Management system for unmanned aerial vehicle test data storage and intelligent analysis
By designing a drone test data management system, including preliminary division of test data, simulation storage modules and actual storage execution modules, the problem of single data acquisition and storage methods in traditional drone test data management is solved, and systematic data acquisition and efficient storage are realized.
Patent Information
- Application Number
- CN202510281436.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-11
- Publication Date
- 2025-06-27
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional drone test data management has problems such as lack of systematic data collection, single storage methods, and not conducive to subsequent storage and analysis.
A management system for drone test data storage and intelligent analysis is designed, including a preliminary division module for test data, a simulation storage module for test data, and a practical storage and execution module for test data. By collecting drone parameter data in real time, combining deep learning models for data level division, and using simulation technology to generate parameter simulation storage models, simulate different storage solutions to select the best solution for data storage.
It realizes systematic collection and efficient storage of drone test data, which facilitates subsequent classified data reading according to different needs, and improves the effect of data storage and reading.
Smart Images
Figure CN120216490A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data management, and more specifically, it relates to a management system for storing and intelligently analyzing UAV test data. Background Art
[0002] With the rapid development of UAV technology, UAVs are increasingly widely used in various fields, including aerial photography, environmental monitoring, agricultural plant protection, topographic mapping, military reconnaissance, etc. The performance and stability of UAVs are directly related to their application effects and safety. Therefore, comprehensive testing of UAVs is a key link to ensure their performance meets the standards. During the UAV testing process, a large amount of test data is generated, which covers the flight state of UAVs, sensor data, control system parameters, etc., and is of great value for evaluating UAV performance, optimizing design schemes, and troubleshooting faults.
[0003] However, there are many deficiencies in traditional UAV test data management. First of all, the collection of test data often lacks systematicness, resulting in chaotic data, which is not conducive to subsequent storage and analysis. Secondly, the data storage method is single, using a unified data indexing mechanism, data storage level, and data storage layout for all test data, lacking the setting of targeted data storage schemes, resulting in poor data storage and subsequent data reading effects.
[0004] Therefore, the present invention proposes a management system for storing and intelligently analyzing UAV test data. Summary of the Invention
[0005] Aiming at the deficiencies of the existing technology, the purpose of the present invention is to provide a management system for storing and intelligently analyzing UAV test data.
[0006] To achieve the above purpose, the present invention provides the following technical solutions:
[0007] A management system for storing and intelligently analyzing UAV test data, including a preliminary test data division module, a simulation test data storage module, and an actual test data storage execution module;
[0008] The preliminary test data division module, during the UAV test flight process, real-time collects the parameter data of the UAV. Every time the UAV goes through a flight stage, the parameter data collected in this flight stage is marked as stage data and is assigned a storage log to be stored.
[0009] The simulation test data storage module, every time a test data storage cycle is completed, collects all the stage data collected during the test data storage cycle, determines the terminal model of the storage terminal, constructs a parameter simulation storage model, the parameter simulation storage model generates multiple simulation storage schemes, and determines the storage reading analysis index of each simulation storage scheme.
[0010] The actual storage execution module of the test data selects the actual storage scheme from the simulation storage schemes, and stores all the stage data collected during the test data storage cycle into the storage terminal according to the actual storage scheme.
[0011] Furthermore, the log to be stored includes the flight stage, the stage test parameter level, the start time of the flight stage, and the end time of the flight stage.
[0012] Furthermore, the stage test parameter level of the log to be stored is determined in the following way: obtain the stage test analysis value APs of the stage data corresponding to the log to be stored, and set a stage test parameter level corresponding to the range of each stage test analysis value APs. The ranges of the stage test analysis values APs are [0, AP1], (AP1, AP2], …, (APS-1, APS], and the stage test parameter levels include stage test parameter level 1, stage test parameter level 2, …, stage test parameter level S-1, and stage test parameter level S.
[0013] Furthermore, the stage test analysis value APs of the stage data is determined in the following way: extract the data characteristics of the stage data to obtain parameter characteristics, perform feature fusion processing on the parameter characteristics to obtain stage fusion characteristics, determine the flight stage where the stage data corresponds to the log to be stored, determine the stage test analysis model corresponding to this flight stage, use the stage fusion characteristics as the input data of the stage test analysis model, and the stage test analysis model outputs the stage test analysis value of this flight stage.
[0014] Furthermore, the storage reading analysis index of the simulation storage scheme is determined in the following way: select a simulation storage scheme, and the control parameter simulation storage model performs simulation storage on all the stage data collected during the test data storage cycle according to the simulation storage scheme. During the simulation storage process, the data error rate of the parameter simulation storage model is collected in real time. After the simulation storage is completed, determine the storage analysis index index(storage), set multiple data reading requirements, determine the required reading index for each data reading requirement, compare all the data reading requirements in pairs, calculate the absolute difference between the required reading indices of the two compared data reading requirements to obtain the reading fluctuation index, calculate the sum mean of all the reading fluctuation indices to obtain the average reading fluctuation index index(fluct), calculate the sum mean of the required reading indices of all the data reading requirements to obtain the average required reading index index(read all), and through obtain the storage reading analysis index index(sive) of the simulation storage scheme.
[0015] Further, the storage analysis index is determined as follows: Obtain the average write throughput thr(avj) of the parameter simulation storage model. Construct a rectangular coordinate system with the storage duration as the X-axis and the data error rate as the Y-axis. Plot the collected data error rates in the rectangular coordinate system in the form of a curve to obtain the data error rate curve. Draw perpendicular lines from both ends of the data error rate curve to the X-axis, and mark the total area of the closed figure formed by the X-axis, the two perpendicular lines, and the data error rate curve as index(errorarea). Through the storage analysis index index(storage) is obtained.
[0016] Further, the demand reading index of the data reading demand is determined as follows: Select a data reading demand, input the instruction of the data reading demand into the parameter simulation storage model. The parameter simulation storage model performs simulated reading on the corresponding stage data according to the data reading demand. During the simulated reading process, after the simulated reading is completed, determine whether the read stage data is complete. When it is determined that a stage of the read data is complete, increment the number of complete data by one, and mark the number of complete data as number(plete). After the simulated reading is completed, mark the total duration of the simulated reading as Time(read). Through the demand reading index index(read) of the data reading demand is obtained.
[0017] Further, mark the simulation storage scheme with the largest value of the storage reading analysis index as the actual storage scheme.
[0018] Compared with the prior art, the present invention has the following beneficial effects:
[0019] The system of the present invention includes a test data preliminary division module, a test data simulation storage module, and a test data actual storage execution module. During the test flight of the unmanned aerial vehicle, according to the flight stages experienced by the unmanned aerial vehicle, the test data is preliminarily collected and divided, and combined with a deep learning model, the divided data is classified, which is convenient for classified data reading according to different needs after data storage. Regularly, according to the data generated during the test process, combined with simulation technology, a parameter simulation storage model is generated, which can simulate the storage process of test data under different data indexing mechanisms, data storage levels, and data storage layouts, and set multiple data reading demands to simulate the data reading process under different demands, and deeply analyze the storage process and the reading process, and comprehensively select the storage scheme with the best comprehensive storage and reading effects for actual data storage management. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 It is a flowchart of the operation of the management system for unmanned aerial vehicle test data storage and intelligent analysis;
[0021] Figure 2 is the block diagram of the present invention;
[0022] Figure 3 is the flowchart for determining the stage test analysis value of stage data. Specific Embodiments
[0023] Referring to Figures 1 to 3 , a management system for storing and intelligently analyzing UAV test data, including a preliminary test data division module, a simulation test data storage module, and an actual test data storage execution module.
[0024] Preliminary test data division module: During the test flight of the UAV, parameter data of the UAV is collected in real time (the parameter data includes flight attitude parameter data, flight performance parameter data, power battery parameter data, etc.). Every time the UAV goes through a flight stage (the UAV needs to go through various flight stages during the test flight, including but not limited to takeoff stage, constant speed cruise stage, turning stage, landing stage, hovering stage, acceleration stage, deceleration stage), the parameter data collected in this flight stage is marked as stage data, the stage test analysis value APs of this stage data is determined, and a log to be stored is assigned to this stage data. The log to be stored includes the flight stage it is in, the stage test parameter level, the start time of the flight stage, and the end time of the flight stage.
[0025] The stage test parameter level of the log to be stored is determined in the following way: Obtain the stage test analysis value APs of the stage data corresponding to the log to be stored. Set a stage test parameter level corresponding to the range of each stage test analysis value APs. The range of the stage test analysis value APs is [0, AP1], (AP1, AP2], …, (APS - 1, APS]. The stage test parameter levels include stage test parameter level 1, stage test parameter level 2, …, stage test parameter level S - 1, stage test parameter level S.
[0026] The stage test analysis value APs of the stage data is determined in the following way: Extract data features from the stage data to obtain parameter features, perform feature fusion processing on the parameter features (the way of feature fusion can be weighted fusion), process to obtain stage fusion features, determine the flight stage where the log to be stored corresponding to the stage data is located, determine the stage test analysis model corresponding to this flight stage, use the stage fusion features as the input data of the stage test analysis model, and the stage test analysis model outputs the stage test analysis value of this flight stage.
[0027] Each flight phase corresponds to an independent phase test and analysis model. For example, the takeoff phase and the constant-speed cruise phase respectively correspond to two phase test and analysis models. The phase test and analysis models for all flight phases are constructed based on deep learning models. The phase test and analysis model for a flight phase is determined in the following way: construct a deep learning model, determine a flight phase, collect multiple phase fusion features corresponding to this flight phase, train the deep learning model with the phase fusion features, assign a phase test analysis value to each phase fusion feature, and the exponential range of the phase test analysis value is (1.0 - 30.0). The larger the phase test analysis value, the more the UAV flight state in this flight phase deviates from the normal state. Divide the training data into a training set, a validation set, and a test set in the ratio of 60%:20%:20%. Use the training set to train the model, and by continuously adjusting the parameters of the model, minimize the value of the loss function. During the training process, use the validation set to monitor the performance of the model to avoid overfitting, and use the test set to evaluate the trained model. Finally, determine the phase test and analysis model for the flight phase.
[0028] Test data simulation storage module: Set the cycle duration of the test data storage cycle to T storage , every time a test data storage cycle elapses, collect all the phase data collected during the test data storage cycle, determine the terminal model of the storage terminal, construct a parameter simulation storage model, the parameter simulation storage model generates multiple simulation storage schemes, and determine the storage read and analysis index for each simulation storage scheme.
[0029] The parameter simulation storage model is constructed through the following way: Select simulation software, create a storage terminal entity in the simulation software based on the terminal model of the storage terminal, and define the storage medium, interface protocol, and storage architecture of the storage terminal entity, and add performance indicators such as latency, throughput, and reliability to the storage terminal entity to construct the parameter simulation storage model. Import all the phase data into the parameter simulation storage model. The parameter simulation storage model can generate multiple simulation storage schemes. Each simulation storage scheme can store all the phase data in the storage terminal entity according to a certain storage rule. There are differences in data indexing mechanisms, data storage levels, data storage formats, data storage layouts, etc. among different simulation storage schemes.
[0030] The storage read analysis index of the simulation storage scheme is determined in the following way: Select a simulation storage scheme. The control parameter simulation storage model performs simulation storage on all stage data collected during the storage cycle of the test data based on the simulation storage scheme. During the simulation storage process, the data error rate of the parameter simulation storage model is collected in real time. After the simulation storage is completed, the storage analysis index index(storage) is determined. Set multiple data reading requirements (data reading requirements are data reading requirements that may occur later for in-depth analysis of stage data. For example, later it may be necessary to read parameter data of the same stage (such as the takeoff stage, constant speed cruise stage, turning stage, etc.), then it is necessary to read all stage data of the same stage in the storage terminal. For example, later it may be necessary to read parameter data of the same stage test parameter level, then it is necessary to read all stage data of the same stage test parameter level in the storage terminal. For example, later it may be necessary to read parameter data within a period of time, then it is necessary to read all stage data within a period of time). Determine the required reading index for each data reading requirement. Compare all data reading requirements in pairs, calculate the absolute difference between the required reading indexes of the two compared data reading requirements to obtain the reading fluctuation index. Calculate the sum mean of all reading fluctuation indexes to obtain the average reading fluctuation index index(fluct). Calculate the sum mean of the required reading indexes of all data reading requirements to obtain the average required reading index index(read all). Through obtain the storage read analysis index index(sive) of the simulation storage scheme, where a3 is the third coefficient, a4 is the fourth coefficient, the value of a3 is 1.39, and the value of a4 is 0.45.
[0031] The storage analysis index is determined in the following way: Obtain the average write throughput thr(avj) of the parameter simulation storage model. Construct a rectangular coordinate system with the storage duration as the X-axis and the data error rate as the Y-axis. Depict the collected data error rate in the rectangular coordinate system in the form of a curve to obtain the data error rate curve. Draw perpendicular lines from both ends of the data error rate curve to the X-axis. Mark the total area of the closed figure formed by the X-axis, the two perpendicular lines, and the data error rate curve as index(errorarea). Through obtain the storage analysis index index(storage), where a1 is the first coefficient, and the value of a1 is 0.84.
[0032] The demand reading index for data reading requirements is determined as follows: Select a data reading requirement, input the instruction input parameters of the data reading requirement into the parameter simulation storage model. The parameter simulation storage model performs simulated reading of the corresponding stage data according to the data reading requirement. During the simulated reading process, after the simulated reading is completed, it is determined whether the read stage data is complete (determine whether the stage data is complete through the hash value). When it is determined that a stage of the read data is complete, increment the number of complete data by one (when it is determined that a stage of the read data is incomplete, no processing is performed), mark the number of complete data as number(plete). After the simulated reading is completed, mark the total duration of the simulated reading as Time(read). Through obtain the demand reading index index(read) of the data reading requirement, where a2 is the second coefficient and the value of a2 is 0.91.
[0033] Test data actual storage execution module: Mark the simulation storage scheme with the largest value of the storage reading analysis index as the actual storage scheme, and store all the stage data collected during the test data storage period into the storage terminal according to the actual storage scheme.
[0034] The main operation process of this system is as follows:
[0035] S1: During the test flight of the drone, the parameter data of the drone is collected in real time.
[0036] S2: Every time the drone goes through a flight stage, mark the parameter data collected in this flight stage as stage data and assign a storage log to be processed.
[0037] S3: Every time a test data storage cycle is completed, collect all the stage data collected during the test data storage cycle, determine the terminal model of the storage terminal, and construct a parameter simulation storage model.
[0038] S4: The parameter simulation storage model generates multiple simulation storage schemes and determines the storage reading analysis index of each simulation storage scheme.
[0039] S5: Select the actual storage scheme from the simulation storage schemes, and store all the stage data collected during the test data storage cycle into the storage terminal according to the actual storage scheme.
[0040] Preliminarily divide modules, test data simulation storage module, and test data actual storage execution module through test data. During the test flight of the unmanned aerial vehicle (UAV), according to the flight stages experienced by the UAV, initially collect and divide the test data, and combine with a deep learning model to classify the divided data, facilitating classified data reading according to different requirements after data storage. Regularly generate a parameter simulation storage model based on the data generated during the test process in combination with simulation technology, which can simulate the storage process of test data under different data indexing mechanisms, data storage levels, and data storage layouts, and set multiple data reading requirements to simulate the data reading process under different requirements, and conduct in-depth analysis on the storage process and the reading process, and comprehensively select the storage scheme with the best comprehensive storage and reading effects for actual data storage management.
[0041] The above formulas are all dimensionless and take their numerical calculations. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.
[0042] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that the computer can access or a data storage device such as a server or data center that contains one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.
[0043] It should be understood that in various embodiments of the present application, the magnitudes of the serial numbers of the above processes do not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.
[0044] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.
[0045] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.
[0046] In several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings, direct couplings, or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.
[0047] If the described functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.
[0048] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.
Claims
1. A management system for drone test data storage and intelligent analysis, characterized by: It includes a test data preliminary division module, a test data simulation storage module, and a test data actual storage execution module; The test data preliminary division module collects the parameter data of the drone in real time during the drone test flight. Each time the drone goes through a flight phase, the parameter data collected in the flight phase is marked as phase data and assigned a log to be stored; The test data simulation storage module, each time undergoing a test data storage cycle, collects all stage data collected during the test data storage cycle, determines the terminal model of the storage terminal, constructs a parameter simulation storage model, the parameter simulation storage model generates multiple simulation storage schemes, and determines the storage read analysis index of each simulation storage scheme; The test data actual storage execution module selects an actual storage scheme from the simulation storage schemes, and stores all phase data collected during the test data storage cycle into the storage terminal according to the actual storage scheme.
2. The management system for drone test data storage and intelligent analysis according to claim 1 is characterized in that: The logs to be stored include the flight phase, phase test parameter level, flight phase start time, and flight phase end time.
3. The management system for drone test data storage and intelligent analysis according to claim 2 is characterized in that: The stage test parameter level of the log to be stored is determined in the following way: obtain the stage test analysis value APs of the stage data corresponding to the log to be stored, set the range of each stage test analysis value APs to correspond to a stage test parameter level, the range of the stage test analysis value APs is [0, AP1], (AP1, AP2], …, (APS-1, APS], and the stage test parameter levels include stage test parameter level 1, stage test parameter level 2, …, stage test parameter level S-1, and stage test parameter level S.
4. The management system for drone test data storage and intelligent analysis according to claim 3 is characterized in that: The stage test analysis value APs of the stage data is determined in the following way: extract data features from the stage data to obtain parameter features, perform feature fusion processing on the parameter features to obtain stage fusion features, determine the flight stage of the log to be stored corresponding to the stage data, determine the stage test analysis model corresponding to the flight stage, use the stage fusion features as input data of the stage test analysis model, and the stage test analysis model outputs the stage test analysis value of the flight stage.
5. The management system for drone test data storage and intelligent analysis according to claim 1 is characterized in that: The storage read analysis index of the simulation storage solution is determined in the following way: select a simulation storage solution, control the parameter simulation storage model to simulate and store all stage data collected during the test data storage cycle based on the simulation storage solution, collect the data error rate of the parameter simulation storage model in real time during the simulation storage process, determine the storage analysis index index(storage) after the simulation storage is completed, set multiple data read requirements, determine the demand read index of each data read requirement, compare all data read requirements in pairs, calculate the absolute difference between the demand read indexes of the two compared data read requirements to obtain the read fluctuation index, sum and average all read fluctuation indexes to obtain the average read fluctuation index index(fluct), sum and average the demand read indexes of all data read requirements to obtain the average demand read index index(read all), through Get the storage read analysis index index(sive) of the simulated storage solution.
6. The management system for storage and intelligent analysis of drone test data according to claim 5 is characterized in that: The storage analysis index is determined in the following way: the average write throughput thr(avj) of the parameter simulation storage model is obtained, a rectangular coordinate system is constructed with the storage duration as the X-axis and the data error rate as the Y-axis, the collected data error rate is plotted in the rectangular coordinate system in the form of a curve, and a data error rate curve is obtained by plotting, and perpendicular lines are drawn from both ends of the data error rate curve to the X-axis, and the total area of the closed figure formed by the X-axis, the two perpendicular lines and the data error rate curve is marked as index(errorarea), and the Get the storage analysis index index(storage).
7. The management system for storage and intelligent analysis of drone test data according to claim 5 is characterized in that: The demand reading index of the data reading demand is determined in the following way: a data reading demand is selected, and the instruction of the data reading demand is input into the parameter simulation storage model. The parameter simulation storage model simulates and reads the corresponding stage data according to the data reading demand. During the simulation reading process, after the simulation reading is completed, it is determined whether the read stage data is complete. When it is determined that a stage data is complete, the number of complete data is increased by one, and the number of complete data is marked as number(plete). After the simulation reading is completed, the total duration of the simulation reading is marked as Time(read). Get the demand reading index index(read) of the data reading demand.
8. The management system for drone test data storage and intelligent analysis according to claim 1 is characterized in that: The simulated storage solution with the largest storage read analysis index value is marked as the actual storage solution.