Sensor data repairing method and device, equipment and storage medium

Abstract

The application relates to the technical field of data processing, and discloses a sensor data repairing method, device, equipment and storage medium, the method comprising the following steps: when it is detected that current sensor data of a target vehicle is lost, acquiring historical sensor data of the target vehicle, sensor data of a vehicle model same as the vehicle model of the target vehicle, and sensor data of a region suitable for a current region according to data types; predicting the lost part in the current sensor data by a preset deep learning network model according to the historical sensor data, the sensor data of the vehicle model and the sensor data of the region, so as to obtain target filling data; and filling the current sensor data according to the target filling data, so as to realize the repairing of the current sensor data. Compared with the prior art of repairing data by a singular value decomposition strategy, the accuracy of the repaired sensor data can be effectively improved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to methods, apparatus, devices and storage media for repairing sensor data. Background Technology

[0002] As a crucial component of vehicles, sensors directly impact vehicle safety and driver safety through the quality of their measurement data. More importantly, they are vital for data analysis in research and development departments. However, during vehicle operation, data loss can occur when sensors upload measurement data, significantly affecting data analysis. Therefore, it is urgent to repair lost data using existing data. Currently, a common data repair method is to use singular value decomposition (SVD) to identify missing data from the decomposed data. However, this method only repairs partially lost data. If multiple data entries are missing, the accuracy of this method for repairing lost data is low.

[0003] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0004] The main objective of this invention is to provide a method, apparatus, device, and storage medium for repairing sensor data, aiming to solve the technical problem of low accuracy in repairing lost data in existing technologies.

[0005] To achieve the above objectives, the present invention provides a method for repairing sensor data, the method comprising the following steps:

[0006] When it is detected that the current sensor data of the target vehicle is lost, the vehicle model, the current location where the target vehicle is traveling, and the data type of the current sensor data are determined.

[0007] Based on the data type, obtain the historical sensor data of the target vehicle, the sensor data of the same model as the target vehicle, and the location sensor data adapted to the current location;

[0008] Based on the historical sensor data, vehicle model sensor data, and location sensor data, a preset deep learning network model is used to predict the missing parts in the current sensor data to obtain target filling data;

[0009] The current sensor data is filled in based on the target filling data to achieve the repair of the current sensor data.

[0010] Optionally, when detecting that the current sensor data of the target vehicle is lost, the model of the target vehicle, the current section where the target vehicle travels, and the data type of the current sensor data are determined, comprising:

[0011] When detecting that the current sensor data of the target vehicle is lost, the identification information of the target vehicle is extracted;

[0012] According to the identification information, the model of the target vehicle is determined;

[0013] According to the identification information, the current section where the target vehicle travels is queried;

[0014] The current sensor data of the target vehicle is analyzed to obtain the corresponding data type.

[0015] Optionally, according to the data type, the historical sensor data of the target vehicle, the same model sensor data of the model of the target vehicle, and the section sensor data suitable for the current section are obtained, comprising:

[0016] According to the data type, the historical sensor data of the target vehicle is obtained;

[0017] According to the current section where the target vehicle travels, the energy consumption, the driving mileage, and the latitude of the current section of the target vehicle are obtained;

[0018] According to the energy consumption, the driving mileage, the latitude of the current section, and the data type of the target vehicle, the section sensor data suitable for the current section is obtained;

[0019] According to the identification information and the data type of the target vehicle, the same model sensor data of the model of the target vehicle is queried.

[0020] Optionally, according to the energy consumption, the driving mileage, the latitude of the current section, and the data type of the target vehicle, the section sensor data suitable for the current section is obtained, comprising:

[0021] The starting latitude and the ending latitude of the latitude of the current section are extracted;

[0022] According to the starting latitude, the ending latitude, the energy consumption, and the driving mileage, the driving route of the target vehicle is determined;

[0023] According to the driving route and the data type, the section sensor data suitable for the current section is obtained.

[0024] Optionally, before the predicting the missing part of the current sensor data according to the historical sensor data, the vehicle type sensor data and the section sensor data through a preset deep learning network model to obtain target filling data, the method further comprises:

[0025] extracting upload start time, interval time and upload end time of the current sensor data;

[0026] determining sensor data of each time period according to the upload start time, the interval time and the upload end time;

[0027] generating a corresponding sensor data continuity table according to the sensor data of each time period;

[0028] when there is a preset null value in the sensor data continuity table, obtaining the missing part of the current sensor data.

[0029] Optionally, before the predicting the missing part of the current sensor data according to the historical sensor data, the vehicle type sensor data and the section sensor data through a preset deep learning network model to obtain target filling data, the method further comprises:

[0030] counting the number of missing parts in the current sensor data;

[0031] when the number of missing parts in the current sensor data is greater than a preset number threshold, performing the step of predicting the missing part of the current sensor data according to the historical sensor data, the vehicle type sensor data and the section sensor data through a preset deep learning network model to obtain target filling data.

[0032] Optionally, after the counting the number of missing parts in the current sensor data, the method further comprises:

[0033] when the number of missing parts in the current sensor data is equal to a preset number threshold, obtaining the last sensor data and the next sensor data of the missing part in the current sensor data;

[0034] calculating the last sensor data and the next sensor data according to a preset calculation strategy to obtain target filling data.

[0035] In addition, in order to achieve the above-mentioned purpose, the application further provides a sensor data repairing device, which comprises:

[0036] a determining module, configured to determine a vehicle type of a target vehicle, a current section where the target vehicle travels, and a data type of current sensor data of the target vehicle when it is detected that the current sensor data of the target vehicle has a missing part.

[0037] an acquisition module configured to acquire historical sensor data of the target vehicle, vehicle model sensor data of the same vehicle model as the target vehicle, and region sensor data suitable for the current region according to the data type;

[0038] a prediction module configured to predict missing parts in the current sensor data according to the historical sensor data, the vehicle model sensor data, and the region sensor data through a preset deep learning network model to obtain target filling data;

[0039] a filling module configured to fill the current sensor data according to the target filling data to achieve repair of the current sensor data.

[0040] In addition, to achieve the above object, the present application further provides a sensor data repair device, which comprises a memory, a processor, and a sensor data repair program stored in the memory and executable on the processor, and the sensor data repair program is configured to implement the sensor data repair method as described above.

[0041] In addition, to achieve the above object, the present application further provides a storage medium having a sensor data repair program stored thereon, and the sensor data repair program is executed by a processor to implement the sensor data repair method as described above.

[0042] The sensor data repair method provided by the present application can effectively improve the accuracy of repairing sensor data by determining the vehicle model of a target vehicle, the current region where the target vehicle travels, and the data type of the current sensor data when detecting that the current sensor data of the target vehicle is missing, acquiring historical sensor data of the target vehicle, vehicle model sensor data of the same vehicle model as the target vehicle, and region sensor data suitable for the current region according to the data type, predicting missing parts in the current sensor data according to the historical sensor data, the vehicle model sensor data, and the region sensor data through a preset deep learning network model to obtain target filling data, and filling the current sensor data according to the target filling data to achieve repair of the current sensor data, compared with the prior art of repairing data through a singular value decomposition strategy. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 is a structural schematic diagram of a sensor data repair device of a hardware running environment involved in the embodiment scheme of the present application;

[0044] Figure 2A flowchart of a first embodiment of the sensor data repairing method of the present application is shown in the figure.

[0045] Figure 3 A flowchart of a second embodiment of the sensor data repairing method of the present application is shown in the figure.

[0046] Figure 4 A flowchart of a third embodiment of the sensor data repairing method of the present application is shown in the figure.

[0047] Figure 5 A functional module diagram of a first embodiment of the sensor data repairing apparatus of the present application is shown in the figure.

[0048] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0049] It should be understood that the specific embodiments described herein are merely illustrative of the present application and do not limit the present application.

[0050] Reference Figure 1 , Figure 1 A sensor data repairing device structure diagram of a hardware running environment involved in the embodiment scheme of the present application is shown in the figure.

[0051] As Figure 1 shown, the sensor data repairing device can include a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 can include a display screen, an input unit such as a keyboard, and can also include a standard wired interface, a wireless interface. The network interface 1004 can optionally include a standard wired interface, a wireless interface (such as a wireless fidelity (Wi-Fi) interface). The memory 1005 can be a high-speed random access memory (RAM) memory, or a stable non-volatile memory (NVM), such as a magnetic disk memory. The memory 1005 can also be a storage device independent of the aforementioned processor 1001.

[0052] Those skilled in the art can understand that Figure 1 The structure shown in the figure does not constitute a limitation on the sensor data repairing device, and can include more or fewer components than the figure, or combine certain components, or different component arrangements.

[0053] As Figure 1 shown, the memory 1005 as a storage medium can include an operating system, a network communication module, a user interface module and a sensor data repair program.

[0054] In Figure 1 the sensor data repair device shown, the network interface 1004 is mainly used for data communication with the network integration platform workstation; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the sensor data repair device of the application can be arranged in the sensor data repair device, and the sensor data repair device calls the sensor data repair program stored in the memory 1005 through the processor 1001, and executes the sensor data repair method provided in the embodiment of the application.

[0055] Based on the above hardware structure, the sensor data repair method embodiment of the application is proposed.

[0056] Referring to Figure 2 , Figure 2 is a flowchart of the first embodiment of the sensor data repair method of the application.

[0057] In the first embodiment, the sensor data repair method comprises the following steps:

[0058] Step S10, when it is detected that the current sensor data of the target vehicle is lost, determining the vehicle model of the target vehicle, the current section where the target vehicle travels, and the data type of the current sensor data.

[0059] It should be noted that the execution subject of the present embodiment is a sensor data repair device, and can also be other devices that can realize the same or similar functions, such as a sensor data controller, etc., and the present embodiment does not limit this, and in the present embodiment, the sensor data controller is taken as an example for description.

[0060] It should be understood that the vehicle type refers to the model of the target vehicle, and the vehicle type is divided into small cars, micro cars, luxury cars, three-door cars, Car Derived Van (CDV) cars, Multi-Purpose Vehicles (MPV) cars, and Sport Utility Vehicles (SUV), etc. The current location refers to the location where the target vehicle travels when the current sensor data is missing. The current location can be the location where the target user frequently drives the target vehicle for commuting, or the location where the target user occasionally drives the target vehicle for outing. The data type refers to the data type of the current sensor data of the target vehicle when the current sensor data is missing. The data type includes air conditioning domain data type, chassis data type, and performance integration domain data type, etc.

[0061] Further, in order to effectively improve the accuracy of confirming the vehicle type of the target vehicle, the current location where the target vehicle travels, and the data type of the current sensor data, step S10 comprises: when it is detected that the current sensor data of the target vehicle is missing, extracting the identification information of the target vehicle; determining the vehicle type of the target vehicle according to the identification information; querying the current location where the target vehicle travels according to the identification information; and analyzing the current sensor data of the target vehicle to obtain the corresponding data type.

[0062] It can be understood that the identification information refers to information that can uniquely identify the target vehicle. The identification information can be the engine number of the target vehicle. After obtaining the identification information, the vehicle type of the target vehicle can be confirmed through the identification information. The current location where the target vehicle travels when the current sensor data is missing can be queried through the identification information on the big data platform. After obtaining the current sensor data of the target vehicle, the data type to which the current sensor data belongs can be obtained by analyzing the current sensor data.

[0063] Step S20, according to the data type, obtains the historical sensor data of the target vehicle, the sensor data of the same vehicle type as the vehicle type of the target vehicle, and the location sensor data suitable for the current location.

[0064] It can be understood that the historical sensor data refers to the sensor data of the same type as the current sensor data before the current time, the historical sensor data and the current sensor data are both from the target vehicle, the vehicle type sensor data refers to the sensor data of the vehicle of the same type as the target vehicle, and the site sensor data refers to the sensor data of the site suitable for the current site of the target vehicle. For example, when the current sensor data of the target vehicle is lost, the site on which the target vehicle travels has a slope of 5°, and the site suitable for the current site has a slope of 4°-6°. The site sensor data is the sensor data of the target vehicle passing through the suitable site.

[0065] In step S30, the missing part in the current sensor data is predicted by a preset deep learning network model according to the historical sensor data, the vehicle type sensor data and the site sensor data, and target filling data is obtained.

[0066] It should be understood that the missing part refers to the part of the current sensor data with a preset null value, that is, the missing part is the position to be filled in the current sensor data. When the missing part in the current sensor data is confirmed, that is, the current sensor data is partially lost, including partial segment loss and preset data loss. When the current sensor data is more than the preset data loss, it has lost the meaning of filling. When the current sensor data is the preset data loss in the partial loss, the missing part in the current sensor data is predicted according to the historical sensor data, the vehicle type sensor data and the site sensor data by the preset deep learning network model, and the target filling data is obtained.

[0067] Further, in order to effectively improve the efficiency of repairing the sensor data, before step S30, it further includes: extracting the upload start time, interval time and upload end time of the current sensor data; determining the sensor data of each time period according to the upload start time, interval time and upload end time; generating a corresponding sensor data continuity table according to the sensor data of each time period; when there is a preset null value in the sensor data continuity table, obtaining the missing part in the current sensor data.

[0068] It can be understood that the upload start time refers to the time when each sensor in the target vehicle starts uploading data, the upload end time refers to the time when each sensor in the target vehicle ends uploading data, and the interval time is the time consumed between uploading the current data and the next data. According to the upload start time, interval time and upload end time, the sensor data in each time period can be determined. By merging the sensor data in each time period, a corresponding sensor data continuity table is generated. Through the sensor data continuity table, the missing part in the current sensor data can be determined directly and intuitively, that is, the position of the preset null value is the missing part in the current sensor data.

[0069] Step S40, according to the target filling data to fill in the current sensor data, to achieve the repair of the current sensor data.

[0070] It can be understood that after obtaining the target filling data, the target filling data is filled into the current sensor data, and the filled target filling data is marked in the current sensor data to prompt the marked data as the target filling data, and after the filling is completed, the repair of the current sensor data is realized.

[0071] The embodiment determines the vehicle model of the target vehicle, the current section where the target vehicle travels, and the data type of the current sensor data when detecting that the current sensor data of the target vehicle has loss; obtains the historical sensor data of the target vehicle, the vehicle model sensor data of the same vehicle model as the target vehicle, and the section sensor data suitable for the current section according to the data type; predicts the missing part in the current sensor data through a preset deep learning network model according to the historical sensor data, the vehicle model sensor data and the section sensor data, and obtains the target filling data; fills in the current sensor data according to the target filling data, so as to realize the repair of the current sensor data. Compared with the existing technology of repairing data by singular value decomposition strategy, the accuracy of repairing sensor data can be effectively improved.

[0072] In an embodiment, as Figure 3 The second embodiment of the sensor data repair method of the present application is proposed based on the first embodiment, and the step S20 comprises:

[0073] Step S201, obtaining the historical sensor data of the target vehicle according to the data type.

[0074] It should be understood that the historical sensor data refers to the sensor data of the same type as the current sensor data before the current time, that is, after obtaining the current sensor data, the data type to which the current sensor data belongs needs to be determined, and the historical sensor data of the target vehicle is obtained according to the data type.

[0075] Step S202, obtaining the energy consumption, driving mileage and latitude of the current section where the target vehicle travels.

[0076] It can be understood that the energy consumption refers to the energy consumed by the target vehicle when driving through the current section, the energy consumption refers to the energy consumption and the energy consumption, the driving mileage refers to the mileage change value of the odometer of the target vehicle when driving through the current section, and the latitude refers to the latitude of the target vehicle when starting to drive the current section and ending to drive the current section.

[0077] In step S203, the region sensor data corresponding to the current region is obtained according to the energy consumption, the driving distance, the latitude of the current region, and the data type.

[0078] It should be understood that, when the energy consumption, the driving distance, and the latitude of the current region are obtained, the region corresponding to the current region can be determined according to the energy consumption, the driving distance, and the latitude of the current region, and the region sensor data corresponding to the current region can be obtained based on the region.

[0079] Further, in order to effectively improve the accuracy of the region sensor data corresponding to the current region, in step S203, the starting latitude and the ending latitude of the latitude of the current region are extracted, the driving route of the target vehicle is determined according to the starting latitude, the ending latitude, the energy consumption, and the driving distance, and the region sensor data corresponding to the current region is obtained according to the driving route and the data type.

[0080] It should be understood that the starting latitude refers to the latitude of the target vehicle when the target vehicle starts to drive the current region, and the ending latitude refers to the latitude of the target vehicle when the target vehicle has driven the current region. After the starting latitude, the ending latitude, the energy consumption, and the driving distance are obtained, the driving route of the target vehicle can be obtained based on the above parameters and the road index, and the region sensor data corresponding to the current region can be obtained according to the driving route and the data type.

[0081] In step S204, the vehicle type sensor data corresponding to the same vehicle type as the target vehicle is obtained according to the identification information of the target vehicle and the data type.

[0082] It should be understood that the identification information refers to information that can uniquely identify the target vehicle. After the identification information and the data type are obtained, the vehicle corresponding to the same vehicle type as the target vehicle can be queried in the big data platform, and the vehicle type sensor data corresponding to the vehicle can be obtained based on the vehicle.

[0083] The embodiment obtains historical sensor data of the target vehicle according to the data type, obtains energy consumption, driving mileage and latitude of the current section where the target vehicle travels according to the current section where the target vehicle travels, obtains section sensor data suitable for the current section according to the energy consumption, driving mileage, latitude of the current section where the target vehicle travels and the data type, and queries vehicle type sensor data corresponding to the target vehicle according to the identification information of the target vehicle and the data type. Since the embodiment obtains historical sensor data of the target vehicle through the data type, obtains section sensor data suitable for the current section according to the energy consumption, driving mileage, latitude of the current section where the target vehicle travels and the data type, and queries vehicle type sensor data corresponding to the target vehicle according to the identification information of the target vehicle and the data type, the accuracy of the obtained historical sensor data, section sensor data and vehicle type sensor data can be effectively improved.

[0084] In an embodiment, as Figure 4 The third embodiment of the sensor data repairing method is proposed based on the first embodiment, and before step S30, the method further includes:

[0085] Step S205: counting the number of missing parts in the current sensor data.

[0086] It can be understood that when it is confirmed that the current sensor data is missing, the number of missing parts is counted, and it is confirmed that the missing parts of the current sensor data are partial fragment loss and preset data loss by counting the number of missing parts.

[0087] Further, in order to effectively improve the accuracy of repairing sensor data, after step S205, the method further includes: when the number of missing parts in the current sensor data is equal to a preset number threshold, obtaining the last sensor data and the next sensor data of the missing part in the current sensor data; and calculating the last sensor data and the next sensor data according to a preset calculation strategy to obtain target filling data.

[0088] It should be understood that when it is determined that the number of missing parts in the current sensor data is equal to the preset number threshold, that is, the number of missing parts is 1, the target filling data is calculated by the last sensor data and the next sensor data of the missing part at this time, specifically, the last sensor data and the next sensor data are calculated by a preset calculation strategy, and the preset calculation strategy refers to an average value calculation strategy, for example, the last sensor data is A and the next sensor data is B, and the target filling data is (A+B) / 2.

[0089] In step S206, when the number of missing parts in the current sensor data is greater than a preset number threshold, the step of predicting the missing parts in the current sensor data according to the historical sensor data, the vehicle type sensor data and the region sensor data through a preset deep learning network model to obtain target filling data is performed.

[0090] It should be understood that the preset number threshold refers to the minimum number of missing parts in the current sensor data, which can be set to 1. When the number of missing parts in the current sensor data is greater than the preset number threshold, the number of missing parts in the current sensor data is tens to hundreds, and at this time, the missing parts in the current sensor data need to be predicted according to the historical sensor data, the vehicle type sensor data and the region sensor data through a preset deep learning network model to obtain target filling data. The preset deep learning network model can be a recurrent neural network model (RNN), or other deep learning network models, which are not limited in the embodiment, and are described by taking the RNN network model as an example.

[0091] It can be understood that in the process of predicting according to the historical sensor data, the vehicle type sensor data and the region sensor data, in order to ensure the reliability and accuracy of the data, different weights of the historical sensor data, the vehicle type sensor data and the region sensor data need to be set. Since the historical sensor data is the data of the target vehicle itself, the weight of the historical sensor data is set to be the largest, and the weights of the vehicle type sensor data and the region sensor data are the same, for example, the weight of the historical sensor data is 0.8, and the weights of the vehicle type sensor data and the region sensor data are both 0.1.

[0092] The embodiment counts the number of missing parts in the current sensor data, and when the number of missing parts in the current sensor data is greater than a preset number threshold, the step of predicting the missing parts in the current sensor data according to the historical sensor data, the vehicle type sensor data and the region sensor data through a preset deep learning network model to obtain target filling data is performed. Since the embodiment counts the number of missing parts in the current sensor data, and judges that the number of missing parts in the current sensor data is greater than the preset number threshold, if it is greater, the missing parts in the current sensor data are predicted according to the historical sensor data, the vehicle type sensor data and the region sensor data through a preset deep learning network model to obtain target filling data, so as to effectively improve the accuracy of repairing sensor data.

[0093] In addition, the embodiment of the present application further provides a storage medium, wherein the storage medium stores a sensor data repairing program, and the sensor data repairing program is executed by a processor to implement the steps of the sensor data repairing method.

[0094] Since the storage medium adopts all the technical solutions of the above embodiments, it has all the beneficial effects brought by the technical solutions of the above embodiments, which will not be repeated here.

[0095] In addition, with reference to Figure 5 the embodiment of the present application further provides a sensor data repairing device, which comprises:

[0096] A determining module 10 is configured to determine a vehicle model of a target vehicle, a current section where the target vehicle travels, and a data type of current sensor data of the target vehicle when it is detected that the current sensor data of the target vehicle is missing.

[0097] It should be understood that the vehicle model refers to the model of the target vehicle, and the vehicle model is classified into small cars, micro cars, luxury cars, three-door cars, Car Derived Van (CDV) cars, Multi-Purpose Vehicles (MPV) cars, and Sport Utility Vehicle (SUV) cars, etc. The current section refers to the section where the target vehicle travels when the current sensor data is missing. The current section can be a section where the target user frequently drives the target vehicle for commuting, or a section where the target user occasionally drives the target vehicle for going out. The data type refers to the data type of the current sensor data of the target vehicle when the current sensor data is missing. The data type includes air conditioning domain data type, chassis data type, and performance integration domain data type, etc.

[0098] Further, in order to effectively improve the accuracy of confirming the vehicle model of the target vehicle, the current section where the target vehicle travels, and the data type of the current sensor data, the determining module 10 is further configured to extract identification information of the target vehicle when it is detected that the current sensor data of the target vehicle is missing; determine the vehicle model of the target vehicle according to the identification information; query the current section where the target vehicle travels according to the identification information; and analyze the current sensor data of the target vehicle to obtain the corresponding data type.

[0099] It can be understood that the identification information refers to information capable of uniquely identifying the target vehicle, and the identification information can be the engine number of the target vehicle. After obtaining the identification information, the vehicle model of the target vehicle can be confirmed through the identification information, and the current section traveled by the target vehicle when the current sensor data exists loss can be queried in the big data platform through the identification information. After obtaining the current sensor data of the target vehicle, the data type to which the current sensor data belongs can be obtained by analyzing the current sensor data.

[0100] The acquisition module 20 is configured to acquire historical sensor data of the target vehicle, vehicle model sensor data of the same vehicle model as the vehicle model of the target vehicle, and section sensor data suitable for the current section according to the data type.

[0101] It can be understood that the historical sensor data refers to sensor data of the same type as the current sensor data before the current time, and the historical sensor data and the current sensor data are both from the target vehicle. The vehicle model sensor data refers to sensor data of a vehicle of the same vehicle model as the target vehicle. The section sensor data refers to sensor data of a section suitable for the current section of the target vehicle. For example, when the current sensor data of the target vehicle exists loss, the section traveled by the target vehicle has a slope of 5°, and the section suitable for the current section has a slope of 4°-6°. The section sensor data is the sensor data of the target vehicle passing through the suitable section.

[0102] The prediction module 30 is configured to predict the missing part in the current sensor data through a preset deep learning network model according to the historical sensor data, the vehicle model sensor data, and the section sensor data, to obtain target filling data.

[0103] It should be understood that the missing part refers to a part of the current sensor data that has a preset null value. In other words, the missing part is the position to be filled in the current sensor data. When the missing part in the current sensor data is confirmed, i.e., the current sensor data is partially lost, including partial segment loss and preset data loss. When the loss is greater than the preset data loss, it has lost the meaning of filling. When the current sensor data is the preset data loss in the partial loss, the missing part in the current sensor data is predicted through a preset deep learning network model according to the historical sensor data, the vehicle model sensor data, and the section sensor data, to obtain target filling data.

[0104] The filling module 40 is configured to fill the current sensor data according to the target filling data, to realize repair of the current sensor data.

[0105] It is understandable that after obtaining the target filling data, the target filling data is filled into the current sensor data, and the filled target filling data is marked in the current sensor data. The marked data is used as the target filling data. After the filling is completed, the current sensor data is repaired.

[0106] This embodiment determines the vehicle model, the current location of the target vehicle, and the data type of the current sensor data when a loss of current sensor data is detected. Based on the data type, it obtains historical sensor data of the target vehicle, sensor data of the same vehicle model, and sensor data of the location corresponding to the current location. Using a preset deep learning network model, it predicts the missing portion of the current sensor data based on the historical, vehicle, and location sensor data to obtain target-filled data. The current sensor data is then filled using the target-filled data to repair the current sensor data. Compared to existing technologies that repair data using singular value decomposition strategies, this method effectively improves the accuracy of sensor data repair.

[0107] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of this invention. In practical applications, those skilled in the art can select some or all of the workflow to achieve the purpose of this embodiment according to actual needs, and no restrictions are imposed here.

[0108] In addition, for technical details not described in detail in this embodiment, please refer to the sensor data repair method provided in any embodiment of the present invention, which will not be repeated here.

[0109] In one embodiment, the determining module 10 is further configured to: extract the identification information of the target vehicle when it is detected that the current sensor data of the target vehicle is lost; determine the model of the target vehicle based on the identification information; query the current location of the target vehicle based on the identification information; and analyze the current sensor data of the target vehicle to obtain the corresponding data type.

[0110] In one embodiment, the acquisition module 20 is further configured to acquire historical sensor data of the target vehicle according to the data type; obtain the energy consumption, mileage, and latitude of the target vehicle based on the current location of the target vehicle; obtain location sensor data adapted to the current location based on the energy consumption, mileage, latitude, and data type of the target vehicle; and query vehicle model sensor data with the same model as the target vehicle based on the identification information and data type of the target vehicle.

[0111] In one embodiment, the acquisition module 20 is further configured to extract the starting latitude and ending latitude of the current location; determine the driving route of the target vehicle based on the starting latitude, ending latitude, energy consumption, and driving mileage; and obtain location sensor data adapted to the current location based on the driving route and data type.

[0112] In one embodiment, the prediction module 30 is further configured to extract the upload start time, interval time, and upload end time of the current sensor data; determine the sensor data for each time period based on the upload start time, interval time, and upload end time; generate a corresponding continuous sensor data table based on the sensor data for each time period; and, when a preset null value exists in the continuous sensor data table, obtain the missing portion of the current sensor data.

[0113] In one embodiment, the prediction module 30 is further configured to count the number of missing parts in the current sensor data; when the number of missing parts in the current sensor data is greater than a preset threshold, the step of predicting the missing parts in the current sensor data based on historical sensor data, vehicle model sensor data, and location sensor data through a preset deep learning network model to obtain target filling data is executed.

[0114] In one embodiment, the prediction module 30 is further configured to, when the number of missing parts in the current sensor data is equal to a preset number threshold, obtain the previous sensor data and the next sensor data of the missing parts in the current sensor data; and calculate the previous sensor data and the next sensor data according to a preset calculation strategy to obtain target filling data.

[0115] Other embodiments or implementation methods of the sensor data repair device of the present invention can be referred to the above-described method embodiments, and will not be repeated here.

[0116] Furthermore, it should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0117] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0118] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory (ROM) / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, all-in-one platform workstation, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0119] The above are merely preferred embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for repairing sensor data, characterized in that, The method for repairing the sensor data includes the following steps: When it is detected that the current sensor data of the target vehicle is lost, the vehicle model, the current location where the target vehicle is traveling, and the data type of the current sensor data are determined. Based on the data type, obtain the historical sensor data of the target vehicle, the vehicle model sensor data of the same type as the target vehicle, and the location sensor data adapted to the current location, wherein the location sensor data refers to the sensor data of the location adapted to the current location of the target vehicle. Based on the historical sensor data, vehicle model sensor data, and location sensor data, a preset deep learning network model is used to predict the missing parts in the current sensor data to obtain target filling data; The current sensor data is filled in according to the target filling data to achieve the repair of the current sensor data; The step of acquiring historical sensor data of the target vehicle, sensor data of the same vehicle type as the target vehicle, and location sensor data adapted to the current location based on the data type includes: Obtain historical sensor data of the target vehicle based on the data type described; The energy consumption, mileage, and latitude of the target vehicle are obtained based on the current location of the target vehicle. Based on the target vehicle's energy consumption, mileage, latitude of the current location, and data type, obtain location sensor data adapted to the current location; Based on the identification information and data type of the target vehicle, query the sensor data of the vehicle model that is the same as the target vehicle model.

2. The sensor data repair method as described in claim 1, characterized in that, When the loss of current sensor data of the target vehicle is detected, determining the vehicle model, the current location of the target vehicle, and the data type of the current sensor data includes: When the current sensor data of the target vehicle is detected to be lost, the identification information of the target vehicle is extracted; The vehicle model of the target vehicle is determined based on the identification information; The location where the target vehicle is currently traveling is queried based on the identification information; The current sensor data of the target vehicle is analyzed to obtain the corresponding data type.

3. The sensor data repair method as described in claim 1, characterized in that, The step of obtaining location sensor data adapted to the current location based on the target vehicle's energy consumption, mileage, latitude of the current location, and data type includes: Extract the starting and ending latitudes of the current location; The travel route of the target vehicle is determined based on the starting latitude, ending latitude, energy consumption, and travel distance. Based on the driving route and data type, location sensor data adapted to the current location is obtained.

4. The sensor data repair method as described in claim 1, characterized in that, Before the step of predicting the missing portion in the current sensor data using a preset deep learning network model based on the historical sensor data, vehicle model sensor data, and location sensor data to obtain the target filling data, the method further includes: Extract the start time, interval time, and end time of the current sensor data upload; The sensor data for each time period is determined based on the upload start time, interval time, and upload end time. Generate a corresponding continuous table of sensor data based on the sensor data for each time period; If a preset null value exists in the continuous sensor data table, the missing portion of the current sensor data is retrieved.

5. The sensor data repair method as described in claim 4, characterized in that, Before the step of predicting the missing portion in the current sensor data using a preset deep learning network model based on the historical sensor data, vehicle model sensor data, and location sensor data to obtain the target filling data, the method further includes: Count the number of missing parts in the current sensor data; When the number of missing parts in the current sensor data exceeds a preset threshold, the step of predicting the missing parts in the current sensor data using a preset deep learning network model based on historical sensor data, vehicle model sensor data, and location sensor data to obtain target filling data is executed.

6. The sensor data repair method as described in claim 5, characterized in that, After calculating the number of missing portions in the current sensor data, the method further includes: When the number of missing parts in the current sensor data is equal to a preset threshold, the previous and next sensor data of the missing parts in the current sensor data are obtained. The target filling data is obtained by calculating the previous and next sensor data according to the preset calculation strategy.

7. A sensor data repair device, characterized in that, The sensor data repair device includes: The determination module is used to determine the vehicle model, the current location where the target vehicle is traveling, and the data type of the current sensor data when the current sensor data of the target vehicle is detected to be lost. The acquisition module is used to acquire, according to the data type, the historical sensor data of the target vehicle, the vehicle model sensor data of the same type as the target vehicle, and the location sensor data adapted to the current location, wherein the location sensor data refers to the sensor data of the location adapted to the current location of the target vehicle. The prediction module is used to predict the missing parts in the current sensor data based on the historical sensor data, vehicle model sensor data, and location sensor data through a preset deep learning network model, so as to obtain target filling data. The filling module is used to fill in the current sensor data according to the target filling data, so as to repair the current sensor data; The acquisition module is further configured to acquire historical sensor data of the target vehicle according to the data type; obtain the energy consumption, mileage, and latitude of the target vehicle based on the current location of the target vehicle; obtain location sensor data adapted to the current location based on the energy consumption, mileage, latitude, and data type of the target vehicle; and query vehicle model sensor data with the same model as the target vehicle based on the identification information and data type of the target vehicle.

8. A sensor data repair device, characterized in that, The sensor data repair device includes: a memory, a processor, and a sensor data repair program stored in the memory and executable on the processor, wherein the sensor data repair program is configured to implement the sensor data repair method as described in any one of claims 1 to 6.

9. A storage medium, characterized in that, The storage medium stores a sensor data repair program, which, when executed by a processor, implements the sensor data repair method as described in any one of claims 1 to 6.

Images