Unmanned aerial vehicle anti-interference control device and control method

Abstract

The present application relates to the technical field of unmanned aerial vehicle anti-interference, and discloses an unmanned aerial vehicle anti-interference control device and control method, comprising: the data module is used for acquiring a plurality of groups of parameters of a navigation unit, a plurality of groups of parameters of a communication unit and a plurality of groups of parameters of an environment perception unit of the unmanned aerial vehicle in real time; the matrix module is used for establishing a navigation feature matrix through the plurality of groups of parameters of the navigation unit, establishing a communication feature matrix through the plurality of groups of parameters of the communication unit and establishing an environment feature matrix through the plurality of groups of parameters of the environment perception unit; the calculation module is used for establishing a matrix fusion unit; the navigation feature matrix, the communication feature matrix and the environment feature matrix are fused into a comprehensive state matrix through the matrix fusion unit; and a singular value is calculated and output from the comprehensive state matrix as an unmanned aerial vehicle anti-interference decision characteristic value, so that high-precision multi-modal intelligent anti-interference of the unmanned aerial vehicle is realized.

Description

Technical Field

[0001] This invention relates to the technical field of anti-interference for unmanned aerial vehicles (UAVs), and discloses an anti-interference control device and control method for UAVs. Background Technology

[0002] With the widespread application of unmanned aerial vehicles (UAVs) in surveying, logistics, reconnaissance, and other fields, the electromagnetic spectrum of their flight environment is becoming increasingly complex, and the interference threats they face are becoming increasingly severe. Existing anti-interference solutions mostly focus on single-dimensional interference suppression. Single-dimensional anti-interference technologies have many inherent limitations, such as being unable to effectively deal with coordinated or compound interference occurring simultaneously across multiple subsystems such as navigation, communication, and sensing; decision-making mechanisms based on a single information source have a high misjudgment rate, and are prone to affecting normal mission execution due to accidental triggering in interference edge scenarios; existing solutions lack a holistic and quantitative assessment of the interference situation, resulting in single and rigid anti-interference strategies that cannot adapt and refine responses to continuous changes in interference intensity. Summary of the Invention

[0003] The purpose of this section is to outline some aspects of embodiments of the present invention and to briefly describe some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be construed as limiting the scope of the invention.

[0004] To address the aforementioned technical problems, this invention provides an anti-interference control device and control method for unmanned aerial vehicles.

[0005] On one hand, the present invention provides an anti-interference control device for unmanned aerial vehicles, comprising a data module, a matrix module, a calculation module, and a decision module; wherein:

[0006] The data module includes a navigation unit, a communication unit, and an environmental perception unit. The data module acquires multiple sets of parameters from the navigation unit, the communication unit, and the environmental perception unit of the unmanned aerial vehicle in real time.

[0007] The matrix module includes a navigation feature matrix, a communication feature matrix, and an environmental feature matrix. The navigation feature matrix is ​​established using multiple sets of parameters from the navigation unit, the communication feature matrix is ​​established using multiple sets of parameters from the communication unit, and the environmental feature matrix is ​​established using multiple sets of parameters from the environmental perception unit.

[0008] The calculation module is used to establish a matrix fusion unit, which fuses the navigation feature matrix, the communication feature matrix, and the environmental feature matrix into a comprehensive state matrix. The comprehensive state matrix is ​​used to calculate and output singular values ​​as anti-interference decision feature values ​​for the unmanned aerial vehicle.

[0009] The decision module receives the anti-interference decision feature value and outputs interference control instructions.

[0010] As a preferred embodiment of the anti-interference control device for unmanned aerial vehicles of the present invention, wherein:

[0011] The navigation unit has multiple sets of parameters, including carrier noise power density ratio, number of visible satellites, and positioning accuracy attenuation factor.

[0012] The communication unit has multiple sets of parameters, including received signal strength indication, signal-to-noise ratio, and channel estimation value;

[0013] The environmental perception unit has multiple parameters, including the average interference energy and the image entropy calculated from images acquired by the visual sensor.

[0014] As a preferred embodiment of the anti-interference control device for unmanned aerial vehicles of the present invention, wherein:

[0015] The numerical sequences of carrier noise power density ratio, number of visible satellites, and positioning accuracy attenuation factor at N consecutive sampling times are used to calculate the navigation Pearson correlation coefficients between each pair of carrier noise power density ratio, number of visible satellites, and positioning accuracy attenuation factor.

[0016] The Pearson correlation coefficients are constructed into a navigation feature matrix with Q rows and Q columns;

[0017] The off-diagonal elements of the navigation feature matrix This is used to represent the dynamic correlation strength and direction between the i-th navigation parameter and the j-th navigation parameter within a time window.

[0018] As a preferred embodiment of the anti-interference control device for unmanned aerial vehicles of the present invention, wherein:

[0019] Received signal strength indication, signal-to-noise ratio and channel estimate are measured and obtained on T different communication sub-channels. The communication Pearson correlation coefficient between each pair of signal strength indication, signal-to-noise ratio and channel estimate is calculated.

[0020] The communication Pearson correlation coefficients are constructed into a communication feature matrix with Q rows and Q columns;

[0021] The off-diagonal elements of the communication feature matrix This is used to represent the distributional correlation characteristics of the i-th communication parameter and the j-th communication parameter in the frequency domain spatial dimension.

[0022] As a preferred embodiment of the anti-interference control device for unmanned aerial vehicles of the present invention, wherein:

[0023] The mean interference energy and image entropy at P consecutive sampling times are standardized respectively;

[0024] The standardized mean sequence of interference energy and the image entropy sequence are used to construct a feature matrix with Z rows and X columns;

[0025] Each row of the environmental feature matrix represents a standardized environmental parameter time-series change, and each column represents a standardized state vector of multi-source environmental sensing parameters at a given time.

[0026] As a preferred embodiment of the anti-interference control device for unmanned aerial vehicles of the present invention, wherein:

[0027] The matrix fusion unit is used to generate a comprehensive state matrix by weighted direct sum fusion of the navigation feature matrix, communication feature matrix and environmental feature matrix;

[0028] The matrix fusion unit also assigns dynamic adaptive weights to the navigation feature matrix, communication feature matrix, and environmental feature matrix by real-time reliability assessment of the navigation feature matrix, communication feature matrix, and environmental feature matrix, and performs weighted fusion after the dynamic adaptive weight allocation is completed.

[0029] The method for determining the dynamic adaptive weights includes:

[0030] The stability of the navigation feature matrix, communication feature matrix, and environmental feature matrix within the most recent time window is used to calculate the initial confidence weights.

[0031] The attitude stability data from the UAV flight control is used as an external verification factor to correct the initial confidence weight;

[0032] The data stability is quantified by calculating the inverse variance of the main diagonal elements of the navigation feature matrix, communication feature matrix, and environmental feature matrix.

[0033] As a preferred embodiment of the anti-interference control device for unmanned aerial vehicles of the present invention, wherein:

[0034] The matrix fusion unit establishes a block-diagonal comprehensive state matrix by using the navigation feature matrix, communication feature matrix and environmental feature matrix as diagonal sub-blocks;

[0035] The calculation module further includes a singular value decomposition unit, which is used to decompose the singular values ​​of the block diagonal form of the comprehensive state matrix to obtain a singular value sequence.

[0036] The singular value decomposition unit is used to extract the maximum singular value from the singular value sequence obtained after decomposition, and output the maximum singular value as an anti-interference decision feature value that characterizes the overall degree of interference of the system.

[0037] As a preferred embodiment of the anti-interference control device for unmanned aerial vehicles of the present invention, the calculation module processes the comprehensive state matrix through singular value decomposition and extracts its maximum singular value as the anti-interference decision feature value.

[0038] The decision feature value serves as a continuous, quantified indicator, which is positively correlated with the overall interference intensity experienced by the unmanned aerial vehicle.

[0039] As a preferred embodiment of the anti-interference control device for unmanned aerial vehicles of the present invention, the anti-interference decision feature value is compared with a plurality of preset interference threshold intervals;

[0040] Based on the interference threshold range into which the feature value falls, select the corresponding combination of control commands from the predefined anti-interference strategy mapping table;

[0041] The control command combination is used to schedule one or more execution subsystems, including a communication unit, a navigation unit, and a flight control unit, to perform coordinated anti-interference actions.

[0042] This invention provides an anti-interference control method for unmanned aerial vehicles, comprising:

[0043] S1. Real-time acquisition of multiple sets of parameters from the navigation unit, communication unit, and environmental perception unit of the unmanned aerial vehicle;

[0044] S2. Construct navigation feature matrix, communication feature matrix and environment feature matrix respectively using the multiple sets of parameters;

[0045] S3. Establish a matrix fusion unit, and fuse the navigation feature matrix, the communication feature matrix, and the environmental feature matrix into a comprehensive state matrix through the matrix fusion unit;

[0046] S4. Calculate the singular values ​​of the integrated state matrix, and use the singular values ​​as the anti-interference decision characteristic values ​​of the unmanned aerial vehicle;

[0047] S5. Output the anti-interference strategy through the anti-interference decision feature value and generate control commands to send to the execution unit.

[0048] The beneficial effects of this invention are as follows:

[0049] This application acquires multiple sets of parameters from navigation, communication, and environmental perception units in real time through a data module, constructs corresponding feature matrices, and finally integrates them into a comprehensive state matrix and extracts decision feature values. This enables a multi-dimensional, global, and quantitative assessment of the interference experienced by unmanned aerial vehicles, overcoming the one-sidedness of single-dimensional perception. It can detect complex interference, especially cooperative and compound interference, earlier and more comprehensively.

[0050] This application uses the anti-interference decision feature value obtained by the singular value decomposition of the comprehensive state matrix to accurately reflect the comprehensive intensity of the interference. The decision unit performs threshold judgment and strategy mapping, which greatly reduces the misjudgment rate.

[0051] This invention enables unmanned aerial vehicles to adaptively adjust their anti-interference strategies based on the dynamic changes in the external interference environment through multimodal perception and processing, thus ensuring flight safety and mission continuity in variable interference environments. Attached Figure Description

[0052] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0053] Figure 1 A schematic diagram of the anti-interference control device for unmanned aerial vehicles provided by the present invention;

[0054] Figure 2 Flowchart of the anti-interference control method for unmanned aerial vehicles provided by the present invention;

[0055] Figure 3 The matrix fusion unit and dynamic adaptive weight allocation method of the anti-interference control device for unmanned aerial vehicles provided by the present invention are shown in the figure.

[0056] Figure 4 This invention provides a comprehensive state matrix representation of the anti-interference control device for unmanned aerial vehicles. Detailed Implementation

[0057] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0058] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0059] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.

[0060] Example 1

[0061] like Figure 1 As shown, the anti-interference control device for unmanned aerial vehicles includes a data module, a matrix module, a calculation module, and a decision module; wherein:

[0062] The data module includes a navigation unit, a communication unit, and an environmental perception unit. The data module acquires multiple sets of parameters from the navigation unit, the communication unit, and the environmental perception unit of the unmanned aerial vehicle in real time.

[0063] The navigation unit has multiple sets of parameters, including carrier noise power density ratio, number of visible satellites, and positioning accuracy attenuation factor.

[0064] Specifically, the carrier-to-noise power density ratio is directly provided by the airborne global navigation satellite receiver, and the comparison between the received satellite signal strength and the background noise is used to determine whether the navigation signal is subject to suppression interference.

[0065] The number of visible satellites is provided by the GNSS receiver and indicates the number of satellites that can currently be locked onto and used to calculate their positions.

[0066] The positioning accuracy attenuation factor is used to reflect the degree of influence of the spatial geometric distribution of currently visible satellites on positioning accuracy. Deceptive interference from abnormal environments can cause abnormal changes in the positioning accuracy attenuation factor.

[0067] The communication unit has multiple sets of parameters, including received signal strength indication, signal-to-noise ratio, and channel estimation value;

[0068] Specifically, the received signal strength indication is provided by the airborne transceiver and is used to reflect the power of the received signal from the ground station or other communication nodes.

[0069] The signal-to-noise ratio (SNR) is the ratio of the received signal strength to the background noise strength.

[0070] Channel estimates are obtained by probing and analyzing the communication channel.

[0071] The environmental perception unit has multiple parameters, including the average interference energy and the image entropy calculated from images acquired by the visual sensor.

[0072] Specifically, the average interference energy is obtained by the airborne spectrum sensing device performing energy detection on preset key anti-interference frequency bands such as the GNSS frequency band and the remote control image transmission frequency band.

[0073] Image entropy is calculated from image data acquired by airborne vision sensors. Image entropy is used as an indicator to measure the complexity and information content of an image texture. When the vision sensor is subjected to strong light interference or electromagnetic interference, causing abnormal noise in the image sensor, the image entropy will change significantly.

[0074] The matrix module includes a navigation feature matrix, a communication feature matrix, and an environmental feature matrix. The navigation feature matrix is ​​established using multiple sets of parameters from the navigation unit, the communication feature matrix is ​​established using multiple sets of parameters from the communication unit, and the environmental feature matrix is ​​established using multiple sets of parameters from the environmental perception unit.

[0075] The numerical sequences of carrier noise power density ratio, number of visible satellites, and positioning accuracy attenuation factor at N consecutive sampling times are used to calculate the navigation Pearson correlation coefficients between each pair of carrier noise power density ratio, number of visible satellites, and positioning accuracy attenuation factor.

[0076] The Pearson correlation coefficients are constructed into a navigation feature matrix with Q rows and Q columns;

[0077] The off-diagonal elements of the navigation feature matrix This is used to represent the dynamic correlation strength and direction between the i-th navigation parameter and the j-th navigation parameter within a time window.

[0078] The navigation parameter correlation coefficient matrix is ​​used to identify interference by monitoring abnormal changes in the inherent correlation between parameters within the navigation subsystem.

[0079] Specifically, the navigation feature matrix is ​​a navigation parameter correlation coefficient matrix, and its construction method includes: data preprocessing and time window delineation, correlation coefficient calculation and matrix construction.

[0080] Furthermore, data preprocessing and time window delineation include: the system sets a sliding time window containing N consecutive sampling times. The length N of the sliding time window needs to be optimized based on the dynamic response time of the unmanned aerial vehicle and the duration of the interference to ensure that dynamic changes can be captured without introducing excessive computational delay.

[0081] Within the sliding time window, the values ​​of carrier noise power density ratio, number of visible satellites, and positioning accuracy attenuation factor corresponding to each moment are extracted to form three time series of length N.

[0082] Furthermore, the correlation coefficient calculation and matrix construction include: the system calculates the Pearson correlation coefficient between every two time series of three time series of length N.

[0083] Furthermore, the Pearson correlation coefficient includes the correlation coefficient between the carrier noise power density ratio and the number of visible satellites.

[0084] The correlation coefficient between carrier noise power density ratio and positioning accuracy attenuation factor.

[0085] The correlation coefficient between the number of satellites and the positioning accuracy attenuation factor can be observed.

[0086] The three calculated correlation coefficients, along with the correlation coefficient between each parameter and itself, are filled into a symmetric matrix of row Q and column Q to form the navigation feature matrix. The diagonal elements of this navigation feature matrix are always 1, and the off-diagonal elements are the correlation coefficients calculated above.

[0087] It should be noted that those skilled in the art can use the technical content disclosed in this application, combined with the well-known Pearson correlation coefficient calculation method, to calculate the correlation coefficient between the carrier noise power density ratio and the number of visible satellites, the correlation coefficient between the carrier noise power density ratio and the positioning accuracy attenuation factor, and the correlation coefficient between the number of visible satellites and the positioning accuracy attenuation factor.

[0088] In this application, a preferred implementation method for a navigation feature matrix includes:

[0089] The rows and columns of the navigation feature matrix represent the following: each row and column corresponds to one of the three navigation parameters: [carrier noise power density ratio, number of visible satellites, and positioning accuracy attenuation factor]. Therefore, the value and sign of the off-diagonal element in the i-th row and j-th column of the matrix represent the dynamic correlation strength and direction between the i-th and j-th parameters within the time window of the N time points.

[0090] Furthermore, under normal flight conditions with no interference or only slight noise, there is a physical correlation between the carrier noise power density ratio, the number of visible satellites, and the positioning accuracy attenuation factor. For example, when a new satellite is captured by the UAV, the number of visible satellites increases, which leads to a decrease in the positioning accuracy attenuation factor, showing a negative correlation between the two. At the same time, the carrier noise power density ratio fluctuates within the normal range, maintaining a relatively stable relationship with the number of visible satellites and the positioning accuracy attenuation factor.

[0091] When a drone is subjected to specific interference, the physical correlation between the carrier noise power density ratio, the number of visible satellites, and the positioning accuracy attenuation factor is disrupted. For example, spoofing interference may inject a large number of false satellite signals, leading to an abnormal increase in the number of visible satellites. At the same time, due to the poor geometric distribution of false satellites, the positioning accuracy attenuation factor increases, weakening, reversing, or making the original negative correlation weaker, reversed, or irregular.

[0092] Compared to traditional single-parameter thresholding methods, identifying interference by monitoring changes in the off-diagonal elements of the correlation coefficient matrix offers higher reliability and stronger robustness. By detecting anomalies in relational patterns rather than absolute numerical anomalies in individual parameters, false alarms are effectively reduced.

[0093] Received signal strength indication, signal-to-noise ratio and channel estimate are measured and obtained on T different communication sub-channels. The communication Pearson correlation coefficient between each pair of signal strength indication, signal-to-noise ratio and channel estimate is calculated.

[0094] The communication Pearson correlation coefficients are constructed into a communication feature matrix with Q rows and Q columns;

[0095] The off-diagonal elements of the communication feature matrix This is used to represent the distributional correlation characteristics of the i-th communication parameter and the j-th communication parameter in the frequency domain spatial dimension.

[0096] The correlation coefficient matrix of communication parameters is used to evaluate the spectral characteristics of interference by analyzing the cooperative variation patterns among different channel quality parameters.

[0097] Specifically, the construction of the communication feature matrix includes: synchronous acquisition of multi-channel data and calculation of correlation coefficients and matrix construction.

[0098] Furthermore, in this application, a preferred method for multi-channel data synchronization acquisition includes: simultaneously measuring or estimating received signal strength indication, signal-to-noise ratio, and channel estimation values ​​on T different communication sub-channels, for example, different operating frequencies or spatial streams;

[0099] For each communication sub-channel, a parameter vector containing signal strength indication, signal-to-noise ratio, and channel estimate can be obtained;

[0100] The system obtains T parameter vectors, each vector representing a communication quality profile of a sub-channel.

[0101] Furthermore, in this application, a preferred method for calculating the correlation coefficient and constructing the matrix includes:

[0102] The system treats T sub-channels as observation samples and calculates the communication Pearson correlation coefficients between each pair of these T samples for the signal strength indication, signal-to-noise ratio, and channel estimate.

[0103] Specifically, the Pearson correlation coefficient for communication includes the correlation coefficient between the received signal strength indication and the signal-to-noise ratio.

[0104] The correlation coefficient between the received signal strength indication and the channel estimate.

[0105] The correlation coefficient between the signal-to-noise ratio and the channel estimate.

[0106] The correlation coefficients between the received signal strength indication and the signal-to-noise ratio, the correlation coefficients between the received signal strength indication and the channel estimate, and the correlation coefficients between the signal-to-noise ratio and the channel estimate, along with the correlation coefficients of each parameter with itself, are filled into a Q-row, Q-column symmetric matrix to form the communication feature matrix. The diagonal elements of the communication feature matrix are always 1, and the off-diagonal elements are the calculated correlation coefficients.

[0107] The meaning of the rows and columns of the communication feature matrix includes: the rows and columns of the communication feature matrix correspond to the three communication quality parameters [received signal strength indication, signal-to-noise ratio, and channel estimate] respectively.

[0108] The non-diagonal element in the i-th row and j-th column of the communication feature matrix, with its magnitude and sign, precisely represents the correlation characteristics of the numerical distribution of the i-th parameter and the j-th parameter in the frequency domain space dimension composed of T sub-channels. This is used to reflect whether different communication quality indicators show consistency in their coordinated changes across different channels.

[0109] In interference-free or uniform background noise communication environments for unmanned aerial vehicles (UAVs), the communication quality parameters of each sub-channel are mainly affected by the characteristics of the equipment itself and general path loss. The distribution of different quality parameters in the frequency domain exhibits stable correlation or independence. For example, received signal strength indication (RSI) and signal-to-noise ratio (SNR) may show a strong positive correlation on different channels because channels with strong signals usually also have high SNR.

[0110] When unmanned aerial vehicles are subjected to certain types of interference, this normal distribution correlation pattern can be disrupted, such as narrowband interference and broadband interference.

[0111] Narrowband interference only severely affects individual or a small number of consecutive sub-channels. The received signal strength on the interfered channel increases abnormally, but the signal-to-noise ratio deteriorates sharply. This causes the relationship between these two parameters to become abnormal on the interfered and uninterrupted channels, thereby changing the overall correlation coefficient.

[0112] If the interference affects all or most of the sub-channels uniformly, the overall correlation pattern between parameters may not differ much from the normal state, but the absolute values ​​of each parameter will deteriorate.

[0113] By analyzing the correlation coefficients of parameters in the frequency / spatial dimension, this application can effectively distinguish the types of interference, such as narrowband and broadband, and reveal the spectral characteristics of the interference.

[0114] The mean interference energy and image entropy at P consecutive sampling times are standardized respectively;

[0115] The standardized mean sequence of interference energy and the image entropy sequence are used to construct an environment feature matrix with Z rows and X columns;

[0116] Each row of the environmental feature matrix represents a standardized environmental parameter time-series change, and each column represents a standardized state vector of multi-source environmental sensing parameters at a given time.

[0117] The environmental feature matrix is ​​used to characterize the spatiotemporal evolution trend of interference intensity in the external environment after normalization in the multimodal perception dimension.

[0118] Specifically, the construction of the environmental feature matrix includes: multi-source parameter normalization preprocessing and normalized sequence matrixization.

[0119] Furthermore, in the multi-source parameter standardization preprocessing, the UAV performs standardization preprocessing on the mean value of interference energy and image entropy acquired at P consecutive sampling times. The standardization process is used to eliminate the incomparability of the two parameters due to their different physical dimensions and numerical ranges, and to uniformly transform them into dimensionless standard scores with the same numerical scale.

[0120] For parameters such as the mean of interference energy or image entropy, the UAV calculates its arithmetic mean and standard deviation in the P time-series. The original value at each time is subtracted from the mean and divided by the standard deviation to obtain the standardized numerical sequence.

[0121] The normalized mean sequence of interference energy and the image entropy sequence are constructed into a matrix by row.

[0122] Specifically, the normalized mean sequence of interference energy is used as the first row of the matrix, and the normalized sequence of image entropy is used as the second row of the matrix.

[0123] The technical meaning of the rows and columns of the environmental feature matrix includes: each row of the environmental feature matrix represents a specific type of environmental perception parameter, such as electromagnetic interference energy or visual interference level.

[0124] Each column of the environmental feature matrix represents a vector consisting of the standardized values ​​of all environmental sensing parameters at a specific sampling time.

[0125] Furthermore, standardization processing allows interference energy and image entropy, which originally had different dimensions, to be compared and fused for analysis at the same scale.

[0126] The environmental feature matrix simultaneously captures information about environmental disturbances across both the temporal and perceptual modal dimensions for P consecutive time periods. By calculating its singular values ​​or analyzing the correlations between row vectors, the duration, rate of change, and synergy or variability exhibited by the disturbances across different perceptual modalities can be assessed.

[0127] Multi-source environmental sensing parameters are effectively integrated into a unified mathematical representation to accurately describe the comprehensive intensity of external disturbance environments and their spatiotemporal variation patterns.

[0128] It should be noted that those skilled in the art can perform normalization and data standardization by combining the technical content disclosed in this application with common general knowledge in the field.

[0129] The calculation module is used to establish a matrix fusion unit, which fuses the navigation feature matrix, the communication feature matrix, and the environmental feature matrix into a comprehensive state matrix. The comprehensive state matrix is ​​used to calculate and output singular values ​​as anti-interference decision feature values ​​for the unmanned aerial vehicle.

[0130] The matrix fusion unit is used to generate a comprehensive state matrix by weighted direct sum fusion of the navigation feature matrix, communication feature matrix and environmental feature matrix;

[0131] The matrix fusion unit also dynamically and adaptively assigns weights to the navigation feature matrix, communication feature matrix, and environmental feature matrix by real-time reliability assessment, and performs weighted fusion after the dynamic adaptive weight assignment is completed; thus enabling the comprehensive state matrix to more accurately reflect the current dominant system state.

[0132] The method for determining the dynamic adaptive weights includes:

[0133] The stability of the navigation feature matrix, communication feature matrix, and environmental feature matrix within the most recent time window is used to calculate the initial confidence weights.

[0134] The attitude stability data from the UAV flight control is used as an external verification factor to correct the initial confidence weight;

[0135] The data stability is quantified by calculating the inverse variance of the main diagonal elements of the navigation feature matrix, communication feature matrix, and environmental feature matrix.

[0136] like Figure 3 As shown, in this application, a preferred implementation method for matrix fusion unit and dynamic adaptive weight allocation includes:

[0137] Specifically, the methods for constructing the weighted direct sum fusion and comprehensive state matrix include:

[0138] Furthermore, the direct fusion includes the matrix fusion unit treating the navigation feature matrix, communication feature matrix, and environmental feature matrix as three independent diagonal sub-blocks;

[0139] Three independent diagonal sub-blocks are concatenated along the main diagonal, with zero elements filled in the off-diagonal positions, to construct a comprehensive state matrix in block diagonal form. This comprehensive state matrix fully preserves all the internal structural information of each source feature matrix, avoiding structural damage and information confusion that may occur when different modal features are vectorized or simply stacked.

[0140] Furthermore, the weighting includes: before the above direct sum fusion, the matrix fusion unit first multiplies the navigation feature matrix, communication feature matrix and environmental feature moments by their respective dynamic adaptive weights. The weighted navigation feature matrix, communication feature matrix and environmental feature moments then participate in the direct sum fusion, and the final comprehensive state matrix is ​​a weighted block diagonal matrix.

[0141] Each dynamic adaptive weight includes the dynamic adaptive weight of the navigation feature matrix, the dynamic adaptive weight of the communication feature matrix, and the dynamic adaptive weight of the environmental feature matrix.

[0142] A preferred method for determining dynamic adaptive weights includes:

[0143] The dynamic adaptive weights are used to adjust the contribution of each subsystem to the final decision based on the real-time reliability of the information.

[0144] Specifically, the dynamic adaptive weights include the initial confidence weight calculation and the weight adjustment based on external validation factors.

[0145] Furthermore, the initial confidence weight calculation uses the variance of the main diagonal elements of each feature matrix as a quantitative indicator of its data stability. For navigation and communication feature matrices, the main diagonal elements are always 1, representing the correlation coefficient between the parameter and itself. For the environmental feature matrix, the main diagonal elements are the standardized parameter values.

[0146] Furthermore, the calculation method includes: setting a sliding time window for the matrix fusion unit to record the values ​​of the main diagonal elements of each feature matrix in the past K update cycles;

[0147] Calculate the variance of these historical values ​​within the time window. The larger the variance, the more drastic the fluctuations in the system state represented by the feature matrix, the worse the stability of its current data, and the lower its reliability.

[0148] Furthermore, the system takes the reciprocal of the calculated variance and uses it as the weight mapping. The larger the reciprocal of the variance, the higher the stability. The initial confidence weight is proportional to this reciprocal of the variance. Through normalization, the reciprocals of the variances of the three feature matrices are converted into initial weights that sum to 1.

[0149] Furthermore, based on the weighted adjustment of the external verification factor, the system receives attitude stability data from the UAV flight control unit as an external verification factor. Attitude stability data is a physical quantity provided by the UAV inertial measurement unit that reflects the rate of change of the aircraft's attitude angle. It is an objective state indicator independent of navigation, communication, and environmental perception.

[0150] The matrix fusion unit cross-validates the interference assessment results reflected by each feature matrix with the attitude stability. If the feature matrix indicates strong interference, for example, the navigation matrix shows that the positioning information is seriously unreliable, but the attitude stability data at the same time shows that the UAV is flying extremely smoothly and there is no expected control anomaly or attitude jitter, then it is determined that the assessment result of the feature matrix is ​​inconsistent with the objective physical state.

[0151] In this case, the system will adjust the initial confidence weights of the feature matrix downwards.

[0152] Conversely, if the evaluation of the feature matrix matches the state revealed by the attitude stability data, such as abnormal jitter, its weights will be maintained or slightly adjusted upwards.

[0153] Through the aforementioned dynamic adaptive weight allocation mechanism, the system no longer mechanically treats all information equally. Instead, it can intelligently adjust the confidence levels of different information sources based on the stability of the data itself and its consistency with external objective facts. Furthermore, by setting adaptive dynamic weights, it effectively suppresses the negative impact of unreliable data caused by temporary failures of individual sensors or instantaneous strong interference on the overall state assessment. The comprehensive state matrix obtained by weighted sum fusion highlights the state information represented by the subsystem with higher credibility at the current moment, thereby enabling more accurate capture and reflection of the interference patterns that have the most significant impact on system stability.

[0154] The matrix fusion unit establishes a block-diagonal comprehensive state matrix by using the navigation feature matrix, communication feature matrix and environmental feature matrix as diagonal sub-blocks;

[0155] The calculation module further includes a singular value decomposition unit, which is used to decompose the singular values ​​of the block diagonal form of the comprehensive state matrix to obtain a singular value sequence.

[0156] The singular value decomposition unit is used to extract the maximum singular value from the singular value sequence obtained after decomposition, and output the maximum singular value as an anti-interference decision feature value that characterizes the overall degree of interference of the system.

[0157] In this application, a preferred implementation method for constructing a comprehensive state matrix using a block diagonalization approach for the matrix fusion unit includes:

[0158] The navigation feature matrix, communication feature matrix, and environment feature matrix, after dynamic adaptive weight scaling, are treated as three independent submatrices.

[0159] Place the three independent submatrices sequentially along the main diagonal;

[0160] like Figure 4As shown, further, the navigation feature matrix is ​​placed in the upper left corner of the integrated state matrix; the communication feature matrix follows immediately below it, located in the lower right corner of the navigation feature matrix; and the environment feature matrix is ​​placed in the lower right corner of the communication feature matrix. All off-diagonal sub-blocks are filled with zero elements.

[0161] One preferred example of a comprehensive state matrix includes Q11, Q12, Q13, Q21, Q22, Q23, Q31, Q32, and Q33 as navigation feature moments;

[0162] P11, P12, P13, P21, P22, P23, P31, P32, and P33 are the communication feature matrices;

[0163] L11, L12, L13, L21, L22, L23, L31, L32, and L33 are environmental feature matrices.

[0164] The final generated integrated state matrix is ​​a typical block diagonal matrix that fully preserves the internal structure and information integrity of the navigation feature matrix, communication feature matrix, and environmental feature matrix, avoiding unnecessary cross-interference or information loss between data of different properties during the fusion process.

[0165] In a preferred embodiment, the singular value decomposition unit included in the calculation module is responsible for analyzing the comprehensive state matrix. Specific implementation methods include:

[0166] The singular value decomposition unit performs singular value decomposition on the block diagonal form of the synthesized state matrix, decomposing any matrix into the product of three specific matrices. This decomposition yields the singular value sequence (σ1, σ2, σ3, ..., σ) of the original matrix. n The singular values ​​in this sequence are arranged in descending order of their numerical values.

[0167] The singular value decomposition unit extracts the largest singular value σ1 from the resulting singular value sequence. This largest singular value is defined as the anti-interference decision feature value and output to the decision unit.

[0168] It is important to note that in matrix theory, the maximum singular value σ1 represents the magnitude of the principal energy component contained in the matrix, and is used to reflect the intensity of the most important and active change mode in the system represented by the matrix.

[0169] In this application, a preferred principal energy component of the integrated state matrix directly corresponds to the intensity of the dominant disturbance mode that has the most significant impact on system stability. When the disturbance intensity increases or multiple disturbances produce coupling effects, the energy of the integrated state matrix will concentrate on the principal component, resulting in a significant increase in the maximum singular value σ1.

[0170] The anti-interference decision characteristic value λ is a continuous and quantitative indicator that is positively correlated with the overall interference intensity experienced by the unmanned aerial vehicle.

[0171] By extracting the maximum singular value through block diagonalization fusion and singular value decomposition, this application condenses complex multimodal perception information into a scalar index with clear physical meaning, realizing the key transformation from multidimensional data to intelligent decision-making.

[0172] The calculation module processes the comprehensive state matrix through singular value decomposition and extracts its maximum singular value as an anti-interference decision feature value.

[0173] The decision feature value serves as a continuous, quantified indicator, which is positively correlated with the overall interference intensity experienced by the unmanned aerial vehicle.

[0174] In this application, a preferred implementation method for the computation module to extract anti-interference decision feature values ​​includes:

[0175] The calculation module performs singular value decomposition on the block diagonal form of the comprehensive state matrix.

[0176] From the sequence of singular values ​​obtained after decomposition, the singular value with the largest value is extracted as the anti-interference decision feature value of the system.

[0177] The anti-interference decision feature values ​​are output in scalar form and transmitted to the decision unit in real time.

[0178] Furthermore, the anti-interference decision eigenvalues ​​are used to represent the principal component energy intensity of the unmanned aerial vehicle's comprehensive state matrix. The anti-interference decision eigenvalues ​​directly characterize the intensity level of the interference mode that currently has the most significant impact on the overall operating state of the unmanned aerial vehicle.

[0179] Furthermore, there is a positive correlation between the anti-interference decision eigenvalue and the overall interference intensity experienced by the unmanned aerial vehicle (UAV). When the interference intensity increases, the energy of anomalous information contained in the overall state matrix increases, leading to a corresponding increase in its maximum singular value; conversely, when the interference weakens or disappears, the anti-interference decision eigenvalue decreases.

[0180] Further anti-interference decision characteristics are continuously quantified indicators that can accurately reflect the gradual change process of interference intensity and provide the system with state awareness capabilities.

[0181] As a continuously quantified indicator, the anti-interference decision characteristic value can accurately reflect the intensity level and changing trend of the interference situation. This enables the system to distinguish different levels of anti-interference scenarios such as slight interference, moderate interference and severe interference. Compared with the traditional threshold triggering mechanism, continuous perception can detect the cumulative effect of interference earlier and provide the system with more sufficient response time.

[0182] The decision-making unit presets multiple threshold ranges corresponding to different levels of interference, and each range is associated with a specific anti-interference strategy.

[0183] By comparing the real-time acquired anti-interference decision feature values ​​with these preset threshold ranges, the system can determine the current level of interference.

[0184] Based on the assessment of the current interference level, the decision-making unit selects and activates a combination of control commands that matches the interference level from a predefined strategy library, thereby achieving precision and optimization of the anti-interference response.

[0185] The hierarchical decision-making mechanism based on continuous quantitative indicators ensures that the system can implement appropriate countermeasures according to the actual severity of the interference. This avoids both the waste of resources caused by overreacting under minor interference and the security risks caused by insufficient response under severe interference.

[0186] The decision module receives the anti-interference decision feature value and outputs interference control instructions.

[0187] The anti-interference decision feature value is compared with multiple preset interference threshold intervals;

[0188] Based on the interference threshold range into which the feature value falls, select the corresponding combination of control commands from the predefined anti-interference strategy mapping table;

[0189] The control command combination is used to schedule one or more execution subsystems, including a communication unit, a navigation unit, and a flight control unit, to perform coordinated anti-interference actions.

[0190] Based on the specific model of the unmanned aerial vehicle, the mission type, and the typical electromagnetic environment, multiple consecutive interference threshold ranges are preset. For example, multiple levels can be set, including a safe range, a light interference range, a moderate interference range, and a heavy interference range. Each range corresponds to a specific anti-interference response level.

[0191] The decision module receives anti-interference decision feature values ​​from the calculation module in real time, quickly compares the anti-interference decision feature values ​​with preset threshold intervals, determines the specific interval to which the current interference level belongs, and accesses a predefined anti-interference strategy mapping table, which defines the optimal combination of control instructions corresponding to each interference threshold interval.

[0192] The control command combination is a pre-designed sequence of cooperative operations for different subsystems.

[0193] The instruction combination generated by the decision module will simultaneously schedule one or more of the following execution subsystems:

[0194] Instructions may include switching to a backup communication band, enabling spread spectrum communication mode, adjusting transmission power, or switching to a directional communication link to avoid interference.

[0195] Instructions may include switching to a pure inertial navigation mode, a vision-assisted navigation mode, or a terrain-matching-based navigation mode when GNSS signals are unreliable.

[0196] Flight control unit commands may include changing the flight path to avoid interference source areas, adjusting flight altitude, or performing specific anti-interference maneuvers to enhance stability.

[0197] Example 2

[0198] like Figure 2 As shown, the anti-interference control method for unmanned aerial vehicles includes:

[0199] S1. The navigation unit continuously acquires parameters such as carrier noise power density ratio, number of visible satellites, and positioning accuracy attenuation factor from the airborne GNSS receiver; the communication unit synchronously acquires parameters such as received signal strength indication, signal-to-noise ratio, and channel estimation from the airborne communication system; the environmental perception unit acquires the average interference energy through a spectrum analyzer and collects images and calculates image entropy parameters through a visual sensor. All parameters are synchronously collected at a preset sampling frequency and timestamped to ensure data temporal consistency.

[0200] S2. Based on the acquired parameter sequence, three-dimensional feature matrices are constructed: the navigation unit calculates the Pearson correlation coefficient based on the parameter sequence within a continuous time window, forming a navigation parameter correlation coefficient matrix; the communication unit calculates the correlation coefficient based on the distribution of multi-sub-channel parameters, forming a communication parameter correlation coefficient matrix; and the environment perception unit constructs a standardized environment perception matrix after standardizing the mean interference energy and image entropy. Each feature matrix retains the unique feature representation of its corresponding subsystem.

[0201] S3. The matrix fusion unit first evaluates the data stability of each feature matrix within the most recent time window, and combines the attitude stability data provided by the flight control unit as an external verification factor to dynamically calculate the adaptive weights of each feature matrix. Then, a weighted direct sum fusion method is used to concatenate the weighted feature matrices into a block diagonal comprehensive state matrix, which fully preserves the feature structure of each subsystem.

[0202] S4. The calculation module performs singular value decomposition on the integrated state matrix to obtain a sequence of singular values ​​arranged in descending order. The largest singular value is extracted as the anti-interference decision feature value. This feature value mathematically represents the principal energy component of the integrated state matrix and physically is positively correlated with the intensity of the integrated interference experienced by the system, providing a continuous quantitative basis for subsequent decision-making.

[0203] S5. The decision-making unit compares the anti-interference decision feature value with multiple preset interference threshold intervals in real time. Based on the specific interval in which the feature value falls, it selects the corresponding control command combination from a predefined anti-interference strategy mapping table. This command combination coordinates the communication unit, navigation unit, and flight control unit to form a system-level coordinated anti-interference response.

[0204] It is important to note that the constructions and arrangements of this application shown in several different exemplary embodiments are merely illustrative. Although only two embodiments are described in detail in this disclosure, those who consult this disclosure will readily understand that many modifications are possible without substantially departing from the novel teachings and advantages of the subject matter described in this application. These modifications may include, for example, changes in the size, dimensions, structure, shape, and proportions of various elements, as well as parameter values ​​(e.g., temperature, pressure, etc.), mounting arrangements, the use of materials, colors, orientations, etc. For example, an element shown as integrally formed may be composed of multiple parts or elements, the position of elements may be inverted or otherwise altered, and the nature or number or position of discrete elements may be changed or altered. Therefore, all such modifications are intended to be included within the scope of this invention. The order or sequence of any process or method steps may be changed or rearranged according to alternative embodiments. Any "device plus function" clause is intended to cover the structure performing the function described herein, and not only structural equivalents but also equivalent structures. Other substitutions, modifications, alterations, and omissions may be made in the design, operation, and arrangement of the exemplary embodiments without departing from the scope of this invention. Therefore, the present invention is not limited to the specific embodiments, but extends to various modifications that still fall within the scope of the appended claims.

[0205] Furthermore, in order to provide a concise description of exemplary embodiments, not all features of actual embodiments (i.e., those features that are not relevant to the best mode of carrying out the invention as currently considered, or those features that are not relevant to implementing the invention) may be omitted.

[0206] It should be understood that numerous specific implementation decisions can be made during the development of any practical implementation, such as in any engineering or design project. Such development efforts may be complex and time-consuming, but for those of ordinary skill in the art who benefit from this disclosure, the development effort will be a routine task in design, manufacturing, and production without requiring extensive experimentation.

[0207] 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 it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An anti-interference control device for unmanned aerial vehicles, characterized in that, include: Data module, matrix module, calculation module, and decision module; The data module includes a navigation unit, a communication unit, and an environmental perception unit. The data module acquires multiple sets of parameters from the navigation unit, the communication unit, and the environmental perception unit of the unmanned aerial vehicle in real time. The matrix module includes a navigation feature matrix, a communication feature matrix, and an environmental feature matrix. The navigation feature matrix is ​​established using multiple sets of parameters from the navigation unit, the communication feature matrix is ​​established using multiple sets of parameters from the communication unit, and the environmental feature matrix is ​​established using multiple sets of parameters from the environmental perception unit. The calculation module is used to establish a matrix fusion unit, which fuses the navigation feature matrix, the communication feature matrix, and the environmental feature matrix into a comprehensive state matrix. The comprehensive state matrix is ​​used to calculate and output singular values ​​as anti-interference decision feature values ​​for the unmanned aerial vehicle. The decision module receives the anti-interference decision feature value and outputs interference control instructions.

2. The anti-interference control device for unmanned aerial vehicles as described in claim 1, characterized in that: The navigation unit has multiple sets of parameters, including carrier noise power density ratio, number of visible satellites, and positioning accuracy attenuation factor. The communication unit has multiple sets of parameters, including received signal strength indication, signal-to-noise ratio, and channel estimation value; The environmental perception unit has multiple parameters, including the average interference energy and the image entropy calculated from images acquired by the visual sensor.

3. The anti-interference control device for unmanned aerial vehicles as described in claim 2, characterized in that: By using the numerical sequences of carrier noise power density ratio, number of visible satellites, and positioning accuracy attenuation factor at N consecutive sampling times, the navigation Pearson correlation coefficient between each pair of carrier noise power density ratio, number of visible satellites, and positioning accuracy attenuation factor is calculated. The Pearson correlation coefficients are constructed into a navigation feature matrix with Q rows and Q columns; The off-diagonal elements of the navigation feature matrix This is used to represent the dynamic correlation strength and direction between the i-th navigation parameter and the j-th navigation parameter within a time window.

4. The anti-interference control device for unmanned aerial vehicles as described in claim 3, characterized in that: Received signal strength indication, signal-to-noise ratio and channel estimate are measured and obtained on T different communication sub-channels. The communication Pearson correlation coefficient between each pair of signal strength indication, signal-to-noise ratio and channel estimate is calculated. The communication Pearson correlation coefficients are constructed into a communication feature matrix with Q rows and Q columns; The off-diagonal elements of the communication feature matrix This is used to represent the distributional correlation characteristics of the i-th communication parameter and the j-th communication parameter in the frequency domain spatial dimension.

5. The anti-interference control device for unmanned aerial vehicles as described in claim 4, characterized in that: The mean interference energy and image entropy at P consecutive sampling times are standardized respectively; The standardized mean sequence of interference energy and the image entropy sequence are used to construct a feature matrix with Z rows and X columns; Each row of the environmental feature matrix represents a standardized environmental parameter time-series change, and each column represents a standardized state vector of multi-source environmental sensing parameters at a given time.

6. The anti-interference control device for unmanned aerial vehicles as described in claim 5, characterized in that: The matrix fusion unit is used to generate a comprehensive state matrix by weighted direct sum fusion of the navigation feature matrix, communication feature matrix and environmental feature matrix; The matrix fusion unit also assigns dynamic adaptive weights to the navigation feature matrix, communication feature matrix, and environmental feature matrix by real-time reliability assessment of the navigation feature matrix, communication feature matrix, and environmental feature matrix, and performs weighted fusion after the dynamic adaptive weight allocation is completed. The method for determining the dynamic adaptive weights includes: The initial confidence weight is calculated by assessing the stability of the navigation feature matrix, communication feature matrix, and environmental feature matrix within the most recent time window based on their own data. The attitude stability data from the UAV flight control is used as an external verification factor to correct the initial confidence weight; The data stability is quantified by calculating the inverse variance of the main diagonal elements of the navigation feature matrix, communication feature matrix, and environmental feature matrix.

7. The anti-interference control device for unmanned aerial vehicles as described in claim 6, characterized in that: The matrix fusion unit establishes a block-diagonal comprehensive state matrix by using the navigation feature matrix, communication feature matrix and environmental feature matrix as diagonal sub-blocks; The calculation module further includes a singular value decomposition unit, which is used to decompose the singular values ​​of the block diagonal form of the comprehensive state matrix to obtain a singular value sequence. The singular value decomposition unit is used to extract the maximum singular value from the singular value sequence obtained after decomposition, and output the maximum singular value as an anti-interference decision feature value that characterizes the overall degree of interference of the system.

8. The anti-interference control device for unmanned aerial vehicles as described in claim 7, characterized in that: The calculation module processes the comprehensive state matrix through singular value decomposition and extracts its maximum singular value as an anti-interference decision feature value. The decision feature value serves as a continuous, quantified indicator, which is positively correlated with the overall interference intensity experienced by the unmanned aerial vehicle.

9. The anti-interference control device for unmanned aerial vehicles as described in claim 8, characterized in that: The anti-interference decision feature value is compared with multiple preset interference threshold intervals; Based on the interference threshold range into which the feature value falls, select the corresponding combination of control commands from the predefined anti-interference strategy mapping table; The control command combination is used to schedule one or more execution subsystems, including a communication unit, a navigation unit, and a flight control unit, to perform coordinated anti-interference actions.

10. An anti-interference control method for unmanned aerial vehicles, characterized in that... Applied to the anti-interference control device for unmanned aerial vehicles as described in any one of claims 1-9; wherein: S1. Real-time acquisition of multiple sets of parameters from the navigation unit, communication unit, and environmental perception unit of the unmanned aerial vehicle; S2. Construct navigation feature matrix, communication feature matrix and environment feature matrix respectively using the multiple sets of parameters; S3. Establish a matrix fusion unit, and fuse the navigation feature matrix, the communication feature matrix, and the environmental feature matrix into a comprehensive state matrix through the matrix fusion unit; S4. Calculate the singular values ​​of the integrated state matrix, and use the singular values ​​as the anti-interference decision characteristic values ​​of the unmanned aerial vehicle; S5. Output the anti-interference strategy through the anti-interference decision feature value and generate control commands to send to the execution unit.

Images