CN115757782A

CN115757782A - Method and device for updating classification rules, electronic equipment and storage medium

Info

Publication number: CN115757782A
Application number: CN202211442070.3A
Authority: CN
Inventors: 王涵
Original assignee: Apollo Zhilian Beijing Technology Co Ltd; Apollo Zhixing Technology Guangzhou Co Ltd
Current assignee: Apollo Zhilian Beijing Technology Co Ltd; Apollo Zhixing Technology Guangzhou Co Ltd
Priority date: 2022-11-17
Filing date: 2022-11-17
Publication date: 2023-03-07

Abstract

The disclosure provides a classification rule updating method and device, electronic equipment and a storage medium, and relates to the technical field of artificial intelligence, in particular to the technical fields of automatic driving, intelligent transportation and the like. The specific implementation scheme is as follows: acquiring description information of a first path measurement problem; matching the description information through the multiple matching models to obtain first matching results corresponding to the multiple matching models, wherein the first matching results are used for indicating whether the first path test problem belongs to the problem category corresponding to the matching models; when the plurality of first matching results indicate that the first path measurement problem does not belong to the corresponding problem category, acquiring a first target problem category set for the first path measurement problem; and updating the original classification rule based on the first target problem category and the description information of the first test problem to obtain a target classification rule, and synchronizing the target classification rule to a plurality of clients. The method realizes timely and accurate updating of the classification rules and lays a foundation for realizing accurate classification of the drive test problem.

Description

Method and device for updating classification rules, electronic equipment and storage medium

Technical Field

The present disclosure relates to the field of artificial intelligence technologies, and in particular, to the field of automatic driving, intelligent transportation, and the like, and in particular, to a method and an apparatus for updating classification rules, an electronic device, and a storage medium.

Background

Before the automatic driving vehicle is put into use, large-scale road test, namely road test, needs to be carried out to ensure the driving safety of the automatic driving vehicle. And the standardized classification of the problems of the vehicles in the road test is an important prepositive work for solving the problems subsequently, so that the problem solving efficiency can be improved.

With the continuous updating of the scene of the road test, the classification rule according to which the problems of the vehicles in the road test are classified changes in real time. How to realize timely and accurate updating of classification rules is a problem to be solved urgently.

Disclosure of Invention

The disclosure provides a classification rule updating method and device, electronic equipment and a storage medium.

According to an aspect of the present disclosure, there is provided a method for updating a classification rule, the method including: acquiring description information of a first routing problem sent by a first client in a plurality of clients; matching the description information of the first path testing problem through a plurality of matching models to obtain first matching results corresponding to the matching models, wherein the first matching results corresponding to the matching models are used for indicating whether the first path testing problem belongs to the problem category corresponding to the matching models; under the condition that first matching results corresponding to the plurality of matching models indicate that the first routing problem does not belong to the corresponding problem category, acquiring a first target problem category set for the first routing problem based on the description information; updating an original classification rule based on the first target problem category and the description information of the first test problem to obtain a target classification rule, and synchronizing the target classification rule to the plurality of clients.

According to another aspect of the present disclosure, there is provided an apparatus for updating a classification rule, the apparatus including: the first acquisition module is used for acquiring description information of a first routing problem sent by a first client side in the plurality of client sides; the first processing module is configured to perform matching processing on the description information of the first path of test question through a plurality of matching models to obtain first matching results corresponding to the plurality of matching models, where the first matching results corresponding to the matching models are used to indicate whether the first path of test question belongs to a question category corresponding to the matching models; a second obtaining module, configured to obtain a first target problem category set for the first road test problem based on the description information when first matching results corresponding to the multiple matching models all indicate that the first road test problem does not belong to a corresponding problem category; and the updating and synchronizing module is used for updating the original classification rule based on the first target problem category and the description information of the first testing problem to obtain a target classification rule, and synchronizing the target classification rule to the plurality of clients.

According to another aspect of the present disclosure, there is provided an electronic device including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the classification rule updating method of the present disclosure.

According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method of updating classification rules disclosed in the embodiments of the present disclosure.

According to another aspect of the present disclosure, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the steps of the method of updating classification rules of the present disclosure.

It should be understood that the statements in this section are not intended to identify key or critical features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.

Drawings

The drawings are included to provide a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:

fig. 1 is a flowchart illustrating a method for updating a classification rule according to a first embodiment of the present disclosure;

fig. 2 is a flowchart illustrating a method for updating classification rules according to a second embodiment of the present disclosure;

fig. 3 is a flowchart illustrating a method for updating classification rules according to a third embodiment of the present disclosure;

fig. 4 is an architecture diagram of an update method of classification rules according to a fourth embodiment of the present disclosure;

fig. 5 is a schematic structural diagram of a classification rule updating apparatus according to a fifth embodiment of the present disclosure;

fig. 6 is a schematic structural diagram of a classification rule updating apparatus according to a sixth embodiment of the present disclosure;

fig. 7 is a block diagram of an electronic device for implementing a method for updating classification rules according to an embodiment of the present disclosure.

Detailed Description

Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, in which various details of the embodiments of the disclosure are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

With the continuous updating of the scene of the road test, the classification rule according to which the problems of the vehicles in the road test are classified changes in real time.

In the related technology, an original classification rule is generally formulated based on historical experience and used for classifying problems occurring in vehicles in the road test, the classification rule is issued to a vehicle end tester in a form of a local table, when the vehicle end tester drives the vehicle, the problem category to which the problem belongs is selected from known problem categories based on the classification rule locally according to the specific performance of the problem occurring in the vehicle so as to record the problem and the problem category to which the problem belongs, and new problems which do not belong to the known problem category are marked as 'other'. After a period of time, creating a new problem category according to all the problems marked as 'other', updating the classification rule, and then re-issuing the new classification rule to the vehicle end tester in a local table form.

In such a manner, from the occurrence of a new problem which does not belong to a known problem category to the synchronization of a new classification rule to all vehicle end testers after the classification rule is updated, a long time is consumed, and the situation that data is changed easily occurs depending on a local table, so that the problem that the updating process of the classification rule has hysteresis and is prone to errors exists.

How to realize timely and accurate updating of classification rules is a problem to be solved urgently.

The disclosed embodiments address the above issues, and provide a method, an apparatus, an electronic device, a non-transitory computer-readable storage medium, and a computer program product for updating a classification rule. The method for updating the classification rules comprises the following steps: acquiring description information of a first routing problem sent by a first client in a plurality of clients; matching the description information of the first path test problem through a plurality of matching models to obtain first matching results corresponding to the plurality of matching models, wherein the first matching results corresponding to the matching models are used for indicating whether the first path test problem belongs to the problem category corresponding to the matching models; under the condition that first matching results corresponding to the multiple matching models indicate that the first path problem does not belong to the corresponding problem category, acquiring a first target problem category set for the first path problem based on the description information; and updating the original classification rule based on the first target problem category and the description information of the first test problem to obtain a target classification rule, and synchronizing the target classification rule to a plurality of clients. Therefore, the classification rules are timely and accurately updated, and a foundation is laid for realizing accurate classification of the drive test problems.

The method and the device for updating the classification rules, the electronic device, the non-transitory computer readable storage medium and the computer program product relate to the technical field of artificial intelligence, in particular to the technical fields of automatic driving, intelligent transportation and the like.

The artificial intelligence is a subject for researching some thinking process and intelligent behaviors (such as learning, reasoning, thinking, planning and the like) of a human by a computer, and has a hardware level technology and a software level technology. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing, and the like; the artificial intelligence software technology mainly comprises computer vision, a voice recognition technology, a natural language processing technology, machine learning/deep learning, a big data processing technology, a knowledge map technology and the like.

A method, an apparatus, an electronic device, a non-transitory computer-readable storage medium, and a computer program product for updating classification rules of embodiments of the present disclosure are described below with reference to the accompanying drawings.

First, a method for updating a classification rule provided in the embodiment of the present disclosure will be described.

Fig. 1 is a flowchart illustrating a method for updating a classification rule according to a first embodiment of the present disclosure. Note that, in the method for updating a classification rule according to the present embodiment, the execution subject is an updating device for a classification rule, and the updating device for a classification rule may be implemented by software and/or hardware. The updating device of the classification rule may be an electronic device with a calculation processing function, or may be integrated in the electronic device, which is not limited in this disclosure. In the embodiment of the present disclosure, an updating apparatus of a classification rule is taken as an electronic device, and the electronic device is taken as a server for example.

As shown in fig. 1, the method for updating the classification rule may include:

step 101, obtaining description information of a first routing problem sent by a first client in a plurality of clients.

The client may be an application installed on the user terminal or a service number used, and the like, which is not limited in this disclosure.

The first client is any client in the plurality of clients.

The drive test problem is a problem occurring in the drive test process of the vehicle. The first routing problem in the embodiment of the present disclosure is a routing problem for a newly added problem category.

The description information of the first path measurement problem is information for describing the concrete expression of the first path measurement problem. For example, in the process of straight running of the vehicle on a straight road, when the front of the vehicle is cut by a pedestrian, the vehicle does not have an avoiding action, and a problem may occur in a planning module; the vehicle is suddenly anchored in the running process; the vehicle enters a non-motorized lane or the like while traveling.

As an example, a user interaction interface may be displayed in a plurality of clients, and when a driver at any vehicle end finds a drive test problem that does not belong to a known problem category during a drive test of a vehicle, the driver may input description information for a first drive test problem of a newly added category at a corresponding position of a used client, for example, a user interaction interface of a first client, and click a button of the user interaction interface having a function of the newly added problem category to trigger a newly added request of the problem category, where the newly added request may carry the description information of the first drive test problem, so that a server may obtain the description information of the first drive test problem sent by the first client in the plurality of clients. Wherein the problem classes are known, i.e. the problem classes already existing in the current classification rules. Here, the current classification rule is an original classification rule. The original classification rule is a rule used for classifying the drive test problem before updating.

102, matching the description information of the first path measurement problem through the multiple matching models to obtain first matching results corresponding to the multiple matching models, wherein the first matching results corresponding to the matching models are used for indicating whether the first path measurement problem belongs to the problem category corresponding to the matching models.

The matching model can be any neural network model for data matching, and can judge whether different data are matched or not based on the characteristics of different data.

The problem category is a category to which the drive test problem belongs, and may be an algorithm strategy category, a hardware fault category, a poor motion sensing category, a collision receiving pipe category, a non-collision receiving pipe category, and the like, which is not limited by the present disclosure.

In the embodiment of the present disclosure, for each problem category in the original classification rule, a corresponding matching model may be generated in advance, where the matching model is used to perform matching processing on any drive test problem to obtain a matching result corresponding to the matching model, and the matching result indicates whether the any drive test problem belongs to the problem category corresponding to the matching model.

The plurality of matching models in the embodiment of the present disclosure include matching models corresponding to problem categories in the original classification rule.

And matching the description information of the first path of test problems through the plurality of matching models to obtain first matching results corresponding to the plurality of matching models, so that whether the first path of test problems belongs to the known problem category in the original classification rule can be judged.

Step 103, under the condition that the first matching results corresponding to the multiple matching models all indicate that the first routing problem does not belong to the corresponding problem category, acquiring a first target problem category set for the first routing problem based on the description information.

The first target problem category may be manually set for the first path measurement problem based on description information of the first path measurement problem, or may be automatically generated by the server according to the description information of the first path measurement problem by using a preset policy, which is not limited by the present disclosure.

In the embodiment of the present disclosure, in a case that the first matching results corresponding to the multiple matching models all indicate that the first path problem does not belong to the corresponding problem category, it indicates that the first path problem does not belong to the known problem category in the original classification rule, so that the first target problem category set for the first path problem based on the description information may be obtained.

As an example, the server may display description information of the first routing problem on a provided human-computer interaction interface when it is determined that the first matching results corresponding to the multiple matching models all indicate that the first routing problem does not belong to the corresponding problem category, and obtain a first target problem category set for the first routing problem in response to obtaining the description information of the first routing problem input on the human-computer interaction interface.

And step 104, updating the original classification rule based on the first target problem category and the description information of the first testing problem to obtain a target classification rule, and synchronizing the target classification rule to a plurality of clients.

The original classification rule is a rule used for classifying the drive test problem before updating. The original classification rule may include a plurality of problem categories, category description information corresponding to each problem category, description information of the drive test problem as an example, and the like, which is not limited in this disclosure. The category description information corresponding to the question category may describe the question category to which the question category is applicable.

The target classification rule is an updated rule for classifying the drive test problem.

As an example, the server may supplement the description information of the first target problem category and the first test problem to the original classification rule, so as to obtain the target classification rule.

As an example, the server may obtain, while obtaining the first target problem category, category description information set for the first target problem category, where the category description information may describe what problem category the first target problem category is applicable to, and may further supplement, to the original classification rule, the first target problem category, the corresponding category description information, and the description information of the first routing problem, so as to obtain the target classification rule.

After the target classification rule is obtained, the server can synchronize the target classification rule to a plurality of clients in real time.

Therefore, after a vehicle end tester finds a first path test problem which does not belong to a known problem category, description information of the first path test problem can be sent to a server through a first client in real time, the server can judge whether the first path test problem belongs to the known problem category in an original classification rule or not in real time, under the condition that the first path test problem does not belong to the known problem category in the original classification rule, a first target problem category set for the first path test problem is quickly obtained, the original classification rule is updated based on the first target problem category and the description information of the first path test problem, and the updated target classification rule is synchronized to a plurality of clients in real time, so that timely updating of the classification rule can be achieved, and updating of the classification rule based on the description information of the path test problem which substantially belongs to the known problem category is avoided. Through the interaction mode of the client and the server, the accurate transmission of information can be realized, so that the accurate updating of the classification rules is ensured. Moreover, the manual intervention is reduced, and the labor cost is saved. In addition, in the embodiment of the disclosure, when a drive test problem that does not belong to a known problem category occurs, a problem category is newly added in the server, and the updated classification rule is synchronized to all the clients, so that unified definition of the newly added problem category is realized, the problem that categories which are identical in nature but different in description are newly added to the classification rule due to artificial subjective judgment is avoided, and unified management of the classification rule is realized.

In summary, according to the method for updating the classification rule provided by the embodiment of the present disclosure, description information of a first path test problem sent by a first client in a plurality of clients is obtained, and the description information of the first path test problem is matched through a plurality of matching models to obtain first matching results corresponding to the plurality of matching models, where the first matching results corresponding to the matching models are used to indicate whether the first path test problem belongs to a problem category corresponding to the matching models, and in a case where the first matching results corresponding to the plurality of matching models all indicate that the first path test problem does not belong to a corresponding problem category, a first target problem category set for the first path test problem based on the description information is obtained, and an original classification rule is updated based on the first target problem category and the description information of the first path test problem to obtain a target classification rule, and the target classification rule is synchronized to the plurality of clients. Therefore, the classification rules are timely and accurately updated, and a foundation is laid for realizing accurate classification of the drive test problems.

As can be seen from the above analysis, in the embodiment of the present disclosure, the description information of the first path test problem may be subjected to matching processing by using a plurality of matching models, so as to obtain a first matching result corresponding to the plurality of matching models. With reference to fig. 2, a process of performing matching processing on description information of a first path test problem through multiple matching models in the method for updating a classification rule provided in the embodiment of the present disclosure is further described below.

Fig. 2 is a flowchart illustrating a method for updating a classification rule according to a second embodiment of the present disclosure. As shown in fig. 2, the method for updating the classification rule may include the following steps:

step 201, obtaining description information of a first path test problem sent by a first client in a plurality of clients.

For a specific implementation process and principle of step 201, reference may be made to descriptions of other embodiments, which are not described herein again.

Step 202, for each matching model, inputting description information of the first road test problem into the matching model, so as to perform feature extraction on the description information based on a plurality of preset road test problem feature items, and obtain target feature values of the plurality of road test problem feature items.

The plurality of drive test problem feature items are feature items for feature extraction of description information of the drive test problem, and may be set according to the needs of an application scenario, which is not limited by the present disclosure.

In the embodiment of the disclosure, for each matching model, description information of a first drive test problem may be input into the matching model, so as to perform feature extraction on the description information based on a plurality of preset drive test problem feature items, and obtain target feature values of the plurality of drive test problem feature items.

The plurality of drive test problem characteristic items may include, for example, at least two of the following: the system comprises a main vehicle behavior module, a road structure module, an obstacle type module, an obstacle direction module, an obstacle action module and a problem belonging module.

The behavior of the main vehicle, that is, the behavior of the test vehicle, may include straight traveling, left turning, right turning, reversing, and the like.

The road structure, i.e., the structure of the road on which the vehicle travels at the time of the drive test, may include a straight road, a curved road, an intersection, and the like.

The type of the obstacle, i.e., the type of the obstacle around the vehicle at the time of the drive test, may include the types of a pedestrian, a vehicle, a building, and the like.

The obstacle orientation, that is, the orientation of an obstacle around the vehicle with respect to the vehicle at the time of drive test, may include being located in front of, behind, in front of, and in front of the vehicle.

The obstacle motion, that is, the motion of an obstacle around the vehicle at the time of the drive test, may include cut-in, overtaking, crossing, going straight, turning, and the like.

And the module to which the problem belongs, namely the module to which the problem occurs in the vehicle during the drive test belongs, such as a planning module and a path optimization module of the vehicle.

Taking the example that the plurality of drive test problem characteristic items comprise a main vehicle behavior, a road structure, an obstacle type, an obstacle direction, an obstacle action and a module to which the problem belongs, for any matching model, after the description information of the first drive test problem is input into the any matching model, the description information can be subjected to characteristic extraction by the any matching model, and the characteristic values of the drive test problem characteristic items of the main vehicle behavior, the road structure, the obstacle type, the obstacle direction, the obstacle action and the module to which the problem belongs are obtained.

For example, if the description information of the first path measurement problem is that, in the process of the vehicle going straight on a straight road, the vehicle encounters a cut-in of a pedestrian in front of the vehicle, the vehicle does not have an avoidance action, and possibly a problem occurs in the planning module, the description information of the first path measurement problem is input into a certain matching model, and then the characteristic values "straight going" of the characteristic values "straight road" of the "main vehicle action" and "road structure", the characteristic values "pedestrian" of the "road structure", the characteristic values "front" of the "obstacle orientation", the "cut-in" of the characteristic values "obstacle action" and "planning module" of the characteristic values "of the" obstacle action "of the" main vehicle action "of the characteristic item of the path measurement problem can be obtained.

By setting a plurality of drive test problem characteristic items including drive test problem characteristic items such as main vehicle behaviors, road structures, obstacle types, obstacle directions, obstacle actions and modules to which problems belong, the detailed description of the drive test problems can be detailed in a plurality of aspects such as driving information of vehicles, environmental information around the vehicles and modules to which the problems occurring in the vehicles belong, and a foundation is laid for accurately judging whether the first drive test problem belongs to the problem category corresponding to the matching model.

And 203, matching the target characteristic values with the reference characteristic values of the matching models corresponding to the drive test problem characteristic items to obtain the matching degrees of the target characteristic values and the reference characteristic values.

In the embodiment of the present disclosure, for the matching model corresponding to each problem category in the original classification rule, reference feature values of a plurality of drive test problem feature items may be set for the matching model in advance, so that the plurality of target feature values and the plurality of reference feature values corresponding to the matching model may be matched according to the corresponding drive test problem feature items through each matching model, and a matching degree between the plurality of target feature values and the plurality of reference feature values is obtained.

The matching degree between the target feature values and the reference feature values may represent the matching degree between the target feature values and the reference feature values.

And 204, determining a first matching result corresponding to the matching model according to the matching degree.

And the first matching result corresponding to the matching model is used for indicating whether the first path problem belongs to the problem category corresponding to the matching model.

In the embodiment of the present disclosure, a matching threshold may be set in advance as needed, and when the matching degree between the plurality of target feature values and the plurality of reference feature values is greater than the matching threshold, it may be determined that the plurality of target feature values match the plurality of reference feature values, so as to obtain a first matching result indicating that the first routing problem belongs to a problem category corresponding to the matching model. Under the condition that the matching degree of the target characteristic values and the reference characteristic values is not larger than the matching threshold value, the target characteristic values and the reference characteristic values can be determined to be not matched, and therefore a matching result indicating that the first path measurement problem does not belong to the problem category corresponding to the matching model is obtained.

Through each matching model, based on a plurality of preset drive test problem feature items, feature extraction is carried out on the description information of the first drive test problem to obtain target feature values of the plurality of drive test problem feature items, quantification of the description information of the first drive test problem can be achieved, and detailed description of the drive test problem is achieved from multiple aspects, so that the matching model can accurately judge whether the first drive test problem belongs to the problem category corresponding to the matching model based on the target feature values and the reference feature values of the plurality of drive test problem feature items.

Step 205, under the condition that the first matching results corresponding to the multiple matching models all indicate that the first routing problem does not belong to the corresponding problem category, obtaining a first target problem category set for the first routing problem based on the description information.

Step 206, updating the original classification rule based on the first target problem category and the description information of the first test problem to obtain a target classification rule, and synchronizing the target classification rule to a plurality of clients.

For the specific implementation process and principle of steps 205-206, reference may be made to the description of other embodiments, which are not described herein again.

It can be understood that after the target classification rule is obtained, the target classification rule may need to be further updated, and when the target classification rule is updated again, whether the drive test problem belongs to a known problem category in the target classification rule needs to be determined through a plurality of matching models based on the obtained description information of the drive test problem for the newly added category. Then, in the embodiment of the present disclosure, in order to determine whether the drive test problem belongs to the known first target problem category in the target classification rule, a matching model corresponding to the first target problem category needs to be generated.

That is, after step 205, the method may further include:

and generating a matching model corresponding to the first target problem category based on the description information of the first path measurement problem.

The matching model corresponding to the first target problem category may be used to perform matching processing on description information of any drive test problem sent by any client of the multiple clients to obtain a third matching result, where the third matching result is used to indicate whether the any drive test problem belongs to the first target problem category.

In the embodiment of the disclosure, feature analysis may be performed on the description information of the first road test problem, feature extraction may be performed to obtain feature values of a plurality of road test problem feature items, an occurrence scene of the first road test problem is analyzed based on data features of the feature values, the description information of the plurality of road test problems in the scene is used as a sample, an initial matching model is iteratively updated to obtain a matching model corresponding to the first target problem category, and a reference feature value of the matching model corresponding to the plurality of road test problem feature items is generated.

Therefore, after the description information of any drive test problem is acquired, whether the drive test problem belongs to the first target problem category or not can be judged through the matching model corresponding to the first target problem category.

In addition, in the embodiment of the disclosure, the classification rules can be directly updated at the server side, and the updated classification rules are synchronized to the plurality of clients, so that the latest classification rules are timely and accurately synchronized to the plurality of clients.

Specifically, the server may provide a classification rule configuration page for a maintainer of the classification rule to configure the classification rule. When the classification rule needs to be directly updated at the server side, a maintainer can input the problem category (referred to as a third target problem category in the embodiment of the present disclosure) to be newly added and the description information of the drive test problem (referred to as a third drive test problem in the embodiment of the present disclosure) corresponding to the problem category as an example at a corresponding position of a classification rule configuration page, and click a button with a classification rule updating function to trigger a classification rule updating request, where the classification rule updating request includes the problem category and the description information of the corresponding drive test problem. Correspondingly, the server can respond to a classification rule updating request triggered by the classification rule configuration page, update the current classification rule according to the third target problem category and the description information of the corresponding third testing problem included in the classification rule updating request, and synchronize the updated classification rule to the plurality of clients.

That is, the method for updating the classification rule provided in the embodiment of the present disclosure may further include:

responding to a classification rule updating request triggered by a classification rule configuration page, and updating an original classification rule or a target classification rule according to a third target problem category and the description information of a corresponding third testing problem included in the classification rule updating request; and synchronizing the updated classification rules to the plurality of clients.

It should be noted that, in the above embodiment, the addition of the new problem category is directly performed at the server end, which is taken as an example to illustrate, in practical application, the existing problem category may also be directly modified or deleted at the server end, and correspondingly, the classification rule update request may include the target problem category that needs to be modified or deleted and the description information of the corresponding drive test problem, so that the server may respond to the classification rule update request to modify or delete the target problem category that needs to be modified or deleted and the description information of the corresponding drive test problem. Moreover, after the server modifies or deletes a certain target problem category and the description information of the corresponding drive test problem, the modification or deletion operation can be synchronized to the drive test problem associated with the target problem category, so as to realize the data consistency.

In summary, according to the method for updating the classification rule provided in the embodiment of the present disclosure, description information of a first routing problem sent by a first client in a plurality of clients is obtained, for each matching model, the description information of the first routing problem is input into the matching model, so as to perform feature extraction on the description information based on a plurality of preset routing problem feature items, obtain target feature values of the plurality of routing problem feature items, match the plurality of target feature values with reference feature values of the matching model corresponding to the plurality of routing problem feature items, obtain matching degrees of the plurality of target feature values and the plurality of reference feature values, determine a first matching result corresponding to the matching model according to the matching degrees, obtain a first target problem category set for the first routing problem based on the description information under the condition that the first matching result corresponding to the plurality of matching models indicates that the first routing problem does not belong to the corresponding problem category, update the original classification rule based on the first target problem category and the description information of the first routing problem, obtain the target classification rule, and synchronize the target classification rule to the plurality of clients. Therefore, the classification rules are timely and accurately updated, and a foundation is laid for realizing accurate classification of the drive test problems.

In a possible implementation form, the server may further store the problems occurring in the vehicle drive test process and the problem types to which the problems belong, and record the drive test problems and the problem types to which the problems belong. In view of the above situation, the method for updating the classification rule provided in the embodiment of the present disclosure is further described with reference to fig. 3.

Fig. 3 is a flowchart illustrating a method for updating classification rules according to a third embodiment of the present disclosure. As shown in fig. 3, the method for updating the classification rule may include the following steps:

step 301, obtaining description information of a first routing problem sent by a first client in a plurality of clients.

Step 302, performing matching processing on the description information of the first path measurement problem through the multiple matching models to obtain first matching results corresponding to the multiple matching models, where the first matching results corresponding to the matching models are used to indicate whether the first path measurement problem belongs to a problem category corresponding to the matching models.

Step 303, under the condition that the first matching results corresponding to the multiple matching models all indicate that the first routing problem does not belong to the corresponding problem category, obtaining a first target problem category set for the first routing problem based on the description information.

Step 304, updating the original classification rule based on the first target problem category and the description information of the first test problem to obtain a target classification rule, and synchronizing the target classification rule to a plurality of clients.

For the specific implementation processes and principles of steps 301 to 304, reference may be made to descriptions of other embodiments, which are not described herein again.

Step 305, obtaining description information of the second routing problem sent by the second client in the plurality of clients and a problem category to which the second routing problem belongs.

It should be noted that, steps 305 to 307 may be executed before step 301, may be executed simultaneously with steps 301 to 304, or may be executed after step 304, which is not limited by the present disclosure. The disclosed embodiment exemplifies that steps 305-307 are performed after step 304.

The second client is any client in the plurality of clients.

The description information of the second routing problem is information for describing the concrete performance of the second routing problem. For example, in the process that the vehicle moves straight on a straight road, when a pedestrian cuts into the front of the vehicle, the vehicle does not have an avoiding action, and a problem may occur in a planning module; the vehicle is suddenly anchored in the running process; the vehicle enters a non-motorized lane or the like while traveling.

The problem category to which the second path measurement problem belongs may be an algorithm strategy category, a hardware fault category, a poor motion feeling category, a collision connection pipe category, a non-collision connection pipe category, and the like, which is not limited by the present disclosure.

As an example, when a vehicle-end tester finds a drive test problem belonging to a known problem category in a drive test process of a vehicle, a user interaction interface may be displayed in a plurality of clients, and when a used client, for example, a corresponding position of the user interaction interface of a second client, inputs description information of a second drive test problem to be stored and a problem category to which the second drive test problem belongs, and clicks a button having a problem recording function of the user interaction interface to trigger a recording request of the drive test problem, where the recording request may carry description information of the second drive test problem and a problem category to which the second drive test problem selected by the vehicle-end tester from a plurality of known problem categories displayed by the user interaction interface, so that a server may obtain the description information of the second drive test problem sent by the second client in the plurality of clients and the problem category to which the second drive test problem belongs. Wherein the problem classes are known, i.e. the problem classes already existing in the current classification rules. Wherein the current classification rule is the target classification rule when the steps 305-307 are executed after the step 304, and the current classification rule is the original classification rule when the steps 305-307 are executed simultaneously with the steps 301-304 or before the step 301.

Step 306, performing consistency check on the description information of the second routing problem and the problem category to which the second routing problem belongs.

In the embodiment of the present disclosure, consistency check may be performed on the description information of the second routing problem and the problem category to which the second routing problem belongs, to determine whether the description information of the second routing problem belongs to the problem category, if yes, the consistency check is passed, and otherwise, the consistency check is not passed.

In an embodiment of the present disclosure, step 306 may be implemented by: matching the description information of the second road test problem through the plurality of matching models to obtain second matching results corresponding to the plurality of matching models, wherein the second matching results corresponding to the matching models are used for indicating whether the second road test problem belongs to the problem category corresponding to the matching models; determining a second target problem category to which the second routing test problem belongs according to the problem category corresponding to at least one matching model under the condition that the second matching result corresponding to at least one matching model in the second matching results corresponding to the multiple matching models indicates that the second routing test problem belongs to the problem category corresponding to at least one matching model; when the second target problem type is consistent with the problem type sent by the second client, the consistency check is determined to be passed; and determining that the consistency check is not passed under the condition that the second target question category is inconsistent with the question category sent by the second client.

Therefore, the description information of the second routing problem and the problem category to which the second routing problem belongs can be accurately checked for consistency.

Wherein, when steps 305-307 are executed after step 304, the plurality of matching models includes a matching model corresponding to the first target question category and a matching model corresponding to other question categories in the target classification rule.

The process of matching the description information of the second routing problem through the multiple matching models may refer to the process of matching the description information of the first routing problem through the multiple matching models in other embodiments, and is not described here again.

And matching the description information of the second road test problem through the plurality of matching models to obtain second matching results corresponding to the plurality of matching models, namely judging whether the second road test problem belongs to the known problem category in the target classification rule.

In this embodiment of the present disclosure, in a case that the second matching result corresponding to one or more matching models indicates that the second routing problem belongs to the problem category corresponding to the one or more matching models, the problem category corresponding to the one or more matching models may be determined as a second target problem category to which the second routing problem belongs.

In the embodiment of the present disclosure, when there is a problem category that is the same as the problem category sent by the second client in the one or more second target problem categories, it may be determined that the second target problem category is consistent with the problem category sent by the second client.

In the embodiment of the present disclosure, in a case where one or more second target question categories are all different from the question category sent by the second client, it may be determined that the second target question categories are not consistent with the question categories sent by the second client.

And 307, correspondingly storing the description information of the second routing problem and the problem type sent by the second client under the condition that the consistency check is passed.

In summary, the method for updating the classification rule provided by the embodiment of the present disclosure realizes timely and accurate update of the classification rule, and lays a foundation for realizing accurate classification of the drive test problem. And the description information of the second road test problem and the problem category to which the second road test problem belongs are checked for consistency by acquiring the description information of the second road test problem and the problem category to which the second road test problem belongs, and the description information of the second road test problem and the problem category to which the second road test problem belongs are correspondingly stored under the condition that the consistency check is passed, so that the problem occurring in the vehicle road test process and the problem category to which the problem belongs are correspondingly stored in the server to record the road test problem and the problem category to which the problem belongs, the loss of data is avoided, and the description information of the road test problem and the problem category to which the problem belongs are checked for consistency before storage, so that the description information of the road test problem and the problem category to which the problem belongs are accurately stored.

The following describes an architecture of the method for updating the classification rule provided in the embodiment of the present disclosure with reference to fig. 4.

The method for updating the classification rule provided by the embodiment of the present disclosure may be applied to the server 410, and the server 410 may interact with the client 420. Wherein fig. 4 illustrates an example of one of the plurality of clients.

Referring to fig. 4, a classification rule database 4101, a matching model database 4102, and a drive test problem database 4103 may be included in the server 410. The classification rule database 4101 may store a classification rule for classifying the drive test problem, where the classification rule includes a plurality of known problem categories, and each problem category may have corresponding category description information. The matching model database 4102 may store a plurality of matching models and reference feature values of a plurality of preset drive test problem feature items corresponding to the matching models. The drive test problem database 4103 may store description information of a plurality of drive test problems and problem categories to which each drive test problem belongs.

The client 420 may display a user interaction interface, and when a vehicle end tester using the client 420 finds a drive test problem that does not belong to a known problem category during a drive test of a vehicle, the vehicle end tester may input description information for a first drive test problem of a new category at a corresponding position of the user interaction interface of the client 420, and click a button of the user interaction interface having a function of the new category to trigger a category new module 401 in the client 420, and send a new request of the problem category to the server 410, where the new request may carry drive test problem data 4104 including description information of the first drive test problem.

Further, after the server 410 acquires the drive test problem data 4104 sent by the client 420, the checksum matching model generating module 402 may perform matching processing on the description information of the first drive test problem in the drive test problem data 4104 through a plurality of matching models in the matching model database 4102 to obtain first matching results corresponding to the plurality of matching models, and in a case that the first matching results corresponding to the plurality of matching models all indicate that the first drive test problem does not belong to the corresponding problem category, the server 410 may acquire a first target problem category set for the first drive test problem based on the description information, and update the classification rule module 403 based on the first target problem category and the description information of the first drive test problem to obtain a target classification rule, and synchronize the target classification rule to the client 420 and other clients through the classification rule synchronizing module 404. In addition, after the first target problem category is obtained, a matching model corresponding to the first target problem category may be generated by the checksum matching model generation module 402, and the matching model and related data may be stored in the matching model database 4102 for subsequent verification of the drive test problem and the problem category to which the drive test problem belongs.

Therefore, timely and accurate updating of the classification rules is achieved, a foundation is laid for accurate classification of the drive test problems, and updating of the classification rules based on description information of the drive test problems which substantially belong to known problem categories is avoided. Moreover, manual intervention is reduced, and labor cost is saved. When the drive test problem which does not belong to the known problem category occurs, the problem category is newly added in the server, and the updated classification rule is synchronized to all the client sides, so that the unified definition of the newly added problem category is realized, the phenomenon that the categories which are identical in nature but different in description are newly added into the classification rule due to artificial subjective judgment is avoided, and the unified management of the classification rule is realized.

When a vehicle end tester using the client 420 finds a drive test problem belonging to a known problem category during a drive test of a vehicle, the vehicle end tester may input description information of a second drive test problem to be stored at a corresponding position of a user interface of the client 420, and select a problem category to which the second drive test problem belongs from a plurality of known problem categories displayed on the user interface. Accordingly, the drive test problem recording module 405 in the client 420 may obtain the description information of the second drive test problem, and the category recording module 406 may obtain the category of the problem to which the second drive test problem belongs. Further, a vehicle end tester using the client 420 may click a button with a problem recording function of the user interaction interface to trigger the data upload module 407 of the client 420 to trigger a recording request of the drive test problem, where the recording request may carry description information of the second drive test problem and a problem category to which the second drive test problem belongs.

Further, after the server 410 obtains the description information of the second routing problem and the problem category to which the second routing problem belongs, which are sent by the client 420, the checking module 408 may perform consistency check on the description information of the second routing problem and the problem category to which the second routing problem belongs, and correspondingly store the description information of the second routing problem and the problem category sent by the second client in the routing problem database 4103 when the consistency check passes.

Therefore, the problems occurring in the vehicle drive test process and the types of the problems belonging to the problems are stored in the server 410, the drive test problems and the types of the problems belonging to the problems are recorded, data loss is avoided, the description information of the drive test problems and the types of the problems belonging to the problems are verified before storage, and therefore the description information of the drive test problems and the types of the problems belonging to the problems are accurately stored.

Next, an updating apparatus of the classification rule provided in the present disclosure will be described with reference to fig. 5.

Fig. 5 is a schematic structural diagram of a classification rule updating apparatus according to a fifth embodiment of the present disclosure.

As shown in fig. 5, the present disclosure provides an apparatus 500 for updating a classification rule, including: a first obtaining module 501, a first processing module 502, a second obtaining module 503, and an updating and synchronizing module 504.

The first obtaining module 501 is configured to obtain description information of a first routing problem sent by a first client in the multiple clients;

the first processing module 502 is configured to perform matching processing on the description information of the first road test question through the multiple matching models to obtain first matching results corresponding to the multiple matching models, where the first matching results corresponding to the matching models are used to indicate whether the first road test question belongs to a question category corresponding to the matching models;

a second obtaining module 503, configured to obtain a first target problem category set for the first routing problem based on the description information when the first matching results corresponding to the multiple matching models all indicate that the first routing problem does not belong to the corresponding problem category;

an updating and synchronizing module 504, configured to update the original classification rule based on the first target problem category and the description information of the first testing problem, to obtain a target classification rule, and synchronize the target classification rule to multiple clients.

It should be noted that the updating apparatus 500 for the classification rule provided in this embodiment may execute the method for updating the classification rule of the foregoing embodiment. Wherein, the updating means 500 of the classification rule can be implemented by software and/or hardware. The updating apparatus 500 of the classification rule may be an electronic device with a calculation processing function, or may be integrated in the electronic device, which is not limited by the present disclosure. In the embodiment of the present disclosure, the updating apparatus 500 of the classification rule is taken as an electronic device, and the electronic device is taken as a server for example.

It should be noted that the foregoing description of the embodiment of the method for updating the classification rule is also applicable to the apparatus for updating the classification rule provided in the present disclosure, and is not repeated herein.

The updating device for the classification rules, provided by the embodiment of the present disclosure, obtains description information of a first path measurement problem sent by a first client in a plurality of clients, performs matching processing on the description information of the first path measurement problem through a plurality of matching models, and obtains first matching results corresponding to the plurality of matching models, where the first matching results corresponding to the matching models are used to indicate whether the first path measurement problem belongs to a problem category corresponding to the matching models, and when the first matching results corresponding to the plurality of matching models all indicate that the first path measurement problem does not belong to a corresponding problem category, obtains a first target problem category set for the first path measurement problem based on the description information, and updates an original classification rule based on the first target problem category and the description information of the first path measurement problem, so as to obtain a target classification rule, and synchronizes the target classification rule to the plurality of clients. Therefore, timely and accurate updating of the classification rules is realized, and a foundation is laid for accurate classification of the drive test problem.

The updating device of the classification rule provided by the present disclosure is further described below with reference to fig. 6.

Fig. 6 is a schematic structural diagram of an updating apparatus for classification rules according to a sixth embodiment of the present disclosure.

As shown in fig. 6, the present disclosure provides an apparatus 600 for updating a classification rule, including: a first obtaining module 601, a first processing module 602, a second obtaining module 603, and an updating and synchronizing module 604. The first obtaining module 601, the first processing module 602, the second obtaining module 603, and the updating and synchronizing module 604 in fig. 6 have the same functions and structures as the first obtaining module 501, the first processing module 502, the second obtaining module 503, and the updating and synchronizing module 504 in fig. 5.

It should be noted that, for the detailed description of the first obtaining module 601, the first processing module 602, the second obtaining module 603, and the updating and synchronizing module 604, reference may be made to the description of the first obtaining module 501, the first processing module 502, the second obtaining module 503, and the updating and synchronizing module 504 in fig. 5, and the description is not repeated here.

In an implementation of the present disclosure, the first processing module 602 includes:

the characteristic extraction unit is used for inputting the description information of the first road test problem into the matching models for each matching model so as to extract the characteristics of the description information based on a plurality of preset road test problem characteristic items and obtain target characteristic values of the plurality of road test problem characteristic items;

the first matching unit is used for matching the target characteristic values with the reference characteristic values of the matching models corresponding to the drive test problem characteristic items to obtain the matching degrees of the target characteristic values and the reference characteristic values;

and the first determining unit is used for determining a first matching result corresponding to the matching model according to the matching degree.

In implementations of the present disclosure, the plurality of drive test problem characteristic items includes at least two of: the system comprises a main vehicle behavior module, a road structure module, an obstacle type module, an obstacle direction module, an obstacle action module and a problem belonging module.

In an implementation of the present disclosure, the apparatus 600 for updating the classification rule further includes:

a third obtaining module 605, configured to obtain description information of a second routing problem sent by a second client in the multiple clients and a problem category to which the second routing problem belongs;

a second processing module 606, configured to perform consistency check on the description information of the second routing problem and the problem category to which the second routing problem belongs;

the storage module 607 is configured to, in a case that the consistency check passes, correspondingly store the description information of the second routing problem and the problem category sent by the second client.

In implementations of the present disclosure, a second processing module includes:

the second matching unit is used for matching the description information of the second road test problem through the multiple matching models to obtain second matching results corresponding to the multiple matching models, wherein the second matching results corresponding to the matching models are used for indicating whether the second road test problem belongs to the problem category corresponding to the matching models;

a second determining unit, configured to determine, according to the problem category corresponding to the at least one matching model, a second target problem category to which the second routing problem belongs, when, among second matching results corresponding to the multiple matching models, the second matching result corresponding to the at least one matching model indicates that the second routing problem belongs to the problem category corresponding to the at least one matching model;

a third determining unit, configured to determine that the consistency check passes when the second target problem category is consistent with the problem category sent by the second client;

and the fourth determining unit is used for determining that the consistency check is not passed under the condition that the second target question category is inconsistent with the question category sent by the second client.

the updating module is used for responding to a classification rule updating request triggered by a classification rule configuration page, and updating the original classification rule or the target classification rule according to a third target problem category and the description information of a corresponding third testing problem included in the classification rule updating request;

and the synchronization module is used for synchronizing the updated classification rules to the plurality of clients.

The updating device for the classification rules, provided by the embodiment of the present disclosure, obtains description information of a first routing problem sent by a first client in a plurality of clients, performs matching processing on the description information of the first routing problem through a plurality of matching models, and obtains first matching results corresponding to the plurality of matching models, where the first matching results corresponding to the matching models are used to indicate whether the first routing problem belongs to a problem category corresponding to the matching model, and when the first matching results corresponding to the plurality of matching models all indicate that the first routing problem does not belong to a corresponding problem category, obtains a first target problem category set for the first routing problem based on the description information, updates an original classification rule based on the first target problem category and the description information of the first routing problem, obtains a target classification rule, and synchronizes the target classification rule to the plurality of clients. Therefore, the classification rules are timely and accurately updated, and a foundation is laid for realizing accurate classification of the drive test problems.

Based on the above embodiment, the present disclosure also provides an electronic device, including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of updating classification rules of the present disclosure.

Based on the above embodiments, the present disclosure also provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the method for updating the classification rules disclosed in the embodiments of the present disclosure.

Based on the above embodiments, the present disclosure also provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the method for updating classification rules of the present disclosure.

The present disclosure also provides an electronic device and a readable storage medium and a computer program product according to embodiments of the present disclosure.

FIG. 7 shows a schematic block diagram of an example electronic device 700 that may be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processors, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 7, the electronic device 700 may include a computing unit 701 that may perform various appropriate actions and processes according to a computer program stored in a read-only memory (RO) 702 or a computer program loaded from a storage unit 708 into a Random Access Memory (RAM) 703. In the RAM703, various programs and data required for the operation of the device 700 can also be stored. The computing unit 701, the ROM702, and the RAM703 are connected to each other by a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.

Various components in the device 700 are connected to the I/O interface 705, including: an input unit 706 such as a keyboard, a mouse, or the like; an output unit 707 such as various types of displays, speakers, and the like; a storage unit 708 such as a magnetic disk, optical disk, or the like; and a communication unit 709 such as a network card, modem, wireless communication transceiver, etc. The communication unit 709 allows the device 700 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.

Computing unit 701 may be a variety of general purpose and/or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various dedicated Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, and so forth. The calculation unit 701 performs the respective methods and processes described above, such as the update method of the classification rule. For example, in some embodiments, the method of updating classification rules may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as storage unit 708. In some embodiments, part or all of a computer program may be loaded onto and/or installed onto device 700 via ROM702 and/or communications unit 709. When the computer program is loaded into the RAM703 and executed by the computing unit 701, one or more steps of the method for updating classification rules described above may be performed. Alternatively, in other embodiments, the computing unit 701 may be configured in any other suitable way (e.g. by means of firmware) to perform the update method of the classification rules.

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), system on a chip (SOCs), complex Programmable Logic Devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on 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 or flash memory), an optical fiber, a compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a 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 acoustic, speech, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), the internet, and blockchain networks.

The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The Server may be a cloud Server, also called a cloud computing Server or a cloud host, and is a host product in a cloud computing service system, so as to solve the defects of high management difficulty and weak service extensibility in a traditional physical host and VPS service ("Virtual Private Server", or "VPS" for short). The server may be a cloud server, a server of a distributed system, or a server with a combined blockchain.

It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present disclosure may be executed in parallel, sequentially, or in different orders, as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved, and the present disclosure is not limited herein.

The above detailed description should not be construed as limiting the scope of the disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present disclosure should be included in the scope of protection of the present disclosure.

Claims

1. A method for updating classification rules, wherein the method comprises:

acquiring description information of a first path problem sent by a first client in a plurality of clients;

matching the description information of the first path testing problem through a plurality of matching models to obtain first matching results corresponding to the matching models, wherein the first matching results corresponding to the matching models are used for indicating whether the first path testing problem belongs to the problem category corresponding to the matching models;

under the condition that first matching results corresponding to the plurality of matching models indicate that the first path testing problem does not belong to the corresponding problem category, acquiring a first target problem category set for the first path testing problem based on the description information;

updating an original classification rule based on the first target problem category and the description information of the first test problem to obtain a target classification rule, and synchronizing the target classification rule to the plurality of clients.

2. The method according to claim 1, wherein the matching the description information of the first path problem through a plurality of matching models to obtain first matching results corresponding to the plurality of matching models includes:

for each matching model, inputting description information of the first routing problem into the matching model, and performing feature extraction on the description information based on a plurality of preset routing problem feature items to obtain target feature values of the plurality of routing problem feature items;

matching the target characteristic values with the reference characteristic values of the matching models corresponding to the drive test problem characteristic items to obtain matching degrees of the target characteristic values and the reference characteristic values;

and determining a first matching result corresponding to the matching model according to the matching degree.

3. The method of claim 2, wherein the plurality of drive test problem signature terms includes at least two of: the system comprises a main vehicle behavior module, a road structure module, an obstacle type module, an obstacle direction module, an obstacle action module and a problem belonging module.

4. The method according to any one of claims 1-3, wherein the method further comprises:

acquiring description information of a second routing problem sent by a second client in the plurality of clients and a problem category to which the second routing problem belongs;

carrying out consistency check on the description information of the second routing problem and the problem category to which the second routing problem belongs;

and correspondingly storing the description information of the second routing test problem and the problem category sent by the second client under the condition that the consistency check is passed.

5. The method of claim 4, wherein the performing consistency check on the description information of the second routing problem and the problem category to which the second routing problem belongs comprises:

matching the description information of the second routing problem through the plurality of matching models to obtain second matching results corresponding to the plurality of matching models, wherein the second matching results corresponding to the matching models are used for indicating whether the second routing problem belongs to the problem category corresponding to the matching models;

determining a second target problem category to which the second routing problem belongs according to the problem category corresponding to at least one matching model under the condition that the second matching result corresponding to at least one matching model in the second matching results corresponding to the matching models indicates that the second routing problem belongs to the problem category corresponding to the at least one matching model;

determining that the consistency check passes when the second target problem category is consistent with the problem category sent by the second client;

and determining that the consistency check fails when the second target problem category is inconsistent with the problem category sent by the second client.

6. The method according to any one of claims 1-3, wherein the method further comprises:

responding to a classification rule updating request triggered by a classification rule configuration page, and updating the original classification rule or the target classification rule according to a third target problem category and the description information of a corresponding third testing problem included in the classification rule updating request;

synchronizing the updated classification rules to the plurality of clients.

7. An apparatus for updating classification rules, wherein the apparatus comprises:

the first acquisition module is used for acquiring description information of a first routing problem sent by a first client side in the plurality of client sides;

the first processing module is used for matching the description information of the first path measurement problem through a plurality of matching models to obtain first matching results corresponding to the plurality of matching models, wherein the first matching results corresponding to the matching models are used for indicating whether the first path measurement problem belongs to the problem category corresponding to the matching models;

a second obtaining module, configured to obtain a first target problem category set for the first routing problem based on the description information, when first matching results corresponding to the multiple matching models all indicate that the first routing problem does not belong to a corresponding problem category;

and the updating and synchronizing module is used for updating the original classification rule based on the first target problem category and the description information of the first testing problem to obtain a target classification rule, and synchronizing the target classification rule to the plurality of clients.

8. The apparatus of claim 7, wherein the first processing module comprises:

a feature extraction unit, configured to, for each matching model, input description information of the first road test problem into the matching model, so as to perform feature extraction on the description information based on a plurality of preset road test problem feature items, and obtain target feature values of the plurality of road test problem feature items;

the first matching unit is used for matching the target characteristic values with the reference characteristic values of the matching models corresponding to the drive test problem characteristic items to obtain matching degrees of the target characteristic values and the reference characteristic values;

9. The apparatus of claim 8, wherein the plurality of drive test problem signature terms comprises at least two of: the system comprises a main vehicle behavior module, a road structure module, an obstacle type module, an obstacle direction module, an obstacle action module and a problem belonging module.

10. The apparatus of any one of claims 7-9, wherein the apparatus further comprises:

a third obtaining module, configured to obtain description information of a second routing problem sent by a second client in the multiple clients and a problem category to which the second routing problem belongs;

the second processing module is used for carrying out consistency check on the description information of the second routing problem and the problem category to which the second routing problem belongs;

and the storage module is used for correspondingly storing the description information of the second routing problem and the problem category sent by the second client under the condition that the consistency check is passed.

11. The apparatus of claim 10, wherein the second processing module comprises:

a second matching unit, configured to perform matching processing on the description information of the second routing test problem through the multiple matching models to obtain second matching results corresponding to the multiple matching models, where the second matching result corresponding to the matching model is used to indicate whether the second routing test problem belongs to a problem category corresponding to the matching model;

a second determining unit, configured to determine, according to a problem category corresponding to at least one matching model, a second target problem category to which the second way testing problem belongs, when a second matching result corresponding to at least one matching model indicates that the second way testing problem belongs to the problem category corresponding to the at least one matching model, among second matching results corresponding to multiple matching models;

a fourth determining unit, configured to determine that the consistency check fails when the second target problem category is inconsistent with the problem category sent by the second client.

12. The apparatus of any of claims 7-9, wherein the apparatus further comprises:

13. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.

14. A non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method of any one of claims 1-6.

15. A computer program product comprising a computer program which, when executed by a processor, carries out the steps of the method of any one of claims 1-6.