CN115407679A

CN115407679A - Test method, device, equipment and storage medium of automatic driving algorithm

Info

Publication number: CN115407679A
Application number: CN202211042616.6A
Authority: CN
Inventors: 龙鹏宇
Original assignee: Chongqing Changan Automobile Co Ltd
Current assignee: Chongqing Changan Automobile Co Ltd
Priority date: 2022-08-29
Filing date: 2022-08-29
Publication date: 2022-11-29

Abstract

The method comprises the steps of obtaining an automatic driving algorithm to be tested, compiling a source code of the automatic driving algorithm to be tested to obtain a compiled product, constructing to obtain an algorithm mirror to be tested, pushing the algorithm mirror to be tested to a cloud simulation platform, determining a target scene group corresponding to the algorithm mirror to be tested constructed by the automatic driving algorithm to be tested based on an algorithm characteristic identifier and a characteristic identifier-scene group mapping relation of the automatic driving algorithm to be tested, executing a simulation task based on the algorithm mirror to be tested and the target scene group through the cloud simulation platform to obtain a simulation result, testing the automatic driving algorithm to be tested, communicating an algorithm compiling process, a mirror constructing process and a pushing process, communicating the processes with the cloud simulation platform, reducing the workload of testing or a developer for verifying the function of the automatic driving algorithm, improving the working efficiency and reducing the cost.

Description

Test method, device, equipment and storage medium of automatic driving algorithm

Technical Field

The embodiment of the invention relates to the technical field of testing, in particular to a method, a device, equipment and a storage medium for testing an automatic driving algorithm.

Background

Computer simulation of an automatic driving system is a basic key technology for testing and experimenting automatic driving vehicles and is also a basic tool for defining relevant development and admission technical standards of the automatic driving vehicles in the future industry.

The automatic driving simulation test is to establish a mathematical model of a real static environment and a dynamic traffic scene through a computer simulation technology, so that an automatic driving vehicle and an algorithm carry out driving test in a virtual traffic scene. In the process of research and development, the automatic driving algorithm is continuously subjected to iterative upgrade, the automatic driving algorithm can be tested and verified by relying on computer simulation of an automatic driving system, and meanwhile, the automatic driving algorithm is tested to be a work consuming a large amount of manpower and time. In the related art, the simulation test of the automatic driving algorithm usually needs a tester to manually realize the test of the automatic driving algorithm based on the cloud simulation platform in a manual mode, and the method has the advantages of large manual workload, low efficiency and high cost.

Disclosure of Invention

In view of the above-mentioned shortcomings in the prior art, embodiments of the present invention provide a method, an apparatus, a device and a storage medium for testing an autopilot algorithm, so as to solve the above-mentioned technical problems.

The embodiment of the invention provides a method for testing an automatic driving algorithm, which comprises the following steps:

acquiring an automatic driving algorithm to be tested;

compiling the source code of the automatic driving algorithm to be tested to obtain a compiled product, and constructing to obtain an algorithm mirror image to be tested;

pushing the algorithm mirror image to be tested to a cloud simulation platform;

determining a target scene group corresponding to the to-be-tested algorithm mirror image constructed by the to-be-tested automatic driving algorithm based on the algorithm feature identification and the feature identification-scene group mapping relation of the to-be-tested automatic driving algorithm;

and executing a simulation task based on the algorithm mirror image to be tested and the target scene group through the cloud simulation platform to obtain a simulation result so as to test the automatic driving algorithm to be tested.

In an embodiment of the present invention, before pushing the algorithm image to be tested to the cloud simulation platform, the method includes:

acquiring an algorithm mirror image uploading instruction, wherein the algorithm mirror image uploading instruction comprises platform mirror image warehouse access account information and platform mirror image warehouse access passing information;

logging in a platform mirror image warehouse of the cloud simulation platform through the platform mirror image warehouse access account information and the platform mirror image warehouse access pass information;

and if the login is successful, pushing the algorithm mirror image to be tested to the platform mirror image warehouse of the cloud simulation platform.

In an embodiment of the present invention, before logging in a platform mirror image repository of the cloud simulation platform through the platform mirror image repository access account information and the platform mirror image repository access pass information, the method includes:

establishing a platform algorithm mirror image and a platform mirror image identifier of the platform algorithm mirror image in the platform mirror image warehouse;

and generating the algorithm mirror image uploading instruction based on the platform mirror image identification, the platform mirror image warehouse access account information and the platform mirror image warehouse access passing information.

In an embodiment of the present invention, if the login is successful, before pushing the algorithm image to be tested to the platform image repository of the cloud simulation platform, the method includes:

associating the algorithm mirror to be tested with the platform mirror identification;

and if the login is successful, pushing the associated algorithm mirror image to be tested to the platform mirror image warehouse of the cloud simulation platform so as to update the platform algorithm mirror image through the algorithm mirror image to be tested.

In an embodiment of the present invention, after the algorithm mirror to be tested is constructed, the method further includes:

and storing the algorithm mirror to be tested in a temporary mirror warehouse so as to pull the algorithm mirror to be tested from the temporary mirror warehouse and push the algorithm mirror to be tested to the cloud simulation platform.

In an embodiment of the present invention, determining, based on the algorithm feature identifier of the to-be-tested autopilot algorithm and the feature identifier-scene group mapping relationship, a target scene group corresponding to the to-be-tested algorithm image constructed by the to-be-tested autopilot algorithm includes:

obtaining a plurality of test scenes of an automatic driving algorithm to be tested;

configuring algorithm characteristic identification of the automatic driving algorithm to be tested based on the test scene of each automatic driving algorithm to be tested;

respectively creating test scenes of the automatic driving algorithms to be tested into scene groups in the cloud simulation platform;

establishing a mapping relation between the feature identifier and the scene group based on the corresponding relation between the algorithm feature identifier and the scene group corresponding to the test scene of the automatic driving algorithm to be tested;

and determining at least one scene group as the target scene group based on the algorithm characteristic identification and the characteristic identification-scene group mapping relation of the automatic driving algorithm to be tested.

In an embodiment of the present invention, after the cloud simulation platform executes a simulation task based on the algorithm image to be tested and the target scene group to obtain a simulation result, the method further includes:

and generating an evaluation report according to the simulation result, and displaying the evaluation report.

The embodiment of the invention also provides a device for testing the automatic driving algorithm, which comprises:

the acquisition module is used for acquiring an automatic driving algorithm to be tested;

the compiling assembly line is used for compiling the source code of the automatic driving algorithm to be tested to obtain a compiled product;

the algorithm mirror image production line is used for constructing and obtaining an algorithm mirror image to be tested based on the compiled product;

the pushing mirror image assembly line is used for pushing the algorithm mirror image to be tested to a cloud simulation platform;

and the cloud simulation platform is used for executing a simulation task based on the algorithm mirror image to be tested and the target scene group to obtain a simulation result so as to test the automatic driving algorithm to be tested, wherein the determination mode of the target scene group comprises the step of determining the target scene group corresponding to the algorithm mirror image to be tested, which is constructed by the automatic driving algorithm to be tested, based on the algorithm characteristic identification and the characteristic identification-scene group mapping relation of the automatic driving algorithm to be tested.

An embodiment of the present invention provides an electronic device, where the electronic device includes:

one or more processors;

a storage device for storing one or more programs that, when executed by the one or more processors, cause the electronic device to implement the method for testing an autopilot algorithm of any of the embodiments described above.

An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor of a computer, the computer program causes the computer to execute the method for testing an autopilot algorithm according to any one of the embodiments.

The embodiment of the invention has the following beneficial effects: the method comprises the steps of compiling a source code of the automatic driving algorithm to be tested by obtaining the automatic driving algorithm to be tested to obtain a compiled product, constructing to obtain an algorithm mirror image to be tested, pushing the algorithm mirror image to be tested to a cloud simulation platform, determining a target scene group corresponding to the algorithm mirror image to be tested constructed by the automatic driving algorithm to be tested based on an algorithm characteristic identifier and a characteristic identifier-scene group mapping relation of the automatic driving algorithm to be tested, executing a simulation task based on the algorithm mirror image to be tested and the target scene group by the cloud simulation platform to obtain a simulation result, testing the automatic driving algorithm to be tested, communicating an algorithm compiling process, a mirror image constructing process and a pushing process, and communicating the processes with the cloud simulation platform, so that the function verification of the automatic driving algorithm can be realized in a testing mode provided by the embodiment. Therefore, the function of automatically and efficiently simulating on the cloud simulation platform after the automatic driving algorithm is updated is realized, the workload of testing or developing personnel for verifying the automatic driving algorithm function is reduced, the working efficiency is improved, and the cost is reduced.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and, together with the description, serve to explain the principles of the application. It is obvious that the drawings in the following description are only some embodiments of the application, and that for a person skilled in the art, other drawings can be derived from them without inventive effort. In the drawings:

FIG. 1 is a flow chart diagram illustrating a method of testing an autopilot algorithm according to an exemplary embodiment of the present application;

fig. 2 is a flowchart illustrating a pushing process of an algorithm image to be tested according to an exemplary embodiment of the present application;

FIG. 3 is a block diagram of a test setup for an autopilot algorithm shown in an exemplary embodiment of the present application;

FIG. 4 is a block diagram of a test setup for an autonomous driving algorithm shown in another exemplary embodiment of the present application;

FIG. 5 illustrates a schematic structural diagram of a computer system suitable for use to implement the electronic device of the embodiments of the subject application.

Detailed Description

Other advantages and effects of the present invention will become apparent to those skilled in the art from the disclosure herein, wherein the embodiments of the present invention are described in detail with reference to the accompanying drawings and preferred embodiments. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention, and are not intended to limit the scope of the present invention.

It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the drawings only show the components related to the present invention rather than being drawn according to the number, shape and size of the components in actual implementation, and the type, amount and proportion of each component in actual implementation can be changed freely, and the layout of the components can be more complicated.

In the following description, numerous details are set forth to provide a more thorough explanation of embodiments of the present invention, however, it will be apparent to one skilled in the art that embodiments of the present invention may be practiced without these specific details, and in other embodiments, well-known structures and devices are shown in block diagram form, rather than in detail, in order to avoid obscuring embodiments of the present invention.

The automatic driving simulation test is to establish a mathematical model of a real static environment and a dynamic traffic scene through a computer simulation technology, so that an automatic driving vehicle and an algorithm carry out driving test in a virtual traffic scene. In the process of research and development, the automatic driving algorithm is continuously subjected to iterative upgrade, the automatic driving algorithm can be tested and verified by relying on computer simulation of an automatic driving system, and meanwhile, the automatic driving algorithm is tested to be a work consuming a large amount of manpower and time. In the related art, the simulation test of the automatic driving algorithm usually needs a tester to manually test the automatic driving algorithm based on the cloud simulation platform in a manual mode, and the method is large in manual workload, low in efficiency and high in cost.

Tests for the autopilot algorithm can be classified into two categories, one is a test for whether the code can run, and the other is a test for whether the algorithm functions normally. In conventional persistent integration, the main code is concerned about being able to run, and whether the function is normal requires manual verification by a tester or verification using some automatic verification tool. However, for the automatic driving algorithm, the manual verification function of a tester needs to be performed on a real vehicle, which is long in time consumption and not safe enough; if the tester uses the simulation tool for verification, a large amount of time is consumed for adapting the automatic driving algorithm and uploading the adaptive automatic driving algorithm to the cloud simulation platform to select a corresponding scene for simulation.

In order to solve the above problems, embodiments of the present application respectively provide a method for testing an autopilot algorithm, an apparatus for testing an autopilot algorithm, an electronic device, a computer-readable storage medium, and a computer program product, and the embodiments will be described in detail below.

According to the testing method provided by the embodiment, the continuous integration tool is communicated with the cloud simulation platform, so that the functional verification of the automatic driving algorithm can be triggered through the continuous integration tool. Therefore, the function of automatically and efficiently simulating on the cloud simulation platform after the automatic driving algorithm is updated is realized, and the workload of testing or developing personnel for verifying the automatic driving algorithm function is reduced.

Referring to fig. 1, fig. 1 is a flowchart illustrating a testing method of an autopilot algorithm according to an exemplary embodiment of the present application. As shown in fig. 1, in an exemplary embodiment, the testing method of the automatic driving algorithm at least includes steps S101 to S104, which are described in detail as follows:

and S101, acquiring an automatic driving algorithm to be tested.

In one embodiment, the to-be-tested automatic driving algorithm is stored in the code warehouse, and when the code warehouse is monitored to have the newly-added to-be-tested automatic driving algorithm, the step of obtaining the to-be-tested automatic driving algorithm is triggered. Therefore, once the to-be-tested automatic driving algorithm is updated or newly added, the to-be-tested automatic driving algorithm can be tested, manual operation is not needed, and the method is convenient and quick. The monitoring of whether the newly-added automatic driving algorithm to be tested exists in the code warehouse can be determined by comparing the automatic driving algorithms to be tested at the first time and the second time in the code warehouse, or can be performed by monitoring a write operation log of the code warehouse, and once the log is updated, the updated or newly-added automatic driving algorithm to be tested is determined based on the log and is tested.

The number of the automatic driving algorithms to be tested may be one or more, and is not limited herein. The automated driving algorithm to be tested may include one or more modules, such as a prediction module, a decision module, a regulatory module, and the like. The creation of the autopilot algorithm to be tested can be implemented in a manner known to those skilled in the art and is not limited herein.

The automatic driving algorithm to be tested may have continuous iterative upgrade, and the method provided by the embodiment can be adopted for testing the automatic driving algorithm to be tested after each upgrade.

In an embodiment, the step of obtaining the automated driving algorithm to be tested may be performed by modifying the code repository storing the automated driving algorithm to be tested, and then automatically using the modified automated driving algorithm to be tested as the obtained automated driving algorithm to be tested. That is, the step of obtaining the automated driving algorithm to be tested is triggered based on the update of the automated driving algorithm to be tested in the code repository. Therefore, the newly added or newly revised automatic driving algorithm to be detected can be comprehensively, timely and automatically tested, and the process of manpower screening and execution is saved.

And S102, compiling the source code of the automatic driving algorithm to be tested to obtain a compiled product, and constructing to obtain an algorithm mirror image to be tested.

In one embodiment, an exemplary implementation of step S102 may be by building a pipeline of automated driving algorithm source code compilation, testing (compilation pipeline). And a compiling production line is built through a continuous integration tool, so that the function of compiling the to-be-tested automatic driving algorithm from the source code to obtain a corresponding product is realized, and an environment during compiling is required to be built before the production line is executed.

In an embodiment, after the mirror image of the algorithm to be tested is constructed, the method further includes:

and storing the mirror image of the algorithm to be tested in a temporary mirror image warehouse so as to pull the mirror image of the algorithm to be tested from the temporary mirror image warehouse and push the mirror image of the algorithm to be tested to the cloud simulation platform.

In one embodiment, an exemplary implementation of step S102 includes:

pulling Ubuntu18.04 as the base image, which is only an example here, the base image may also be other images known to those skilled in the art;

installing various dependency libraries required by a source code of a compiling algorithm in a mirror image;

pulling codes of each code warehouse;

executing the compiling script or the compiling instruction of each module in the container to generate a corresponding product and a unit test module;

executing the unit test module and collecting an execution result;

and uploading the compiled product to a product storage platform.

The unit testing module is used for detecting the product, when the execution result is a preset result, the product is qualified, and the compiled product can be uploaded to a product storage platform.

By the aid of the method, the source code of the automatic driving algorithm can be compiled in advance regardless of whether the automatic driving algorithm is tested or not, the compiled product is uploaded to a product storage platform through the unit testing module, and the compiled product is directly acquired from the product storage platform when the product is needed, so that the method is more convenient and rapid.

In one embodiment, the method for constructing the algorithm mirror to be tested based on the algorithm product can be realized by a pre-constructed algorithm mirror production line, and the algorithm compiling product is constructed into the algorithm mirror which can be started by a sentence of command according to a specific directory structure by constructing the production line through a continuous integration tool. An exemplary algorithm image construction pipeline building step is as follows:

the compiled products in the above embodiments are pulled, and the products are organized according to the step of manual integration in the real vehicle or simulation environment, and the algorithm products may be stored in a certain directory, format, or configuration file manner to obtain a plurality of files.

And writing an algorithm starting script and modifying the executable authority of each file.

And (3) constructing a mirror image (algorithm mirror image) through the Dockerfile, and uploading the mirror image to a temporary mirror image warehouse after the construction is finished.

By the method, the compiled product can be automatically converted into the algorithm mirror image to be tested.

And step S103, pushing the mirror image of the algorithm to be tested to a cloud simulation platform.

The method comprises the steps that a pushing assembly line is constructed in advance, the to-be-tested algorithm mirror image is pushed to the cloud simulation platform, and therefore continuity can be achieved, and the to-be-tested algorithm mirror image is automatically pushed to the cloud simulation platform.

In one embodiment, before pushing the algorithm image to be tested to the cloud simulation platform, the method includes:

acquiring an algorithm mirror image uploading instruction, wherein the algorithm mirror image uploading instruction comprises platform mirror image warehouse access account information and platform mirror image warehouse access traffic information;

and if the login is successful, pushing the algorithm mirror image to be tested to a platform mirror image warehouse of the cloud simulation platform.

The access account information of the platform image warehouse can be a temporary user name for accessing the cloud simulation platform, the access passing information of the platform image warehouse can be a temporary password corresponding to the temporary user name, and whether a user in the current execution step has the authority for accessing the platform image warehouse in the cloud simulation platform can be proved through the verification of the temporary user name and the temporary password.

In one embodiment, before logging in a platform mirror repository of a cloud simulation platform through platform mirror repository access account information and platform mirror repository access pass information, the method includes:

establishing a platform algorithm mirror image and a platform mirror image identifier of the platform algorithm mirror image in a platform mirror image warehouse;

and generating an algorithm mirror image uploading instruction based on the platform mirror image identifier, the platform mirror image warehouse access account information and the platform mirror image warehouse access pass information.

In an embodiment, if the login is successful, before pushing the algorithm image to be tested to a platform image repository of the cloud simulation platform, the method includes:

associating the algorithm mirror image to be tested with the platform mirror image identification;

and if the login is successful, pushing the associated algorithm mirror image to be tested to a platform mirror image warehouse of the cloud simulation platform so as to update the platform algorithm mirror image through the algorithm mirror image to be tested.

In an embodiment, the pushing of the algorithm image to be tested may be based on a pre-constructed push autopilot algorithm image to a push image pipeline implementation on the cloud simulation platform. Because the cloud simulation platform maintains the algorithm mirror image independently, the algorithm mirror image to be tested needs to be pushed according to the requirements of the cloud simulation platform. Referring to fig. 2, fig. 2 is a flowchart illustrating a pushing process of an algorithm image to be tested according to an exemplary embodiment of the present application, and as shown in fig. 2, an exemplary manner is as follows:

after a push mirror image production line (push mirror image production line) starts, a new platform algorithm and a new platform algorithm mirror image are created through an Application Programming Interface (API) of a cloud simulation platform, the platform algorithm and the platform algorithm mirror image are null information, after the platform algorithm mirror image to be tested is pulled, the platform algorithm mirror image is updated by the algorithm mirror image to be tested, namely the algorithm mirror image to be tested is used as the platform algorithm mirror image.

And pulling an algorithm mirror (to-be-tested algorithm mirror) from the temporary mirror image warehouse, wherein the to-be-tested algorithm mirror mentioned in the embodiment is stored in the temporary mirror image warehouse, a source code of the to-be-tested autopilot algorithm is compiled to obtain a compiled product without waiting for an algorithm mirror uploading instruction, the to-be-tested algorithm mirror is constructed based on the compiled product to obtain the to-be-tested algorithm mirror, and the to-be-tested algorithm mirror is stored in the temporary mirror image warehouse and waits to be pulled.

And calling the cloud simulation platform API to acquire the label, the temporary user name and the password. For example, an instruction (algorithm image uploading instruction) required for uploading an image is obtained through the API, and the instruction includes a tag (platform image identifier) of an algorithm in the cloud simulation platform, a temporary user name (platform image warehouse access account information) and a password (platform image warehouse access pass information) of a platform image warehouse in the cloud simulation platform. The temporary user name and the password are used for determining whether the current user has access authority for accessing a platform image warehouse in the cloud simulation platform.

And if the API call fails, recording error information.

If the API is successfully called, the constructed to-be-tested algorithm mirror image is pulled from the temporary mirror image warehouse, and the label of the mirror image is marked as a label in the cloud simulation platform (namely, the to-be-tested algorithm mirror image is associated with the platform mirror image identification).

Logging in a mirror image warehouse of the cloud simulation platform through a temporary user name and a temporary password, and recording error information if the logging fails.

And if the login is successful, pushing the algorithm mirror to be tested to an algorithm mirror warehouse (platform mirror warehouse) of the cloud simulation platform, judging whether the pushing is successful, and if the pushing is failed, recording error information. And if the pushing is successful, ending the pushing mirror image assembly line.

And S104, determining a target scene group corresponding to the mirror image of the algorithm to be tested constructed by the automatic driving algorithm to be tested based on the algorithm characteristic identification and the characteristic identification-scene group mapping relation of the automatic driving algorithm to be tested.

In one embodiment, determining a target scene group corresponding to a mirror image of an algorithm to be tested constructed by the automatic driving algorithm to be tested based on the algorithm feature identifier and the feature identifier-scene group mapping relationship of the automatic driving algorithm to be tested comprises:

establishing a feature identifier-scene group mapping relation based on the corresponding relation between the algorithm feature identifier and the scene group corresponding to the test scene of the automatic driving algorithm to be tested;

and determining at least one scene group as a target scene group based on the algorithm characteristic identification and the characteristic identification-scene group mapping relation of the automatic driving algorithm to be tested.

Simulation scene combinations corresponding to different automatic driving algorithm modules are preset. Since the automatic driving algorithm is generally composed of a plurality of modules, for example: a prediction module, a decision module, a regulation module, etc. The scenes used for verification of each module are different, so that corresponding scene groups need to be created in the cloud simulation platform, and related scenes need to be associated to the scene groups.

In an embodiment, before step S105, the method further includes:

and building a cloud simulation platform. The cloud emulation platform needs to contain at least the following functions: the method supports uploading of the algorithm mirror to be tested, operation of the simulation task in the algorithm mirror mode, uploading of the simulation scene, management of the simulation scene group, output of an evaluation report after the simulation task is completed, and calling of the simulation task through an API (application program interface).

And S105, executing a simulation task based on the algorithm mirror image to be tested and the target scene group through the cloud simulation platform to obtain a simulation result so as to test the automatic driving algorithm to be tested.

Taking the algorithm feature identification as the name of the algorithm image to be tested as an example, for example, calling a cloud simulation platform interface through an API. The name of the algorithm mirror image to be tested is analyzed to obtain the name of the changed algorithm module, so that the corresponding scene group is found. And transmitting the ID of the corresponding scene group and the ID of the algorithm mirror image through the API, and executing the simulation task.

In an embodiment, after the cloud simulation platform executes a simulation task based on the algorithm image to be tested and the target scene group to obtain a simulation result, the method further includes:

The cloud simulation platform generates an evaluation report after the simulation task is completed, and can call the API to obtain the evaluation report and display the evaluation report in the production line.

The evaluation report may be generated by means of a related art, which is not limited herein.

According to the test method of the automatic driving algorithm, the cloud simulation platform is built, the cloud simulation platform supports simulation through algorithm mirroring, and a development interface is exposed. And setting simulation scene combinations corresponding to different automatic driving algorithm modules. And (3) building an automatic driving algorithm source code compiling and testing assembly line, compiling the source code of the automatic driving algorithm to be tested automatically to obtain a compiled product, and carrying out legality inspection on the compiled product. And building a production line for building an algorithm mirror image by using a product compiled by the automatic driving algorithm code so as to build an algorithm mirror image to be tested based on the compiled product. And establishing a flow line for pushing the automatic driving algorithm mirror image to the cloud simulation platform so as to push the algorithm mirror image to be tested to the cloud simulation platform. Calling a cloud simulation platform interface through an API (application program interface), operating an automatic driving algorithm by using a corresponding scene, namely determining a scene group corresponding to the automatic driving algorithm to be tested, testing the automatic driving algorithm to be tested by using the scene group, operating simulation by using the cloud simulation platform, and outputting an evaluation report. Communicating a continuous integration tool (communicating an algorithm compiling process, a mirror image constructing process and a pushing process) with the cloud simulation platform, so that the functional verification of the automatic driving algorithm can be triggered through the continuous integration tool. Therefore, the function of automatically and efficiently simulating on the cloud simulation platform after the automatic driving algorithm is updated is realized, and the workload of testing or developing personnel for verifying the automatic driving algorithm function is reduced.

According to the method for testing the automatic driving algorithm, the automatic driving algorithm to be tested is obtained, the source code of the automatic driving algorithm to be tested is compiled to obtain a compiled product, the algorithm mirror to be tested is constructed to obtain an algorithm mirror to be tested, the algorithm mirror to be tested is pushed to the cloud simulation platform, the target scene group corresponding to the algorithm mirror to be tested constructed by the automatic driving algorithm to be tested is determined based on the algorithm characteristic identification and characteristic identification-scene group mapping relation of the automatic driving algorithm to be tested, the cloud simulation platform executes a simulation task based on the algorithm mirror to be tested and the target scene group to obtain a simulation result, the automatic driving algorithm to be tested is tested, the algorithm compiling process, the mirror constructing process and the pushing process are communicated, the process is communicated with the cloud simulation platform, and the function verification of the automatic driving algorithm can be achieved through the testing method provided by the embodiment. Therefore, the function of automatically and efficiently simulating on the cloud simulation platform after the automatic driving algorithm is updated is realized, the workload of testing or developing personnel for verifying the automatic driving algorithm function is reduced, the working efficiency is improved, and the cost is reduced.

Fig. 3 is a block diagram of a test apparatus of an autonomous driving algorithm shown in an exemplary embodiment of the present application. As shown in fig. 3, the exemplary autopilot algorithm testing apparatus 300 includes:

an obtaining module 301, configured to obtain an automatic driving algorithm to be tested;

a compiling pipeline 302, configured to compile a source code of the to-be-tested autopilot algorithm to obtain a compiled product;

an algorithm mirror flow line 303, configured to construct an algorithm mirror to be tested based on the compiled product;

a mirror image pushing pipeline 304, configured to push the mirror image of the algorithm to be tested to a cloud simulation platform;

the cloud simulation platform 305 is configured to execute a simulation task based on the to-be-tested algorithm image and a target scene group to obtain a simulation result, so as to test the to-be-tested automatic driving algorithm, where the determination manner of the target scene group includes determining, based on an algorithm feature identifier and a feature identifier-scene group mapping relationship of the to-be-tested automatic driving algorithm, the target scene group corresponding to the to-be-tested algorithm image that is constructed by the to-be-tested automatic driving algorithm.

In one embodiment, referring to fig. 4, fig. 4 is a block diagram of a test device of an autopilot algorithm shown in another exemplary embodiment of the present application. The testing device of the automatic driving algorithm further comprises a code warehouse, and the automatic driving algorithm to be tested is obtained from the code warehouse through the obtaining module. The acquisition module is omitted from the drawing. The whole framework is divided into three parts: a code warehouse, a continuous integration tool (a compiling pipeline, an algorithm mirror pipeline and a push mirror pipeline), and a cloud simulation platform. In the continuous integration tool, functions of compiling, testing, mirror image construction, mirror image pushing and the like are realized by building a plurality of flow lines. The cloud simulation platform provides an API interface for the pipeline to call.

It should be noted that the test apparatus for an autopilot algorithm provided in the foregoing embodiment and the test method for an autopilot algorithm provided in the foregoing embodiment fig. 2 belong to the same concept, and specific ways in which the respective modules and units perform operations have been described in detail in the method embodiment, and are not described herein again. In practical applications, the testing device for an autopilot algorithm provided in the above embodiment may distribute the functions to different functional modules as needed, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above, which is not limited herein.

An embodiment of the present application further provides an electronic device, including: one or more processors; a storage device for storing one or more programs, which when executed by the one or more processors, cause the electronic device to implement the test method of the automated driving algorithm provided in the above-described embodiments.

FIG. 5 illustrates a schematic structural diagram of a computer system suitable for use in implementing the electronic device of an embodiment of the present application. It should be noted that the computer system 1100 of the electronic device shown in fig. 5 is only an example, and should not bring any limitation to the functions and the application scope of the embodiments of the present application.

As shown in fig. 5, the computer system 1100 includes a Central Processing Unit (CPU) 1101, which can perform various appropriate actions and processes, such as performing the methods described in the above embodiments, according to a program stored in a Read-Only Memory (ROM) 1102 or a program loaded from a storage section 1108 into a Random Access Memory (RAM) 1103. In the RAM1103, various programs and data necessary for system operation are also stored. The CPU 1101, ROM 1102, and RAM1103 are connected to each other by a bus 1104. An Input/Output (I/O) interface 1105 is also connected to bus 1104.

The following components are connected to the I/O interface 1105: an input portion 1106 including a keyboard, mouse, and the like; an output section 1107 including a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, as well as a speaker and the like; a storage section 1108 including a hard disk and the like; and a communication section 1109 including a Network interface card such as a LAN (Local Area Network) card, a modem, or the like. The communication section 1109 performs communication processing via a network such as the internet. A driver 1110 is also connected to the I/O interface 1105 as necessary. A removable medium 1111, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like, is installed on the drive 1110 as needed, so that the computer program read out therefrom is installed into the storage section 1108 as needed.

In particular, according to embodiments of the present application, the processes described above with reference to the flow diagrams may be implemented as computer software programs. For example, embodiments of the present application include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising a computer program for performing the method illustrated by the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication portion 1109 and/or installed from the removable medium 1111. When the computer program is executed by a Central Processing Unit (CPU) 1101, various functions defined in the system of the present application are executed.

It should be noted that the computer readable medium shown in the embodiments of the present application may be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium may be, for example, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a Read-Only Memory (ROM), an Erasable Programmable Read-Only Memory (EPROM), a flash Memory, an optical fiber, a portable Compact Disc Read-Only Memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present application, a computer-readable signal medium may comprise a propagated data signal with a computer-readable computer program embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. The computer program embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the foregoing.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. Each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The units described in the embodiments of the present application may be implemented by software or hardware, and the described units may also be disposed in a processor. Wherein the names of the elements do not in some way constitute a limitation on the elements themselves.

Another aspect of the present application also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor of a computer, causes the computer to execute the method of testing an autopilot algorithm as described above. The computer-readable storage medium may be included in the electronic device described in the above embodiment, or may exist separately without being incorporated in the electronic device.

Another aspect of the application also provides a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the test method of the autopilot algorithm provided in the above-described embodiments.

The foregoing embodiments are merely illustrative of the principles of the present invention and its efficacy, and are not to be construed as limiting the invention. Those skilled in the art can modify or change the above-described embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention are covered by the claims of the present invention.

Claims

1. A method for testing an autonomous driving algorithm, the method comprising:

acquiring an automatic driving algorithm to be tested;

pushing the algorithm mirror image to be tested to a cloud simulation platform;

determining a target scene group corresponding to the mirror image of the algorithm to be tested constructed by the automatic driving algorithm to be tested based on the algorithm characteristic identification and the characteristic identification-scene group mapping relation of the automatic driving algorithm to be tested;

2. The method for testing an autonomous driving algorithm of claim 1 wherein prior to pushing the algorithm image to be tested to a cloud simulation platform, the method comprises:

3. The method for testing the autopilot algorithm of claim 2 wherein prior to logging in a platform image repository of the cloud simulation platform via the platform image repository access account information and the platform image repository access pass information, the method comprises:

4. The method for testing an automated driving algorithm of claim 3, wherein if the login is successful, before pushing the algorithm image to be tested to the platform image repository of the cloud simulation platform, the method comprises:

associating the algorithm mirror to be tested with the platform mirror identifier;

5. The method for testing an autopilot algorithm of any one of claims 1-4 wherein after constructing the mirror image of the algorithm to be tested, the method further comprises:

6. The method for testing the automatic driving algorithm as claimed in any one of claims 1 to 4, wherein determining the target scene group corresponding to the mirror image of the algorithm to be tested constructed by the automatic driving algorithm to be tested based on the algorithm feature identifier and the feature identifier-scene group mapping relationship of the automatic driving algorithm to be tested comprises:

configuring algorithm characteristic identification of the automatic driving algorithms to be tested based on the test scenes of the automatic driving algorithms to be tested;

7. The method for testing an automated driving algorithm of any one of claims 1 to 4, wherein after the cloud simulation platform executes a simulation task based on the algorithm image to be tested and the target scene group to obtain a simulation result, the method further comprises:

8. A device for testing an autopilot algorithm, the device comprising:

the compiling assembly line is used for compiling the source code of the to-be-tested automatic driving algorithm to obtain a compiling product;

9. An electronic device, characterized in that the electronic device comprises:

one or more processors;

storage means for storing one or more programs which, when executed by the one or more processors, cause the electronic device to carry out the method of any one of claims 1 to 7.

10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor of a computer, causes the computer to carry out the method of any one of claims 1 to 7.