A processing system and method for multi-source heterogeneous test data of an inertial navigation system

Abstract

The application relates to a data processing system and method of an inertial navigation system, in particular to a processing system and method for multi-source heterogeneous test data of the inertial navigation system; and is used for solving the problems that the test data of the inertial navigation system is dispersed in storage positions, the formats are different, the data processing efficiency is low, and data standardization is difficult. The processing system comprises a signal capturing module, a standardization conversion module and a data storage processing module; the signal capturing module automatically traverses and captures multi-format test files stored in multiple levels of nested folders in non-fixed positions through a breadth-first search algorithm. The standardization conversion module parses and converts the captured files into standardized data in a unified format according to a predefined three-layer index system of'single machine, component and index', and dynamically generates database operation statements. The data storage processing module executes the warehousing operation, and automatically completes the calculation and associated storage of characteristics such as maximum value, minimum value, average value and variance.

Description

Technical Field

[0001] This invention relates to a data processing system and method for inertial navigation systems, and more specifically to a processing system and method for multi-source heterogeneous test data of inertial navigation systems. Background Technology

[0002] Inertial navigation systems (INS) are core devices for high-precision navigation and positioning. Their performance test data forms the basis for assessing system health, diagnosing faults, and developing maintenance strategies. This test data exhibits typical "multi-source heterogeneous" characteristics during its generation, storage, and management: diverse data sources (from different test units, sensors, and logs), scattered and unfixed storage locations (distributed across infinitely nested folders), and complex formats (including TXT, Excel, binary, and direct database exports). Currently, processing this type of data primarily relies on manual or semi-automatic scripts, often encountering the following challenges:

[0003] First, the automation and intelligence levels of data acquisition are low, making it difficult to handle complex storage structures. Current technologies typically rely on manual searching or writing scripts to find fixed file paths when processing test files scattered in non-fixed locations and deeply nested folders. This approach is inefficient, prone to missing files, and unable to adapt to dynamically changing directory structures. Especially during large-scale, periodic data collection, the lack of intelligent mechanisms to resume interrupted tasks and capture only synchronous incremental data leads to poor reliability and significant resource waste in the data acquisition process.

[0004] Second, data standardization relies on manual processing, resulting in low conversion efficiency and poor consistency. Faced with raw data in different formats, existing methods often employ manual parsing or conversion tools customized for a single format. This process requires significant manual intervention to identify data meaning and align indicator fields, which is not only inefficient but also prone to introducing errors, leading to inconsistent data quality and preventing the formation of a unified and standardized dataset for subsequent analysis.

[0005] Third, the data processing chain is broken, and feature extraction is not deeply integrated with storage. Existing technologies often treat data storage and feature calculation as two separate steps. After data is stored, additional tools or scripts are needed to extract features (such as maximum, minimum, average, and variance), which is cumbersome. Furthermore, the association between feature data and original data is loosely managed, which is not conducive to subsequent traceability and analysis. Summary of the Invention

[0006] The purpose of this invention is to solve the technical problems of low data processing efficiency and difficulty in data standardization caused by the dispersed storage locations and inconsistent formats of test data in inertial navigation systems, and to provide a system and method for processing multi-source heterogeneous test data of inertial navigation systems.

[0007] To achieve the above objectives, the technical solution provided by this invention is as follows:

[0008] A processing system for multi-source heterogeneous test data of inertial navigation systems is characterized by including a signal acquisition module, a standardization conversion module, and a data storage and processing module.

[0009] The data input terminal of the signal acquisition module is communicatively connected to the data output terminal of at least one inertial navigation system, and is used to acquire multi-source heterogeneous test data from the inertial navigation system.

[0010] The data input terminal of the standardization conversion module is communicatively connected to the data output terminal of the signal capture module, and is used to convert the multi-source heterogeneous test data into standardized data.

[0011] The data input terminal of the data storage and processing module is communicatively connected to the data output terminal of the standardization conversion module, and is used to store the standardized data and perform feature calculation processing on the stored standardized data.

[0012] Furthermore, the signal capture module, standardization conversion module, and data storage processing module are deployed independently and communicate with each other through an API gateway.

[0013] Furthermore, the signal capture module, standardization conversion module, and data storage and processing module are integrated and deployed in a central server, and exchange data through an inter-process communication mechanism.

[0014] Meanwhile, the present invention also provides a method for processing multi-source heterogeneous test data of inertial navigation systems, which is characterized by including the following steps:

[0015] S1. Build the above-mentioned system for processing multi-source heterogeneous test data of inertial navigation system;

[0016] S2. Construct an inertial navigation system index definition file based on the logical structure of the single-machine layer, component layer, and index layer; the index definition file contains positioning information and conversion rules for converting multi-source heterogeneous test data into standardized data; the single-machine layer is used to identify the inertial navigation system; the component layer is used to identify the internal components of the inertial navigation system, including gyroscopes and accelerometers; the index layer is used to define the physical quantities that need to be extracted from the internal components; the physical quantities include at least acceleration, angular velocity, and attitude angle;

[0017] S3. Capture multi-source heterogeneous test data of the inertial navigation system through the signal capture module, and transmit the multi-source heterogeneous test data to the standardization conversion module;

[0018] S4. The standardization conversion module constructs an index mapping relationship between the multi-source heterogeneous test data and the standardized data according to the index definition file; then, it converts the multi-source heterogeneous test data into standardized data according to the index mapping relationship.

[0019] S5. The data storage and processing module receives and stores the standardized data; then it extracts features from the standardized data stored in the storage and processing module, and associates and stores the extraction results with the standardized data.

[0020] S6. Repeat steps S3-S5 until the processing of multi-source heterogeneous test data of the inertial navigation system is completed.

[0021] Furthermore, step S3 specifically includes:

[0022] The signal acquisition module uses a breadth-first search algorithm to traverse the inertial navigation system's file storage system to locate and acquire multi-source heterogeneous test data stored in different locations and formats within the inertial navigation system's file storage system, and then transmits the multi-source heterogeneous test data to the standardization conversion module.

[0023] Furthermore, in step S2, the multi-source heterogeneous test data includes acceleration, angular velocity, and attitude angle data;

[0024] In step S3, the formats of the multi-source heterogeneous test data include TXT format, Excel format, binary format, and database format.

[0025] Furthermore, in step S3, the signal acquisition module uses a multi-threaded concurrent mechanism to execute the breadth-first search algorithm in order to traverse and read the file storage systems of multiple inertial navigation systems in parallel.

[0026] The signal capture module supports breakpoint resume function and incremental update mode.

[0027] Furthermore, step S4 specifically includes:

[0028] S4.1 The standardization conversion module obtains and parses the indicator definition file, and constructs the indicator mapping relationship between the multi-source heterogeneous test data and the standardized data;

[0029] S4.2 The standardization conversion module locates and extracts the corresponding indicator values ​​from the multi-source heterogeneous test data according to the indicator mapping relationship;

[0030] S4.3. The extracted indicator values ​​from the multi-source heterogeneous test data are encapsulated into persistent objects and temporarily stored in a collection. The collection is traversed and concatenated to generate batch insertion SQL statements, which are then output as standardized data to the data storage and processing module.

[0031] Further, in step S4.2, when the multi-source heterogeneous test data is in TXT format, each line of data is extracted as a string, and the string is parsed using a space separator to obtain the index value of the multi-source heterogeneous test data;

[0032] When the multi-source heterogeneous test data is in Excel format, the cell data is read row by row to obtain the index values ​​of the multi-source heterogeneous test data;

[0033] When the multi-source heterogeneous test data is in binary format, the index value of the multi-source heterogeneous test data is parsed out according to the byte offset and data type in the index mapping relationship.

[0034] When the multi-source heterogeneous test data is in a non-standard format, it is converted and adapted according to a custom rule to obtain the index value of the multi-source heterogeneous test data; the non-standard format is any format other than TXT format, Excel format, binary format and database format.

[0035] Furthermore, step S5 specifically includes:

[0036] S5.1 The data storage processing module receives and executes SQL statements, thereby storing the standardized data into the data storage processing module;

[0037] S5.2 The data storage and processing module extracts the maximum, minimum, average and variance features of the standardized data stored in it according to the logical structure of the single machine layer, component layer and index layer.

[0038] S5.3. The extraction results are correlated with the standardized data and stored in the data storage and processing module.

[0039] Compared with the prior art, the present invention has the following beneficial technical effects:

[0040] 1. This invention provides a processing system for multi-source heterogeneous test data of inertial navigation systems. It constructs a complete automated data processing pipeline, realizing the centralized capture, unified standardization, and deep feature extraction of scattered and heterogeneous multi-source heterogeneous test data. This provides a high-quality and well-organized data foundation for upper-level analysis, fundamentally solving the problem of the difficulty in uniformly utilizing multi-source heterogeneous test data.

[0041] 2. The present invention provides a processing system for multi-source heterogeneous test data of inertial navigation systems. It adopts a microservice architecture and standardized interfaces, which realizes the decoupling and independent deployment of the core functional modules of the system, significantly improving the system's flexibility, maintainability and module-level scalability.

[0042] 3. The present invention provides a processing system for multi-source heterogeneous test data of inertial navigation systems, which provides an integrated deployment scheme, reduces the complexity of distributed management of the system and network communication overhead, and is suitable for application scenarios with higher requirements for deployment simplicity, internal communication efficiency and data security.

[0043] 4. The present invention provides a method for processing multi-source heterogeneous test data of inertial navigation systems, which provides a standardized and repeatable data processing flow to ensure efficient and accurate conversion from messy raw files to structured feature data that can be used for in-depth analysis, thereby improving the reliability and efficiency of the entire data preprocessing process.

[0044] 5. This invention provides a method for processing multi-source heterogeneous test data for inertial navigation systems. It clarifies that the system can handle several of the most common data formats in the current inertial navigation testing field, demonstrating its strong format compatibility and wide applicability. Simultaneously, by utilizing a breadth-first search algorithm, it systematically solves the problem of locating files in deep, irregular directory structures, achieving comprehensive and complete automatic acquisition of test files.

[0045] 6. This invention provides a method for processing multi-source heterogeneous test data of inertial navigation systems. Through multi-threaded concurrency, it significantly improves file retrieval efficiency; breakpoint resumption enhances the robustness of large-scale data transmission; and the incremental update mode avoids repetitive operations on already processed data, significantly improving the efficiency of subsequent data processing. Simultaneously, the standardization conversion process is decomposed into three logically clear sub-steps. In particular, through a three-layer modeling approach of "single machine, component, and index," the physical composition of the inertial navigation system is accurately mapped, giving the data conversion process clear semantics and good configurability.

[0046] 7. The present invention provides a method for processing multi-source heterogeneous test data of inertial navigation systems. It provides specific parsing strategies for different file formats and introduces custom mapping rules, which greatly enhances the system's ability to process non-standard and customized data formats and improves its adaptability to diverse data sources.

[0047] 8. This invention provides a method for processing multi-source heterogeneous test data of inertial navigation systems. It clarifies the types of extracted statistical features, which form the basis for subsequent health assessments and trend analyses. By organizing the calculation results by indicators and storing them in a structured manner, it greatly facilitates subsequent querying, analysis, and modeling, directly enhancing the usability of the data. Attached Figure Description

[0048] Figure 1 This is a schematic diagram of an embodiment of a system for processing multi-source heterogeneous test data of an inertial navigation system according to the present invention;

[0049] Figure 2This is a flowchart illustrating an embodiment of a method for processing multi-source heterogeneous test data of an inertial navigation system according to the present invention.

[0050] The attached figures are labeled as follows:

[0051] 1-Signal capture module, 2-Standardization conversion module, 3-Data storage and processing module. Detailed Implementation

[0052] To make the objectives, advantages, and features of the present invention clearer, the following detailed description of a system and method for processing multi-source heterogeneous test data of an inertial navigation system, in conjunction with the accompanying drawings and specific embodiments, is provided. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0053] This invention discloses a processing system for multi-source heterogeneous test data of inertial navigation systems, used for acquiring and processing data from inertial navigation systems, and an intelligent processing system such as... Figure 1 As shown, it includes: signal capture module 1, standardization conversion module 2, and data storage and processing module 3.

[0054] The data input terminal of the signal acquisition module 1 is communicatively connected to the data output terminal of at least one inertial navigation system, and is used to acquire multi-source heterogeneous test data from the inertial navigation system. The multi-source heterogeneous test data consists of test data stored in different locations and in different formats (such as TXT and Excel formats) within the inertial navigation system.

[0055] The data input terminal of the standardization conversion module 2 is connected to the data output terminal of the signal capture module 1 to convert multi-source heterogeneous test data into standardized data with a unified standard.

[0056] The data input terminal of the data storage and processing module 3 is communicatively connected to the data output terminal of the standardization conversion module 2, and is used to store standardized data and perform feature calculation processing on the stored standardized data.

[0057] The signal capture module 1, standardization conversion module 2, and data storage and processing module 3 are deployed independently and communicate with each other through an API gateway.

[0058] In other embodiments, the signal capture module 1, the standardization conversion module 2, and the data storage and processing module 3 are integrated and deployed in a central server, and exchange data through an inter-process communication mechanism.

[0059] Based on the aforementioned intelligent processing system, a method for processing multi-source heterogeneous test data of inertial navigation systems is described as follows: Figure 2 The process shown is implemented with the following specific steps:

[0060] S1. Build the above-mentioned processing system for multi-source heterogeneous test data of inertial navigation system.

[0061] S2. Based on the logical structure of the single-machine layer (i.e., defining which device under test), component layer (i.e., defining which sensor the data comes from, such as gyroscope or accelerometer), and index layer (i.e., defining the main content of the file, specifically acceleration, angular velocity, or temperature, etc.), the index definition file of the inertial navigation system is constructed. The index definition file is the positioning information and conversion rules for converting multi-source heterogeneous test data into standardized data. Multi-source heterogeneous test data includes acceleration, angular velocity, and attitude angle data.

[0062] The index definition file is based on the physical structure of the inertial navigation system and is organized in a three-layer logic of "single machine, component and index": single machine layer, which identifies the device under test (such as a certain type of inertial navigation system); component layer, which identifies the sub-units inside the device (such as gyroscope, accelerometer, temperature sensor, etc.); index layer, which specifies the physical quantities to be extracted (such as X-axis angular velocity, Y-axis acceleration, attitude angle, etc.).

[0063] S3. The signal acquisition module 1 acquires multi-source heterogeneous test data of the inertial navigation system and transmits the multi-source heterogeneous test data to the standardization conversion module 2; specifically:

[0064] The signal acquisition module 1 uses a breadth-first search algorithm to traverse the directory structure of the inertial navigation system's file storage system in order to locate and acquire multi-source heterogeneous test data stored in different locations and formats in the inertial navigation system's file storage system, and then transmits the acquired multi-source heterogeneous test data to the standardization conversion module 2.

[0065] The heterogeneous test data from multiple sources includes TXT, Excel, binary, and database formats. Signal capture module 1 employs a multi-threaded concurrent mechanism to execute a breadth-first search algorithm, traversing and reading the file storage systems of multiple inertial navigation systems in parallel. Signal capture module 1 supports breakpoint resumption functionality after network interruption or system anomaly, as well as an incremental update mode for capturing newly added or changed files since the last operation.

[0066] The process of the breadth-first search algorithm is illustrated below:

[0067] (1) Initialize the queue: Put the root directory data into a first-in-first-out (FIFO) queue.

[0068] (2) Loop processing: Take a directory item (e.g., data) from the head of the queue and traverse all entries under that directory.

[0069] If a subdirectory is encountered (such as 2025-01-01, 2025-01-02, archive), then these subdirectory paths are added to the end of the queue.

[0070] If a file is encountered (such as data / 2025-01-01 / IMU1 / accel.txt), it is determined whether it is a test signal file to be captured based on the preset file extension filtering rules (such as txt, csv, bin, xlsx, dat, etc.). If so, the file path is added to the capture list.

[0071] (3) Multi-threaded concurrency: To improve efficiency, the system starts multiple worker threads. Each thread obtains the file path from the shared list of files to be crawled and performs reading operations in parallel (e.g., reading file content via FTP, SMB, or local file system). At the same time, the file traversal process can also be performed in multiple threads, with each thread processing different directories in the queue, thus speeding up the directory scanning.

[0072] (4) Resume interrupted download: The system records the metadata of the captured files (such as file path, modification time, file size) and the current traversal progress (queue status) locally. If the task is terminated due to network interruption or system abnormality, after restarting, the system first reads the last saved status, restores the queue, and continues traversing and capturing from the breakpoint to avoid repetitive work.

[0073] (5) Incremental Update: When the system is set to incremental update mode, it first reads the file metadata list saved during the last capture. During the traversal, for each file, it checks whether its modification time or file size has changed. If it has not changed, it skips it; it only captures newly appearing or changed files.

[0074] S4. Standardization conversion module 2 constructs an indicator mapping relationship between multi-source heterogeneous test data and standardized data based on the indicator definition file; then, it converts the multi-source heterogeneous test data into standardized data based on the indicator mapping relationship.

[0075] Indicator mapping refers to the rules that establish correspondence between specific fields, column positions, or data identifiers in the original test data and standard data during the data standardization process. It is equivalent to a "translator" that tells the system which standard indicator (such as "X-axis acceleration") corresponds to a certain element in the original data (such as data in a column named "Acc_X"), and how to convert the data into a standard format (such as unit conversion, data type conversion, etc.).

[0076] S4.1 Standardization Conversion Module 2 acquires and parses the indicator definition file, and constructs the indicator mapping relationship between multi-source heterogeneous test data and standardized data.

[0077] The standardization conversion module 2 first reads the pre-configured indicator definition file. It then parses each definition in the file, extracting the location information of each standard indicator in the original data (e.g., column number in a TXT file, column in an Excel sheet, starting byte and data type in a binary file), as well as any necessary format conversion rules (e.g., unit conversion factors, string truncation methods). All these correspondences are organized into a mapping table for use in subsequent steps.

[0078] S4.2 Standardization conversion module 2 locates and extracts the corresponding indicator values ​​from multi-source heterogeneous test data according to the indicator mapping relationship;

[0079] When the multi-source heterogeneous test data is in TXT format, each line of data is extracted as a string, and the string is parsed using a space separator to obtain the index value of the multi-source heterogeneous test data.

[0080] When the multi-source heterogeneous test data is in Excel format, the cell data is read row by row to obtain the index values ​​of the multi-source heterogeneous test data.

[0081] When the multi-source heterogeneous test data is in binary format, the index values ​​of the multi-source heterogeneous test data are parsed out based on the byte offset and data type in the index mapping relationship.

[0082] When the multi-source heterogeneous test data is in a non-standard format (non-standard format is any format other than TXT, Excel, binary, and database formats), it is converted and adapted according to custom rules to obtain the index values ​​of the multi-source heterogeneous test data; custom rules (such as converting the unit from "degrees / second" to "radians / second", or mapping non-standard index names, such as GyrX, to standard names, such as Gyro_X) are used to convert and adapt index names or data units in non-standard formats.

[0083] S4.3, Standardization and Transformation Module 2 encapsulates the extracted indicator values ​​into persistent objects (e.g., a Java POJO object containing information such as single machine ID, component ID, indicator ID, timestamp, and numerical value) and temporarily stores them in a collection. It then iterates through the collection and concatenates the data to generate batch insert SQL statements (e.g., using INSERT INTO and VALUES syntax), and outputs them as standardized data to Data Storage and Processing Module 3.

[0084] S5. Data storage and processing module 3 receives and stores standardized data; then it extracts features from the standardized data stored in storage and processing module 3 and associates the extraction results with the standardized data for storage.

[0085] S5.1 The data storage processing module 3 receives and executes SQL statements, thereby storing standardized data into the data storage processing module 3.

[0086] S5.2, Data storage and processing module 3 extracts features of maximum, minimum, average, and variance from the standardized data stored within it, according to the logical structure of the single-machine layer, component layer, and indicator layer. Assume that the standardized data without feature extraction contains the following fields: id (primary key), device_id (device ID), component (component, e.g., "accelerometer"), indicator (indicator, e.g., "Accel_X"), timestamp, and value. The standardized data after feature extraction contains the following fields: id (primary key), device_id (device ID), component (component, e.g., "accelerometer"), indicator (indicator, e.g., "Accel_X"), stat_type (statistical type, e.g., "max", "min", "avg", "var"), start_time, end_time, and result. S5.3 The extracted results are then correlated with the standardized data and stored in data storage and processing module 3.

[0087] Feature extraction examples are as follows:

[0088] (1) Determine the calculation dimensions: The preset dimensions can be a combination of "by device, component, and indicator" and may be grouped by time window (such as hour, day, or each test task). For example, this calculation requires statistical analysis of the various characteristics of the X-axis of the accelerometer of device "IMU-01" throughout the day on January 1, 2025.

[0089] (2) Data query: Data storage and processing module 3 initiates a query to the database and filters out data that meets the conditions: The query result returns a set of values ​​(for example, containing 86,400 values ​​sampled per second).

[0090] (3) Traverse and count: Traverse the set of values ​​and calculate:

[0091] Maximum value: Records the maximum value and its timestamp (optional); Minimum value: Records the minimum value and its timestamp; Average value: Sum of all values ​​and divide by the total; Variance: Calculated according to the variance formula, or a recursive algorithm can be used to reduce memory usage.

[0092] For example, for a set of acceleration data [0.01, 0.02, -0.01, 0.03, 0.00], the following can be calculated:

[0093] Maximum value = 0.03, minimum value = -0.01, average value = 0.01, variance = 0.0002.

[0094] (4) Result storage: Associate the calculation results with the original indicators and insert them into the feature table.

[0095] (5) Batch processing and scheduling: Feature calculation can be performed by scheduled tasks (such as calculating the previous day's data at midnight every day) or event triggering (incremental features are calculated in real time after new data is entered into the database). For large-scale data, SQL statements for grouping and aggregation can be used to calculate directly, reducing data transmission, and then the results can be directly inserted into the feature table.

[0096] Through the above process, the system transforms the raw time-series data into statistically significant feature data, providing directly usable analytical dimensions for the health assessment and fault diagnosis of inertial navigation systems.

[0097] S6. Repeat steps S3-S5 until the processing of multi-source heterogeneous test data of the inertial navigation system is completed.

[0098] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the present invention.

Claims

1. A system for processing multi-source heterogeneous test data of inertial navigation systems, characterized in that: It includes a signal capture module (1), a standardization conversion module (2), and a data storage and processing module (3); The data input terminal of the signal capture module (1) is communicatively connected to the data output terminal of at least one inertial navigation system, and is used to capture multi-source heterogeneous test data from the inertial navigation system; The data input terminal of the standardization conversion module (2) is communicatively connected to the data output terminal of the signal capture module (1) to convert the multi-source heterogeneous test data into standardized data. The data input terminal of the data storage processing module (3) is communicatively connected to the data output terminal of the standardization conversion module (2) for storing the standardized data and performing feature calculation processing on the stored standardized data.

2. The system for processing multi-source heterogeneous test data of inertial navigation systems according to claim 1, characterized in that: The signal capture module (1), standardization conversion module (2) and data storage processing module (3) are deployed independently and communicate with each other through an API gateway.

3. The system for processing multi-source heterogeneous test data of inertial navigation systems according to claim 1, characterized in that: The signal capture module (1), standardization conversion module (2) and data storage processing module (3) are integrated and deployed in the central server, and exchange data through inter-process communication mechanism.

4. A method for processing multi-source heterogeneous test data of inertial navigation systems, characterized in that, Includes the following steps: S1. Construct the processing system for multi-source heterogeneous test data of inertial navigation system as described in any one of claims 1-3; S2. Construct an inertial navigation system index definition file based on the logical structure of the single-machine layer, component layer, and index layer; the index definition file contains the positioning information and conversion rules for converting multi-source heterogeneous test data into standardized data. S3. The signal acquisition module (1) acquires the multi-source heterogeneous test data of the inertial navigation system and transmits the multi-source heterogeneous test data to the standardization conversion module (2). S4. Standardization conversion module (2) constructs the index mapping relationship between the multi-source heterogeneous test data and the standardized data according to the index definition file; and then converts the multi-source heterogeneous test data into standardized data according to the index mapping relationship. S5. The data storage and processing module (3) receives and stores the standardized data; then it extracts features from the standardized data stored in the storage and processing module (3) and associates the extraction results with the standardized data for storage. S6. Repeat steps S3-S5 until the processing of multi-source heterogeneous test data of the inertial navigation system is completed.

5. The method for processing multi-source heterogeneous test data of inertial navigation systems according to claim 4, characterized in that, Step S3 is as follows: The signal capture module (1) uses a breadth-first search algorithm to traverse the file storage system of the inertial navigation system in order to locate and capture multi-source heterogeneous test data stored in different locations and formats in the file storage system of the inertial navigation system, and transmit the captured multi-source heterogeneous test data to the standardization conversion module (2).

6. The method for processing multi-source heterogeneous test data of an inertial navigation system according to claim 4 or 5, characterized in that: In step S2, the multi-source heterogeneous test data includes acceleration, angular velocity, and attitude angle data; In step S3, the formats of the multi-source heterogeneous test data include TXT format, Excel format, binary format, and database format.

7. The method for processing multi-source heterogeneous test data of inertial navigation systems according to claim 5, characterized in that: In step S3, the signal capture module (1) uses a multi-threaded concurrent mechanism to execute the breadth-first search algorithm to traverse and read the file storage system of multiple inertial navigation systems in parallel; The signal capture module (1) supports breakpoint resume function and incremental update mode.

8. The method for processing multi-source heterogeneous test data of an inertial navigation system according to claim 4 or 5, characterized in that, Step S4 is as follows: S4.1, Standardization conversion module (2) acquires and parses the indicator definition file, and constructs the indicator mapping relationship between the multi-source heterogeneous test data and the standardized data; S4.

2. Standardization conversion module (2) locates and extracts the corresponding index values ​​from the multi-source heterogeneous test data according to the index mapping relationship; S4.3, Standardization conversion module (2) encapsulates the extracted index values ​​into persistent objects and temporarily stores them in a collection, traverses the collection and concatenates them to generate batch insertion SQL statements, and outputs them as standardized data to the data storage and processing module (3).

9. The method for processing multi-source heterogeneous test data of inertial navigation systems according to claim 8, characterized in that: In step S4.2, when the multi-source heterogeneous test data is in TXT format, each line of data is extracted as a string, and the string is parsed using a space separator to obtain the index value of the multi-source heterogeneous test data. When the multi-source heterogeneous test data is in Excel format, the cell data is read row by row to obtain the index values ​​of the multi-source heterogeneous test data; When the multi-source heterogeneous test data is in binary format, the index value of the multi-source heterogeneous test data is parsed according to the byte offset and data type in the index mapping relationship; when the multi-source heterogeneous test data is in non-standard format, it is converted and adapted according to custom rules to obtain the index value of the multi-source heterogeneous test data; the non-standard format is other formats besides TXT format, Excel format, binary format and database format.

10. The method for processing multi-source heterogeneous test data of an inertial navigation system according to claim 8, characterized in that, Step S5 is as follows: S5.1 The data storage processing module (3) receives and executes the SQL statement, thereby storing the standardized data into the data storage processing module (3). S5.2, Data storage and processing module (3) extracts the maximum value, minimum value, average value and variance of the standardized data stored in it according to the logical structure of single machine layer, component layer and index layer; S5.

3. The extraction results are associated with the standardized data and stored in the data storage processing module (3).

Images