CN115269573A - Method and device for complementing missing vehicle data, vehicle and storage medium - Google Patents
Method and device for complementing missing vehicle data, vehicle and storage medium Download PDFInfo
- Publication number
- CN115269573A CN115269573A CN202210858665.0A CN202210858665A CN115269573A CN 115269573 A CN115269573 A CN 115269573A CN 202210858665 A CN202210858665 A CN 202210858665A CN 115269573 A CN115269573 A CN 115269573A
- Authority
- CN
- China
- Prior art keywords
- data
- sampling
- actual
- value
- moment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
The application relates to the technical field of vehicles, in particular to a method and a device for completing missing of vehicle data, a vehicle and a storage medium, wherein the method comprises the following steps: identifying the actual sampling time and the actual sampling period of any data value in the vehicle data; judging whether data are sampled at a preset sampling moment in an actual sampling period or not, wherein the time difference value between the preset sampling moment and the actual sampling moment is smaller than a preset value; and if no sampling data exists at the preset sampling moment, inputting any data value into a prediction model matched with the actual sampling period, outputting the prediction data at the preset sampling moment, and filling the prediction data to the data position at the sampling moment. Therefore, the problems that in the related technology, the completion of the missing value has strong dependence on the un-missing data, the requirement is high, the data completion range is limited, the application scene is limited, the reliability of the filling data is poor, the completion result precision is low, the actual use requirement cannot be met and the like are solved.
Description
Technical Field
The present application relates to the field of vehicle technologies, and in particular, to a method and an apparatus for completing a vehicle data loss, a vehicle, and a storage medium.
Background
For data required by machine learning, not all data can be complete and uniform in specification, so missing value completion is a very important link in machine learning.
In the related art, the missing value may be supplemented by directly deleting the entire sample of the missing data, filling the missing value with a fixed value, hot card filling, regression interpolation, multiple interpolation, or the like. However, the above method does not meet the actual situation in some scenarios, the data to be filled is limited, only the values in the middle of the missing data can be filled, the edges of the data cannot be filled in an expanding manner, and the entire missing value filling scheme cannot be completed in a scenario where the data is extremely sparse, so that the problem of data missing still remains to be solved.
Disclosure of Invention
The application provides a method and a device for completing missing data of a vehicle, the vehicle and a storage medium, and aims to solve the problems that in the related technology, the dependency of missing value completion on un-missing data is high, the requirement is high, the data completion range is limited, the application scene is limited, the reliability of filling data is poor, the completion result precision is low, the actual use requirement cannot be met, and the like.
The embodiment of the first aspect of the application provides a method for completing the missing vehicle data, which comprises the following steps: identifying the actual sampling time and the actual sampling period of any data value in the vehicle data; judging whether data are sampled at a preset sampling moment in the actual sampling period, wherein a time difference value between the preset sampling moment and the actual sampling moment is smaller than a preset value; and if no sampling data exists at the preset sampling moment, inputting the arbitrary data value into a prediction model matched with the actual sampling period, outputting the prediction data at the preset sampling moment, and filling the prediction data to the data position at the sampling moment.
According to the technical means, the embodiment of the application can judge whether the data are missing or not, and the radiation filling is carried out on the surrounding data by traversing the actually collected data values and utilizing the prediction data model so as to fill the missing data; therefore, the embodiment of the application can complement the missing data value to the greatest extent, can fill data except the edge data, eliminates the influence of the missing data on the vehicle, improves safety, meets the requirement of actual use, and improves the use experience of a user.
Optionally, the prediction model is pre-constructed based on training data carrying a sampling period label, and includes: acquiring training data carrying a sampling period label; performing curve fitting according to the training data to obtain a data change curve of the training data; and generating the prediction models with different sampling periods according to the data change curves and the sampling periods of the training data.
According to the technical means, the prediction model matched with any automobile data can be obtained through training fitting by using real sampling data, validity and reliability of the data are guaranteed, accuracy of results is improved, and requirements of actual use are met.
Optionally, identifying an actual sampling period in which any data value in the vehicle data is located includes: acquiring at least one data value near the actual sampling moment; and matching the actual sampling period of the arbitrary data value according to the data change trend of the arbitrary data value and the at least one data value.
According to the technical means, the embodiment of the application can sample at least one nearby data value, combine the random data and nearby sampled data, and match the actual sampling period of the data according to the change trend of the data.
Optionally, the identifying the actual sampling time and the actual sampling period of any data value in the vehicle data are preceded by: judging an actual sampling type of the vehicle data, wherein the actual sampling type is a periodic sampling type and a variable sampling type; if the actual sampling type is the periodic sampling type, identifying the actual sampling time and the actual sampling period; and if the actual sampling type is the change sampling type, filling a data value at a previous moment before any moment in the vehicle data to a data position at any moment when the fact that no data is sampled at any moment in the vehicle data is identified.
According to the technical means, each effective data can be periodically filled according to different types of sampling data, different periods of the effective data are divided according to adjacent data of the effective data or other characteristic data at the same time, the data has strong fitting property in a data scene with periodic change, and guiding significance is provided for filling the periodic change data; in addition, the embodiment of the application does not need to wait until the prediction model is matched and then judge, so that the intellectualization is improved, the redundancy of the data processing process is effectively avoided, the operation cost is saved, and the efficiency is improved.
The second aspect of the present application provides a device for completing the missing vehicle data, including: the identification module is used for identifying the actual sampling time and the actual sampling period of any data value in the vehicle data; the first judgment module is used for judging whether data are sampled at a preset sampling moment in the actual sampling period or not, wherein the time difference value between the preset sampling moment and the actual sampling moment is smaller than a preset value; and the execution module is used for inputting the arbitrary data value into a prediction model matched with the actual sampling period if no sampling data exists at the preset sampling moment, outputting the prediction data at the preset sampling moment and filling the prediction data to the data position at the sampling moment.
Optionally, the prediction model is pre-constructed based on training data carrying a sampling period label, and the execution module is configured to: acquiring training data carrying a sampling period label; performing curve fitting according to the training data to obtain a data change curve of the training data; and generating the prediction models with different sampling periods according to the data change curves and the sampling periods of the training data.
Optionally, the identification module is configured to: acquiring at least one data value near the actual sampling moment; and matching the actual sampling period of the arbitrary data value according to the data change trend of the arbitrary data value and the at least one data value.
Optionally, the apparatus for supplementing missing vehicle data further includes a second determining module, configured to: judging an actual sampling type of the vehicle data, wherein the actual sampling type is a periodic sampling type and a variable sampling type; if the actual sampling type is the periodic sampling type, identifying the actual sampling time and the actual sampling period; and if the actual sampling type is the change sampling type, filling a data value at a previous moment before any moment in the vehicle data to a data position at any moment when the fact that no data is sampled at any moment in the vehicle data is identified.
An embodiment of a third aspect of the present application provides a vehicle, comprising: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor executing the program to implement the method for vehicle data miss completion as described in the above embodiments.
A fourth aspect of the present application provides a computer-readable storage medium, on which a computer program is stored, the program being executed by a processor for implementing the method for supplementing missing vehicle data as described in the above embodiments.
Therefore, the application has at least the following beneficial effects:
(1) The method and the device can judge whether the data are missing or not, and utilize the prediction data model to perform radiation filling on the surrounding data by traversing the actually acquired data values so as to fill the missing data; therefore, the data missing value can be supplemented to the maximum extent, data except the edge data can be filled, the influence of the data missing on the vehicle is eliminated, the safety is improved, the requirement of actual use is met, and the use experience of a user is improved;
(2) According to the embodiment of the application, the prediction model matched with any automobile data can be obtained through training fitting by using real sampling data, so that the validity and the reliability of the data are ensured, the accuracy of the result is improved, and the requirement of actual use is met;
(3) The embodiment of the application can sample at least one nearby data value, combine any data with nearby sampled data, and match the actual sampling period of the data according to the change trend of the data;
(4) According to the embodiment of the application, each effective data can be periodically filled according to different types of sampling data, different periods of the effective data are divided according to adjacent data of the effective data or other characteristic data at the same time, the effective data has strong fitting property in a data scene with periodic variation, and guiding significance is provided for filling of the periodic variation data; in addition, the embodiment of the application does not need to wait until the prediction model is matched for judgment, so that the intellectualization is improved, the redundancy of the data processing process is effectively avoided, the operation cost is saved, and the efficiency is improved.
Additional aspects and advantages of the present application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the present application.
Drawings
The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a schematic diagram of a missing data distribution of data samples provided according to an embodiment of the present application;
FIG. 2 is a flow chart of a method for supplementing missing vehicle data according to an embodiment of the present application;
FIG. 3 is a graph showing temperature rise in an un-turned on air-conditioned vehicle as a function of time according to an embodiment of the present application;
FIG. 4 is a graph showing the temperature drop in an air-conditioned vehicle as a function of time according to an embodiment of the application;
FIG. 5 is a graph of outdoor temperature versus time provided in accordance with an embodiment of the present application;
FIG. 6 is a flowchart illustrating missing values filling process for patent data according to an embodiment of the present application;
FIG. 7 is an exemplary diagram of a missing replenishment arrangement for vehicle data provided in accordance with an embodiment of the present application;
FIG. 8 is a schematic structural diagram of a vehicle according to an embodiment of the present application.
Detailed Description
Reference will now be made in detail to the embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are illustrative and intended to explain the present application and should not be construed as limiting the present application.
For data collected by an automobile sensor, two collection modes, namely variable sampling and periodic sampling, are mainly adopted in the related art. Wherein, the change sampling is to record the change information and the change time when the value or the state changes; the periodic sampling is to trigger an acquisition request according to a fixed frequency and record the current value or state and time information. The application scenes of the variable sampling and the periodic sampling are different, and the variable sampling can be adopted for some state signals, such as normal or abnormal signals, opening or closing signals and the like; for a value having a continuous change, since it is a continuous value, even if a change sample is used, it eventually becomes a periodic sample, and thus a periodic sample may be employed; the use of the variation sampling and the period sampling may be specifically selected according to actual situations, and is not limited to the above.
Since the systems on the vehicle are very complex and involve many devices, it may be necessary to use the data of different sensors or even different recording devices of the vehicle when performing some analyses; when all data are gathered together, due to the difference of sampling modes and sampling periods, sample data acquired at the same time point is not necessarily complete, wherein the missing data distribution of the data samples can be shown in fig. 1; therefore, filling in missing values becomes an essential step.
For data required by machine learning, not all data can be complete and uniform in specification, but data loss has great influence on the result of the whole machine learning or deep learning in a scene with a small data sample; for the case that part of the sample data contains missing values, if the missing values are deleted in a rough manner, the related information of the data that is not missing in the missing information sample may be lost. Therefore, missing value completion is a very important link in machine learning.
In the related art, the deficiency value completion method is roughly as follows: (1) The method can be used for the conditions that the data sample set is large and the proportion of the missing samples is controlled within a certain threshold value; (2) The missing values are filled in using fixed values, which are commonly used and can be: 0. mean, median, mode, etc.; (3) Hot card filling, namely filling values of the missing samples by using samples similar to the missing samples or adjacent sample data; (4) Performing regression interpolation, namely establishing a regression equation for the characteristics of the missing data, and predicting filling values by using the regression equation; (5) Multiple interpolation can be understood as a more complex and more considered regression interpolation method.
The missing value filling method of the related art is mainly based on the existing data in the missing data set, and uses a pre-estimation or statistical means to fill the missing value; however, for data with a relatively sparse Matrix, the FM (Matrix Factorization) technique and the dimensionality reduction technique used in the related art, such as PCA (principal component Analysis) and LDA (Linear Discriminant Analysis), cannot solve the problems encountered in the present scenario. Moreover, in the missing value filling schemes listed above, any filling scheme cannot represent the change rule in the actual environment, for example, a certain numerical value such as the change of the temperature in the vehicle is taken as an example, as shown in fig. 1, only a few values exist in a certain time period interval, and the few values obviously cannot correctly simulate a correct curve to predict the missing data, and at this time, the missing value filling of the parameters cannot perform interpolation or use a fixed value.
A method, an apparatus, a vehicle, and a storage medium for complementing missing vehicle data according to embodiments of the present application will be described below with reference to the accompanying drawings.
Specifically, fig. 2 is a schematic flowchart of a method for supplementing missing vehicle data according to an embodiment of the present disclosure.
As shown in fig. 2, the method for complementing the missing vehicle data includes the steps of:
in step S101, the actual sampling timing and the actual sampling period of any data value in the vehicle data are identified.
The vehicle data may be any data value of various types of the vehicle, such as temperature data, etc., and the actual sampling period may correspond to the data type of the sampled data, which is obtained according to the actual situation, which is not specifically limited.
For convenience of description, in the following embodiments, temperature data will be exemplified in the examples of the present application. When the vehicle data is temperature data, the actual sampling period of the present application may include four types: starting an automobile engine under the condition of not starting an air conditioner, and obtaining a change curve (temperature rise or temperature drop) of the temperature in the automobile; under the condition of not starting the air conditioner, the automobile engine is not started, and the change curve (temperature rise or temperature drop) of the temperature in the automobile is obtained; starting an automobile engine under the condition of starting an air conditioner, and obtaining a change curve (temperature rise or temperature drop) of the temperature in the automobile; when the air conditioner is started, the engine of the automobile is not started, and the temperature in the automobile changes (increases or decreases).
It can be understood that, in the embodiment of the present application, a data value of any vehicle data may be selected for identification, an actual sampling time of the data is obtained, and an actual sampling period of the data when the data is sampled is obtained, so as to be used for generating the possibly existing prediction data in the subsequent step.
In the embodiment of the present application, identifying the actual sampling period in which any data value in the vehicle data is located includes: acquiring at least one data value near the actual sampling moment; and matching the actual sampling period of the arbitrary data value according to the data change trend of the arbitrary data value and the at least one data value.
It can be understood that, the embodiment of the application can acquire at least one other data value near the identified actual sampling time of any vehicle data, and match the actual sampling period of the data value by using the change trend embodied by the data.
For example, taking the temperature in the vehicle as an example, a value containing temperature data and a time position where the value is located are found, a relevant state of the value near the value and the current temperature value is detected, for example, a driver can check the relevant state of the temperature and the current temperature value, the state of whether the air conditioner is started or not, the state of the air conditioner compressor, the outdoor environment temperature value and the like, and then a model period (cooling by turning on the air conditioner, heating by turning on the air conditioner, dimension reduction by not turning on the air conditioner, heating by not turning on the air conditioner and the like) where the temperature value is located is judged according to a data change trend embodied by the temperature value and at least one relevant state value.
In the embodiment of the present application, identifying the actual sampling time and the actual sampling period of any data value in the vehicle data, further includes: judging the actual sampling type of the vehicle data, wherein the actual sampling type is a periodic sampling type and a variable sampling type; if the actual sampling type is a periodic sampling type, identifying the actual sampling moment and the actual sampling period; and if the actual sampling type is the variation sampling type, when the data which is not sampled at any time in the vehicle data is identified, filling the data value at the previous time at any time into the data position at any time.
It can be understood that, according to the embodiment of the application, the actual sampling type of the vehicle data can be judged firstly before the actual sampling time and the actual sampling period of any data of the vehicle are identified, further operation on the data is selected according to the sampling type of the data, and judgment is not needed before the prediction model is matched, so that the scheme is more intelligent, the redundancy of the data processing process is effectively avoided, the operation cost is saved, and the efficiency is improved.
Specifically, the actual sampling type of any vehicle data can be obtained and recognized, when the data is change sampling, the data is recorded only when the change sampling is changed every time, a temperature field change model is not needed to be used for predicting the data, and the middle vacant data can be filled by using the last data; when the data is a periodic sample, the above-described step S101 is executed, and the subsequent steps S102 and S103 are executed.
In step S102, it is determined whether data is sampled at a preset sampling time within an actual sampling period, where a time difference between the preset sampling time and the actual sampling time is smaller than a preset value.
The preset sampling time may be set according to the actual sampling time of any data acquired in step S101, for example, the preset sampling time may be a target sampling time that is closest to any data time within a sampling period in which any data is located, and the like; the preset value may be specifically set according to actual conditions, and is not specifically limited.
It can be understood that, in the embodiment of the present application, a preset sampling time and a preset value can be set, and whether the data is sampled at the preset sampling time is judged, and if the data is sampled at the preset sampling time, it can be understood that there is no data vacancy at the time; if no sample data exists at the preset moment, the data is understood to be vacant at the moment.
It should be noted that when the time difference between the actual sampling time and the preset sampling time of any data is smaller than the preset value, the data can be understood as the data of which the any data is similar to the preset sampling time, and the variation trend represented by the data at the preset sampling time can also be reflected, so that the method can be applied to the subsequent steps; when the time difference between the actual sampling time of any data and the preset sampling time is greater than the preset value, the method and the device for identifying the vehicle data can return to the step S101 to continuously identify any other vehicle data until the data meeting the requirements are obtained.
In step S103, if there is no sampling data at the preset sampling time, an arbitrary data value is input to the prediction model matching the actual sampling period, prediction data at the preset sampling time is output, and the prediction data is filled to the data position at the sampling time.
It can be understood that if there is no sampling data at the preset sampling time, the data vacancy at the preset sampling time can be understood, and therefore, in the embodiment of the present application, after the cycle state of the current data is determined, missing data of a time near the data value is predicted and filled, any data that satisfies the step S102 is input into a corresponding prediction model to obtain predicted data at the predicted sampling time, and the data vacancy at the time is filled with the predicted data.
In the embodiment of the present application, the prediction model is pre-constructed based on training data carrying a sampling period label, and includes: acquiring training data carrying a sampling period label; performing curve fitting according to the training data to obtain a data change curve of the training data; and generating prediction models with different sampling periods according to the data change curve and the sampling period of the training data.
It can be understood that, in the embodiments of the present application, a data variation curve may be generated by fitting training data with a sampling period label, and then a prediction model of a sampling period may be generated based on the variation curve and the sampling period of the training data.
For example, taking temperature data as an example, the period of the embodiment of the present application may be as shown in the above four types of variation periods, respectively sampling data of temperature variation data in an automobile in four periods, and fitting the sampled data by using a deep learning model to obtain 8 fitting models (four scenes have temperature rise or temperature fall respectively) about the time variation of a temperature field in the automobile, where the obtained curve of the temperature rise in the automobile without turning on an air conditioner along with time may be as shown in fig. 3, and the curve of the temperature fall in the automobile with air conditioner along with time may be as shown in fig. 4; similarly, the embodiment of the application can acquire the temperature of the environment temperature change, fit the data of the temperature rise or the temperature drop of the environment temperature along with the time, and fit the two models by using deep learning; wherein, the time-varying outdoor temperature curve can be shown in fig. 5.
The missing data filling process of the present application will be described with an embodiment, in which the vehicle data is temperature data, as shown in fig. 6, specifically as follows:
(1) The flow begins, and the sampling type of the data is determined, namely the data belongs to a change sampling type or a periodic sampling type;
(2) Judging whether the data is periodically sampled data, if so, continuing the step (3); if the data is not the periodic sampling data, judging that the data is the variable sampling data, and using the last effectively acquired data as the subsequent data filling value;
(3) When the data is the periodic sampling data, judging which period the current data belongs to according to the adjacent data of the current data or other signal values of the current data;
(4) After the period is determined, finding an estimated model in the period, and taking the estimated model as the data and the reverse filling value of the vacant time point in the data radiation range;
(5) The data are input into a driver air conditioner temperature setting recommendation system, so that a more comprehensive and accurate data set is provided for the system; the flow ends.
In summary, the missing temperature values in the same period can be predicted in the same period, and the missing data is filled by using different filling value methods according to the mode of automobile data sampling, so that the embodiment of the application can maximize the data missing values and obtain the most accurate filling effect by traversing all the actually acquired data values and performing radiation filling on the surrounding data; data except the edge data can be filled and input into a subsequent deep learning model, so that the learning effect of the deep learning model is improved; the change rule of data loss in an actual scene is met, the accuracy of the obtained data result is improved, and the user experience is improved.
According to the method for completing the missing of the vehicle data, the beneficial effects are as follows:
(1) The method and the device can judge whether the data are missing or not, and utilize the prediction data model to perform radiation filling on the surrounding data by traversing the actually acquired data values so as to fill the missing data; therefore, the data missing value can be supplemented to the maximum extent, data except the edge data can be filled, the influence of the data missing on the vehicle is eliminated, the safety is improved, the requirement of actual use is met, and the use experience of a user is improved;
(2) According to the embodiment of the application, the prediction model matched with any automobile data can be obtained through training fitting by using real sampling data, so that the validity and the reliability of the data are ensured, the accuracy of the result is improved, and the actual use requirement is met;
(3) The embodiment of the application can sample at least one nearby data value, combine any data with nearby sampled data, and match the actual sampling period of the data according to the change trend of the data;
(4) According to the embodiment of the application, each effective data can be periodically filled according to different types of sampling data, different periods of the effective data are divided according to adjacent data of the effective data or other characteristic data at the same time, the effective data has strong fitting property in a data scene with periodic variation, and a guiding significance is provided for filling the periodic variation data; in addition, the embodiment of the application does not need to wait until the prediction model is matched for judgment, so that the intellectualization is improved, the redundancy of the data processing process is effectively avoided, the operation cost is saved, and the efficiency is improved.
Next, a missing complementing device of vehicle data proposed according to an embodiment of the present application is described with reference to the drawings.
Fig. 7 is a block diagram schematically illustrating a vehicle data missing completion apparatus according to an embodiment of the present application.
As shown in fig. 7, the vehicle data missing completion apparatus 10 includes: an identification module 100, a first judgment module 200 and an execution module 300.
The identification module 100 is used for identifying the actual sampling time and the actual sampling period of any data value in the vehicle data; the first judging module 200 is configured to judge whether data is sampled at a preset sampling time in an actual sampling period, where a time difference between the preset sampling time and the actual sampling time is smaller than a preset value; the execution module 300 is configured to, if there is no sampling data at the preset sampling time, input any data value into the prediction model matched with the actual sampling period, output prediction data at the preset sampling time, and fill the prediction data in the data position at the sampling time.
In this embodiment of the present application, the prediction model is obtained by pre-constructing based on training data carrying a sampling period label, and the execution module 300 is configured to: acquiring training data carrying a sampling period label; performing curve fitting according to the training data to obtain a data change curve of the training data; and generating prediction models with different sampling periods according to the data change curves and the sampling periods of the training data.
In an embodiment of the present application, the identification module 100 is configured to: acquiring at least one data value near the actual sampling moment; and matching the actual sampling period of the arbitrary data value according to the data change trend of the arbitrary data value and the at least one data value.
In the embodiment of the present application, the apparatus 10 for supplementing missing vehicle data further includes a second determination module, configured to: judging the actual sampling type of the vehicle data, wherein the actual sampling type is a periodic sampling type and a variable sampling type; if the actual sampling type is a periodic sampling type, identifying the actual sampling moment and the actual sampling period; and if the actual sampling type is the change sampling type, when the data which is not sampled at any time in the vehicle data is identified, filling the data value at the previous time at any time to the data position at any time.
It should be noted that the explanation of the embodiment of the method for supplementing missing vehicle data is also applicable to the device for supplementing missing vehicle data of this embodiment, and details are not repeated here.
According to the missing completion device for the vehicle data, the missing data can be filled by using different filling methods according to the sampling mode of the vehicle data in the same period. Therefore, the embodiment of the application can achieve the maximization of the data missing value and obtain the most accurate filling effect by traversing all the actually acquired data values and carrying out radiation filling on the surrounding data; data except the edge data can be filled and input into a subsequent deep learning model, so that the learning effect of the deep learning model is improved; the change rule of data loss in an actual scene is met, the accuracy of the obtained data result is improved, and the user experience is improved.
Fig. 8 is a schematic structural diagram of a vehicle according to an embodiment of the present application. The vehicle may include:
a memory 801, a processor 802, and a computer program stored on the memory 801 and executable on the processor 802.
The processor 802 executes the program to implement the method for complementing the missing vehicle data provided in the above-described embodiment.
Optionally, the vehicle further comprises:
a communication interface 803 for communicating between the memory 801 and the processor 802.
A memory 801 for storing computer programs operable on the processor 802.
The Memory 801 may include a high-speed RAM (Random Access Memory) Memory, and may also include a non-volatile Memory, such as at least one disk Memory.
If the memory 801, the processor 802 and the communication interface 803 are implemented independently, the communication interface 803, the memory 801 and the processor 802 may be connected to each other via a bus and communicate with each other. The bus may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown in FIG. 8, but this is not intended to represent only one bus or type of bus.
Optionally, in a specific implementation, if the memory 801, the processor 802, and the communication interface 803 are integrated on one chip, the memory 801, the processor 802, and the communication interface 803 may complete mutual communication through an internal interface.
The processor 802 may be a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), or one or more Integrated circuits configured to implement embodiments of the present Application.
Embodiments of the present application also provide a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to implement the above method for supplementing missing vehicle data.
In the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or N embodiments or examples. Moreover, various embodiments or examples and features of various embodiments or examples described in this specification can be combined and combined by one skilled in the art without being mutually inconsistent.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "N" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more N executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of implementing the embodiments of the present application.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a programmable gate array, a field programmable gate array, or the like.
It will be understood by those skilled in the art that all or part of the steps carried out in the method of implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and the program, when executed, includes one or a combination of the steps of the method embodiments.
While embodiments of the present application have been shown and described above, it will be understood that the above embodiments are exemplary and should not be construed as limiting the present application and that changes, modifications, substitutions and alterations in the above embodiments may be made by those of ordinary skill in the art within the scope of the present application.
Claims (10)
1. A method for complementing missing vehicle data is characterized by comprising the following steps:
identifying the actual sampling time and the actual sampling period of any data value in the vehicle data;
judging whether data are sampled at a preset sampling moment in the actual sampling period, wherein a time difference value between the preset sampling moment and the actual sampling moment is smaller than a preset value;
and if no sampling data exists at the preset sampling moment, inputting the arbitrary data value into a prediction model matched with the actual sampling period, outputting the prediction data at the preset sampling moment, and filling the prediction data to the data position at the sampling moment.
2. The method of claim 1, wherein the prediction model is pre-constructed based on training data carrying sampling period labels, and comprises:
acquiring training data carrying a sampling period label;
performing curve fitting according to the training data to obtain a data change curve of the training data;
and generating the prediction models with different sampling periods according to the data change curve and the sampling period of the training data.
3. The method of claim 1, wherein identifying an actual sampling period in which any data value in the vehicle data is located comprises:
acquiring at least one data value near the actual sampling moment;
and matching the actual sampling period of the arbitrary data value according to the data change trend of the arbitrary data value and the at least one data value.
4. The method of any one of claims 1-3, wherein the identifying the actual sampling time and the actual sampling period for any data value in the vehicle data is preceded by:
judging an actual sampling type of the vehicle data, wherein the actual sampling type is a periodic sampling type and a variable sampling type;
if the actual sampling type is the periodic sampling type, identifying the actual sampling time and the actual sampling period;
and if the actual sampling type is the change sampling type, filling a data value at a previous moment before any moment in the vehicle data to a data position at any moment when the fact that no data is sampled at any moment in the vehicle data is identified.
5. A missing complementing device for vehicle data, characterized by comprising:
the identification module is used for identifying the actual sampling time and the actual sampling period of any data value in the vehicle data;
the first judgment module is used for judging whether data are sampled at a preset sampling moment in the actual sampling period, wherein a time difference value between the preset sampling moment and the actual sampling moment is smaller than a preset value;
and the execution module is used for inputting the arbitrary data value into a prediction model matched with the actual sampling period if no sampling data exists at the preset sampling moment, outputting the prediction data at the preset sampling moment and filling the prediction data to the data position at the sampling moment.
6. The apparatus of claim 5, wherein the prediction model is pre-constructed based on training data carrying sampling period labels, and the execution module is configured to:
acquiring training data carrying a sampling period label;
performing curve fitting according to the training data to obtain a data change curve of the training data;
and generating the prediction models with different sampling periods according to the data change curve and the sampling period of the training data.
7. The apparatus of claim 5, wherein the identification module is configured to:
acquiring at least one data value near the actual sampling moment;
and matching the actual sampling period of the arbitrary data value according to the data change trend of the arbitrary data value and the at least one data value.
8. The apparatus according to any one of claims 5 to 7, wherein the apparatus for complementing the missing vehicle data further comprises a second determining module configured to:
judging an actual sampling type of the vehicle data, wherein the actual sampling type is a periodic sampling type and a variable sampling type;
if the actual sampling type is the periodic sampling type, identifying the actual sampling moment and the actual sampling period;
and if the actual sampling type is the variation sampling type, when the fact that no data is sampled at any time in the vehicle data is identified, filling a data value at the previous time of the any time to a data position at the any time.
9. A vehicle, characterized by comprising: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor executing the program to implement the method of missing replenishment of vehicle data according to any one of claims 1 to 4.
10. A computer-readable storage medium, on which a computer program is stored, characterized in that the program is executed by a processor for implementing the method for missing-supplement of vehicle data according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210858665.0A CN115269573A (en) | 2022-07-20 | 2022-07-20 | Method and device for complementing missing vehicle data, vehicle and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210858665.0A CN115269573A (en) | 2022-07-20 | 2022-07-20 | Method and device for complementing missing vehicle data, vehicle and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115269573A true CN115269573A (en) | 2022-11-01 |
Family
ID=83767303
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210858665.0A Pending CN115269573A (en) | 2022-07-20 | 2022-07-20 | Method and device for complementing missing vehicle data, vehicle and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115269573A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117874435A (en) * | 2024-03-11 | 2024-04-12 | 中国电子科技集团公司第十五研究所 | Distributed edge data acquisition method and device, electronic equipment and storage medium |
-
2022
- 2022-07-20 CN CN202210858665.0A patent/CN115269573A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117874435A (en) * | 2024-03-11 | 2024-04-12 | 中国电子科技集团公司第十五研究所 | Distributed edge data acquisition method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111401609B (en) | Prediction method and prediction device for traffic flow time series | |
| US6829515B2 (en) | Method and device for determining changes in technical systems such as electric motors caused by ageing | |
| EP1835426A1 (en) | Estimating software power consumption | |
| US20080059020A1 (en) | Data Recorder For Vehicle | |
| CN115269573A (en) | Method and device for complementing missing vehicle data, vehicle and storage medium | |
| CN112271359A (en) | Heating system control method and device based on battery and electronic equipment | |
| US11199419B2 (en) | Method for reducing exhaust gas emissions of a drive system of a vehicle including an internal combustion engine | |
| CN112149908A (en) | Vehicle driving prediction method, system, computer device and readable storage medium | |
| US20210390802A1 (en) | Method, Computer Program And Device For Processing Signals | |
| CN116821638B (en) | Data analysis method and system for AI chip application optimization design | |
| CN112560974B (en) | Information fusion and vehicle information acquisition method and device | |
| EP2778640B1 (en) | Outside air temperature measurement device and method | |
| JP6758403B2 (en) | Deterioration diagnosis system, deterioration diagnosis method and deterioration diagnosis program | |
| CN114312213B (en) | Method, device, equipment and medium for predicting abnormal refrigerant capacity of vehicle | |
| US20150226808A1 (en) | Method of Use for Electric Energy Stores, Arrangement for carrying out such a Method of Use, Battery and Motor Vehicle having such a Battery | |
| CN115655753A (en) | Electric automobile endurance mileage testing method, system, device and storage medium | |
| CN113378304B (en) | Method and device for determining vehicle performance target, storage medium and equipment | |
| CN116257437A (en) | ADAS system defect verification method and device based on real vehicle data reinjection | |
| US20220215138A1 (en) | Method for Validating System Parameters of an Energy System, Method for Operating an Energy System, and Energy Management System for an Energy System | |
| CN114327557A (en) | Vehicle OTA (over the air) upgrade time evaluation method and device | |
| CN111340349B (en) | Product reliability evaluation method, device, computer equipment and storage medium | |
| CN111143423A (en) | Dynamic scene labeling data mining method and device and terminal | |
| WO2022180681A1 (en) | Data generation system, data generation method, and data generation program | |
| US20240142916A1 (en) | Method and control unit for automatically selecting data sets for a method for machine learning | |
| CN112906953B (en) | People flow prediction method, device, computer equipment and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |