A six-axis sensor acquisition system with anti-interference

Abstract

The present invention relates to the technical field of data processing, and specifically provides an anti-interference six-dimensional sensor acquisition system, including a multimodal signal fusion module for receiving the original acquisition data of the six-dimensional sensor and realizing the fusion of the acquisition data of all sensors based on an adaptive dynamic weight allocation algorithm; an intelligent error compensation module for positively compensating the errors in the six-dimensional sensor data to improve the accuracy of the data; and a calibration feedback module for self-calibrating the data acquired by the six-dimensional sensor at the application end and performing feedback adjustment of the sampling parameters according to the calibration results. The present invention adopts an adaptive covariance matching filtering algorithm, combines Kalman filtering to dynamically adjust the measurement noise covariance matrix, improves the denoising effect, and adaptively adjusts the data smoothing process by setting a dynamic smoothing factor to ensure the optimal signal quality under different noise levels.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and specifically to an anti-interference six-dimensional sensor acquisition system. Background Art

[0002] Six-dimensional sensors are mainly used for high-precision measurement, such as in the fields of industrial robot control, UAV attitude detection, autonomous driving systems, etc. However, in complex environments, the data acquisition of six-dimensional sensors faces many challenges, mainly including problems such as signal interference, error accumulation, and noise influence, resulting in a decline in data reliability and stability.

[0003] Therefore, anti-interference design needs to fully consider the stability of signals and the reliability of data. However, at the current stage, during the data acquisition process of six-dimensional sensors, the anti-interference design still faces the following problems:

[0004] Sensors are vulnerable to noise influence in high electromagnetic interference and high mechanical vibration environments, resulting in data jitter or distortion;

[0005] Traditional weighted fusion methods use fixed weights and it is difficult to adjust the weights in real time according to the sensor state, resulting in unstable data fusion effects;

[0006] Most existing systems adopt fixed compensation models and cannot perform precise compensation for different types of errors (such as sensor zero-bias drift, electromagnetic interference, temperature drift, etc.);

[0007] Traditional filtering methods (such as mean filtering and low-pass filtering) have good effects in fixed noise environments, but the filtering effect decreases in the face of complex environments (such as dynamic noise and sudden interference);

[0008] Traditional systems lack a self-correction mechanism based on feedback control and cannot automatically adjust sampling parameters and filtering strategies when the environment changes, resulting in a decline in data stability. Summary of the Invention

[0009] In view of the above existing problems, the present invention is proposed.

[0010] To solve the above technical problems, the present invention provides the following technical solution: An anti-interference six-dimensional sensor acquisition system, including,

[0011] A multi-modal signal fusion module, further including,

[0012] For the original acquisition data received by the six-dimensional sensor, through an adaptive dynamic weight allocation algorithm, adaptive weight allocation is performed on different sensors, and the data collected by all sensors is fused according to the allocated weights. At the same time, an adaptive covariance allocation algorithm is used to denoise the fused data, and by setting a filtering threshold, it is judged whether the data filtering is complete;

[0013] The intelligent error compensation module further includes

[0014] Based on the historical distribution of sensor data, a dynamic error model is constructed. Using the constructed dynamic error model, the error type is automatically identified, and according to the identified error type, the compensation strategy data error is adaptively adjusted for positive compensation.

[0015] And the calibration feedback module further includes

[0016] By performing real-time self-calibration on the collected data at the application end, a double-layer feedback control mechanism is constructed to dynamically adjust the sampling parameters, and the filtering threshold is adaptively adjusted based on environmental changes.

[0017] As a preferred solution of the anti-interference six-dimensional sensor acquisition system described in the present invention, wherein: the adaptive dynamic weight allocation algorithm is specifically implemented as follows:

[0018] Set the confidence evaluation index, and calculate the confidence corresponding to each sensor according to the standard deviation of the data collected by the sensor ;

[0019] According to the calculated sensor confidence, perform adaptive weight allocation for the sensors, then there is

[0020]

[0021] Wherein, represents the confidence corresponding to the th sensor, represents the th sensor corresponding weight coefficient;

[0022] According to the allocated weights, perform data fusion, then there is

[0023]

[0024] Wherein, represents the th sensor corresponding weight coefficient, represents the total number of sensors, represents the th sensor collected data, represents the data after fusion of all sensors.

[0025] As a preferred solution of the anti-interference six-dimensional sensor acquisition system described in the present invention, wherein: an anomaly detection factor is introduced to realize secondary regulation of the sensor weight coefficient, specifically:

[0026] Calculate the mutation rate of the data collected within the time window, then there is

[0027]

[0028] Among them, represents the data collected by the -th sensor within the current time window, represents the data collected by the -th sensor within the previous time window, represents the anomaly detection factor of the data collected by the -th sensor, which is used to determine whether the data collected by the current sensor has an abnormal mutation. Specifically:

[0029] Set the anomaly detection factor threshold . If the anomaly detection factor corresponding to the sensor satisfies the formula when compared with the set anomaly detection factor threshold, it indicates that the data collected by the current sensor has an abnormal mutation, and the weight corresponding to the current sensor is reduced.

[0030] As a preferred solution of the six-dimensional sensor acquisition system with anti-interference described in the present invention, among them: The specific implementation of the adaptive covariance allocation algorithm is as follows:

[0031] Calculate the measurement noise variance at moments and the state estimation error covariance ;

[0032] Set the smoothing factor to implement the update of the measurement noise covariance matrix and the process noise covariance matrix. Then,

[0033]

[0034] Among them, represents the smoothing factor, represents the measurement noise variance, represents the state estimation error covariance, , respectively represent the measurement noise covariance matrix and the process noise covariance matrix at the -th moment, , respectively represent the measurement noise covariance matrix and the process noise covariance matrix at the -th moment.

[0035] As a preferred solution of the six-dimensional sensor acquisition system with anti-interference described in the present invention, among them: For the denoised data , by setting the filtering threshold , judge the fused data Whether the filtering is complete, specifically:

[0036] If the data after denoising meets the formula when compared with the filtering threshold , it indicates that the filtering of the current data is incomplete and the abnormal data is not completely filtered. The data is filtered twice by adjusting the smoothing factor until the formula is satisfied, which means that the current data filtering is complete.

[0037] As a preferred solution of the anti-interference six-dimensional sensor acquisition system described in the present invention, among them: the forward compensation for adaptively adjusting and compensating the data error is specifically implemented as follows:

[0038] Predict the error types, including the error caused by linear interference and the error caused by non-linear interference;

[0039] For the error caused by linear interference, use the direct compensation strategy to compensate for the error, then

[0040]

[0041] Among them, represents the predicted error value at the th moment, represents the sensor measurement data at the th moment, represents the compensated sensor measurement data;

[0042] For the error caused by non-linear interference, use the exponential smoothing compensation strategy to compensate for the error, then

[0043]

[0044] Among them, represents the predicted error value at the th moment, represents the sensor measurement data at the th moment, represents the sensor measurement data at the th moment, represents the compensated sensor measurement data, represents the dynamic adjustment factor.

[0045] As a preferred solution of the anti-interference six-dimensional sensor acquisition system described in the present invention, among them: the double-layer feedback control mechanism includes local feedback correction and global feedback correction, and the specific implementation is as follows:

[0046] Local feedback correction, using the sampling value of the current sensor and the mean value of the most recent historical data , calculate the data deviation ;

[0047] According to the calculated real-time deviation, determine whether sudden noise occurs, specifically:

[0048] Setting the local deviation threshold , if the calculated data deviation satisfies the formula , indicating the occurrence of burst noise, self-correcting the data, we have,

[0049]

[0050] in, represents the dynamic adjustment factor, Indicates recent The mean of historical data, Represents the sensor sampling value after self-calibration and calculates the data deviation twice , and then compare it with the local deviation threshold twice. If the comparison result satisfies the formula , it means that the self-calibration is accurate, otherwise it will be re-calibrated until the formula is satisfied until;

[0051] Global feedback correction uses a sliding window to calculate the long-term data error change trend, then we have:

[0052]

[0053] in, Indicates the current sensor sampling value, Indicates the sensor reference value at the current moment. Indicates the set sliding window size. Indicates the error within the current window size, represents the error within the previous window size, Indicates the data error change in the current window, which is used for global error compensation. Specifically:

[0054] Setting the error change threshold , if the calculated error change satisfies the formula, , indicating that the data error in the current window exceeds the threshold, by adjusting the dynamic adjustment factor in the exponential smoothing compensation strategy until the error changes to satisfy the formula until.

[0055] As a preferred solution of the anti-interference six-dimensional sensor acquisition system described in the present invention, the adaptive adjustment of the filtering threshold based on environmental changes is specifically as follows:

[0056] Ambient noise levels are based on recent The standard deviation of the data, specifically:

[0057]

[0058] Among them, represents the sampling value of the current sensor, represents the most recent mean value of historical data, represents the environmental noise level, used to dynamically adjust the filtering threshold size, then there is

[0059]

[0060] Among them, represents the set filtering threshold size, represents the smoothing factor during the filtering process, represents the environmental noise level, represents the adjusted filtering threshold, used for data filtering under the current environmental noise level.

[0061] A computer device, including a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, an anti-interference six-dimensional sensor acquisition system is implemented.

[0062] A computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, an anti-interference six-dimensional sensor acquisition system is implemented.

[0063] Advantages of the present invention:

[0064] The present invention adopts an adaptive dynamic weight allocation algorithm, dynamically adjusts the weight according to the confidence of sensor data, makes the data fusion process have stronger anti-interference ability, improves data reliability, and at the same time introduces an anomaly detection factor to identify and adjust burst noise, reducing the impact of abnormal data on the data fusion result;

[0065] Adopts an adaptive covariance matching filtering algorithm, combines Kalman filtering to dynamically adjust the measurement noise covariance matrix, improves the denoising effect, and adaptively adjusts the data smoothing process by setting a dynamic smoothing factor to ensure the optimal signal quality under different noise levels;

[0066] Adopts an LSTM-based time series error prediction model to dynamically learn the error change trend, improves the intelligence of error compensation, and adopts a linear interference compensation strategy (zero bias drift, electromagnetic interference) and a non-linear interference compensation strategy (temperature drift, low-frequency interference) for different types of errors, greatly reducing the system error;

[0067] Local feedback correction is adopted to quickly correct short-term burst noise and improve the stability of data in a short time. Global feedback correction is adopted to calculate the long-term error trend using a sliding window and correct the long-term error accumulation to improve the long-term reliability of data. Brief Description of the Drawings

[0068] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. Among them:

[0069] Figure 1 It is a schematic diagram of the overall system step structure of a six-dimensional sensor acquisition system with anti-interference of the present invention. Detailed Description of the Embodiments

[0070] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will describe the specific embodiments of the present invention in detail with reference to the drawings of the specification. Obviously, the described embodiments are some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0071] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0072] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation manner of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that excludes other embodiments.

[0073] The present invention is described in detail in conjunction with the schematic diagrams. When describing the embodiments of the present invention in detail, for the convenience of explanation, the cross-sectional views showing the device structures will be locally enlarged out of the general scale, and the schematic diagrams are only examples and should not limit the protection scope of the present invention herein. In addition, in actual production, three-dimensional spatial dimensions including length, width, and depth should be included.

[0074] At the same time, in the description of the present invention, it should be noted that the terms "first, second, or third" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance.

[0075] Unless otherwise clearly specified and defined in the present invention, the terms "installation, connection, and coupling" shall be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may also be a mechanical connection, an electrical connection, or a direct connection, and may also be indirectly connected through an intermediate medium, or may be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0076] Example 1: Refer to Figure 1 , which is an embodiment of the present invention, and provides an anti-interference six-dimensional sensor acquisition system, including a multi-modal signal fusion module, an intelligent error compensation module, and a calibration feedback module;

[0077] Specifically, the multi-modal signal fusion module is used to receive the original acquisition data of the six-dimensional sensor and realize the fusion of the acquisition data of all sensors based on the adaptive dynamic weight allocation algorithm; the intelligent error compensation module is used to perform positive compensation on the errors in the six-dimensional sensor data to improve the accuracy of the data; the calibration feedback module is used to perform self-calibration on the data acquired by the six-dimensional sensor at the application end and perform feedback adjustment on the sampling parameters according to the calibration result.

[0078] Furthermore, the multi-modal signal fusion module is used to receive the original acquisition data of the six-dimensional sensor and realize the fusion of the acquisition data of all sensors based on the adaptive dynamic weight allocation algorithm. At the same time, it realizes data denoising based on the adaptive covariance allocation algorithm to ensure the accuracy of the acquired data.

[0079] Furthermore, the adaptive dynamic weight allocation algorithm adaptively adjusts its contribution weight in the data fusion process based on the real-time stability and reliability of each sensor. The specific implementation is as follows:

[0080] Before data fusion, a confidence evaluation index is set, which represents the reliability of the data acquired by the th sensor. Specifically:

[0081]

[0082] Among them, represents the standard deviation of the data acquired by the th sensor, which is used to reflect the stability of the data acquired by the sensor. The smaller the standard deviation, the more stable the data and the higher the confidence. On the contrary, it means that the data is unstable and the confidence is low. represents the confidence corresponding to the th sensor;

[0083] Based on the calculated confidence, an adaptive weight allocation mechanism is constructed as follows:

[0084]

[0085] Among them, represents the confidence corresponding to the -th sensor, represents the weight coefficient corresponding to the -th sensor, which is used to realize the fusion of data collected by six-dimensional sensors. Then,

[0086]

[0087] Among them, represents the weight coefficient corresponding to the -th sensor, represents the total number of sensors, represents the data collected by the -th sensor, represents the data after fusion of all sensors.

[0088] It should be noted that an anomaly detection factor is introduced simultaneously to realize the secondary regulation of the sensor weight coefficient, specifically as follows:

[0089] The time of the data collected by the sensor is divided into multiple time windows with the same time interval length. According to the mutation rate of the data collected within the time window, the secondary regulation of the weight coefficient of the corresponding sensor during data fusion is realized. Then,

[0090]

[0091] Among them, represents the data collected by the -th sensor within the current time window, represents the data collected by the -th sensor in the previous time window, The anomaly detection factor of the data collected by the -th sensor is used to judge whether the data collected by the current sensor has an abnormal mutation. Specifically as follows:

[0092] Set the anomaly detection factor threshold . If the anomaly detection factor corresponding to the sensor satisfies the formula when compared with the set anomaly detection factor threshold, it means that the data collected by the current sensor has an abnormal mutation, and the weight corresponding to the current sensor is reduced.

[0093] Further, the adaptive covariance allocation algorithm is for the Kalman filtering algorithm. According to the noise characteristics of the sensor, it dynamically adjusts the process noise covariance matrix and the measurement noise covariance matrix in the denoising process of the Kalman filtering algorithm to improve the data denoising effect. The specific implementation is as follows:

[0094] Set a sliding window , and calculate respectively the measurement noise variance and the state estimation error covariance at

[0095] measurement noise variance,

[0096]

[0097] where, represents the measurement data at the -th moment, represents the lower time limit within the current sliding window, represents the mean value of the measurement data within the sliding window, represents the measurement noise variance, which is used to update the measurement noise covariance matrix;

[0098] state estimation error covariance,

[0099]

[0100] where, represents the state estimation value at the -th moment, the mean value of the state estimation within the sliding window, represents the state estimation error covariance, which is used to update the process noise covariance matrix;

[0101] Update the measurement noise covariance matrix and the process noise covariance matrix, specifically:

[0102] Set a smoothing factor , and implement the update of the measurement noise covariance matrix and the process noise covariance matrix, then there is,

[0103]

[0104] where, represents the smoothing factor, which is set by the implementer according to the actual application scenario, represents the measurement noise variance, represents the state estimation error covariance, , respectively represent the measurement noise covariance matrix and the process noise covariance matrix at the -th moment, , respectively represent the measurement noise covariance matrix and the process noise covariance matrix at the th moment, and are the adjusted measurement noise covariance matrix and the process noise covariance matrix.

[0105] It should be noted that for the denoised data , by setting the filtering threshold , it is determined whether the fused data is completely filtered. Specifically:

[0106] If the denoised data compared with the filtering threshold satisfies the formula , it means that the current data is not completely filtered and the abnormal data is not completely filtered. The data is filtered twice by adjusting the smoothing factor until the formula is satisfied, then it means that the current data is completely filtered.

[0107] It should be noted that through the adaptive dynamic weight allocation algorithm, when a certain sensor is interfered, its contribution weight is automatically reduced to avoid the influence of wrong data on the overall system. Combining with the Kalman filter of adaptive covariance matching, the filtering parameters can be adaptively adjusted so that the system can obtain the optimal data fusion effect in different noise environments. Through the adaptive covariance matching filter, the system can dynamically adjust the filtering strategy according to different external interference situations to improve the adaptability.

[0108] Further, the intelligent error compensation module is used to perform positive compensation on the errors in the six-dimensional sensor data to improve the accuracy of the data. It constructs a dynamic error model based on the historical distribution of the sensor data, uses the constructed dynamic error model to automatically identify the error type, and dynamically adjusts the compensation strategy.

[0109] Further, the dynamic error model is a time series prediction model based on LSTM, which learns the dynamic change trend of the error in real time and adjusts the adaptive compensation strategy. The specific implementation is as follows:

[0110] According to the sensor measurement data and the true reference value , the historical error is calculated. Then,

[0111]

[0112] where represents the calculated historical error value, represents the sensor measurement data, represents the true reference value;

[0113] Calculate the error values at the past moments, and use the calculation results as the input data of the time series prediction model , and the input data satisfies the formula ;

[0114] Based on the input data, predict the future error value, then

[0115]

[0116] where represents the set of predicted error values, which satisfies the formula , and is as the prediction result of the error value of the input data, represents the LSTM network, including an input layer, an LSTM layer, a fully connected layer, and a loss function. The LSTM network is a well-known technology that is easily associated by those skilled in the art and will not be elaborated in this embodiment;

[0117] Based on the set of predicted error values , including the error caused by linear interference, including sensor zero bias drift and electromagnetic interference, and the error caused by non-linear interference, including temperature offset and low-frequency interference, adaptively adjust the compensation strategy according to the error type, specifically:

[0118] When the LSTM network determines that the error type is the error caused by linear interference, adopt a direct compensation strategy for error compensation, then

[0119]

[0120] where represents the error value predicted at the th moment, represents the sensor measurement data at the th moment, represents the compensated sensor measurement data;

[0121] When the LSTM network determines that the error type is the error caused by non-linear interference, adopt an exponential smoothing compensation strategy for error compensation, then

[0122]

[0123] where represents the error value predicted at the th moment, represents the sensor measurement data at the th moment, represents the sensor measurement data at the th moment, represents the compensated sensor measurement data, represents the dynamic adjustment factor, which is set by the implementer according to the actual application scenario.

[0124] It should be noted that by predicting the error with LSTM and combining multiple compensation strategies, the anti-interference ability and accuracy of six-dimensional sensor data are effectively improved, which is particularly suitable for precision measurement and control systems in complex environments.

[0125] Furthermore, the calibration feedback module is used to perform self-calibration on the data collected by the six-dimensional sensor at the application end and adjust the sampling parameters according to the calibration results. It achieves this by performing real-time self-calibration on the collected data at the application end, constructing a two-layer feedback control mechanism to dynamically adjust the sampling parameters, and adaptively adjusting the filtering threshold based on environmental changes to improve the data stability and accuracy of the entire system in complex environments.

[0126] Furthermore, the two-layer feedback control mechanism introduces two-layer feedback control to dynamically adjust the sampling parameters of the sensor, improving the system's adaptability to different environments. It includes local feedback calibration and global feedback calibration, and the specific implementation is as follows:

[0127] Local feedback calibration is used to detect and suppress the interference of burst noise for short-term calibration. Specifically:

[0128] Using the sampling value of the current sensor and the mean value of the nearest historical data , calculate the data deviation

[0129] According to the calculated real-time deviation, determine whether burst noise occurs. Specifically:

[0130] Set the local deviation threshold , if the calculated data deviation satisfies the formula , it indicates that burst noise appears, and the data is self-calibrated. Then,

[0131]

[0132] where represents the dynamic adjustment factor, represents the mean value of the nearest data, represents the sensor sampling value after self-calibration, and the data deviation is calculated again and compared with the local deviation threshold for the second time. If the comparison result satisfies the formula , it indicates that the self-calibration is accurate; otherwise, the self-calibration is performed again until the formula

[0133] Global feedback calibration is used to detect and suppress non-linear interference for long-term calibration. Specifically:

[0134] Using a sliding window to calculate the long-term data error change trend, we have

[0135]

[0136] where represents the sampling value of the current sensor, represents the sensor reference value at the current moment, which is set by the implementer according to the actual application scenario, represents the size of the set sliding window, which is set by the implementer according to the actual application scenario, represents the error within the current window size, represents the error within the previous window size, represents the data error change within the current window, which is used for global error compensation. Specifically:

[0137] Set the error change threshold , if the calculated error change satisfies the formula , it means that the data error within the current window exceeds the threshold. By adjusting the size of the dynamic adjustment factor in the exponential smoothing compensation strategy until the error change satisfies the formula .

[0138] Furthermore, the adaptive adjustment of the filtering threshold based on environmental changes is achieved by calculating the environmental noise level and dynamically adjusting the size of the filtering threshold to achieve signal quality changes in different environments and avoid over-filtering or false detection. The specific implementation is as follows:

[0139] The environmental noise level is reflected by the standard deviation of the most recent data. Specifically:

[0140]

[0141] where represents the sampling value of the current sensor, represents the mean of the most recent historical data, represents the environmental noise level, which is used to dynamically adjust the size of the filtering threshold , then we have

[0142]

[0143] where represents the set size of the filtering threshold, represents the smoothing factor during the filtering process, represents the environmental noise level, Represents the adjusted filtering threshold for data filtering under the current ambient noise level.

[0144] It should be noted that by introducing local feedback (short-term correction) + global feedback (long-term trend correction), the sampling parameters of the sensor are dynamically adjusted to improve the system's adaptability to different environments; based on the real-time calculated ambient noise level, the filtering threshold is dynamically adjusted to enable the system to maintain the optimal sampling accuracy under different interference conditions.

[0145] Furthermore, if the described function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, 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 system described in various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical disks, etc., which can store program codes.

[0146] The logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a defined sequence list of executable instructions for implementing logical functions, which can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or used in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device.

[0147] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection (electronic device) having one or more wirings, a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable media can even be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other media, then editing, interpreting, or otherwise processing it as appropriate, and then storing it in a computer memory.

[0148] Moreover, to provide a concise description of the exemplary embodiments, all features of the actual embodiments may not be described (i.e., those features that are not relevant to the currently contemplated best mode of carrying out the invention or those features that are not relevant to the implementation of the invention).

[0149] It should be understood that in the development of any actual implementation, as in any engineering or design project, numerous specific implementation decisions may be made. Such development efforts may be complex and time-consuming, but for those of ordinary skill in the art who benefit from this disclosure, without undue experimentation, such development efforts would be a routine task of design, fabrication, and production.

[0150] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. An anti-interference six-dimensional sensor acquisition system, characterized in that: include, The multimodal signal fusion module also includes: For the original data collected by the six-dimensional sensor, the adaptive dynamic weight allocation algorithm is used to adaptively allocate weights to different sensors, and the data collected by all sensors are fused according to the allocated weights. At the same time, the adaptive covariance allocation algorithm is used to denoise the fused data, and the filtering threshold is set to determine whether the data is completely filtered. Intelligent error compensation module, also includes, Based on the historical distribution of sensor data, a dynamic error model is constructed. The error type is automatically identified using the constructed dynamic error model. According to the identified error type, the compensation strategy is adaptively adjusted to achieve positive compensation of the error. The adaptive adjustment compensation strategy to achieve forward compensation of errors is specifically implemented as follows: The types of prediction errors include errors caused by linear disturbances and errors caused by nonlinear disturbances; For the error caused by linear interference, the direct compensation strategy is used to compensate for the error, then, in, Indicates The error value of the prediction at the moment, Indicates The sensor measures data at each moment. Represents the compensated sensor measurement data; For the error caused by nonlinear interference, the exponential smoothing compensation strategy is used to compensate for the error, then, in, Indicates The error value of the prediction at the moment, Indicates The sensor measures data at each moment. Indicates The sensor measures data at each moment. represents the compensated sensor measurement data, represents the dynamic adjustment factor; And the correction feedback module also includes, By performing real-time self-correction on the collected data at the application end, and building a two-layer feedback control mechanism to dynamically adjust the sampling parameters, the filtering threshold is adaptively adjusted based on environmental changes.

2. The anti-interference six-dimensional sensor acquisition system according to claim 1, characterized in that: The adaptive dynamic weight allocation algorithm is specifically implemented as follows: Set the confidence evaluation index and calculate the confidence corresponding to each sensor based on the standard deviation of the data collected by the sensor ; According to the calculated sensor confidence, the adaptive weight allocation of the sensor is performed, then, in, Indicates The confidence level corresponding to each sensor is Indicates The weight coefficient corresponding to each sensor; According to the assigned weights, data fusion is performed, then we have: in, Indicates The weight coefficient corresponding to each sensor is: Represents the total number of sensors, Indicates The data collected by the sensors, Represents the data after all sensors are fused.

3. The anti-interference six-dimensional sensor acquisition system according to claim 2, characterized in that: The anomaly detection factor is introduced to achieve secondary regulation of the sensor weight coefficient, specifically: Calculate the mutation rate of the data collected in the time window, then we have, in, Indicates The data collected by each sensor in the current time window is Indicates The data collected by the sensor in the previous time window, Indicates The anomaly detection factor of the data collected by the sensor is used to determine whether the data collected by the current sensor has an abnormal mutation. Specifically: Setting anomaly detection factor threshold , if the abnormal detection factor corresponding to the sensor is compared with the set abnormal detection factor threshold and satisfies the formula , indicating that the data collected by the current sensor has an abnormal mutation, reducing the corresponding weight of the current sensor.

4. The anti-interference six-dimensional sensor acquisition system according to claim 3, characterized in that: The adaptive covariance allocation algorithm is specifically implemented as follows: calculate The measurement noise variance at each moment and the state estimation error covariance ; Set the smoothing factor , to update the measurement noise covariance matrix and the process noise covariance matrix, we have, in, represents the smoothing factor, represents the measurement noise variance, represents the state estimation error covariance, , Respectively represent The measurement noise covariance matrix and process noise covariance matrix at each moment, , Respectively represent The measurement noise covariance matrix and process noise covariance matrix at each moment.

5. The anti-interference six-dimensional sensor acquisition system according to claim 4, characterized in that: For the denoised data , by setting the filtering threshold , judge the fused data Whether filtering is complete, specifically: If the denoised data meets the filter threshold, the formula , indicating that the current data is not completely filtered, and the abnormal data is not completely filtered. The data is filtered again by adjusting the smoothing factor until the formula is satisfied. , it means that the current data is completely filtered.

6. The anti-interference six-dimensional sensor acquisition system according to claim 5, characterized in that: The dual-layer feedback control mechanism includes local feedback correction and global feedback correction, which is specifically implemented as follows: Local feedback correction, using the current sensor sampling value and recently The average of historical data , calculate the data deviation ; According to the calculated real-time deviation, determine whether sudden noise occurs, specifically: Setting the local deviation threshold , if the calculated data deviation satisfies the formula , indicating the occurrence of burst noise, self-correcting the data, we have, in, represents the dynamic adjustment factor, Indicates recent The mean of historical data, Represents the sensor sampling value after self-calibration and calculates the data deviation twice , and then compare it with the local deviation threshold twice. If the comparison result satisfies the formula , it means that the self-calibration is accurate, otherwise it will be re-calibrated until the formula is satisfied until; Global feedback correction uses a sliding window to calculate the long-term data error change trend, then we have: in, Indicates the current sensor sampling value, Indicates the sensor reference value at the current moment. Indicates the set sliding window size. Indicates the error within the current window size, represents the error within the previous window size, Indicates the data error change in the current window, which is used for global error compensation. Specifically: Setting the error change threshold , if the calculated error change satisfies the formula , indicating that the data error in the current window exceeds the threshold, by adjusting the dynamic adjustment factor in the exponential smoothing compensation strategy until the error changes to satisfy the formula until.

7. The anti-interference six-dimensional sensor acquisition system according to claim 6, characterized in that: The adaptive adjustment of the filtering threshold based on environmental changes is specifically as follows: Ambient noise levels are based on recent The standard deviation of the data is shown as follows: in, Indicates the current sensor sampling value, Indicates recent The mean of historical data, Indicates the ambient noise level, used to dynamically adjust the filtering threshold The size of , then, in, Indicates the set filtering threshold size. represents the smoothing factor in the filtering process, Indicates the ambient noise level, Represents the adjusted filtering threshold used for data filtering under the current environmental noise level.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the system according to any one of claims 1 to 7 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the system according to any one of claims 1 to 7 is implemented.

Images