivity can be considered the contribution of a single scenario variable to the variance of the functional relationship. The ratio of the variance generated by the scenario variable to the total variance of the functional relationship can be used as the first-order sensitivity, as shown in the following formula:

[0091] S i =D i / D;

[0092] Among them, S i D represents the first-order sensitivity corresponding to the i-th scene variable. i Let represent the variance produced by the individual effect of the i-th scenario variable, and D represent the total variance of the functional relationship. D can be calculated using the following formula:

[0093]

[0094] In another implementation, determining the first-order sensitivity corresponding to the scenario variable based on its contribution to the variance of the functional relationship can also satisfy the following formula:

[0095]

[0096] in, The output in the functional relationship is determined solely by the scene variable x. i The resulting variance, V(Y), represents the unconditional variance of the output in the functional relationship.

[0097] In this embodiment, the total sensitivity corresponding to each scenario variable can also be calculated. For example, optionally, determining the sensitivity corresponding to each scenario variable based on the functional relationship further includes: for each scenario variable, determining the total sensitivity corresponding to the scenario variable based on the contribution of the scenario variable to the variance of the functional relationship, and the contribution of the interaction between the scenario variable and other scenario variables other than the scenario variable to the variance.

[0098] Specifically, the sensitivity of individual scene variables and the interactions between scene variables in a functional relationship can be defined to satisfy the following formula:

[0099]

[0100] Among them, S i S represents the first-order sensitivity corresponding to the i-th scene variable. ij S represents secondary sensitivity. 1,2,…n This represents the nth order of sensitivity. For second-order sensitivity, third-order sensitivity, etc., taking nth-order sensitivity as an example, it can be calculated using the following formula:

[0101]

[0102] Therefore, for the i-th scenario variable x i The corresponding total sensitivity can be defined as:

[0103] S Tj =ΣS i ;

[0104] Among them, S Tj Represents the scene variable x i The total sensitivity, expressed by the formula, is for all variables involving S. i The summation of terms. That is, the total sensitivity corresponding to the scenario variable, including the contribution of a single scenario variable to the variance of the functional relationship, and the contribution of the interaction between the scenario variable and other scenario variables to the variance.

[0105] In another implementation, the total sensitivity of the scenario variables can also be calculated based on the following formula:

[0106]

[0107] in, The output in the functional relationship is determined by factors other than the scene variable x. i Variance caused by other scenario variables besides those in the scenario.

[0108] By using the above methods, the sensitivity of all scene variables can be accurately determined. Then, by combining individual scene variables and the interactions between scene variables, the impact of scene variables on scene analysis indicators can be analyzed, thus achieving accurate analysis of the sensitivity of each scene variable and improving the accuracy of the determined specific scene generation strategy.

[0109] S240. Based on the sensitivity of each of the aforementioned scene variables, determine the specific scene generation strategy corresponding to each of the aforementioned scene variables.

[0110] Specifically, the specific scene generation strategy for each scene variable can be determined based on its primary sensitivity. Alternatively, it can be determined based on all the sensitivities of the scene variables. Or, it can be determined based on both the primary and all the sensitivities of the scene variables.

[0111] In this embodiment, functional relationships can be established and the sensitivity of various scenario variables can be analyzed in mathematical analysis software. Specifically, a data interaction channel can be constructed between the autonomous driving simulation software and the mathematical analysis software. Through this channel, the extracted scenario variables and simulation results corresponding to scenario analysis indicators from the autonomous driving simulation software are transmitted to the mathematical analysis software, enabling the software to construct functional relationships and analyze the sensitivity of each scenario variable. By combining autonomous driving simulation testing with mathematical analysis, the dimensionality of scenario variables in the logical scenario is reduced. After expressing the simulation test scenario mathematically, a functional relationship between scenario variables and simulation output is established using mathematical methods, and sensitivity analysis is performed to reduce the number of specific scenarios, thereby mitigating the risk of scenario explosion.

[0112] The technical solution of this embodiment determines the first-order sensitivity of each scenario variable by measuring the contribution of scenario variables to the variance of the functional relationship. This achieves accurate determination of the first-order sensitivity of each scenario variable, and thus accurate analysis of the sensitivity of each scenario variable. Furthermore, based on the first-order sensitivity of the scenario variable, a specific scenario generation strategy is determined for each scenario variable to reduce the dimensionality of each scenario variable in the logical scenario, thereby reducing the number of generalized specific scenarios.

[0113] Example 3

[0114] Figure 3A This is a flowchart illustrating a method for reducing the number of simulation test scenarios provided in Embodiment 3 of the present invention. Based on the above embodiments, this embodiment further describes the simulation process of an autonomous driving system. Figure 3A As shown, the method includes:

[0115] S310: Obtain the pre-built vehicle model, sensor model, and control model corresponding to the autonomous driving system.

[0116] S320. Obtain the scene variables, scene analysis indicators and simulation scene templates corresponding to the autonomous driving system.

[0117] Specifically, for test scenarios in autonomous driving systems, we can define scenario variables and scenario analysis indicators, as well as construct simulation scenario templates.

[0118] The simulation scenario template can be a template describing a test scenario, including settings for various scenario variables and related elements. For example, the construction steps for a simulation scenario template can be as follows:

[0119] Step 1: Construct static scene elements, which include at least one of road information, lane information, and environmental information;

[0120] Step 2: Construct dynamic scene elements, which include at least one of traffic characteristic information, vehicle information, target vehicle information, and other traffic participant information.

[0121] Step 3: Set the simulation duration, trigger conditions, and simulation termination conditions.

[0122] In the above steps, road information can include road geometry (such as road length and width), road surface material, and the number of lanes; lane information can include lane length, lane width, and lane markings. Environmental information can include traffic lights, road markings, traffic signs, roadblocks, fences, and road structures (such as bridges and tunnels). Traffic characteristic information can include vehicle density, vehicle speed, and pedestrian / vehicle distribution; vehicle information can include its own geometric model information, motion parameters, and trajectory; target vehicle information can include its own geometric model information, motion parameters, and trajectory; other traffic participant information can include the motion information and geometric model information of pedestrians, animals, and other participants. Simulation triggering conditions can be conditions that trigger the start of the simulation test, such as the vehicle starting to brake. Simulation termination conditions can be conditions that end the simulation test, such as the moment the vehicle's speed decreases to 0; of course, to ensure sufficient scene simulation results are collected, the simulation termination condition can be set to the vehicle's speed decreasing to 0 after a set delay time.

[0123] S330. The scenario variables are sampled based on a preset sampling strategy. Based on the sampling results, the simulation scenario template, the vehicle model, the sensor model, and the control model, simulation results corresponding to the scenario analysis indicators are generated.

[0124] The preset sampling strategy can be determined based on the type and value range of the scenario variables. Specifically, after sampling the scenario variables, the sampling results can be substituted into the simulation scenario template to generate a scenario model. Then, based on the scenario model, vehicle model, sensor model, and control model, the simulation of the autonomous driving system can be achieved.

[0125] For example, generating the simulation results corresponding to the scene analysis indicators based on the sampling results, the simulation scene template, the vehicle model, the sensor model, and the control model includes: generating a scene model based on the sampling results and the simulation scene template; and determining the simulation results corresponding to the scene analysis indicators based on the target environment information transmitted from the scene model to the sensor model, the target information transmitted from the sensor model to the control model, the control information transmitted from the control model to the vehicle model, the motion state information transmitted from the vehicle model to the control model, and the position information transmitted from the vehicle model to the scene model.

[0126] The target environment information may include lane information, road information, and traffic participant information. Target information may include the position, speed, acceleration, relative distance, relative speed, and relative angle of the target vehicle or object. Control information may include the calculated vehicle motion information for the next moment.

[0127] Specifically, the control model can calculate the vehicle's motion information at the next moment based on the received target information and motion state information, and send it to the vehicle model so that the vehicle model can brake the vehicle according to the information. A closed loop is formed between the control model, vehicle model, scene model, and sensor model to realize autonomous driving simulation.

[0128] S340. Based on the simulation results and the sampled values ​​of each scene variable, construct the functional relationship between each scene variable and the scene analysis index, determine the sensitivity corresponding to each scene variable based on the functional relationship, and determine the specific scene generation strategy corresponding to each scene variable based on the sensitivity corresponding to each scene variable.

[0129] The technical solution of this embodiment acquires the constructed vehicle model, sensor model, control model, simulation scene template, and defined scene variables and scene analysis indicators. It then samples the scene variables according to a preset sampling strategy. Based on the sampling results, simulation scene template, vehicle model, sensor model, and control model, it generates simulation results corresponding to the scene analysis indicators, thereby realizing automatic simulation of the autonomous driving system and providing data support for the sensitivity analysis of various scene variables in the autonomous driving system.

[0130] In this embodiment, an exemplary method for reducing the number of simulation test scenarios can also be demonstrated, the method comprising the following steps:

[0131] Step 1: Establish a data exchange channel between autonomous driving simulation testing software and mathematical analysis software;

[0132] Step 2: Build vehicle and sensor models in the autonomous driving simulation testing software;

[0133] Step 3: Integrate the autonomous driving control algorithm to obtain the control model;

[0134] Specifically, autonomous driving control algorithms can be integrated with other models in the Simulink environment. The vehicle model inputs information such as its speed, acceleration, steering wheel angle, and accelerator and brake pedal openings to the control model; the sensor model inputs information such as the relative distance, relative speed, and relative angle between the target object and the vehicle. Connecting the autonomous driving control algorithm and the vehicle model forms a closed loop.

[0135] Step 4: Define the variables and analysis metrics for each scenario;

[0136] Step 5: Build a simulation scene template;

[0137] Step 6: Sampling of each scenario variable according to its type and value range; generating a scenario model based on the sampling results and simulation scenario template; and obtaining simulation results corresponding to the scenario analysis indicators based on the scenario model, vehicle model, sensor model, and control model.

[0138] Specifically, such as Figure 3B The diagram illustrates the process of reducing the number of simulation test scenarios, where data is transmitted between the scenario model, vehicle model, control model, and sensor model to form a closed loop.

[0139] Step 7: Based on the simulation results corresponding to the scene analysis indicators and the sampled values ​​of each scene variable, construct the functional relationship between each scene variable and the scene analysis indicators, determine the sensitivity of each scene variable based on the functional relationship, and select the corresponding specific scene generation strategy based on the sensitivity.

[0140] Specifically, such as Figure 3B As shown, the autonomous driving simulation testing software sends the simulation results corresponding to the scene analysis indicators and the sampled values ​​of each scene variable to the mathematical analysis software. Further, the mathematical analysis software constructs a relationship based on the simulation results and the sampled values, and performs sensitivity analysis on each scene variable through the functional relationship. Of course, the autonomous driving simulation testing software can also feed back the sensitivity analysis results to the autonomous driving simulation testing software, enabling the autonomous driving simulation testing software to generalize to specific scenarios based on the sensitivity analysis results.

[0141] This method combines autonomous driving simulation testing with sensitivity analysis in mathematics. By mathematically expressing scenario variables and simulation outputs, it analyzes and quantifies the impact of the uncertainty of scenario variables on the simulation output. Based on the sensitivity analysis results, when conducting simulation testing, for scenario variables with low or extremely low sensitivity, a strategy of larger dispersion or fixed values ​​can be adopted during scenario generalization to reduce the number of specific scenarios that need to be tested and improve the efficiency of simulation testing.

[0142] Example 4

[0143] Figure 4 This is a schematic diagram of a device for reducing the number of simulation test scenarios provided in Embodiment 4 of the present invention. Figure 4 As shown, the device includes: a simulation module 410, a function construction module 420, and a sensitivity analysis module 430.

[0144] The simulation module 410 is used to obtain the simulation results corresponding to the scene analysis indicators of the autonomous driving system based on the pre-built vehicle model, sensor model, control model corresponding to the autonomous driving system and sampled values ​​of various scene variables corresponding to the autonomous driving system.

[0145] The function construction module 420 is used to construct the functional relationship between each of the scenario variables and the scenario analysis index based on the simulation results and the sampled values ​​of each of the scenario variables;

[0146] Sensitivity analysis module 430 is used to determine the sensitivity of each scene variable based on the functional relationship, and to determine the specific scene generation strategy corresponding to each scene variable based on the sensitivity of each scene variable.

[0147] Optionally, the sensitivity analysis module 430 includes a sensitivity determination unit, which is used to determine the first-order sensitivity of each scenario variable based on the contribution of the scenario variable to the variance of the functional relationship.

[0148] Optionally, the sensitivity determination unit is further configured to determine, for each scenario variable, the total sensitivity corresponding to the scenario variable based on the contribution of the scenario variable to the variance of the functional relationship, and the contribution of the interaction between the scenario variable and other scenario variables other than the scenario variable to the variance.

[0149] Optionally, the sensitivity analysis module 430 further includes a strategy determination unit, which is used to determine the sensitivity level of each scene variable based on the primary sensitivity and / or the total sensitivity corresponding to the scene variable; and to determine the specific scene generation strategy corresponding to the scene variable based on the sensitivity level and the value range corresponding to the scene variable.

[0150] Optionally, the strategy determination unit is further configured to determine the value step size and number of values ​​corresponding to the scene variable based on the sensitivity level and the value range corresponding to the scene variable; or, based on the sensitivity level and the value range corresponding to the scene variable, determine the fixed value corresponding to the scene variable.

[0151] Optionally, the simulation module 410 includes a model acquisition unit, a variable template definition unit, and an output unit;

[0152] The model acquisition unit is used to acquire pre-built vehicle models, sensor models, and control models corresponding to the autonomous driving system.

[0153] The variable template definition unit is used to obtain the scene variables, scene analysis indicators and simulation scene templates corresponding to the autonomous driving system.

[0154] The output unit is used to sample the scenario variables based on a preset sampling strategy, and generate simulation results corresponding to the scenario analysis indicators based on the sampling results, the simulation scenario template, the vehicle model, the sensor model, and the control model.

[0155] Optionally, the output unit is specifically used to: generate a scene model based on the sampling results and the simulation scene template; and determine the simulation results corresponding to the scene analysis indicators based on the target environment information transmitted from the scene model to the sensor model, the target information transmitted from the sensor model to the control model, the control information transmitted from the control model to the vehicle model, the motion state information transmitted from the vehicle model to the control model, and the position information transmitted from the vehicle model to the scene model.

[0156] In this embodiment, by using a pre-built vehicle model, sensor model, control model corresponding to the autonomous driving system, and sampled values ​​of various scenario variables corresponding to the autonomous driving system, simulation results corresponding to the scenario analysis indicators of the autonomous driving system are obtained. Based on the simulation results and the sampled values ​​of each scenario variable, a functional relationship between the scenario variables and the scenario analysis indicators is constructed. The sensitivity of each scenario variable is determined through the functional relationship to analyze the magnitude of the impact of each scenario variable on the autonomous driving system. Then, based on the sensitivity, the specific scenario generation strategy of each scenario variable is determined. This reduces the number of specific scenarios that the autonomous driving system needs to test, thereby greatly reducing the risk of scenario explosion, accelerating the testing process, and improving simulation testing efficiency.

[0157] The apparatus for reducing the number of simulation test scenarios provided in the embodiments of the present invention can execute the method for reducing the number of simulation test scenarios provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0158] Example 5

[0159] Figure 5 This is a schematic diagram of the structure of an electronic device provided in Embodiment 5 of the present invention. Figure 5 A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0160] like Figure 5As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0161] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0162] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as methods for reducing the number of simulation test scenarios.

[0163] In some embodiments, the method for reducing the number of simulation test scenarios can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method for reducing the number of simulation test scenarios described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the method for reducing the number of simulation test scenarios by any other suitable means (e.g., by means of firmware).

[0164] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0165] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0166] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0167] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0168] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0169] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0170] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0171] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for reducing the number of simulation test scenarios, characterized in that, include: Based on the pre-built vehicle model, sensor model, control model corresponding to the autonomous driving system, and sampled values ​​of various scenario variables corresponding to the autonomous driving system, the simulation results corresponding to the scenario analysis indicators of the autonomous driving system are obtained. Based on the simulation results and the sampled values ​​of each scenario variable, a functional relationship between each scenario variable and the scenario analysis index is constructed. Based on the functional relationship, the sensitivity corresponding to each scene variable is determined, and based on the sensitivity corresponding to each scene variable, the specific scene generation strategy corresponding to each scene variable is determined. The determination of the sensitivity corresponding to each scenario variable based on the functional relationship includes: For each of the aforementioned scenario variables, the first-order sensitivity corresponding to the scenario variable is determined based on the contribution of the scenario variable to the variance of the functional relationship. The determination of the sensitivity corresponding to each scenario variable based on the functional relationship further includes: For each scenario variable, the total sensitivity corresponding to the scenario variable is determined based on the contribution of the scenario variable to the variance of the functional relationship, and the contribution of the interaction between the scenario variable and other scenario variables other than the scenario variable to the variance.

2. The method according to claim 1, characterized in that, The step of determining the specific scene generation strategy corresponding to each scene variable based on the sensitivity of each scene variable includes: For each of the aforementioned scenario variables, a sensitivity level corresponding to the scenario variable is determined based on the first sensitivity and / or the total sensitivity corresponding to the scenario variable. Based on the sensitivity level and the value range of the scene variable, the specific scene generation strategy corresponding to the scene variable is determined.

3. The method according to claim 2, characterized in that, The step of determining the specific scene generation strategy corresponding to the scene variable based on the sensitivity level and the value range of the scene variable includes: Based on the sensitivity level and the value range corresponding to the scene variable, determine the value step size and the number of values ​​corresponding to the scene variable; or, Based on the sensitivity level and the value range corresponding to the scene variable, a fixed value corresponding to the scene variable is determined.

4. The method according to claim 1, characterized in that, The simulation results corresponding to the scene analysis indicators of the autonomous driving system are obtained based on the pre-built vehicle model, sensor model, control model corresponding to the autonomous driving system, and sampled values ​​of various scene variables corresponding to the autonomous driving system, including: Obtain pre-built vehicle models, sensor models, and control models corresponding to the autonomous driving system; Obtain the scene variables, scene analysis indicators, and simulation scene templates corresponding to the autonomous driving system; The scenario variables are sampled based on a preset sampling strategy. Based on the sampling results, the simulation scenario template, the vehicle model, the sensor model, and the control model, simulation results corresponding to the scenario analysis indicators are generated.

5. The method according to claim 4, characterized in that, The process of generating simulation results corresponding to the scenario analysis indicators based on the sampling results, the simulation scenario template, the vehicle model, the sensor model, and the control model includes: A scene model is generated based on the sampling results and the simulation scene template. Based on the target environment information transmitted from the scene model to the sensor model, the target information transmitted from the sensor model to the control model, the control information transmitted from the control model to the vehicle model, the motion state information transmitted from the vehicle model to the control model, and the position information transmitted from the vehicle model to the scene model, the simulation results corresponding to the scene analysis indicators are determined.

6. A device for reducing the number of simulation test scenarios, characterized in that, The device includes: The simulation module is used to obtain the simulation results corresponding to the scene analysis indicators of the autonomous driving system based on the pre-built vehicle model, sensor model, control model corresponding to the autonomous driving system, and sampled values ​​of various scene variables corresponding to the autonomous driving system. The function construction module is used to construct the functional relationship between each of the scenario variables and the scenario analysis index based on the simulation results and the sampled values ​​of each of the scenario variables; The sensitivity analysis module is used to determine the sensitivity of each scene variable based on the functional relationship, and to determine the specific scene generation strategy corresponding to each scene variable based on the sensitivity of each scene variable. The sensitivity analysis module includes a sensitivity determination unit, which is used to determine the first-order sensitivity of each scenario variable based on the contribution of the scenario variable to the variance of the functional relationship. The sensitivity determination unit is further configured to, for each scenario variable, determine the total sensitivity corresponding to the scenario variable based on the contribution of the scenario variable to the variance of the functional relationship, and the contribution of the interaction between the scenario variable and other scenario variables other than the scenario variable to the variance.

7. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the method for reducing the number of simulation test scenarios as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the method for reducing the number of simulation test scenarios as described in any one of claims 1-5.

Images