High-precision attitude sensor dynamic calibration system and method thereof

Abstract

The invention discloses a high-precision attitude sensor dynamic calibration system and a method thereof, and belongs to the technical field of attitude sensors. The invention discloses a dynamic calibration system and method for a high-precision attitude sensor. The dynamic calibration system comprises a data acquisition unit, a dynamic calibration unit and a calibration execution unit. The problem that the calibration effect is limited in the prior art is solved, the state of the sensor can be known more comprehensively by collecting the output data of the high-precision attitude sensor, the environmental data and the multi-source data of the accelerometer, the gyroscope and the magnetometer in real time, so that more accurate error analysis and calibration are carried out, and the calibration accuracy is improved. Modeling and error analysis are carried out by utilizing a machine learning algorithm, parameters are automatically adjusted through a model adaptive updating mechanism, error analysis can be carried out on data collected in real time so as to provide reference for subsequent dynamic calibration, and dynamic calibration operation of the high-precision attitude sensor is accurately executed according to an analysis result of the model. And high precision and high stability of attitude measurement are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of attitude sensors, and in particular to a high-precision attitude sensor dynamic calibration system and method. Background Art

[0002] High-precision attitude sensors are increasingly used in aerospace, autonomous driving, robot control, virtual reality and many other fields. These fields have extremely high requirements for the accuracy and stability of attitude sensors, because even a small attitude error may lead to serious consequences. However, in actual use, high-precision attitude sensors are often affected by various factors, such as temperature changes, humidity fluctuations, air pressure differences, magnetic field interference, etc. These factors may cause errors in the sensor output data.

[0003] Most traditional calibration methods use static calibration, that is, calibrating the sensor under fixed environment and conditions. However, this method cannot fully reflect the performance of the sensor in the actual dynamic environment, so the calibration effect is limited; therefore, it does not meet the existing needs. In this regard, we propose a high-precision attitude sensor dynamic calibration system and method. Summary of the invention

[0004] The purpose of the present invention is to provide a high-precision attitude sensor dynamic calibration system and method, which can more comprehensively understand the state of the sensor by collecting the output data of the high-precision attitude sensor in real time and fusing multi-source data of accelerometers, gyroscopes and magnetometers, as well as environmental data. The data is analyzed in detail through a machine learning algorithm, and a model is established to identify errors. Based on the error analysis results of the model, the dynamic calibration operation of the high-precision attitude sensor is accurately performed, thereby solving the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: a high-precision attitude sensor dynamic calibration system, the system comprising a data acquisition unit, a dynamic calibration unit and a calibration execution unit;

[0006] The data acquisition unit is configured to collect output data of a high-precision attitude sensor in real time, and combine multi-source data of an accelerometer, a gyroscope, and a magnetometer, and collect environmental data in real time;

[0007] The dynamic calibration unit is configured to use a machine learning algorithm to model and analyze the collected data based on the data collected by the data acquisition unit, identify potential errors in the data, and at the same time, add a model adaptive update mechanism to automatically adjust the parameters of the model according to the newly collected data and the error analysis results;

[0008] The calibration execution unit is configured to implement dynamic calibration of the high-precision attitude sensor according to the analysis result of the dynamic calibration unit.

[0009] Furthermore, the dynamic calibration unit comprises:

[0010] a feature extraction module configured to extract features related to the error from the collected data, including error size, error frequency, and environmental factors;

[0011] A model building module, configured to build a machine learning model based on a regression algorithm, a classification algorithm, and a clustering algorithm according to the characteristics of the error data, train the machine learning model using the historical error data, and evaluate the performance of the machine learning model through cross-validation, accuracy, and recall indicators;

[0012] A model deployment module is configured to use the trained machine learning model to perform error analysis on the data collected in real time by the data acquisition unit, including the error size and type, and generate calibration reference information required by the calibration execution unit based on the error analysis results;

[0013] An adaptive update module is configured to automatically adjust the parameters of the machine learning model based on newly acquired data and error analysis results.

[0014] Furthermore, the feature extraction module is specifically:

[0015] Error size: The distribution of error size is described by statistical methods, including mean, median and standard deviation;

[0016] Error frequency: record the number of errors and calculate the frequency of errors, including hourly, daily or weekly error rates, and identify the periodicity of error occurrence through time series analysis;

[0017] Environmental factors: Based on the collected environmental data, correlation analysis is used to determine the correlation between environmental factors and error size or frequency, and the environmental factors are input as features into the subsequent machine learning model.

[0018] Furthermore, the adaptive update module specifically performs the following process:

[0019] Use online learning algorithms, including incremental learning and online gradient descent, to update the machine learning model in real time, so that the machine learning model can adapt to the newly collected data and its characteristics;

[0020] Dynamically adjust the hyperparameters and structure of the machine learning model based on the error analysis results;

[0021] Regularly evaluate the performance of updated machine learning models and feed the evaluation results back to the model building module to further optimize the machine learning model.

[0022] Furthermore, the calibration execution unit includes:

[0023] A dynamic calibration module is configured to dynamically select different calibration methods and parameter adjustment ranges according to the error size and type, and dynamically calibrate the high-precision attitude sensor according to the selected calibration method and parameter adjustment range;

[0024] The monitoring interaction module is configured to provide a user interaction interface to monitor the calibration execution process in real time, including real-time display of status information during the calibration execution process, and after the calibration is completed, display the calibration results to the user, including error comparison before and after calibration and calibration parameter values.

[0025] Furthermore, the dynamic calibration module includes:

[0026] The calibration evaluation module is configured to compare the calibrated high-attitude sensor data with known standard data, analyze the error changes before and after calibration, evaluate the calibrated high-attitude sensor data, and verify whether the calibration effect reaches the expected goal.

[0027] Furthermore, the data acquisition unit includes:

[0028] A data acquisition module is configured to acquire output data of the high-precision attitude sensor in real time, including acquiring raw data of an accelerometer, a gyroscope, and a magnetometer, and acquiring environmental data of an environment in which the high-precision attitude sensor is located in real time, including temperature, humidity, air pressure, and magnetic field interference;

[0029] A data processing module configured to pre-process all data collected by the data collection module, including filtering, denoising, normalization, formatting and data compression;

[0030] The data fusion module is configured to use a fusion algorithm of a Kalman filter or an extended Kalman filter to fuse the pre-processed data into unified posture information and transmit it to the dynamic calibration unit for analysis.

[0031] Furthermore, the data acquisition module includes:

[0032] A data synchronization module configured to add a unified timestamp to the collected output data and environmental data;

[0033] The quality monitoring module is configured to monitor the quality of the collected data in real time and detect abnormal values ​​and missing data.

[0034] Furthermore, the quality monitoring module specifically performs the following process:

[0035] Preset threshold range, compare the collected data with the preset threshold range, if the data exceeds the threshold range, mark it as an outlier;

[0036] Check the continuity of the collected data to determine whether there is any data loss. If data loss is found, record the time range and data type of the loss;

[0037] When an abnormal value or data loss is detected, an alarm signal is issued and the abnormal information is recorded in the log file, including the timestamp, data type, abnormal type and specific value.

[0038] A high-precision attitude sensor dynamic calibration method is applied to a high-precision attitude sensor dynamic calibration system, comprising the following steps:

[0039] Real-time collection of high-precision attitude sensor output data, environmental data, and multi-source data from accelerometers, gyroscopes, and magnetometers;

[0040] Adopt Kalman filter or extended Kalman filter fusion algorithm to fuse all collected data into unified posture information;

[0041] Extract error-related features from the fused data, including error size, error frequency, and environmental factors;

[0042] Build a machine learning model and use historical error data for training. Use the trained machine learning model to perform error analysis on real-time collected data and output the error analysis results, including error size and type.

[0043] According to the error size and type, different calibration methods and parameter adjustment ranges are dynamically selected to dynamically calibrate the high-precision attitude sensor. The calibration execution process is monitored in real time to analyze the error changes before and after calibration.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] The present invention can more comprehensively understand the status of the sensor by real-time acquisition of output data of the high-precision attitude sensor, environmental data, and multi-source data of the accelerometer, gyroscope, and magnetometer, thereby performing more accurate error analysis and calibration. The present invention uses a machine learning algorithm to perform modeling and error analysis, and automatically adjusts parameters through a model adaptive update mechanism. It can perform error analysis on the real-time acquired data to provide a reference for subsequent dynamic calibration. According to the analysis results of the model, the dynamic calibration operation of the high-precision attitude sensor is accurately performed, and errors are compensated by adjusting sensor parameters in real time or using advanced calibration algorithms, thereby ensuring high accuracy and high stability of attitude measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is the overall structure diagram of the high-precision attitude sensor dynamic calibration system of the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] In order to solve the technical problem that most existing calibration methods use static calibration, that is, the sensor is calibrated in a fixed environment and conditions, however, this method cannot fully reflect the performance of the sensor in an actual dynamic environment, so the calibration effect is limited, please refer to Figure 1 , this embodiment provides the following technical solutions:

[0049] A high-precision attitude sensor dynamic calibration system, the system comprising a data acquisition unit, a dynamic calibration unit and a calibration execution unit;

[0050] The data acquisition unit is configured to collect output data of a high-precision attitude sensor in real time, and combine multi-source data of an accelerometer, a gyroscope, and a magnetometer, and collect environmental data in real time;

[0051] The dynamic calibration unit is configured to use a machine learning algorithm to model and analyze the collected data based on the data collected by the data acquisition unit, identify potential errors in the data, and at the same time, add a model adaptive update mechanism to automatically adjust the parameters of the model according to the newly collected data and the error analysis results;

[0052] The calibration execution unit is configured to implement dynamic calibration of the high-precision attitude sensor according to the analysis result of the dynamic calibration unit.

[0053] The technical effect of the above content is: the data acquisition unit collects the output data of the high-precision attitude sensor, environmental data, and multi-source data of the accelerometer, gyroscope and magnetometer in real time to provide data support for subsequent analysis, so that more accurate error analysis and calibration can be performed. The dynamic calibration unit uses machine learning algorithms to perform modeling and error analysis, and automatically adjusts parameters through the model adaptive update mechanism. It can perform error analysis on the real-time collected data. The output analysis results provide a reference for subsequent dynamic calibration. The calibration execution unit accurately performs the dynamic calibration operation of the high-precision attitude sensor based on the output analysis results, thereby ensuring high precision and high stability of attitude measurement. Through the data acquisition unit, the dynamic calibration unit and the calibration execution unit, the sensor error can be identified and corrected in real time and accurately, and the accuracy and reliability of attitude measurement can be improved. At the same time, it has adaptive capabilities to adapt to complex and changing environments and dynamic conditions.

[0054] Dynamic calibration unit, comprising:

[0055] a feature extraction module configured to extract features related to the error from the collected data, including error size, error frequency, and environmental factors;

[0056] A model building module, configured to build a machine learning model based on a regression algorithm, a classification algorithm, and a clustering algorithm according to the characteristics of the error data, train the machine learning model using the historical error data, and evaluate the performance of the machine learning model through cross-validation, accuracy, and recall indicators;

[0057] A model deployment module is configured to use the trained machine learning model to perform error analysis on the data collected in real time by the data acquisition unit, including the error size and type, and generate calibration reference information required by the calibration execution unit based on the error analysis results;

[0058] An adaptive update module is configured to automatically adjust the parameters of the machine learning model based on newly acquired data and error analysis results.

[0059] The technical effects of the above content are as follows: the feature extraction module can extract error-related features such as error size, frequency and environmental factors from the collected data, providing an accurate basis for subsequent calibration. The model building module combines regression, classification and clustering algorithms to build a machine learning model, and evaluates the model performance through indicators such as cross-validation, accuracy and recall rate to ensure that the model can accurately analyze errors. The model deployment module uses the trained model to perform error analysis on real-time collected data and generates calibration reference information to effectively reduce measurement errors and improve measurement accuracy. The adaptive update module can automatically adjust the model parameters according to new data and error analysis results to ensure that the calibration unit always maintains a high calibration accuracy during long-term operation.

[0060] Feature extraction module, specifically:

[0061] Error size: The distribution of error size is described by statistical methods, including mean, median and standard deviation;

[0062] Error frequency: record the number of errors and calculate the frequency of errors, including hourly, daily or weekly error rates, and identify the periodicity of error occurrence through time series analysis;

[0063] Environmental factors: Based on the collected environmental data, correlation analysis is used to determine the correlation between environmental factors and error size or frequency, and the environmental factors are input as features into the subsequent machine learning model.

[0064] The technical effect of the above content is: the feature extraction module provides high-quality feature input for the model building module by accurately quantifying the error size, analyzing the error frequency and related environmental factors, which significantly improves the accuracy, adaptability and stability of the calibration system.

[0065] The adaptive update module specifically performs the following process:

[0066] Use online learning algorithms, including incremental learning and online gradient descent, to update the machine learning model in real time, so that the machine learning model can adapt to the newly collected data and its characteristics;

[0067] Dynamically adjust the hyperparameters and structure of the machine learning model based on the error analysis results;

[0068] Regularly evaluate the performance of updated machine learning models and feed the evaluation results back to the model building module to further optimize the machine learning model.

[0069] The technical effect of the above content is: through online learning algorithms (such as incremental learning and online gradient descent), the adaptive update module can update the machine learning model in real time so that it can quickly adapt to newly collected data and its characteristics. This real-time update mechanism ensures that the model is always in the latest state and can promptly reflect the changing trends of the data, thereby improving the dynamic performance of the system. According to the error analysis results, the adaptive update module can dynamically adjust the hyperparameters and structure of the machine learning model. This dynamic adjustment capability enables the model to automatically optimize its own performance according to changes in actual data, further improving the adaptability and accuracy of the model. Regularly evaluating the performance of the updated machine learning model and feeding back the evaluation results to the model building module can continuously improve the generalization ability and stability of the model, ensuring that the model always maintains a high performance during long-term operation.

[0070] Calibration execution unit, comprising:

[0071] A dynamic calibration module is configured to dynamically select different calibration methods and parameter adjustment ranges according to the error size and type, and dynamically calibrate the high-precision attitude sensor according to the selected calibration method and parameter adjustment range;

[0072] The monitoring interaction module is configured to provide a user interaction interface to monitor the calibration execution process in real time, including real-time display of status information during the calibration execution process, and after the calibration is completed, display the calibration results to the user, including error comparison before and after calibration and calibration parameter values.

[0073] The technical effect of the above content is: the dynamic calibration module can dynamically select different calibration methods and parameter adjustment ranges according to the error size and type, so that the calibration process can adopt the most suitable calibration strategy for different error characteristics, thereby improving the calibration accuracy. By dynamically adjusting the amplitude of the calibration parameters, the calibration execution unit can calibrate the high-precision attitude sensor more accurately, avoiding calibration errors caused by mismatch of calibration parameters, and further improving the calibration effect. The monitoring and interaction module allows users to clearly understand the calibration progress and enhance the user's control over the calibration process. After the calibration is completed, the monitoring and interaction module displays the calibration results to the user, including error comparison and calibration parameter values ​​before and after calibration. The visual display method not only allows users to intuitively understand the calibration effect, but also helps users evaluate whether the calibration has achieved the expected goals, thereby enhancing the user experience.

[0074] Dynamic calibration module, including:

[0075] The calibration evaluation module is configured to compare the calibrated high-attitude sensor data with known standard data, analyze the error changes before and after calibration, evaluate the calibrated high-attitude sensor data, and verify whether the calibration effect reaches the expected goal.

[0076] The technical effect of the above content is: by comparing the calibrated high-precision attitude sensor data with the known standard data, the calibration evaluation module can clearly analyze the error changes before and after calibration, and clarify whether the calibration has effectively reduced the error, thereby verifying whether the calibration has achieved the expected goal, and ensuring that the high-precision attitude sensor can operate stably and reliably after calibration. Only when the calibration effect reaches the expected goal can the sensor be put into use, thereby improving the overall reliability of the system. If the calibration effect does not reach the expected goal, the calibration evaluation module can promptly discover and feedback the problem, which is convenient for further adjustment of the calibration strategy or recalibration, avoiding the accumulation of system errors caused by improper calibration, and further enhancing the stability of the system.

[0077] The data acquisition unit comprises:

[0078] A data acquisition module is configured to acquire output data of the high-precision attitude sensor in real time, including acquiring raw data of an accelerometer, a gyroscope, and a magnetometer, and acquiring environmental data of an environment in which the high-precision attitude sensor is located in real time, including temperature, humidity, air pressure, and magnetic field interference;

[0079] A data processing module configured to pre-process all data collected by the data collection module, including filtering, denoising, normalization, formatting and data compression;

[0080] The data fusion module is configured to use a fusion algorithm of a Kalman filter or an extended Kalman filter to fuse the pre-processed data into unified posture information and transmit it to the dynamic calibration unit for analysis.

[0081] The technical effect of the above content is: the data acquisition module can more comprehensively understand the status of the sensor by real-time acquisition of the output data of the high-precision attitude sensor, environmental data, and multi-source data of the accelerometer, gyroscope and magnetometer, so as to perform more accurate error analysis and calibration. A series of preprocessing operations such as filtering, denoising, normalization, formatting and data compression are performed through the data processing model, which is conducive to improving the accuracy of the data, thereby improving the accuracy of the model analysis. Data fusion is achieved through the data fusion module, and high-precision attitude information is output to provide basic data for subsequent error analysis and calibration. The data acquisition unit provides high-quality data support for the dynamic calibration of high-precision attitude sensors through real-time data acquisition, efficient data preprocessing and advanced data fusion algorithms.

[0082] Data acquisition module, including:

[0083] A data synchronization module configured to add a unified timestamp to the collected output data and environmental data;

[0084] The quality monitoring module is configured to monitor the quality of the collected data in real time and detect abnormal values ​​and loss in the data. Specifically, the following process is performed:

[0085] Preset threshold range, compare the collected data with the preset threshold range, if the data exceeds the threshold range, mark it as an outlier;

[0086] Check the continuity of the collected data to determine whether there is any data loss. If data loss is found, record the time range and data type of the loss;

[0087] When an abnormal value or data loss is detected, an alarm signal is issued and the abnormal information is recorded in the log file, including the timestamp, data type, abnormal type and specific value.

[0088] The technical effects of the above content are as follows: the data synchronization module adds a unified timestamp to the collected output data and environmental data to ensure that all data are consistent and traceable in time, so that data from different sources can be aligned in the time dimension, which is convenient for subsequent data processing and analysis, and the unified timestamp provides an accurate time reference for the data fusion module, so that data from accelerometers, gyroscopes, magnetometers and environmental sensors can be accurately fused together to avoid errors caused by time deviation, thereby improving the overall accuracy of the system. The quality monitoring module can monitor the quality of the collected data in real time, detect abnormal values ​​and loss in the data, so that problems that may occur in the data collection process can be discovered in time to ensure the integrity and reliability of the data. When abnormal values ​​or data loss are detected, the quality monitoring module can issue an alarm signal and record the abnormal information in the log file. In this way, the system administrator or user can be notified in time, which is convenient for taking measures quickly and reducing the impact of abnormal data on the system. The data acquisition module significantly improves the time consistency, integrity and reliability of the data through data synchronization and quality monitoring mechanisms, enhances the stability and fault tolerance of the system, reduces maintenance costs, and improves the intelligence level and user experience of the system.

[0089] Specifically, this embodiment further proposes a high-precision attitude sensor dynamic calibration method, which is applied to a high-precision attitude sensor dynamic calibration system, and includes the following steps:

[0090] Real-time collection of high-precision attitude sensor output data, environmental data, and multi-source data from accelerometers, gyroscopes, and magnetometers;

[0091] Adopt Kalman filter or extended Kalman filter fusion algorithm to fuse all collected data into unified posture information;

[0092] Extract error-related features from the fused data, including error size, error frequency, and environmental factors;

[0093] Build a machine learning model and use historical error data for training. Use the trained machine learning model to perform error analysis on real-time collected data and output the error analysis results, including error size and type.

[0094] According to the error size and type, different calibration methods and parameter adjustment ranges are dynamically selected to dynamically calibrate the high-precision attitude sensor. The calibration execution process is monitored in real time to analyze the error changes before and after calibration.

[0095] Working principle: By collecting the output data of high-precision attitude sensors, environmental data, and multi-source data of accelerometers, gyroscopes, and magnetometers in real time, we can have a more comprehensive understanding of the status of the sensor, thereby performing more accurate error analysis and calibration. We build a machine learning model to perform error analysis on the real-time collected data, which provides a reference for subsequent dynamic calibration. According to the error size and type, we dynamically select different calibration methods and parameter adjustment ranges to dynamically calibrate the high-precision attitude sensor. After the calibration is completed, the monitoring interaction module displays the calibration results to the user, including the error comparison before and after calibration and the calibration parameter values, which can help users evaluate whether the calibration has achieved the expected goals, thereby enhancing the user experience.

[0096] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0097] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. A high-precision attitude sensor dynamic calibration system, characterized in that: The system includes a data acquisition unit, a dynamic calibration unit and a calibration execution unit; The data acquisition unit is configured to collect output data of a high-precision attitude sensor in real time, and combine multi-source data of an accelerometer, a gyroscope, and a magnetometer, and collect environmental data in real time; The dynamic calibration unit is configured to use a machine learning algorithm to model and analyze the collected data based on the data collected by the data acquisition unit, identify potential errors in the data, and at the same time, add a model adaptive update mechanism to automatically adjust the parameters of the model according to the newly collected data and the error analysis results; The calibration execution unit is configured to implement dynamic calibration of the high-precision attitude sensor according to the analysis result of the dynamic calibration unit.

2. A high-precision attitude sensor dynamic calibration system according to claim 1, characterized in that: The dynamic calibration unit comprises: a feature extraction module configured to extract features related to the error from the collected data, including error size, error frequency, and environmental factors; A model building module, configured to build a machine learning model based on a regression algorithm, a classification algorithm, and a clustering algorithm according to the characteristics of the error data, train the machine learning model using the historical error data, and evaluate the performance of the machine learning model through cross-validation, accuracy, and recall indicators; A model deployment module is configured to use the trained machine learning model to perform error analysis on the data collected in real time by the data acquisition unit, including the error size and type, and generate calibration reference information required by the calibration execution unit based on the error analysis results; An adaptive update module is configured to automatically adjust the parameters of the machine learning model based on newly acquired data and error analysis results.

3. A high-precision attitude sensor dynamic calibration system according to claim 2, characterized in that: The feature extraction module is specifically: Describe the distribution of error magnitudes using statistical methods, including mean, median, and standard deviation; Record the number of errors and calculate their frequency, including hourly, daily or weekly error rates, and identify the periodicity of error occurrence through time series analysis; Based on the collected environmental data, correlation analysis is used to determine the correlation between environmental factors and error size or frequency, and the environmental factors are input as features into the subsequent machine learning model.

4. A high-precision attitude sensor dynamic calibration system according to claim 2, characterized in that: The adaptive update module specifically performs the following process: Use online learning algorithms, including incremental learning and online gradient descent, to update the machine learning model in real time, so that the machine learning model can adapt to the newly collected data and its characteristics; Dynamically adjust the hyperparameters and structure of the machine learning model based on the error analysis results; Regularly evaluate the performance of updated machine learning models and feed the evaluation results back to the model building module to further optimize the machine learning model.

5. A high-precision attitude sensor dynamic calibration system according to claim 1, characterized in that: The calibration execution unit comprises: A dynamic calibration module is configured to dynamically select different calibration methods and parameter adjustment ranges according to the error size and type, and dynamically calibrate the high-precision attitude sensor according to the selected calibration method and parameter adjustment range; The monitoring interaction module is configured to provide a user interaction interface to monitor the calibration execution process in real time, including real-time display of status information during the calibration execution process, and after the calibration is completed, display the calibration results to the user, including error comparison before and after calibration and calibration parameter values.

6. A high-precision attitude sensor dynamic calibration system according to claim 5, characterized in that: The dynamic calibration module comprises: The calibration evaluation module is configured to compare the calibrated high-attitude sensor data with known standard data, analyze the error changes before and after calibration, evaluate the calibrated high-attitude sensor data, and verify whether the calibration effect reaches the expected goal.

7. The high-precision attitude sensor dynamic calibration system according to claim 1, characterized in that: The data acquisition unit comprises: A data acquisition module is configured to acquire output data of the high-precision attitude sensor in real time, including acquiring raw data of an accelerometer, a gyroscope, and a magnetometer, and acquiring environmental data of an environment in which the high-precision attitude sensor is located in real time, including temperature, humidity, air pressure, and magnetic field interference; A data processing module configured to pre-process all data collected by the data collection module, including filtering, denoising, normalization, formatting and data compression; The data fusion module is configured to use a fusion algorithm of a Kalman filter or an extended Kalman filter to fuse the pre-processed data into unified posture information and transmit it to the dynamic calibration unit for analysis.

8. A high-precision attitude sensor dynamic calibration system according to claim 7, characterized in that: The data acquisition module comprises: A data synchronization module configured to add a unified timestamp to the collected output data and environmental data; The quality monitoring module is configured to monitor the quality of the collected data in real time and detect abnormal values ​​and missing data.

9. A high-precision attitude sensor dynamic calibration system according to claim 8, characterized in that: The quality monitoring module specifically performs the following process: Preset threshold range, compare the collected data with the preset threshold range, if the data exceeds the threshold range, mark it as an outlier; Check the continuity of the collected data to determine whether there is any data loss. If data loss is found, record the time range and data type of the loss; When an abnormal value or data loss is detected, an alarm signal is issued and the abnormal information is recorded in the log file, including the timestamp, data type, abnormal type and specific value.

10. A high-precision attitude sensor dynamic calibration method, applied to the high-precision attitude sensor dynamic calibration system according to any one of claims 1 to 9, characterized in that: The following steps are involved: Real-time collection of high-precision attitude sensor output data, environmental data, and multi-source data from accelerometers, gyroscopes, and magnetometers; Adopt Kalman filter or extended Kalman filter fusion algorithm to fuse all collected data into unified posture information; Extract error-related features from the fused data, including error size, error frequency, and environmental factors; Build a machine learning model and use historical error data for training. Use the trained machine learning model to perform error analysis on real-time collected data and output the error analysis results, including error size and type. According to the error size and type, different calibration methods and parameter adjustment ranges are dynamically selected to dynamically calibrate the high-precision attitude sensor. The calibration execution process is monitored in real time to analyze the error changes before and after calibration.

Images