CN118132387A

CN118132387A - Method, device, equipment, storage medium and program product for determining target vehicle data quality

Info

Publication number: CN118132387A
Application number: CN202410537004.7A
Authority: CN
Inventors: 褚佳鑫; 何雯; 张天雷; 王晓东
Original assignee: Beijing Zhuxian Technology Co Ltd
Current assignee: Beijing Zhuxian Technology Co Ltd
Priority date: 2024-04-30
Filing date: 2024-04-30
Publication date: 2024-06-04

Abstract

The application provides a method, a device, equipment, a storage medium and a program product for determining the data quality of a target vehicle, belongs to the technical field of data processing, and can be applied to scenes such as driving vehicles, automatic driving, intelligent parking and the like. The method comprises the following steps: when determining that the determining condition of the data quality is met currently, if determining that the vehicle is running currently, acquiring target real-time monitoring data, target time series data, target system log data and a plurality of target data quality determining models established in advance according to different quality indexes, and respectively inputting the target real-time monitoring data, the target time series data and the target system log data into each target data quality determining model, so that corresponding data quality parameters are output based on each target data quality determining model, and further determining whether the data quality of the target vehicle is abnormal according to each data quality parameter. The accuracy of determining the data quality is improved, and the driving experience is improved.

Description

Method, device, equipment, storage medium and program product for determining target vehicle data quality

Technical Field

The present application relates to the field of data processing technologies, and in particular, to a method, an apparatus, a device, a storage medium, and a program product for determining a target vehicle data quality.

Background

With the rapid development of automatic driving technology, the dependence of vehicles on data is gradually deepened. These data cover several core links in the driving process of the vehicle, and the quality of these data needs to be determined to ensure the stable operation and safety of the automatic driving technology.

The current data quality determination of the vehicle is generally to set a fixed threshold for various parameters of the vehicle according to human experience or industry rules, then compare various parameter data collected during the running process of the vehicle with corresponding fixed thresholds, and directly determine that the data quality of the vehicle is abnormal when certain parameter data is determined to not meet the corresponding fixed threshold.

However, the method for directly determining the data quality of the vehicle according to the comparison result is single in reference basis, so that the accuracy of determining the data quality is poor, and the driving experience is affected.

Disclosure of Invention

The application provides a method, a device, equipment, a storage medium and a program product for determining the data quality of a target vehicle, which are used for solving the technical problems that the accuracy of determining the data quality is poor and the driving experience is influenced in the prior art.

In a first aspect, the present application provides a method for determining a target vehicle data quality, including: responding to the determination condition of meeting the data quality, and acquiring target real-time monitoring data, target time sequence data and target system log data corresponding to a target vehicle;

acquiring a target data quality determination model set; each target data quality determining model in the target data quality determining model set is obtained by training a training sample marked with data quality parameters until convergence, and the data quality parameters are parameters representing the quality of certain data;

respectively inputting the target real-time monitoring data, the target time sequence data and the target system log data into each target data quality determining model;

determining and outputting corresponding data quality parameters based on each target data quality determination model;

and determining whether the data quality of the target vehicle is abnormal according to the data quality parameters.

Optionally, the data quality parameter is a data accuracy rate, a data integrity score, or a data consistency score; the target data quality determining model set comprises a target data accuracy determining model, a target data integrity determining model and a target data consistency determining model; the outputting of the corresponding data quality parameters based on the target data quality determination models comprises the following steps: determining the data accuracy corresponding to the model output based on the target data accuracy; and outputting a corresponding data integrity score based on the target data integrity determination model; and outputting the corresponding data consistency score based on the target data consistency determination model.

Optionally, the determining whether the data quality of the target vehicle is abnormal according to the data quality parameter includes: acquiring current parameter thresholds corresponding to the target data quality determination models, and judging whether the data quality parameters are equal to or larger than the corresponding current parameter thresholds; if yes, determining that the data quality of the target vehicle is normal; if not, determining that the data quality of the target vehicle is abnormal.

Optionally, after determining that the data quality of the target vehicle is abnormal, the method further includes: generating and outputting first alarm information based on the data quality parameters smaller than the corresponding current parameter threshold values; acquiring adjustment data according to the first alarm information; the adjustment data are adjustment data sent by the cloud device, or adjustment data obtained based on a preset threshold prediction model, or adjustment data obtained based on a preset sliding window algorithm; and adjusting the current parameter threshold by adopting the adjustment data to obtain an adjusted parameter threshold.

Optionally, before the acquiring the target real-time monitoring data, the target time series data and the target system log data corresponding to the target vehicle in response to the meeting of the determining condition of the data quality, the method further includes: collecting real-time data of a target vehicle, and preprocessing the real-time data; and analyzing the preprocessed real-time data by adopting a preset time sequence analysis strategy to obtain target time sequence data.

Optionally, the analyzing the preprocessed real-time data by using a preset time sequence analysis strategy includes: checking whether the preprocessed real-time data has a deletion or not; if not, carrying out time sequence arrangement on the preprocessed real-time data, and determining that the arranged real-time data is target time sequence data; if yes, acquiring the undelayed real-time data, carrying out time sequence arrangement on the undelayed real-time data, and determining the arranged real-time data as target time sequence data.

Optionally, the target real-time monitoring data includes first real-time monitoring data, second real-time monitoring data and third real-time monitoring data; the method further comprises the steps of: the acquisition process, the preprocessing operation process and the analysis process are respectively monitored to respectively obtain first real-time monitoring data, second real-time monitoring data and third real-time monitoring data.

Optionally, monitoring the acquisition process includes: judging whether the acquisition process is abnormal or not, and judging whether the state of the target vehicle is abnormal or not according to the real-time data; and generating and outputting second alarm information according to the abnormal result in response to at least one judgment result being yes.

Optionally, the method further comprises: receiving an alarm request sent by cloud equipment; the alarm request is used for indicating the vehicle control equipment to generate and output third alarm information; and generating and outputting third alarm information according to the alarm request.

Optionally, monitoring the pretreatment operation includes: acquiring a current caching parameter threshold value, and judging whether an abnormality occurs in a preprocessing operation process according to the current caching parameter threshold value; the current buffer parameter threshold is a threshold set for parameters related to data buffer in the current preprocessing operation process; if yes, fourth alarm information is generated and output according to the abnormal result.

Optionally, monitoring the analytical process includes: acquiring a current flow rate threshold value, and judging whether the data flow rate in the analysis process is smaller than the current flow rate threshold value; the current flow rate threshold value is a threshold value set for the data flow rate in the current analysis process; if yes, determining that the analysis process is abnormal, and generating and outputting fifth alarm information according to the abnormal result.

Optionally, the method further comprises: and generating and outputting sixth alarm information based on the missing real-time data in response to the detection of the missing real-time data.

In a second aspect, the present application provides a target vehicle data quality determining apparatus, the apparatus comprising: the acquisition module is used for responding to the determination condition of the data quality and acquiring target real-time monitoring data, target time sequence data and target system log data corresponding to the target vehicle;

the acquisition module is also used for acquiring a target data quality determination model set; each target data quality determining model in the target data quality determining model set is obtained by training a training sample marked with data quality parameters until convergence, and the data quality parameters are parameters representing the quality of certain data;

The input module is used for respectively inputting the target real-time monitoring data, the target time sequence data and the target system log data into each target data quality determination model;

the determining module is used for determining corresponding data quality parameters based on each target data quality determining model;

The output module is used for outputting the corresponding data quality parameters;

the determining module is further used for determining whether the data quality of the target vehicle is abnormal according to the data quality parameters.

In a third aspect, the present application provides a vehicle control apparatus comprising: a processor, and a memory communicatively coupled to the processor;

The memory stores computer-executable instructions;

the processor executes computer-executable instructions stored in the memory to implement the method as described in the first aspect.

In a fourth aspect, the present application provides a computer-readable storage medium having stored therein computer-executable instructions for performing the method according to the first aspect when executed by a processor.

In a fifth aspect, the application provides a computer program product comprising a computer program for implementing the method according to the first aspect when the computer program is executed by a processor.

The application provides a method, a device, equipment, a storage medium and a program product for determining the data quality of a target vehicle, wherein target real-time monitoring data, target time sequence data and target system log data corresponding to the target vehicle are obtained in response to the condition for determining the data quality; acquiring a target data quality determination model set; each target data quality determining model in the target data quality determining model set is obtained by training a training sample marked with data quality parameters until convergence, and the data quality parameters are parameters representing the quality of certain data; respectively inputting the target real-time monitoring data, the target time sequence data and the target system log data into each target data quality determining model; determining and outputting corresponding data quality parameters based on each target data quality determination model; and determining whether the data quality of the target vehicle is abnormal according to the data quality parameters. Because the target real-time monitoring data, the target time sequence data and the target system log data corresponding to the target vehicle are stored in advance, and a plurality of target data quality determining models, namely a target data quality determining model set, are established based on training samples marked with data quality parameters representing whether certain data quality is good or bad, when the current state of the target vehicle is determined to meet the determining condition of the data quality, the target real-time monitoring data, the target time sequence data and the target system log data corresponding to the target vehicle are acquired, and the target data quality determining model set is acquired, so that the target real-time monitoring data, the target time sequence data and the target system log data can be respectively input into each target data quality determining model. And by determining and outputting the corresponding data quality parameters based on each target data quality determination model, it is possible to determine whether the data quality of the target vehicle is abnormal or not based on the data quality parameters. Therefore, the basis for determining whether the data quality is abnormal is increased, the accuracy for determining the data quality is improved, and the driving experience is improved.

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.

FIG. 1 is an application scenario diagram of a method for determining a target vehicle data quality according to an embodiment of the present application;

FIG. 2 is a flow chart of a method for determining a target vehicle data quality according to an embodiment of the present application;

FIG. 3 is a flow chart of a method for determining a target vehicle data quality according to another embodiment of the present application;

FIG. 4 is a complete flow chart of a method for determining the quality of data of a target vehicle according to yet another embodiment of the present application;

FIG. 5 is a schematic structural diagram of a target vehicle data quality determining apparatus according to an embodiment of the present application;

fig. 6 is a schematic structural diagram of a vehicle control apparatus according to an embodiment of the present application.

Specific embodiments of the present application have been shown by way of the above drawings and will be described in more detail below. The drawings and the written description are not intended to limit the scope of the inventive concepts in any way, but rather to illustrate the inventive concepts to those skilled in the art by reference to the specific embodiments.

Detailed Description

Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the application. Rather, they are merely examples of apparatus and methods consistent with aspects of the application as detailed in the accompanying claims.

For a clear understanding of the technical solutions of the present application, the prior art solutions will be described in detail first.

Currently, when determining the data quality of a vehicle, a fixed threshold is generally set in advance to determine whether the data quality of the vehicle is abnormal. Specifically, a series of fixed thresholds are set for various parameters of the vehicle, such as the vehicle speed, acceleration, steering angle, etc., according to human experience or industry rules in advance, then when the vehicle is in a driving state, the vehicle control device collects various parameter data of the vehicle, such as sensor data, vehicle state data, etc., in real time, and then compares the collected parameter data with a preset fixed threshold, such as comparing a specific vehicle speed with a preset maximum speed threshold, so as to directly determine that the data quality of the vehicle is abnormal when it is determined that certain parameter data of the vehicle does not meet the corresponding fixed threshold. The reference basis is single, so that the accuracy of determining the data quality is poor, and the driving experience is affected.

In order to improve accuracy in determining the quality of data and to improve driving experience, therefore, in order to face the problems in the prior art, instead of directly determining that the quality of data of a vehicle is abnormal when certain parameter data of the vehicle does not satisfy a corresponding fixed threshold, multifaceted vehicle data including monitoring data, time series data and overall logs are considered, and on the basis, a deep learning model of a plurality of quality index dimensions is adopted to integrate the multifaceted vehicle data to determine whether the data quality of the vehicle is abnormal. When the vehicle controller determines that the determination condition of the data quality is currently met, if the vehicle is monitored to be running, multi-aspect vehicle data and models which are established in advance according to different quality indexes are acquired, and the multi-aspect vehicle data are respectively input into each model, so that corresponding quality index data are output based on each model, and whether the data quality of the vehicle is abnormal or not is determined according to each quality index data. Therefore, the basis for determining whether the data quality is abnormal is increased, the accuracy for determining the data quality is improved, and the driving experience is improved.

Fig. 1 is an application scenario diagram of a method for determining a target vehicle data quality according to an embodiment of the present application, where as shown in fig. 1, the present application provides a system for determining a target vehicle data quality, where the system for determining a target vehicle data quality includes: a target vehicle 1, a vehicle control device 2, a database 3. The target vehicle 1 is a vehicle capable of executing a method for determining the data quality of the target vehicle, the vehicle control device 2 is an electronic device for controlling the target vehicle 1, and is specifically located in the target vehicle 1, and the database 3 is a database corresponding to the vehicle control device 2. Wherein the vehicle control device 2 is in communication with the database 3. First, the target vehicle 1 is started by the driver so that the target vehicle 1 is in a running state, and then the vehicle control apparatus 2 acquires target real-time monitoring data, target time series data, and target system log data corresponding to the target vehicle based on the database 3 and acquires a target data quality determination model set in response to satisfaction of the determination condition of the data quality. And respectively inputting the target real-time monitoring data, the target time sequence data and the target system log data into each target data quality determining model, determining and outputting corresponding data quality parameters based on each target data quality determining model, and determining whether the data quality of the target vehicle is abnormal or not according to the data quality parameters. Each target data quality determining model in the target data quality determining model set is obtained by training a training sample marked with data quality parameters until convergence, and the data quality parameters are parameters representing the quality of certain data.

The application provides a method for determining the data quality of a target vehicle, which aims to solve the technical problems in the prior art.

The following describes the technical scheme of the present application and how the technical scheme of the present application solves the above technical problems in detail with specific embodiments. The following embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments. Embodiments of the present application will be described below with reference to the accompanying drawings.

Fig. 2 is a flowchart of a method for determining a target vehicle data quality according to an embodiment of the present application, as shown in fig. 2, where an execution subject of the embodiment is a device for determining a target vehicle data quality, and the device for determining a target vehicle data quality may be located in a vehicle control device, and the method for determining a target vehicle data quality according to the embodiment includes the following steps:

S201, acquiring target real-time monitoring data, target time sequence data and target system log data corresponding to a target vehicle in response to the condition for determining the data quality.

Wherein the target vehicle is a vehicle having a determined data quality requirement.

The target real-time monitoring data are data obtained by monitoring various data generated by the target vehicle when the target vehicle is in a certain state. The target time-series data is data obtained by performing time-series analysis on various data generated by the target vehicle when the target vehicle is in a certain state. The target system log data is various vehicle-related data recorded based on the target vehicle.

In this embodiment, the vehicle control device monitors the current state of the target vehicle in real time, and obtains target real-time monitoring data, target time series data and target system log data corresponding to the target vehicle stored in advance in response to the fact that the current state of the target vehicle meets a determination condition of the target vehicle data quality, for example, the target vehicle is currently in a driving state.

S202, acquiring a target data quality determination model set. Each target data quality determining model in the target data quality determining model set is obtained by training to convergence by adopting a training sample marked with data quality parameters, wherein the data quality parameters are parameters representing the quality of certain data.

The target data quality determining model set is a set comprising a plurality of target data quality determining models, wherein the target data quality determining models are models for determining the data quality of a target vehicle based on certain data quality parameters, and specifically are obtained by training to convergence by adopting training samples marked with the data quality parameters.

Wherein the data quality parameter is a parameter that characterizes the quality of certain data. The training samples are samples of various data representing the target vehicle, and can be specifically existing real-time monitoring data, time series data and system log data.

For example, when a training sample marked with a data quality parameter is used to obtain a target data quality determination model, the method specifically may be: and (3) pre-establishing an initial model, marking the existing real-time monitoring data, time sequence data and system log data by adopting the value when the target data quality determining model is determined to be a model for determining the data accuracy and the corresponding data quality parameter is a value for representing the data accuracy, and training the initial model to be converged based on the marking result to obtain the target data quality determining model.

It should be understood that the above manner of obtaining the target data quality determination model is merely an example, and should not be construed as limiting the present application in any way.

Based on the above, in order to multi-dimensionally determine the data quality of the target vehicle, different target data quality determination models are established based on different data quality parameters, a target data quality determination model set is formed, and after target real-time monitoring data, target time series data and target system log data corresponding to the target vehicle are acquired, a pre-stored target data quality determination model set is acquired.

S203, the target real-time monitoring data, the target time series data and the target system log data are respectively input into each target data quality determination model.

In the present embodiment, since the target real-time monitoring data, the target time series data, and the target system log data characterize the multifaceted data of the target vehicle, after each target data quality determination model is acquired, the target real-time monitoring data, the target time series data, and the target system log data are taken as a whole, and the whole is input into each target data quality determination model, respectively.

S204, corresponding data quality parameters are determined and output based on each target data quality determination model.

It can be understood that, since each target data quality determination model is trained based on the training samples of the labeling data quality parameters until convergence is obtained, after the target real-time monitoring data, the target time series data and the target system log data are respectively input into each target data quality determination model, each target data quality determination model can be calculated based on the target real-time monitoring data, the target time series data and the target system log data to determine the data quality parameters corresponding to each target data quality determination model. Based on this, the corresponding data quality parameter is output according to the determination result.

S205, determining whether the data quality of the target vehicle is abnormal according to the data quality parameters.

In this embodiment, based on the data quality parameters corresponding to the obtained target data quality determination models, whether the data quality of the target vehicle is abnormal is determined according to the data quality parameters.

According to the method for determining the data quality of the target vehicle, provided by the embodiment, the target real-time monitoring data, the target time sequence data and the target system log data corresponding to the target vehicle are obtained in response to the condition that the data quality is determined; acquiring a target data quality determination model set; each target data quality determining model in the target data quality determining model set is obtained by training a training sample marked with data quality parameters until convergence, and the data quality parameters are parameters representing the quality of certain data; respectively inputting target real-time monitoring data, target time sequence data and target system log data into each target data quality determining model; determining and outputting corresponding data quality parameters based on each target data quality determination model; and determining whether the data quality of the target vehicle is abnormal according to the data quality parameters. Because the target real-time monitoring data, the target time sequence data and the target system log data corresponding to the target vehicle are stored in advance, and a plurality of target data quality determining models, namely a target data quality determining model set, are established based on training samples marked with data quality parameters representing whether certain data quality is good or bad, when the current state of the target vehicle is determined to meet the determining condition of the data quality, the target real-time monitoring data, the target time sequence data and the target system log data corresponding to the target vehicle are acquired, and the target data quality determining model set is acquired, so that the target real-time monitoring data, the target time sequence data and the target system log data can be respectively input into each target data quality determining model. And by determining and outputting the corresponding data quality parameters based on each target data quality determination model, it is possible to determine whether the data quality of the target vehicle is abnormal or not based on each data quality parameter. Therefore, the basis for determining whether the data quality is abnormal is increased, the accuracy for determining the data quality is improved, and the driving experience is improved.

As an alternative embodiment, the data quality parameter is a data accuracy rate, a data integrity score, or a data consistency score based on the above embodiment, the set of target data quality determining models includes a target data accuracy determining model, a target data integrity determining model, and a target data consistency determining model, and when outputting the corresponding data quality parameter based on each target data quality determining model, S204 specifically includes the following steps:

and determining the data accuracy corresponding to the model output based on the target data accuracy. And outputting the corresponding data integrity score based on the target data integrity determination model. And outputting the corresponding data consistency score based on the target data consistency determination model.

In this embodiment, the target data quality determining model set may include a target data accuracy determining model, a target data integrity determining model, and a target data consistency determining model, and correspondingly, the data quality parameter may be a data accuracy, a data integrity score, or a data consistency score. The data accuracy is an index for representing the data accuracy. The data integrity score is an indicator that characterizes the integrity of the data. Data consistency is an indicator that characterizes data consistency.

It can be appreciated that when the data quality parameter is the data accuracy, the target data accuracy determination model can be obtained based on the training sample of the labeled data accuracy. Similarly, when the data quality parameter is a data integrity score, a target data integrity determination model can be obtained based on training samples labeled with the data integrity score. Similarly, when the data quality parameter is a data integrity score, a target data consistency determination model can be obtained based on training samples labeled with the data integrity score.

Thus, in response to determining that the model outputs corresponding data quality parameters based on each target data quality, determining that the model outputs corresponding data accuracy based on the target data accuracy, and determining that the model outputs corresponding data integrity scores based on the target data integrity, and determining that the model outputs corresponding data consistency scores based on the target data consistency.

According to the method for determining the data quality of the target vehicle, when corresponding data quality parameters are output based on each target data quality determination model, the corresponding data accuracy rate of the model output is determined based on the target data accuracy; and outputting a corresponding data integrity score based on the target data integrity determination model; and outputting the corresponding data consistency score based on the target data consistency determination model. Because three indexes of determining the data quality, namely accuracy, integrity and consistency are established in advance, the corresponding data accuracy can be output based on the target data accuracy determining model, the corresponding data integrity score can be output based on the target data integrity determining model, and the corresponding data consistency score can be output based on the target data consistency determining model, so that multidimensional parameters affecting the data quality of the vehicle are considered, and the accuracy of determining the data quality of the vehicle is further improved.

Fig. 3 is a flowchart of a method for determining the data quality of a target vehicle according to another embodiment of the present application, as shown in fig. 3, where, based on the corresponding embodiment of fig. 2, whether the data quality of the target vehicle is abnormal or not is determined according to the data quality parameter, the method specifically includes the following steps when determining whether the data quality of the target vehicle is abnormal or not according to the data quality parameter:

S301, obtaining current parameter thresholds corresponding to the target data quality determination models, and judging whether the data quality parameters are equal to or larger than the corresponding current parameter thresholds.

The current parameter threshold is a threshold determined by the data quality parameters corresponding to each target data quality determination model.

For example, if the target data quality determination model is a target data accuracy determination model, and the data quality parameter is a data accuracy rate, the corresponding current parameter threshold, such as 90%, may be set based on the data accuracy rate. If the target data quality determination model is a target data integrity determination model, and the data quality parameters are data integrity scores, a corresponding current parameter threshold, such as 90 points, may be set based on the data integrity. If the target data quality determination model is a target data consistency determination model, and the data quality parameters are data consistency scores, a corresponding current parameter threshold, such as 95 points, may be set based on the data consistency.

Based on the above, in order to determine whether each data quality parameter reaches a preset standard, the current parameter threshold corresponding to each target data quality determination model is used as a comparison parameter, and whether each data quality parameter obtained by each target data quality determination model is equal to or greater than the corresponding current parameter threshold is determined.

And S302, if yes, determining that the data quality of the target vehicle is normal.

In this embodiment, based on determining whether the data quality parameters are equal to or greater than the corresponding current parameter thresholds, if yes, it is determined that the data quality parameters of the target vehicle all reach the preset standard, and it is determined that the data quality of the target vehicle is normal.

S303, if not, determining that the data quality of the target vehicle is abnormal.

In this embodiment, based on determining whether the data quality parameters are equal to or greater than the corresponding current parameter thresholds, if not, it is indicated that at least one of the data quality parameters does not reach the preset standard, and it is determined that the data quality of the target vehicle is abnormal.

According to the method for determining the data quality of the target vehicle, when determining whether the data quality of the target vehicle is abnormal or not according to the data quality parameters, the current parameter threshold corresponding to each target data quality determination model is obtained, and whether the data quality parameters are equal to or larger than the corresponding current parameter threshold is judged; if yes, determining that the data quality of the target vehicle is normal; if not, determining that the data quality of the target vehicle is abnormal. Since the corresponding current parameter threshold value is set in advance based on the data quality parameters of each target data quality determination model, by acquiring the current parameter threshold value corresponding to each target data quality determination model, it is possible to determine whether the data quality parameters are equal to or greater than the corresponding current parameter threshold values by taking the current parameter threshold value as a comparison condition, thereby determining that the data quality of the target vehicle is normal when it is determined that the data quality parameters are equal to or greater than the corresponding current parameter threshold values, and determining that the data quality of the target vehicle is abnormal when it is determined that there is at least one data quality parameter less than the corresponding current parameter threshold value. Thereby improving the accuracy of judging whether the data quality is abnormal.

As an alternative embodiment, the present embodiment further includes, on the basis of the above embodiment, after determining that the data quality of the target vehicle is abnormal, the steps of:

s401, generating and outputting first alarm information based on the data quality parameters smaller than the corresponding current parameter threshold.

The first alarm information is information for alarming when the data quality parameter is smaller than the current parameter threshold value.

In this embodiment, in order to ensure that a driver and a related manager in the target vehicle can learn the result of the data quality abnormality of the current target vehicle, after determining that the data quality of the target vehicle is abnormal, first alarm information is generated and output based on the data quality parameter smaller than the corresponding current parameter threshold.

When the first alarm information is output, the method specifically may be: corresponding mail, text message or system notification is formed based on the first alert information using related interfaces and tools, such as simple mail transfer protocol SMTP service or short message service provider's interface, to inform the driver of the first alert information and related management personnel.

It should be understood that the above output modes are only examples, and should not be construed as limiting the present application.

S402, acquiring adjustment data according to the first alarm information. The adjustment data are adjustment data sent by the cloud device, or adjustment data obtained based on a preset threshold prediction model, or adjustment data obtained based on a preset sliding window algorithm.

The adjustment data is data for adjusting the threshold, specifically, adjustment data sent by the cloud device, or adjustment data obtained based on a preset threshold prediction model, or adjustment data obtained based on a preset sliding window algorithm. The preset threshold prediction model is a model for predicting adjustment data, and the preset sliding window algorithm is an algorithm for calculating statistical characteristic values of the data, such as calculating a mean value, a standard deviation and the like of the data.

It will be appreciated that, after the first alarm information is output, in order to ensure that the current parameter threshold value matches the state in which the target vehicle is located, it is necessary to check, according to the alarm result, whether the state in which the target vehicle is located is indeed abnormal, that is, whether the state in which the target vehicle is located is indeed abnormal according to the result that there is at least one data quality parameter less than the corresponding current parameter threshold value. If so, the state of the target vehicle is truly fed back by the first alarm information, and the fact that the current parameter threshold value reflected in the first alarm information is matched with the state of the target vehicle is determined. If not, the state of the first alarm information and the state of the target vehicle are inconsistent, and the current parameter threshold value reflected in the first alarm information is not matched with the state of the target vehicle, and adjustment data are required to be acquired according to the first alarm information so as to adjust the current parameter threshold value reflected in the first alarm information.

Based on the adjustment data can be obtained through the cloud device according to the first alarm information. Specifically, based on the unmatched result, the relevant manager selects a new value for the current parameter threshold value of the alarm, forms adjustment data, and sends the adjustment data to the vehicle control device through the cloud device. Correspondingly, the vehicle control equipment receives the adjustment data sent by the cloud equipment and acquires the adjustment data.

And adjusting data can be obtained through a preset threshold prediction model according to the first alarm information. Specifically, training a newly built deep learning model to be converged by adopting a training sample marked with adjustment data in advance to obtain a preset threshold prediction model. And after the first alarm information is output, acquiring subsequent real-time monitoring data, time sequence data and system log data, and inputting the subsequent data into a preset threshold prediction model to obtain predicted adjustment data. The training samples may be existing real-time monitoring data, time series data, and system log data.

The adjustment data can also be obtained through a preset sliding window algorithm according to the first alarm information. Specifically, subsequent real-time monitoring data, time sequence data and system log data are obtained, a preset sliding window algorithm is adopted to calculate corresponding statistical characteristic values according to the subsequent data, corresponding adjustment data are determined according to the change of the statistical characteristic values, if standard deviation is increased, the increase of data fluctuation is indicated, and corresponding adjustment data are correspondingly determined according to fluctuation amplitude.

S403, adjusting the current parameter threshold by adopting the adjustment data to obtain an adjusted parameter threshold.

In this embodiment, based on the obtained adjustment data, a replacement operation is performed on the current parameter threshold value reflected in the first alarm information by using the adjustment data, so as to obtain an adjusted parameter threshold value, thereby taking the adjusted parameter threshold value as a new current parameter threshold value, and a subsequent process of determining whether the data quality of the target vehicle is abnormal is performed based on the new current parameter threshold value.

According to the method for determining the data quality of the target vehicle, after the data quality abnormality of the target vehicle is determined, first alarm information is generated and output based on the data quality parameter smaller than the corresponding current parameter threshold; acquiring adjustment data according to the first alarm information; the adjustment data are adjustment data sent by the cloud device, or adjustment data obtained based on a preset threshold prediction model, or adjustment data obtained based on a preset sliding window algorithm; and adjusting the current parameter threshold by adopting the adjustment data to obtain an adjusted parameter threshold. Because the current parameter threshold value may not be in line with the actual system state of the target vehicle, the driver and the related manager can timely learn the result of the abnormal data quality by generating and outputting the first alarm information according to the data quality parameter smaller than the corresponding current parameter threshold value, so that the driving process is prevented from being influenced by the abnormal data quality, and the driving experience is further improved. After the first alarm information is output, the current parameter threshold value can be adjusted by adopting the adjustment data by acquiring the adjustment data obtained based on the cloud device or a preset threshold value prediction model or a preset sliding window algorithm according to the first alarm information, so that the adjusted parameter threshold value is obtained. Therefore, the parameter threshold value is dynamically adjusted, the matching property of the current parameter threshold value and the actual state of the vehicle is improved, and the accuracy of determining the data quality is further improved.

As an alternative embodiment, the present embodiment further includes, on the basis of the corresponding embodiment of fig. 2, before acquiring the target real-time monitoring data, the target time series data, and the target system log data corresponding to the target vehicle in response to the satisfaction of the determination condition of the data quality, the following steps:

s501, collecting real-time data of a target vehicle, and preprocessing the real-time data.

The real-time data are data generated by the target vehicle in the driving process. The preprocessing operation is an operation of performing a series of processes on real-time data, and may specifically include a cleaning process, a feature extraction process, a normalization and normalization process, and a caching and backup process.

It can be understood that, before the implementation of the scheme, the real-time monitoring data, the time series data and the log data of the target system corresponding to the target vehicle need to be determined, so that one premise of the implementation of the scheme is that the time series data of the target corresponding to the target vehicle is determined currently.

In this embodiment, the sensor may be integrated with a sensor on the target vehicle, such as a laser radar, a camera, a millimeter wave radar, etc., and a unified data format standard may be designed, and then the sensor may be communicated with using a synchronization mechanism, a controller area network protocol, or a dedicated sensor communication protocol. Based on this, real-time data generated by the target vehicle is monitored in real time by the relevant sensors and acquired by the above configuration.

Then, based on the above, the above obtained real-time data is sequentially subjected to the cleaning process, the feature extraction process, the normalization and normalization process, and the caching and backup process operations according to the pre-stored process sequence. Specifically, noise, repeated or invalid data are removed from the collected real-time data, and the cleaned or real-time data in different formats are uniformly converted into real-time data in a format which can be processed by the system by adopting a data conversion algorithm. And extracting key features from the real-time data in the processable format of the system by adopting a feature extraction technology, and carrying out standardization and normalization processing on the real-time data after extraction so that the real-time data with different sources and dimensions can be compared and analyzed on the same scale. And finally, a data backup function is adopted to ensure the reliability and the restorability of the real-time data.

S502, analyzing the preprocessed real-time data by adopting a preset time sequence analysis strategy to obtain target time sequence data.

The preset time sequence analysis strategy is a strategy for analyzing the change process of data in a certain time, wherein the strategy is stored in advance.

In this embodiment, preprocessing operation is performed on real-time data, preprocessed real-time data and a pre-stored preset time sequence analysis strategy are obtained, and the preprocessed real-time data is analyzed by adopting the preset time sequence analysis strategy to obtain target time sequence data.

According to the method for determining the data quality of the target vehicle, the real-time data of the target vehicle is collected and preprocessed before the real-time monitoring data, the target time sequence data and the target system log data corresponding to the target vehicle are obtained in response to the condition that the data quality is determined; and analyzing the preprocessed real-time data by adopting a preset time sequence analysis strategy to obtain target time sequence data. Because the target time sequence data corresponding to the target vehicle is required to be determined first, the real-time data can be preprocessed by collecting the real-time data of the target vehicle, and the preprocessed real-time data can be analyzed by adopting a preset time sequence analysis strategy, so that the target time sequence data is obtained, the target time sequence data can be successfully obtained in the subsequent execution process, and the success rate of determining the data quality is improved.

As an optional embodiment, in this embodiment, when the pre-processed real-time data is analyzed by using a preset time sequence analysis policy based on the above embodiment, S502 specifically includes the following steps:

s5021, checking whether the preprocessed real-time data has a defect.

In this embodiment, in response to analyzing the preprocessed real-time data by using a preset time sequence analysis strategy, traversing the preprocessed real-time data based on the preset time sequence analysis strategy, and detecting abnormal fluctuation and trend change in the preprocessed real-time data, thereby checking whether there is a defect in the preprocessed real-time data.

S5022, if not, carrying out time sequence arrangement on the preprocessed real-time data, and determining that the arranged real-time data is target time sequence data.

In this embodiment, based on checking whether the preprocessed real-time data has a defect, if it is determined that the preprocessed real-time data has no defect, it is indicated that there is no abnormality in the preprocessed real-time data, all the preprocessed real-time data is obtained, and the data are arranged according to a time sequence, so that it is determined that the arranged real-time data is the target time sequence data.

And S5023, if yes, acquiring the undelayed real-time data, arranging the undelayed real-time data on a time sequence, and determining the arranged real-time data as target time sequence data.

In this embodiment, based on whether the preprocessed real-time data is missing or not, if it is determined that the preprocessed real-time data is missing, it is indicated that there is an abnormality in the preprocessed real-time data, and the real-time data that is not missing is obtained and is arranged according to a time sequence, so that it is determined that the arranged real-time data is the target time sequence data.

According to the method for determining the target vehicle data quality, when the pre-processed real-time data is analyzed by adopting a preset time sequence analysis strategy, whether the pre-processed real-time data has a defect or not is checked; if not, carrying out time sequence arrangement on the preprocessed real-time data, and determining that the arranged real-time data is target time sequence data; if yes, acquiring the undelayed real-time data, arranging the undelayed real-time data on a time sequence, and determining the arranged real-time data as target time sequence data. Because the preset time sequence analysis strategy is stored in advance, by checking the preprocessed real-time data based on the strategy, the preprocessed real-time data can be directly arranged on the time sequence when the fact that the preprocessed real-time data is not missing is determined, and the non-missing real-time data can be arranged on the time sequence when the fact that the preprocessed real-time data is missing is determined, so that the arranged real-time data is determined to be target time sequence data, the target time sequence data can be accurately determined, the data integrity can be preliminarily checked when the time sequence analysis is performed, and the accuracy of determining the data quality is further improved.

As an optional embodiment, in this embodiment, on the basis of the foregoing embodiment or the foregoing previous embodiment, the target real-time monitoring data includes first real-time monitoring data, second real-time monitoring data, and third real-time monitoring data, and the method further includes:

The acquisition process, the preprocessing operation process and the analysis process are respectively monitored to respectively obtain first real-time monitoring data, second real-time monitoring data and third real-time monitoring data.

The first real-time monitoring data are data obtained by monitoring the acquisition process, the second real-time monitoring data are data obtained by monitoring the preprocessing operation process, and the third real-time monitoring data are data obtained by monitoring the analysis process.

In this embodiment, another premise of the implementation of the scheme is that the real-time target monitoring data corresponding to the target vehicle is determined currently. Based on the above, the acquisition process, the preprocessing operation process and the analysis process are respectively monitored to respectively obtain first real-time monitoring data, second real-time monitoring data and third real-time monitoring data, namely target real-time monitoring data.

The method for determining the data quality of the target vehicle provided in the embodiment further includes: the acquisition process, the preprocessing operation process and the analysis process are respectively monitored to respectively obtain first real-time monitoring data, second real-time monitoring data and third real-time monitoring data. Because the real-time monitoring data of the target corresponding to the target vehicle is required to be determined first, the first real-time monitoring data, the second real-time monitoring data and the third real-time monitoring data can be respectively obtained by respectively monitoring the acquisition process, the preprocessing operation process and the analysis process. The target real-time monitoring data can be obtained, so that the target real-time monitoring data can be successfully obtained in the subsequent execution process, and the success rate of determining the data quality is further improved.

As an alternative embodiment, this embodiment, based on the above embodiment, specifically includes the following steps when monitoring the acquisition process:

S601, judging whether the acquisition process is abnormal or not, and judging whether the state of the target vehicle is abnormal or not according to the real-time data.

The state of the target vehicle is the current state of the target vehicle. For example, the current environmental state of the target vehicle.

In this embodiment, based on determining the requirement of the first real-time monitoring data, the current ongoing collection behavior is monitored in real time, and whether the collection behavior is abnormal is determined, so as to determine whether the collection process is abnormal. Meanwhile, whether the state of the target vehicle is abnormal or not is judged according to the acquired real-time data.

When judging whether the acquisition behavior is abnormal based on the monitoring result, the method specifically may be: and setting a corresponding parameter threshold value based on the parameter representing the acquisition behavior in advance, then responding to the monitoring of the acquisition behavior, acquiring the parameter representing the acquisition behavior, and comparing the parameter with the corresponding parameter threshold value, so as to judge whether the acquisition behavior is abnormal or not according to a comparison result. For example, if the parameter characterizing the acquisition behavior is the acquisition frequency and the corresponding parameter threshold is set to be 10 frames/0.1 seconds, then when the acquisition frequency is determined not to reach the threshold, for example, 10 frames/0.2 seconds, it may be determined that the acquisition process is abnormal, and when the acquisition frequency is determined to reach the threshold, it may be determined that the acquisition process is normal.

When judging whether the state of the target vehicle is abnormal according to the real-time data, the method specifically can be as follows: the method comprises the steps of constructing a deep learning model in advance according to existing real-time data of a marked vehicle state, or adopting a method of manually setting rules to establish a mapping relation between the real-time data and a target vehicle state. And then, in response to the monitoring of the acquisition behavior, acquiring real-time data, and determining a target vehicle state corresponding to the real-time data according to the deep learning model or the mapping relation. For example, the environmental state of the target vehicle is determined according to the real-time data, and if the state is determined to be capillary rain, the target vehicle can be determined to be normal. If the state is determined to be heavy rain, the state of the target vehicle can be determined to be abnormal.

S602, generating and outputting second alarm information according to the abnormal result in response to at least one judgment result being yes.

The second alarm information is alarm information when the acquisition process is abnormal.

In this embodiment, whether the state of the target vehicle is abnormal is determined based on the determination of whether the acquisition process is abnormal or not, and based on the real-time data. And responding to at least one judgment result in the two judgment processes, generating and outputting second alarm information according to the abnormal result. For example, if it is determined that the acquisition process is abnormal and it is determined that the target vehicle state is abnormal, second alarm information may be generated and output according to the two abnormal results. If the acquisition process is abnormal but the target vehicle state is not abnormal, generating and outputting second alarm information according to the abnormal result of the acquisition process. If the state of the target vehicle is abnormal but the acquisition process is not abnormal, generating and outputting second alarm information according to the result of the abnormal state of the target vehicle. The method for outputting the second alarm information is similar to the method for outputting the first alarm information in S401, and will not be described herein.

After the second alarm information is generated and output, the adjustment data can be obtained according to the second alarm information, and the parameter threshold value related in the acquisition process is adjusted by adopting the adjustment data so as to obtain the adjusted parameter threshold value. The specific implementation is similar to S402, and will not be described here again. Therefore, the parameter threshold value related in the acquisition process is dynamically adjusted, the matching property between the acquisition process and the related parameter threshold value is improved, and the accuracy of determining the data quality is further improved.

It can be understood that the adjustment data in S402 is adjustment data for the current parameter threshold, so the subsequent data related to the adjustment data is subsequent real-time monitoring data, time series data and system log data, while the adjustment data in the present embodiment is adjustment data for the parameter threshold related to the acquisition process, so the subsequent data related to the present embodiment is subsequent real-time data acquired.

It will be appreciated that after the acquisition process is monitored, the first real-time monitoring data may be obtained based on the monitoring results.

According to the method for determining the quality of the target vehicle data, when the acquisition process is monitored, whether the acquisition process is abnormal or not is judged, and whether the state of the target vehicle is abnormal or not is judged according to real-time data; and generating and outputting second alarm information according to the abnormal result in response to at least one judgment result being yes. Because the first real-time monitoring data is required to be acquired based on the acquisition process, by judging whether the acquisition process is abnormal or not and judging whether the state of the target vehicle is abnormal or not according to the real-time data, when at least one judgment result is yes, second alarm information can be generated and output according to the abnormal result, and the first real-time monitoring data can be acquired based on the monitoring result. Therefore, alarming is carried out when abnormality occurs in the acquisition process, and driving experience is further improved.

As an alternative embodiment, this embodiment further includes the following steps on the basis of the above embodiment:

s701, receiving an alarm request sent by cloud equipment; the alarm request is used for instructing the vehicle control device to generate and output third alarm information.

The alarm request is a request triggered by a related manager based on a cloud, and is specifically used for indicating the vehicle control equipment to generate and output third alarm information. The third alarm information is alarm information sent by the cloud device when the state of the target vehicle is abnormal.

It will be appreciated that heavy rain may be suddenly poured due to a certain sudden occurrence of the target vehicle condition, for example, the environmental condition in which the target vehicle is located. Therefore, when the related manager determines that the current target vehicle state is abnormal but the vehicle control device does not trigger an alarm, the related manager can trigger an alarm request at the cloud device and send the alarm request to the vehicle control device by the cloud device so as to instruct the vehicle control device to generate and output third alarm information. For example, the vehicle control device determines that the environmental state of the target vehicle is a capillary rain according to the real-time data, but the current sudden basin-tilting heavy rain, and receives an alarm request triggered by a related manager at the cloud device.

S702, generating and outputting third alarm information according to the alarm request.

In this embodiment, the vehicle control device generates and outputs third alarm information according to alarm content reflected by the alarm request based on receiving the alarm request sent by the cloud device, for example, generates and outputs third alarm information for heavy rain tilting basin according to an environmental state where the target vehicle is located.

According to the method for determining the target vehicle data quality, an alarm request sent by cloud equipment is received; the alarm request is used for indicating the vehicle control equipment to generate and output third alarm information; and generating and outputting third alarm information according to the alarm request. Because the vehicle state has certain burst and the vehicle control equipment cannot alarm under the burst, the third alarm information can be generated and output based on the alarm request by receiving the alarm request sent by the cloud equipment. The warning information can still be triggered in time when the vehicle control equipment cannot monitor the correct vehicle state, and driving experience is further improved.

As an alternative embodiment, the present embodiment, based on the seventh embodiment, specifically includes the following steps when monitoring the preprocessing operation procedure:

s801, a current caching parameter threshold is obtained, and whether the preprocessing operation process is abnormal or not is judged according to the current caching parameter threshold. The current buffer parameter threshold is a threshold set for parameters related to data buffering during the current preprocessing operation.

The current buffer parameter threshold is a preset threshold set for a parameter involved in the preprocessing operation, and specifically may be a threshold set for a parameter related to data buffer in the current preprocessing operation.

In this embodiment, based on determining the requirement of the second real-time monitoring data, the current ongoing preprocessing operation behavior is monitored in real time, and the parameters characterizing the data buffering behavior are obtained based on the preprocessing operation behavior. The pre-stored current caching parameter threshold value is obtained, and the parameter representing the data caching behavior is compared with the current caching parameter threshold value, so that whether the pre-processing operation process is abnormal or not is judged according to a comparison result. For example, if the parameter characterizing the data buffering behavior is the ratio of the buffering space and the current buffering parameter threshold is 90%, it may be determined that the preprocessing operation process is abnormal when the ratio of the buffering space is equal to or greater than 90%, and that the preprocessing operation process is normal when the ratio of the buffering space is less than 90%.

S802, if yes, fourth alarm information is generated and output according to the abnormal result.

The fourth alarm information is alarm information when the pretreatment operation process is abnormal.

In this embodiment, whether the preprocessing operation process is abnormal is determined based on the threshold value according to the current cache parameter, and if yes, fourth alarm information is generated and output according to the abnormal result of the preprocessing operation process. The method for outputting the fourth alarm information is similar to the method for outputting the first alarm information in S401, and will not be described herein.

After the fourth alarm information is generated and output, the adjustment data sent by the cloud device can be obtained according to the fourth alarm information, and the current cache parameter threshold is adjusted by adopting the adjustment data to obtain the adjusted parameter threshold. The specific implementation is similar to S402, and will not be described here again. Therefore, the current buffer parameter threshold value is dynamically adjusted, the matching performance of the preprocessing process and the current buffer parameter threshold value is improved, and the accuracy of determining the data quality is further improved.

It can be understood that the adjustment data in S402 is adjustment data for the current parameter threshold, so the subsequent data related to the adjustment data is subsequent real-time monitoring data, time series data and system log data, while the adjustment data in the present embodiment is adjustment data for the current buffer parameter threshold, so the subsequent data related to the present embodiment is subsequent real-time data for preprocessing.

It will be appreciated that after the pretreatment process is monitored, second real-time monitoring data can be obtained based on the monitoring results.

According to the method for determining the target vehicle data quality, when the preprocessing operation process is monitored, a current buffer parameter threshold value is obtained, and whether the preprocessing operation process is abnormal or not is judged according to the current buffer parameter threshold value; the current buffer parameter threshold is a threshold set for parameters related to data buffer during the current preprocessing operation; if yes, fourth alarm information is generated and output according to the abnormal result. Because the current buffer parameter threshold value related to the data buffer in the pretreatment operation process is stored in advance, whether the pretreatment operation process is abnormal or not can be judged according to the current buffer parameter threshold value by acquiring the current buffer parameter threshold value, and fourth alarm information can be generated and output according to the abnormal result when the occurrence of the abnormality is determined. The abnormal condition occurring in the preprocessing operation process can be timely known, the alarm is triggered, and the driving experience is further improved.

As an alternative embodiment, the present embodiment, based on the seventh embodiment, specifically includes the following steps when monitoring the analysis process:

S901, acquiring a current flow rate threshold value, and judging whether the data flow rate in the analysis process is smaller than the current flow rate threshold value. The current flow rate threshold is a threshold set for the data flow rate during the current analysis.

The current flow rate threshold is a preset threshold set for parameters involved in the analysis process, and specifically may be a threshold set for a data flow rate in the current analysis process.

In this embodiment, based on determining the requirement of the third real-time monitoring data, the analysis behavior currently in progress is monitored in real time, and the parameter characterizing the flow rate of the data is obtained based on the analysis behavior. And acquiring a pre-stored current flow rate threshold, comparing the parameter of the characterization data flow rate with the current flow rate threshold, and judging whether the parameter of the characterization data flow rate is smaller than the current flow rate threshold.

S902, if yes, determining that an analysis process is abnormal, and generating and outputting fifth alarm information according to an abnormal result.

The fifth alarm information is alarm information when the analysis process is abnormal.

In this embodiment, based on determining whether the data flow rate in the analysis process is smaller than the current flow rate threshold, if yes, it is indicated that the current data flow rate does not meet the preset flow rate requirement, and if the continuous low flow rate condition may affect the analysis process, a fifth alarm message is generated and output according to an abnormal result of the analysis process. The method for outputting the fifth alarm information is similar to the method for outputting the first alarm information in S401, and will not be described herein.

After the fifth alarm information is generated and output, the adjustment data sent by the cloud device can be obtained according to the fifth alarm information, and the current flow rate threshold value is adjusted by adopting the adjustment data to obtain an adjusted threshold value. The specific implementation is similar to S402, and will not be described here again. Therefore, the current flow speed threshold value is dynamically adjusted, the matching performance of the analysis process and the current flow speed threshold value is improved, and the accuracy of determining the data quality is further improved.

It can be understood that the adjustment data in S402 is the adjustment data for the current parameter threshold, so the subsequent data related to the adjustment data is the subsequent real-time monitoring data, the time series data and the system log data, while the adjustment data in the present embodiment is the adjustment data for the current flow rate threshold, so the subsequent data related to the present embodiment is the real-time data for subsequent analysis.

It will be appreciated that after monitoring the analysis process, third real-time monitoring data may be obtained based on the monitoring results.

According to the method for determining the target vehicle data quality, when monitoring an analysis process, a current flow rate threshold value is obtained, and whether the data flow rate in the analysis process is smaller than the current flow rate threshold value is judged; the current flow rate threshold is a threshold set for the data flow rate during the current analysis; if yes, determining that the analysis process is abnormal, and generating and outputting fifth alarm information according to the abnormal result. Since the current flow rate threshold value related to the data flow rate in the analysis process is stored in advance, by acquiring the current flow rate threshold value, it can be judged whether the data flow rate in the analysis process is smaller than the current flow rate threshold value, and it can be determined that the analysis process is abnormal when it is determined that the data flow rate is smaller than the current flow rate threshold value, so that fifth alarm information is generated and output according to the abnormal result. The method and the device can timely learn the abnormality in the analysis process and trigger the alarm, and further improve the driving experience.

And generating and outputting sixth alarm information based on the missing real-time data in response to the detection of the missing real-time data.

The sixth alarm information is alarm information when the preprocessed real-time data is missing.

It is understood that, according to the analysis processes performed in S5021 to S5023, when it is determined that there is a loss of the preprocessed real-time data, in order to avoid the loss from affecting the target vehicle, it is necessary to inform the driver and the related manager of the loss result.

Based on this, in response to detecting that the preprocessed real-time data is missing, sixth alarm information is generated and output based on the result of the missing real-time data. The method for outputting the sixth alarm information is similar to the method for outputting the first alarm information in S401, and will not be described herein.

According to the method for determining the target vehicle data quality, which is provided by the embodiment, in response to the fact that the preprocessed real-time data is missing, sixth alarm information is generated and output based on the missing real-time data. When the real-time monitoring is performed on the analysis process, a process of checking whether the preprocessed real-time data is missing or not in the analysis process is monitored, so that when the fact that the preprocessed real-time data is missing is monitored, sixth alarm information can be generated and output based on the missing real-time data. The method and the device can timely acquire the result of the data loss and trigger an alarm, and further improve driving experience.

It should be noted that, for the implementation process of the present solution, after determining whether the data quality of the target vehicle is abnormal, a corresponding data quality report may also be generated according to the abnormal or normal result, so that the relevant manager may further determine the quality condition, problem and detail of the batch of data according to the data quality report, thereby adjusting the relevant threshold more finely according to the determination result. The data quality report may specifically include data processing frequency, image data exposure condition, low brightness scene proportion, central processor and graphic processor temperature, peak occupation condition, data acquisition time, etc.

It should be noted that, for the execution process of the present solution, factors such as the execution frequency, the update frequency related to the execution process, and the storage space may be comprehensively considered, and a cache policy, for example, a FIFO (first in first out) policy, LRU (least recently used policy), etc., may be preset to store logic for executing the present solution. Specifically, the size of the buffer memory can be set according to the condition of the system hardware resource, and the storage position of the buffer memory can be determined, for example, the buffer memory is set in the memory, or the buffer memory is set in the disk. Based on this, a cache policy is obtained and written into the above-described cache. When the cache policy is read, the cache policy is searched based on the cache, if the search is successful, the cache policy is directly read, and if the search is failed, the cache policy is read from the original data source. When the cache policy is updated, the cache policy in the cache is updated accordingly based on the update result of the original data source. In addition, the expiration time or expiration condition of the cache may be set to ensure that the cache policy in the cache is not outdated or expired.

Fig. 4 is a complete flowchart of a method for determining a target vehicle data quality according to still another embodiment of the present application, as shown in fig. 4, where the method for determining a target vehicle data quality according to the present embodiment specifically includes the following steps:

s1001, collecting real-time data of a target vehicle, and preprocessing the real-time data.

S1002, analyzing the preprocessed real-time data by adopting a preset time sequence analysis strategy to obtain target time sequence data.

Specifically, when the pre-processed real-time data is analyzed by adopting a preset time sequence analysis strategy, whether the pre-processed real-time data has a deletion or not is checked. If not, carrying out time sequence arrangement on the preprocessed real-time data, and determining that the arranged real-time data is the target time sequence data. If yes, acquiring the undelayed real-time data, arranging the undelayed real-time data on a time sequence, and determining the arranged real-time data as target time sequence data.

S1003, respectively monitoring an acquisition process, a preprocessing operation process and an analysis process to obtain target real-time monitoring data.

The target real-time monitoring data comprises first real-time monitoring data, second real-time monitoring data and third real-time monitoring data.

Wherein, when monitoring the acquisition process, the method specifically comprises the following steps: judging whether the acquisition process is abnormal or not, and judging whether the state of the target vehicle is abnormal or not according to the real-time data. And generating and outputting second alarm information according to the abnormal result in response to at least one judgment result being yes. On the basis, an alarm request sent by the cloud device can be received. The alarm request is used for instructing the vehicle control device to generate and output third alarm information. And generating and outputting third alarm information according to the alarm request.

When monitoring the preprocessing operation process, the method specifically may be: and acquiring a current caching parameter threshold, and judging whether the preprocessing operation process is abnormal or not according to the current caching parameter threshold. The current buffer parameter threshold is a threshold set for parameters related to data buffering during the current preprocessing operation. If yes, fourth alarm information is generated and output according to the abnormal result.

Wherein, when monitoring the analysis process, the method specifically comprises the following steps: and acquiring a current flow rate threshold value, and judging whether the data flow rate in the analysis process is smaller than the current flow rate threshold value. The current flow rate threshold is a threshold set for the data flow rate during the current analysis. If yes, determining that the analysis process is abnormal, and generating and outputting fifth alarm information according to the abnormal result. On the basis, the sixth alarm information can be generated and output based on the missing real-time data in response to the fact that the missing real-time data is detected to exist in the preprocessed real-time data.

S1004, acquiring target real-time monitoring data, target time sequence data and target system log data corresponding to the target vehicle in response to the condition for determining the data quality.

S1005, acquiring a target data quality determination model set.

Each target data quality determining model in the target data quality determining model set is obtained by training a training sample marked with data quality parameters until convergence, and the data quality parameters are parameters representing the quality of certain data.

S1006, respectively inputting the target real-time monitoring data, the target time series data and the target system log data into each target data quality determination model.

S1007, corresponding data quality parameters are determined and output based on each target data quality determination model.

S1008, determining whether the data quality of the target vehicle is abnormal according to the data quality parameters.

Wherein the data quality parameter is a data accuracy rate, a data integrity score, or a data consistency score. The set of target data quality determination models includes a target data accuracy determination model, a target data integrity determination model, and a target data consistency determination model.

Specifically, when the model output corresponding data quality parameters are determined based on each target data quality, the model output corresponding data accuracy is determined based on the target data accuracy. And outputting the corresponding data integrity score based on the target data integrity determination model. And outputting the corresponding data consistency score based on the target data consistency determination model.

Specifically, when determining whether the data quality of the target vehicle is abnormal according to the data quality parameters, acquiring current parameter thresholds corresponding to the target data quality determination models, and judging whether the data quality parameters are equal to or greater than the corresponding current parameter thresholds. If yes, determining that the data quality of the target vehicle is normal. If not, determining that the data quality of the target vehicle is abnormal.

S1009, in response to determining that the data quality of the target vehicle is abnormal, generates and outputs first warning information based on the data quality parameter that is less than the corresponding current parameter threshold.

S1010, acquiring adjustment data according to the first alarm information.

The adjustment data are adjustment data sent by the cloud device, or adjustment data obtained based on a preset threshold prediction model, or adjustment data obtained based on a preset sliding window algorithm.

S1011, adjusting the current parameter threshold by adopting the adjustment data to obtain an adjusted parameter threshold.

Fig. 5 is a schematic structural diagram of a target vehicle data quality determining device according to an embodiment of the present application, and as shown in fig. 5, the target vehicle data quality determining device according to the embodiment is located in a vehicle control apparatus. The target vehicle data quality determination device 50 provided in the present embodiment includes: the acquisition module 51, the input module 52, the determination module 53, the output module 54.

The acquiring module 51 is configured to acquire target real-time monitoring data, target time sequence data and target system log data corresponding to a target vehicle in response to a determination condition of data quality being satisfied; the obtaining module 51 is further configured to obtain a set of target data quality determination models; each target data quality determining model in the target data quality determining model set is obtained by training a training sample marked with data quality parameters until convergence, and the data quality parameters are parameters representing the quality of certain data; the input module 52 is configured to input the target real-time monitoring data, the target time series data, and the target system log data into each target data quality determination model; a determining module 53, configured to determine corresponding data quality parameters based on each target data quality determining model; an output module 54, configured to output a corresponding data quality parameter; the determining module 53 is further configured to determine whether the data quality of the target vehicle is abnormal according to the data quality parameter.

The determining device for the target vehicle data quality provided in this embodiment may execute the method embodiment shown in fig. 2, and the specific implementation principle and technical effects are similar, and are not repeated here.

Optionally, the data quality parameter is a data accuracy rate, a data integrity score, or a data consistency score; the target data quality determining model set comprises a target data accuracy determining model, a target data integrity determining model and a target data consistency determining model;

Accordingly, the output module 54 is specifically configured to, when outputting the corresponding data quality parameter based on each target data quality determination model:

Determining the data accuracy corresponding to the model output based on the target data accuracy; and outputting a corresponding data integrity score based on the target data integrity determination model; and outputting the corresponding data consistency score based on the target data consistency determination model.

Optionally, the determining module 53 is specifically configured to, when determining whether the data quality of the target vehicle is abnormal according to the data quality parameter:

Acquiring current parameter thresholds corresponding to the target data quality determination models, and judging whether the data quality parameters are equal to or larger than the corresponding current parameter thresholds; if yes, determining that the data quality of the target vehicle is normal; if not, determining that the data quality of the target vehicle is abnormal.

The target vehicle data quality determining apparatus provided in this embodiment further includes: the generation module and the adjustment module.

Wherein, the generating module is configured to generate, after the determining module 53 determines that the data quality of the target vehicle is abnormal, first alarm information based on the data quality parameter being smaller than the corresponding current parameter threshold, and the output module 54 is further configured to output the first alarm information; the obtaining module 51 is further configured to obtain adjustment data according to the first alarm information, where the adjustment data is adjustment data sent by the cloud device, or adjustment data obtained based on a preset threshold prediction model, or adjustment data obtained based on a preset sliding window algorithm. And the adjusting module is used for adjusting the current parameter threshold by adopting the follow-up real-time monitoring data and the adjusting data so as to obtain the adjusted parameter threshold.

The target vehicle data quality determining apparatus provided in this embodiment further includes: the system comprises an acquisition module, a processing module and an analysis module.

The acquisition module is used for acquiring real-time data of the target vehicle before the acquisition module 51 responds to the determination condition of meeting the data quality to acquire the target real-time monitoring data, the target time sequence data and the target system log data corresponding to the target vehicle, and the processing module is used for preprocessing the real-time data; and the analysis module is used for analyzing the preprocessed real-time data by adopting a preset time sequence analysis strategy so as to obtain target time sequence data.

Optionally, the analysis module is specifically configured to, when analyzing the preprocessed real-time data by using a preset time sequence analysis strategy:

Checking whether the preprocessed real-time data has a deletion or not; if not, carrying out time sequence arrangement on the preprocessed real-time data, and determining that the arranged real-time data is target time sequence data; if yes, acquiring the undelayed real-time data, arranging the undelayed real-time data on a time sequence, and determining the arranged real-time data as target time sequence data.

Optionally, the target real-time monitoring data includes first real-time monitoring data, second real-time monitoring data and third real-time monitoring data. The target vehicle data quality determining apparatus provided in this embodiment further includes: and a monitoring module.

The monitoring module is used for respectively monitoring the acquisition process, the preprocessing operation process and the analysis process so as to respectively obtain first real-time monitoring data, second real-time monitoring data and third real-time monitoring data.

Optionally, the monitoring module is specifically configured to, when monitoring the acquisition process:

judging whether the acquisition process is abnormal or not, and judging whether the state of the target vehicle is abnormal or not according to the real-time data; and generating and outputting second alarm information according to the abnormal result in response to at least one judgment result being yes.

The target vehicle data quality determining apparatus provided in this embodiment further includes: and a receiving module.

The receiving module is used for receiving an alarm request sent by the cloud device; the alarm request is used for indicating the vehicle control equipment to generate and output third alarm information; the generating module is further configured to generate third alarm information according to the alarm request, and the output module 54 is further configured to output the third alarm information.

Optionally, the monitoring module is specifically configured to, when monitoring the preprocessing operation procedure:

Acquiring a current caching parameter threshold, and judging whether the preprocessing operation process is abnormal or not according to the current caching parameter threshold; the current buffer parameter threshold is a threshold set for parameters related to data buffer during the current preprocessing operation; if yes, fourth alarm information is generated and output according to the abnormal result.

Optionally, the monitoring module is specifically configured to, when monitoring the analysis process:

Acquiring a current flow rate threshold value, and judging whether the data flow rate in the analysis process is smaller than the current flow rate threshold value; the current flow rate threshold is a threshold set for the data flow rate during the current analysis; if yes, determining that the analysis process is abnormal, and generating and outputting fifth alarm information according to the abnormal result.

Optionally, the generating module is further configured to generate, in response to detecting that the preprocessed real-time data is missing, sixth alarm information based on the missing real-time data, and the output module 54 is further configured to output the sixth alarm information.

The determining device for the target vehicle data quality provided in this embodiment may execute any one of the above method embodiments, and specific implementation principles and technical effects are similar, and are not repeated here.

Fig. 6 is a schematic structural diagram of a vehicle control apparatus according to an embodiment of the present application, and as shown in fig. 6, a vehicle control apparatus 60 according to the present embodiment includes: a processor 61 and a memory 62 communicatively coupled to the processor.

Wherein the memory 62 stores computer-executable instructions; the processor 61 executes computer-executable instructions stored in the memory 62 to implement the method of determining the quality of target vehicle data provided by any one of the embodiments described above. The related descriptions and effects corresponding to the steps in the drawings can be understood correspondingly, and are not repeated here.

Wherein the program may comprise program code comprising computer-executable instructions. The memory 62 may comprise high-speed RAM memory or may further comprise non-volatile memory (non-volatile memory), such as at least one disk memory.

In this embodiment, the memory 62 and the processor 61 are connected via a bus. The bus may be an industry standard architecture (Industry Standard Architecture, abbreviated ISA) bus, an external device interconnect (PERIPHERAL COMPONENT INTERCONNECT, abbreviated PCI) bus, or an extended industry standard architecture (Extended Industry Standard Architecture, abbreviated EISA) bus, among others. The buses may be divided into address buses, data buses, control buses, etc. For ease of illustration, only one thick line is shown in fig. 6, but not only one bus or one type of bus.

The embodiment of the application also provides a computer readable storage medium, wherein computer executable instructions are stored in the computer readable storage medium, and the computer executable instructions are used for realizing the method for determining the data quality of the target vehicle provided by any one of the embodiments when being executed by a processor. For example, the computer readable storage medium may be ROM, random Access Memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.

The embodiments of the present application also provide a computer program product comprising a computer program for implementing the method for determining a target vehicle data quality provided in any of the above embodiments when executed by a processor.

It should be noted that, for simplicity of description, the foregoing method embodiments are all described as a series of acts, but it should be understood by those skilled in the art that the present application is not limited by the order of acts described, as some steps may be performed in other orders or concurrently in accordance with the present application. Further, those skilled in the art will also appreciate that the embodiments described in the specification are alternative embodiments, and that the acts and modules referred to are not necessarily required for the present application.

It should be further noted that, although the steps in the flowchart are sequentially shown as indicated by arrows, the steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least a portion of the steps in the flowcharts may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, the order in which the sub-steps or stages are performed is not necessarily sequential, and may be performed in turn or alternately with at least a portion of the sub-steps or stages of other steps or other steps.

It will be appreciated that the device embodiments described above are merely illustrative and that the device of the application may be implemented in other ways. For example, the division of the units/modules in the above embodiments is merely a logic function division, and there may be another division manner in actual implementation. For example, multiple units, modules, or components may be combined, or may be integrated into another system, or some features may be omitted or not performed.

In addition, each functional unit/module in each embodiment of the present application may be integrated into one unit/module, or each unit/module may exist alone physically, or two or more units/modules may be integrated together, unless otherwise specified. The integrated units/modules described above may be implemented either in hardware or in software program modules.

The integrated units/modules, if implemented in hardware, may be digital circuits, analog circuits, etc. Physical implementations of hardware structures include, but are not limited to, transistors, memristors, and the like. The artificial intelligence processor may be any suitable hardware processor, such as CPU, GPU, FPGA, DSP and an ASIC, etc., unless otherwise specified. Unless otherwise indicated, the storage elements may be any suitable magnetic or magneto-optical storage medium, such as resistive Random Access Memory RRAM (Resistive Random Access Memory), dynamic Random Access Memory DRAM (Dynamic Random Access Memory), static Random Access Memory SRAM (Static Random-Access Memory), enhanced dynamic Random Access Memory EDRAM (ENHANCED DYNAMIC Random Access Memory), high-Bandwidth Memory HBM (High-Bandwidth Memory), hybrid storage cube HMC (Hybrid Memory Cube), etc.

The integrated units/modules may be stored in a computer readable memory if implemented in the form of software program modules and sold or used as a stand-alone product. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in whole or in part in the form of a software product stored in a memory, comprising several instructions for causing a computer device (which may be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method of the various embodiments of the present application. And the aforementioned memory includes: a usb disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.

In the foregoing embodiments, the descriptions of the embodiments are emphasized, and for parts of one embodiment that are not described in detail, reference may be made to related descriptions of other embodiments. The technical features of the foregoing embodiments may be arbitrarily combined, and for brevity, all of the possible combinations of the technical features of the foregoing embodiments are not described, however, all of the combinations of the technical features should be considered as being within the scope of the disclosure.

Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. This application is intended to cover any variations, uses, or adaptations of the application following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the application pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.

It is to be understood that the application is not limited to the precise arrangements and instrumentalities shown in the drawings, which have been described above, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the application is limited only by the appended claims.

Claims

1. A method of determining a target vehicle data quality, the method comprising:

Responding to the determination condition of meeting the data quality, and acquiring target real-time monitoring data, target time sequence data and target system log data corresponding to a target vehicle;

2. The method of claim 1, wherein the data quality parameter is a data accuracy rate, a data integrity score, or a data consistency score; the target data quality determining model set comprises a target data accuracy determining model, a target data integrity determining model and a target data consistency determining model;

the outputting of the corresponding data quality parameters based on the target data quality determination models comprises the following steps:

determining the data accuracy corresponding to the model output based on the target data accuracy;

And outputting a corresponding data integrity score based on the target data integrity determination model;

And outputting the corresponding data consistency score based on the target data consistency determination model.

3. The method of claim 1, wherein determining whether the data quality of the target vehicle is abnormal based on the data quality parameter comprises:

Acquiring current parameter thresholds corresponding to the target data quality determination models, and judging whether the data quality parameters are equal to or larger than the corresponding current parameter thresholds;

if yes, determining that the data quality of the target vehicle is normal;

If not, determining that the data quality of the target vehicle is abnormal;

After the data quality abnormality of the target vehicle is determined, the method further comprises:

Generating and outputting first alarm information based on the data quality parameters smaller than the corresponding current parameter threshold values;

Acquiring adjustment data according to the first alarm information; the adjustment data are adjustment data sent by the cloud device, or adjustment data obtained based on a preset threshold prediction model, or adjustment data obtained based on a preset sliding window algorithm;

and adjusting the current parameter threshold by adopting the adjustment data to obtain an adjusted parameter threshold.

4. The method according to claim 1, wherein before the acquiring the target real-time monitoring data, the target time series data, and the target system log data corresponding to the target vehicle in response to the satisfaction of the determination condition of the data quality, further comprises:

collecting real-time data of a target vehicle, and preprocessing the real-time data;

analyzing the preprocessed real-time data by adopting a preset time sequence analysis strategy to obtain target time sequence data;

the target real-time monitoring data comprises first real-time monitoring data, second real-time monitoring data and third real-time monitoring data, and the method further comprises:

5. The method of claim 4, wherein monitoring the acquisition process comprises:

Judging whether the acquisition process is abnormal or not, and judging whether the state of the target vehicle is abnormal or not according to the real-time data;

generating and outputting second alarm information according to the abnormal result in response to at least one judgment result being yes;

The method further comprises the steps of:

receiving an alarm request sent by cloud equipment; the alarm request is used for indicating the vehicle control equipment to generate and output third alarm information;

and generating and outputting third alarm information according to the alarm request.

6. The method of claim 4, wherein monitoring the pretreatment operation comprises:

Acquiring a current caching parameter threshold value, and judging whether an abnormality occurs in a preprocessing operation process according to the current caching parameter threshold value; the current buffer parameter threshold is a threshold set for parameters related to data buffer in the current preprocessing operation process;

if yes, fourth alarm information is generated and output according to the abnormal result.

7. The method of claim 4, wherein monitoring the analysis process comprises:

Acquiring a current flow rate threshold value, and judging whether the data flow rate in the analysis process is smaller than the current flow rate threshold value; the current flow rate threshold value is a threshold value set for the data flow rate in the current analysis process;

If yes, determining that an analysis process is abnormal, and generating and outputting fifth alarm information according to an abnormal result;

The method further comprises the steps of:

8. A target vehicle data quality determination apparatus, characterized in that the apparatus comprises:

the acquisition module is used for responding to the determination condition of the data quality and acquiring target real-time monitoring data, target time sequence data and target system log data corresponding to the target vehicle;

9. A vehicle control apparatus characterized by comprising: a processor, and a memory communicatively coupled to the processor;

The memory stores computer-executable instructions;

the processor executes computer-executable instructions stored in the memory to implement the method of any one of claims 1 to 7.

10. A computer readable storage medium having stored therein computer executable instructions which when executed by a processor are adapted to carry out the method of any one of claims 1 to 7.

11. A computer program product, characterized in that the computer program product comprises a computer program for implementing the method according to any one of claims 1 to 7 when being executed by a processor.