CN114839892A

CN114839892A - Method, device and equipment for generating automatic driving high-value scene case

Info

Publication number: CN114839892A
Application number: CN202210423801.3A
Authority: CN
Inventors: 夏博林
Original assignee: Yunkong Zhixing Technology Co Ltd
Current assignee: Yunkong Zhixing Technology Co Ltd
Priority date: 2022-04-21
Filing date: 2022-04-21
Publication date: 2022-08-02

Abstract

The application discloses a method for generating an automatic driving high-value scene case, which comprises the following steps: acquiring scene characteristic parameters, and constructing a scene use case base according to the scene characteristic parameters; operating the scene use case base by using an automatic driving algorithm to obtain an actual weight ratio of the scene characteristic parameters, wherein the weight ratio is the ratio of the scene characteristic parameter weight which is successfully operated to the scene characteristic parameter weight which is failed to operate; predicting the weight ratio of the scene characteristic parameters to obtain a predicted weight ratio; constructing a characteristic evaluation model according to the actual weight ratio and the prediction weight ratio; and performing targeted optimization on the scene use case library through the characteristic evaluation model to obtain a high-value scene use case.

Description

Method, device and equipment for generating automatic driving high-value scene case

Technical Field

The application relates to the technical field of automatic driving, in particular to a method, a device and equipment for generating an automatic driving high-value scene case.

Background

For the simulation test work of the cloud control network connection automatic driving, a large-scale magnitude scene case which can generate continuous improvement value for the existing algorithm needs to be constructed. Based on the goal, a simulation test platform of the internet cloud control is established at present, and the platform can establish a test scene of a scene vehicle and an internet automatic driving vehicle according to a vehicle running script. Meanwhile, a large number of cases are exhausted through changing script case parameters to verify the specified algorithm. The purpose of large-scale production of test cases can be achieved. However, the existing produced use case has certain homogenization and the quality of the use case is difficult to define.

Disclosure of Invention

The embodiment of the specification provides a method, a device and equipment for generating an automatic driving high-value scene case, and is used for solving the problems that the existing produced case has certain homogenization and the quality of the case is difficult to define.

The embodiment of the specification adopts the following technical scheme:

in a first aspect, an embodiment of the present specification provides a method for generating an autopilot high-value scenario use case, where the method includes the following steps:

acquiring scene characteristic parameters, and constructing a scene use case base according to the scene characteristic parameters;

operating the scene use case base by using an automatic driving algorithm to obtain an actual weight ratio of the scene characteristic parameters, wherein the weight ratio is the ratio of the scene characteristic parameter weight which is successfully operated to the scene characteristic parameter weight which is failed to operate;

predicting the weight ratio of the scene characteristic parameters to obtain a predicted weight ratio;

constructing a characteristic evaluation model according to the actual weight ratio and the predicted weight ratio;

performing targeted optimization on the scene use case library through the characteristic evaluation model to obtain a high-value scene use case;

adjusting the scene characteristic parameters to obtain a new scene use case base;

and repeating the steps on the new scene use case library to obtain a new high-value scene use case again.

In a second aspect, an embodiment of the present specification provides an apparatus for generating an automatic driving high-value scenario use case, including:

the scene module is used for acquiring scene characteristic parameters and constructing a scene use case base according to the scene characteristic parameters;

the execution module is used for operating the scene use case base by utilizing an automatic driving algorithm to obtain an actual weight ratio of the scene characteristic parameters, wherein the weight ratio refers to the ratio of the weight of the scene characteristic parameters which are successfully operated to the weight of the scene characteristic parameters which are failed to operate;

the prediction module is used for predicting the weight ratio of the scene characteristic parameters to obtain a predicted weight ratio;

the model construction module is used for constructing a characteristic evaluation model according to the actual weight ratio and the prediction weight ratio;

the high-quality case module is used for performing targeted optimization on the scene case library through the characteristic evaluation model to obtain a high-value scene case;

the adjusting module is used for adjusting the scene characteristic parameters to obtain a new scene use case base;

and the circulating module is used for repeating the steps on the new scene use case library to obtain a new high-value scene use case again.

In a third aspect, embodiments of the present specification provide an electronic device, including at least one processor and a memory, where the memory stores a program and is configured such that the at least one processor executes a method for generating an autopilot high-value scenario use case according to any one of the embodiments

In a fourth aspect, embodiments of the present specification provide a computer-readable storage medium storing computer instructions for causing a computer to perform a method for generating an autopilot high-value scenario use case as described in any one of the embodiments.

The embodiment of the specification adopts at least one technical scheme which can achieve the following beneficial effects: the method can find the characteristics of the high-quality test cases, and can dynamically and continuously produce high-value scene cases. And predicting whether the new generation case is a high-value case or not according to the execution result serving as a feedback continuous optimization characteristic weight coefficient, so that the purpose of producing a high-quality case by continuously optimizing and improving the generation mode of the case is achieved.

Drawings

The accompanying drawings, which are included to provide a further understanding of the embodiments of the specification and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the specification and together with the description serve to explain the application and not to limit the application. In the drawings:

FIG. 1 is a schematic flow chart of a method for generating an autopilot high-value scenario use case according to an embodiment of the present disclosure;

fig. 2 is a schematic structural diagram of an apparatus for generating an automatic driving high-value scene use case according to an embodiment of the present disclosure.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more clear, the technical solutions of the present application will be clearly and completely described below with reference to the specific embodiments of the present specification and the accompanying drawings. It is to be understood that the embodiments described are only a few embodiments of the present disclosure, and not all embodiments. All other embodiments obtained by a person skilled in the art based on the embodiments in the present specification without any inventive step are within the scope of the present application.

In the prior art, the use cases are generated in a large scale in the prior art, so that the produced use cases have certain homogenization and the quality of the use cases is difficult to define.

Therefore, the embodiment of the specification provides a method, a device and equipment for generating an automatic driving high-value scene case, which can discover the characteristics of the high-quality measurement case and can dynamically and continuously produce the high-value scene case. And predicting whether the new generation case is a high-value case or not according to the execution result serving as a feedback continuous optimization characteristic weight coefficient, so that the purpose of producing a high-quality case by continuously optimizing and improving the generation mode of the case is achieved.

The technical solutions provided by the embodiments of the present description are described in detail below with reference to the accompanying drawings.

Example 1

Fig. 1 is a schematic flowchart of a method for generating an autopilot high-value scenario use case according to an embodiment of the present disclosure.

Referring to fig. 1, in embodiment 1, a method for generating a use case of an autopilot high-value scene is provided, where the method includes the following steps:

s101, obtaining scene characteristic parameters, and constructing a scene use case base according to the scene characteristic parameters;

in particular implementations, the scene characteristic parameters include, but are not limited to, factors on which the driving environment depends and the range supported by the script commands, such as maps, vehicles, group relationships, and other elements. And constructing a scene use case base according to the scene characteristic parameters, including but not limited to inputting the acquired scene characteristic parameters into the existing scene use case base and obtaining a new scene use case base so that the new scene use case base is used for running an automatic driving algorithm.

The map includes, but is not limited to, high speed, ramp, urban road, crossroad, road curvature, gradient, lane number, etc. Vehicles include, but are not limited to, vehicle type, initial position, vehicle speed, lane change, acceleration, angular acceleration, and the like. Group relationships include, but are not limited to, relative speed, speed hold, lane hold, total number of vehicles, number of controlled vehicles, inter-vehicle distance, and the like. Other elements include, but are not limited to, non-motor vehicles, pedestrians, obstacles, and the like. And constructing the scene use case base by adopting a conventional use case base construction mode according to the scene characteristic parameters. The scenario case library comprises a plurality of scenario cases. When the method is used, basic scenes are configured according to the existing scene construction scripts (including 5 types of simulation, control, constraint, tracking and perception) and matching with different characteristic maps, and scene use cases are generated on the basis of the basic scenes according to different scene algorithms and the driving design of the automatic driving controlled vehicle. For example, on the basis, the initial parameters, the operating parameters, the constraint objects, the trigger conditions and other elements of the basic scene vehicle and the automatic driving controlled vehicle are adjusted to generate a large-scale scene case according to the test matrix model.

It should be understood that the specific references listed above are for illustrative purposes only and should not be construed as limiting the application in any way.

S103, operating the scene use case base by using an automatic driving algorithm to obtain an actual weight ratio of the scene characteristic parameters, wherein the weight ratio is the ratio of the scene characteristic parameter weight which is successfully operated to the scene characteristic parameter weight which is failed to operate;

in specific implementation, the method for obtaining the actual weight ratio of the scene characteristic parameters by running the scene case library with an automatic driving algorithm includes, but is not limited to, executing the automatic driving algorithm in the scene case library to obtain the scene case library with failed execution; obtaining the execution failure weight of the scene characteristic parameters by using the example library of the scene with the execution failure through a logistic regression model; and obtaining the actual weight ratio of the scene characteristic parameters according to the execution failure weight.

For example, when the automatic driving algorithm is executed, if multiple scene characteristic parameters in multiple scene use cases cannot be successfully executed in the scene use case library, the scenes which cannot pass through are marked, and then the total number of the scene use cases which are failed to be executed due to the scene characteristic parameters is counted, so that the weight of the execution failure of each scene characteristic parameter is counted. Taking the slope in the scene characteristic as an example, when the system runs to the scene constructed by the slope and cannot pass through the scene constructed by the slope during the execution of the automatic driving algorithm, counting the total number of all scene cases which cannot be successfully executed due to the slope in the scene case library, and thus obtaining the execution failure weight of the slope, which is the scene characteristic parameter. Of course, each scene case contains a plurality of scene features, and each scene case also contains a plurality of scene features of execution failure, and the statistical manner is the same as that. The description is not repeated.

S105, predicting the weight ratio of the scene characteristic parameters to obtain a predicted weight ratio;

in specific implementation, predicting the weight ratio of the scene characteristic parameters may be understood as predicting the weight ratio of the scene characteristic parameters in advance, that is, determining the ratio of the weights of the scene characteristic parameters that are executed in the constructed scene example library and failed in advance. Predicting the weight ratio of the scene characteristic parameters, and obtaining a prediction weight ratio mode, wherein the prediction weight ratio mode comprises but is not limited to the mode of constructing a prediction model; and predicting the weight ratio of the scene characteristic parameters through the prediction model to obtain a predicted weight ratio. The method for constructing the prediction model includes, but is not limited to, combining a logistic regression model with probability through fitting a decision boundary and establishing the decision boundary, and then constructing the prediction model by using a sigmoid function.

For example, the example is continued with the scene feature gradient in S103, and in the scene case library constructed using the scene feature of the gradient, it is predicted how many scene cases that have failed to be executed after the execution of the constructed scene case library, and then it is predicted how many cases that have failed to be executed due to the gradient. Thereby obtaining the prediction failure weight of the scene characteristic parameter of gradient.

S107, constructing a feature evaluation model according to the actual weight ratio and the predicted weight ratio;

in the specific embodiment, a logistic regression model is used for classification, the simulated vehicle running result simulated according to the automatic driving algorithm is compared with the expected result of the scene use case base, and the comparison result is labeled. The label can be understood as a pre-prepared feature, such as failure or success, only the weight is calculated, such as the weight of the failure, the weight of the scene feature is obtained through a logistic regression model, and a feature evaluation model is established.

S109, performing targeted optimization on the scene use case library through the feature evaluation model to obtain a high-value scene use case;

in specific implementation, the manner of obtaining the high-value scene use case by performing the targeted optimization on the scene use case library through the feature evaluation model includes, but is not limited to, adjusting the scene feature parameters of the scene use case library in a direction with high execution failure weight through the feature evaluation model so as to obtain the high-value scene use case.

For example, continuing to use the scene feature of the slope in S103 as an example, in the scene case library constructed by the slope, the slope in the successfully executed scene case is subjected to parameter adjustment, and is adjusted in the direction of execution failure, so that the high-value scene case is obtained.

Further, adjusting the scene characteristic parameters to obtain a new scene use case base;

in specific implementations, the manner of adjusting the scene characteristic parameters includes, but is not limited to, inputting new scene characteristic parameters after manually adjusting the scene characteristic parameters, or inputting new scene characteristic parameters after selecting the characteristic parameters.

In this embodiment, for a scenario where execution fails, for example, when the automatic driving algorithm fails, for example, a vehicle collision occurs, an expected event is not generated, and the like, the automatic driving algorithm performs scenario training again after repairing a related problem. The system can repeat the iterative scene training process periodically, can find the sample characteristics of the high-quality test samples, and can dynamically and continuously produce the high-quality test samples. And predicting whether the new generation case is a high-value case or not according to the execution result serving as a feedback continuous optimization characteristic weight coefficient, so that the purpose of generating a high-quality case capable of challenging the cloud automatic driving algorithm by continuously optimizing and improving the generation mode of the case is achieved.

Further, the adjusting the scene characteristic parameter to obtain a new scene use case further includes: and comparing the actual weight ratio with the predicted weight ratio to obtain the accuracy evaluation of the feature evaluation model.

In specific implementation, an automatic driving algorithm is executed in a scene use case library to obtain a scene use case library with execution failure; and obtaining the execution failure weight of the scene characteristic parameters by using the example library of the scene with the execution failure through a logistic regression model. And comparing the predicted weight ratio with the actual weight ratio, judging whether the comparison result exceeds a preset value, if not, obtaining a higher accurate value of the feature evaluation model, and if so, obtaining a lower accurate value of the feature evaluation model.

Example 2

Referring to fig. 2, embodiment 2 provides an apparatus for generating a case of an autopilot high-value scene, including:

the scene module 301 is configured to obtain scene characteristic parameters, and construct a scene use case base according to the scene characteristic parameters;

an executing module 302, configured to run the scene use case base by using an automatic driving algorithm, to obtain an actual weight ratio of the scene characteristic parameter, where the weight ratio is a ratio of a scene characteristic parameter weight that is successfully run to a scene characteristic parameter weight that is unsuccessfully run;

the prediction module 303 is configured to predict the weight ratio of the scene characteristic parameter to obtain a predicted weight ratio;

a model construction module 304, configured to construct a feature evaluation model according to the actual weight ratio and the predicted weight ratio;

a high-quality use case module 305, configured to perform targeted optimization on the scene use case library through the feature evaluation model to obtain a high-value scene use case;

The scene module 301 is further configured to input the acquired scene characteristic parameters into a scene use case library to obtain a new scene use case library, so that the new scene use case library is used for running the automatic driving algorithm.

The execution module 302 is further configured to execute an automatic driving algorithm in the scene use case base, and obtain a scene use case base with execution failure; obtaining the execution failure weight of the scene characteristic parameters by using the example library of the scene with the execution failure through a logistic regression model; and obtaining the actual weight ratio of the scene characteristic parameters according to the execution failure weight.

The model building module 304 is further configured to build a prediction model.

The adjusting module 306 is further configured to adjust the scene characteristic parameters of the scene use case library through the characteristic evaluation model in a direction in which the execution failure weight is high, so as to obtain a high-value scene use case.

And the accuracy judgment module is used for comparing the actual weight ratio with the predicted weight ratio to obtain the accuracy evaluation of the feature evaluation model.

Example 3

Embodiment 3 provides an electronic device comprising at least one processor and a memory, the memory storing a program and configured to the at least one processor to perform a method of generating an autopilot high value scenario use case as recited in any one of the embodiments.

Example 4

Embodiment 4 provides a computer-readable storage medium storing computer instructions for causing a computer to perform a method of generating an autopilot high-value scenario use case as recited in any one of the embodiments.

In the 90 s of the 20 th century, improvements in a technology could clearly distinguish between improvements in hardware (e.g., improvements in circuit structures such as diodes, transistors, switches, etc.) and improvements in software (improvements in process flow). However, as technology advances, many of today's process flow improvements have been seen as direct improvements in hardware circuit architecture. Designers almost always obtain the corresponding hardware circuit structure by programming an improved method flow into the hardware circuit. Thus, it cannot be said that an improvement in the process flow cannot be realized by hardware physical modules. For example, a Programmable Logic Device (PLD), such as a Field Programmable Gate Array (FPGA), is an integrated circuit whose Logic functions are determined by programming the Device by a user. A digital system is "integrated" on a PLD by the designer's own programming without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Furthermore, nowadays, instead of manually making an integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to a software compiler used in program development and writing, but the original code before compiling is also written by a specific Programming Language, which is called Hardware Description Language (HDL), and HDL is not only one but many, such as abel (advanced Boolean Expression Language), ahdl (alternate Language Description Language), traffic, pl (core unified Programming Language), HDCal, JHDL (Java Hardware Description Language), langue, Lola, HDL, laspam, hardsradware (Hardware Description Language), vhjhd (Hardware Description Language), and vhigh-Language, which are currently used in most common. It will also be apparent to those skilled in the art that hardware circuitry that implements the logical method flows can be readily obtained by merely slightly programming the method flows into an integrated circuit using the hardware description languages described above.

The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic for the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may thus be considered a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.

The systems, apparatuses, modules or units illustrated in the above-described embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

For convenience of description, the above devices are described as being divided into various modules or units by function, respectively. Of course, the functionality of the modules or units may be implemented in the same one or more software and/or hardware when implementing the present application.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the specification. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory (NVM), such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

The above description is only an illustrative example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for generating an autopilot high-value scenario use case, the method comprising the steps of:

and performing targeted optimization on the scene use case library through the characteristic evaluation model to obtain a high-value scene use case.

2. The method for generating an automatic driving high-value scene use case according to claim 1, wherein the obtaining of the scene feature parameters further comprises:

and executing the step of operating the scene case library by using an automatic driving algorithm until the scene case library is subjected to targeted optimization through the characteristic evaluation model on the new scene case library, and obtaining a new high-value scene case again.

3. The method for generating an automatic driving high-value scene use case according to claim 1, wherein the constructing a scene use case library according to the scene feature parameters comprises:

and inputting the acquired scene characteristic parameters into a scene use case library to obtain a new scene use case library so that the new scene use case library is used for running an automatic driving algorithm.

4. The method for generating the automatic driving high-value scene use case according to claim 1, wherein the using the automatic driving algorithm to run the scene use case base to obtain the actual weight ratio of the scene feature parameters comprises:

executing an automatic driving algorithm in the scene use case library to obtain the scene use case library with execution failure;

obtaining the execution failure weight of the scene characteristic parameters by using the example library of the scene with the execution failure through a logistic regression model;

and obtaining the actual weight ratio of the scene characteristic parameters according to the execution failure weight.

5. The method for generating the use case for the automatic driving high-value scene according to claim 1, wherein the predicting the weight ratio of the scene feature parameters and obtaining the predicted weight ratio comprises:

constructing a prediction model;

and predicting the weight ratio of the scene characteristic parameters through the prediction model to obtain a predicted weight ratio.

6. The method for generating cases for use in high-value scenarios for autonomous driving of claim 5, wherein the constructing the prediction model comprises constructing the prediction model using sigmoid function after fitting a logistic regression model to the decision boundary, establishing the decision boundary and combining the probability.

7. The method for generating an automatic driving high-value scenario case according to claim 1, wherein the performing targeted optimization on the scenario case library through the feature evaluation model to obtain a high-value scenario case comprises:

and adjusting the scene characteristic parameters of the scene use case library in the direction with high execution failure weight through the characteristic evaluation model so as to obtain the high-value scene use case.

8. The method according to claim 1, wherein the adjusting the scene feature parameters to obtain a new scene use case further comprises:

and comparing the actual weight ratio with the predicted weight ratio to obtain the accuracy evaluation of the feature evaluation model.

9. An apparatus for generating an autopilot high value scenario use case, comprising:

10. An electronic device comprising at least one processor and a memory, the memory storing a program and configured such that the at least one processor performs a method of generating an autopilot high value scenario use case of any of claims 1-8.