Unmanned aerial vehicle countering analysis method fusing radar-photoelectric data

Abstract

The invention relates to the technical field of space perception monitoring, in particular to an unmanned aerial vehicle countering analysis method fusing radar-photoelectric data. The method comprises the following steps: performing cross-modal coupling sensing feature analysis of unmanned aerial vehicle target monitoring on airspace unmanned aerial vehicle multi-source monitoring data, and generating unmanned aerial vehicle target monitoring coupling sensing feature data; performing unmanned aerial vehicle flight behavior propagation feature analysis based on the unmanned aerial vehicle target monitoring coupling sensing feature data to generate unmanned aerial vehicle flight behavior propagation feature data; performing unmanned aerial vehicle threat propagation field feature analysis based on the unmanned aerial vehicle flight behavior propagation feature data to generate unmanned aerial vehicle threat propagation field feature data; and performing intelligent control strategy analysis of unmanned aerial vehicle countering based on the unmanned aerial vehicle threat propagation field feature data, and generating unmanned aerial vehicle countering intelligent control strategy data. According to the invention, through fusion analysis of radar monitoring data and photoelectric monitoring data, an accurate and efficient unmanned aerial vehicle intelligent countering strategy is realized.

Description

Technical Field

[0001] This invention relates to the field of space perception and monitoring technology, and in particular to a method for analyzing UAV countermeasures by fusing radar and photoelectric data. Background Technology

[0002] With the rapid development of drone technology, the application scenarios of consumer and industrial drones are becoming increasingly widespread, playing an important role in fields such as aerial surveying and mapping, power line inspection, and emergency rescue. However, this has also brought serious airspace security risks. Unauthorized drones entering sensitive airspace can easily cause aviation safety accidents, leak classified information, and disrupt normal public order. Currently, drone countermeasures fall into two main categories: radar monitoring and electro-optical monitoring. Radar monitoring technology suffers from low accuracy in identifying small drones, susceptibility to electromagnetic interference, and difficulty in obtaining detailed drone features. Electro-optical monitoring technology is greatly affected by environmental factors such as weather and lighting, has limited detection range, and struggles to achieve comprehensive coverage of large airspace. Existing methods that fuse radar and electro-optical data for drone monitoring fail to fully leverage the complementarity of these two types of data, resulting in incomplete and inaccurate extraction of drone target features. Furthermore, the analysis of drone flight behavior is often limited to single trajectories or simple dynamic characteristics, lacking in-depth research into the propagation patterns of flight behavior, making it difficult to accurately predict drone flight trends and comprehensively reflect the spread and intensity of drone threats. Summary of the Invention

[0003] Based on this, the present invention provides a method for UAV countermeasure analysis that integrates radar and optoelectronic data to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, a UAV countermeasure analysis method integrating radar and optoelectronic data includes the following steps: Step S1: Perform multi-source monitoring and processing of airspace UAV data through deployed radar-electro-optical monitoring equipment to generate airspace UAV multi-source monitoring data; perform cross-modal coupled sensing feature analysis of UAV target monitoring on the airspace UAV multi-source monitoring data to generate UAV target monitoring coupled sensing feature data. Step S2: Analyze the propagation characteristics of UAV flight behavior based on the coupled sensing feature data of UAV target monitoring, and generate UAV flight behavior propagation feature data; Step S3: Analyze the characteristics of drone threat propagation field based on drone flight behavior propagation feature data, and generate drone threat propagation field feature data; Step S4: Analyze the intelligent control strategy for drone countermeasures based on the drone threat propagation field characteristic data, and generate drone countermeasure intelligent control strategy data.

[0005] Furthermore, step S1 includes the following steps: Step S11: Perform multi-source monitoring and processing of airspace UAV data through deployed radar-electro-optical monitoring equipment to generate multi-source monitoring data of airspace UAVs, wherein the multi-source monitoring data of airspace UAVs includes radar monitoring data and electro-optical monitoring data of airspace UAVs. Step S12: Analyze radar monitoring and perception features based on UAV radar monitoring data in the airspace to generate radar monitoring and perception feature data; Step S13: Perform photoelectric observation feature analysis based on airspace UAV photoelectric monitoring data to generate photoelectric observation feature data; Step S14: Perform radar monitoring and sensing feature analysis on the UAV spatial trajectory features of the radar monitoring and sensing feature data to generate radar monitoring and sensing spatial trajectory feature data; Step S15: Perform photoelectric observation feature analysis on the UAV target features of photoelectric observation to generate photoelectric observation target feature data; Step S16: Based on radar monitoring and sensing spatial trajectory feature data and photoelectric observation target feature data, perform cross-modal coupled sensing feature analysis of UAV target monitoring to generate UAV target monitoring coupled sensing feature data.

[0006] Furthermore, step S12 includes the following steps: Step S121: Analyze the characteristics of radar monitoring echo signals based on the UAV radar monitoring data in the airspace, and generate radar monitoring echo signal characteristic data; Step S122: Parse the radar monitoring echo signal characteristic data into radar monitoring electromagnetic wave sensing spectrum data; Step S123: Analyze the radar monitoring and sensing characteristics based on the radar monitoring electromagnetic wave sensing spectrum data to generate radar monitoring and sensing characteristic data.

[0007] Furthermore, step S13 includes the following steps: Step S131: Extract airspace photoelectric visual data and airspace photoelectric thermal radiation data based on airspace UAV photoelectric monitoring data; Step S132: Perform visual feature analysis on the contour and texture of the spatial optoelectronic visual data to obtain optoelectronic visual feature data; perform optoelectronic thermal radiation distribution feature analysis on the spatial optoelectronic thermal radiation data to generate optoelectronic thermal radiation distribution feature data. Step S133: Map the photoelectric thermal radiation distribution characteristic data to the photoelectric visual characteristic data to perform photoelectric observation characteristic analysis and generate photoelectric observation characteristic data.

[0008] Furthermore, step S16 includes the following steps: Step S161: Perform spatial trajectory prediction processing on the radar monitoring and sensing spatial trajectory feature data to generate radar monitoring and sensing spatial trajectory prediction data; Step S162: Map the radar monitoring and sensing spatial trajectory prediction data to the photoelectric observation target feature data to perform observation neighborhood candidate matching analysis for spatial trajectory prediction, and generate spatial trajectory prediction-observation neighborhood candidate matching data; Step S163: Perform candidate association analysis for UAV cross-modal monitoring based on spatial trajectory prediction-observation neighborhood candidate matching data to generate UAV cross-modal monitoring candidate association data; Step S164: Optimize the dynamic target association of UAVs in cross-modal monitoring based on the cross-modal candidate association data of UAV monitoring, and generate UAV cross-modal dynamic target association data; Step S165: Perform cross-modal coupled sensing feature analysis on the dynamic target association data of UAV cross-modal monitoring to generate UAV target monitoring coupled sensing feature data.

[0009] Furthermore, step S2 includes the following steps: Step S21: Based on the UAV target monitoring coupled sensing feature data, perform UAV spatial energy peak migration feature analysis to generate UAV spatial energy peak migration feature data; Step S22: Analyze the dynamic behavior characteristics of the UAV based on the spatial energy peak migration characteristic data, and generate UAV dynamic behavior characteristic data; Step S23: Perform UAV visual structure state feature analysis based on UAV target monitoring coupled perception feature data to generate UAV visual structure state feature data; Step S24: Analyze the UAV flight behavior pattern using UAV dynamic behavior feature data and UAV visual structural state feature data to generate UAV flight behavior pattern data. Step S25: Analyze the propagation characteristics of UAV flight behavior based on the UAV flight behavior pattern data to generate UAV flight behavior propagation characteristic data.

[0010] Furthermore, step S21 includes the following steps: Step S211: Perform time-series feature analysis on the UAV target monitoring coupled sensing feature data to generate UAV target monitoring coupled sensing time-series feature data; Step S212: Based on the UAV target monitoring coupled sensing time series feature data, perform UAV spatial dynamic energy peak identification processing to generate UAV spatial dynamic energy peak data; Step S213: Perform UAV spatial energy peak migration characteristic analysis on the UAV spatial dynamic energy peak data to generate UAV spatial energy peak migration characteristic data.

[0011] Furthermore, step S3 includes the following steps: Step S31: Perform spatial threat multi-track grid cell analysis on the UAV flight behavior propagation characteristic data to generate spatial threat multi-track grid cell data; Step S32: Analyze the space threat diffusion trend based on the space threat multi-track grid cell data to generate space threat diffusion trend data; Step S33: Perform threat indicator feature analysis on the spatial threat diffusion trend data to generate spatial diffusion trend threat indicator feature data; Step S34: Analyze the associated threat characteristics of the spatial diffusion trend based on the spatial diffusion trend threat index characteristic data, and generate spatial diffusion trend associated threat characteristic data; Step S35: Analyze the threat intensity distribution characteristics of the diffusion trend based on the spatial diffusion trend threat index feature data and the spatial diffusion trend related threat feature data, and generate spatial diffusion trend threat intensity distribution feature data; Step S36: Analyze the UAV threat propagation field characteristics of the spatial threat diffusion trend data using the spatial diffusion trend threat intensity distribution characteristic data, and generate UAV threat propagation field characteristic data.

[0012] Furthermore, step S31 includes the following steps: Step S311: Perform UAV multi-track node feature analysis based on UAV flight behavior propagation feature data to generate UAV multi-track node feature data; Step S312: Perform threat multi-track node feature analysis based on UAV multi-track node feature data to generate UAV threat multi-track node feature data; Step S313: Perform spatial threat multi-track grid cell analysis on the protected area based on the multi-track node feature data of UAV threats, and generate spatial threat multi-track grid cell data.

[0013] Furthermore, step S4 includes the following steps: Step S41: Perform drone countermeasure state characteristic analysis on the drone threat propagation field characteristic data to generate drone countermeasure state characteristic data; Step S42: Utilize the preset UAV prior countermeasure strategy library to perform UAV countermeasure state feature analysis on UAV countermeasure strategy matching feature data, and generate UAV countermeasure strategy matching feature data. Step S43: Design the action space of the UAV countermeasure strategy based on the matching feature data of the UAV countermeasure strategy, and generate the action space data of the UAV countermeasure strategy. Step S44: Analyze the intelligent control strategy for UAV countermeasures using the UAV countermeasures action space data to generate intelligent control strategy data for UAV countermeasures.

[0014] The beneficial effects of this application are as follows: This invention, through layered and progressive feature extraction and coupling analysis, first performs in-depth analysis of the echo signal and electromagnetic spectrum of radar monitoring data to extract radar monitoring perception features and corresponding spatial trajectory features. Then, it decomposes and fuses the visual and thermal radiation features of photoelectric monitoring data to extract photoelectric observation target features. Finally, through processes such as trajectory prediction, neighborhood matching, and correlation optimization, it achieves deep cross-modal coupling of the two types of data, completely eliminating the drawbacks of simple data superposition, fully exploring the complementary advantages of radar and photoelectric data, and effectively solving the problem of incomplete and inaccurate feature extraction from single monitoring technologies. It can generate more comprehensive and accurate UAV target monitoring coupled perception feature data, providing reliable core data support for UAV flight behavior analysis and countermeasure strategy formulation, and improving the accuracy and reliability of UAV target monitoring. By analyzing temporal characteristics to identify the spatial dynamic energy peaks of UAVs and analyzing their migration patterns, and combining dynamic characteristics with visual structural state characteristics to construct a complete flight behavior pattern, further analysis of flight behavior propagation characteristics can accurately capture the changing patterns and propagation trends of UAV flight behavior. This effectively solves the problem of difficulty in accurately predicting UAV flight trends. The generated UAV flight behavior propagation characteristic data can clearly reflect the dynamic changes in UAV flight status, providing accurate behavioral basis for subsequent UAV threat assessment and intelligent countermeasure strategy formulation, further improving the scientificity and foresight of UAV countermeasure analysis. Through multi-track node analysis and grid cell division, the UAV flight behavior propagation characteristics are combined with the protected area space to generate spatial threat multi-track grid cell data. Then, the threat diffusion trend, threat index characteristics, associated threat characteristics, and threat intensity distribution are analyzed step by step. Finally, a complete UAV threat propagation field model is constructed, clearly and comprehensively reflecting the diffusion range, diffusion trend, and intensity distribution of UAV threats. This effectively solves the problem of not being able to fully depict the UAV threat status, providing accurate and comprehensive threat assessment basis for subsequent UAV countermeasure strategy formulation, and improving the scientificity and comprehensiveness of threat assessment. By analyzing the characteristics of UAV countermeasure status, matching strategies with a pre-set prior countermeasure strategy library, and then designing a countermeasure strategy action space based on the matching results, intelligent control strategy data is generated. The entire process achieves precise adaptation between countermeasure strategies and the real-time threat status of UAVs. It can dynamically output the optimal countermeasure scheme according to the spread range, intensity, and flight behavior of UAV threats, avoiding the problems of weak targeting and insufficient flexibility of traditional countermeasure strategies. This significantly improves the efficiency and effectiveness of UAV countermeasures, ensures the scientific, rational, and timely nature of countermeasure actions, and provides reliable countermeasure technology support for airspace security.

[0015] Therefore, the UAV countermeasure analysis method integrating radar and optoelectronic data of this invention addresses the inherent limitations of single monitoring technologies by performing deep cross-modal coupled sensing analysis of radar and optoelectronic monitoring data. This enables comprehensive and accurate extraction of UAV target features, improving the accuracy and reliability of UAV target monitoring in complex environments. The method employs deep cross-modal coupling of radar and optoelectronic data to monitor the UAV airspace range, fully leveraging the complementarity of the two data types to ensure comprehensive and accurate extraction of UAV target features. Furthermore, it deeply analyzes the trajectory, dynamics, and structural state of UAV flight behavior, accurately predicting flight trends. Multiple indicators are used for threat assessment, constructing a comprehensive threat propagation field model that comprehensively and clearly reflects the spread and intensity of UAV threats, providing reliable support for subsequent countermeasure strategy formulation. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the steps of a UAV countermeasure analysis method that integrates radar and optoelectronic data according to the present invention. Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S3. The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0017] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0018] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. Functional entities may be implemented in software, or in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods. The term "and / or" as used herein includes any and all combinations of one or more of the associated items listed.

[0019] To achieve the above objectives, please refer to Figures 1 to 2 This invention provides a method for UAV countermeasure analysis that integrates radar and optoelectronic data. In the embodiments of this invention, please refer to... Figure 1The diagram shown is a flowchart illustrating the steps of a UAV countermeasure analysis method that integrates radar and optoelectronic data according to the present invention. The UAV countermeasure analysis method that integrates radar and optoelectronic data includes the following steps: Step S1: Perform multi-source monitoring and processing of airspace UAV data through deployed radar-electro-optical monitoring equipment to generate airspace UAV multi-source monitoring data; perform cross-modal coupled sensing feature analysis of UAV target monitoring on the airspace UAV multi-source monitoring data to generate UAV target monitoring coupled sensing feature data. In this embodiment of the invention, radar-optoelectronic integrated monitoring equipment is deployed in a fixed layout within a preset protected airspace. The monitoring equipment possesses both radar and optoelectronic detection capabilities, enabling simultaneous monitoring and processing of multi-source data from UAVs in the airspace. The monitoring equipment continuously scans the protected airspace in all directions. The radar detection module emits electromagnetic waves in a specific frequency band and receives the echo signals reflected by the UAVs, collecting and forming airspace UAV radar monitoring data. The optoelectronic detection module captures visual and thermal radiation images of the UAVs using visible light imaging and infrared thermal imaging, collecting and forming airspace UAV optoelectronic monitoring data. Both types of data are collected and aggregated simultaneously to form complete airspace UAV multi-source monitoring data. Subsequently, cross-modal coupled sensing feature analysis of UAV target monitoring is performed. First, features are extracted from the radar monitoring data and the optoelectronic monitoring data. Spatial position and trajectory-related features of the UAV are extracted from the radar monitoring data, while shape contour, surface texture, and thermal radiation distribution-related features are extracted from the optoelectronic monitoring data. Then, the two types of features are deeply fused through cross-modal coupling, achieving complementary correlation between the two types of data, eliminating the limitations of single monitoring data, and generating UAV target monitoring coupled sensing feature data that comprehensively and accurately reflects the characteristics of the UAV target.

[0020] Step S2: Analyze the propagation characteristics of UAV flight behavior based on the coupled sensing feature data of UAV target monitoring, and generate UAV flight behavior propagation feature data; In this embodiment of the invention, UAV flight behavior propagation characteristic analysis is performed based on UAV target monitoring coupled sensing feature data. Relevant information such as the UAV's spatial position, motion state, and energy distribution is extracted from the coupled sensing feature data. This information is then used to track the UAV's complete flight trajectory, which is segmented for analysis to clarify changes in the UAV's flight state at different times. Through dynamic analysis of the UAV's flight trajectory, the inherent patterns of UAV flight behavior are uncovered, including changes in flight direction, switching of flight states, and dynamic migration of energy distribution. Simultaneously, combined with the UAV's target characteristics, the propagation patterns of flight behavior in the airspace are analyzed, clarifying the coverage, diffusion trend, and affected area of ​​the flight behavior. The entire analysis process relies on the comprehensiveness of the coupled sensing feature data to achieve multi-dimensional and in-depth analysis of UAV flight behavior, avoiding analytical biases caused by relying on single data sources. This generates UAV flight behavior propagation characteristic data that clearly reflects the propagation patterns of UAV flight behavior, providing accurate behavioral evidence for subsequent UAV threat propagation field analysis.

[0021] Step S3: Analyze the characteristics of drone threat propagation field based on drone flight behavior propagation feature data, and generate drone threat propagation field feature data; In this embodiment of the invention, drone threat propagation field characteristic analysis is performed based on drone flight behavior propagation characteristic data. First, the scope and boundaries of a pre-defined protection zone are defined, and the protection zone is spatially divided into standardized spatial analysis units. Then, the drone flight behavior propagation characteristic data is correlated with the spatial division results of the protection zone to mark the spatial units covered by drone flight behavior. Combining the propagation patterns of drone flight behavior, the threat diffusion trend of drones within the protection zone is analyzed, clarifying the diffusion direction, diffusion speed, and coverage area. Simultaneously, combining the target characteristics and flight status of drones, the threat intensity of drones in different spatial units is quantified, constructing a complete threat intensity distribution system. Through comprehensive analysis of threat diffusion trends and threat intensity distribution, a drone threat propagation field model is constructed. This model can accurately depict the propagation patterns, intensity distribution, and dynamic changes of threats within the protection zone, ultimately generating drone threat propagation field characteristic data that comprehensively reflects the threat status posed by drones to the protected airspace, providing core threat basis for the design of subsequent countermeasure strategies.

[0022] Step S4: Analyze the intelligent control strategy for drone countermeasures based on the drone threat propagation field characteristic data, and generate drone countermeasure intelligent control strategy data.

[0023] In this embodiment of the invention, intelligent control strategy analysis for drone countermeasures is conducted based on drone threat propagation field characteristic data. First, the threat propagation field characteristic data is comprehensively analyzed to clarify the core area, threat intensity, diffusion pattern, and dynamic change patterns of the drone threat. This information is then used to analyze current countermeasure needs and determine the core targets and priorities of the countermeasures. Relying on a pre-set database of prior drone countermeasure strategies, the current threat state is correlated and matched with prior countermeasure strategies to select the countermeasure strategy type most suitable for the current threat state. Then, based on the specific threat state and protection requirements, the specific action space of the countermeasure strategy is designed, clarifying the type of countermeasure action, execution process, triggering conditions, and correlation relationships. Through intelligent control analysis, countermeasure actions are prioritized and dynamically scheduled, setting the execution order and activation timing of countermeasure actions. Simultaneously, a countermeasure action adjustment mechanism is established to ensure that the countermeasure strategy can be optimized in real time according to the dynamic changes of the threat, generating intelligent control strategy data for drone countermeasures. This achieves accurate and efficient countermeasures against drone threats, providing a complete and implementable intelligent control basis for drone countermeasure execution.

[0024] Furthermore, step S1 includes the following steps: Step S11: Perform multi-source monitoring and processing of airspace UAV data through deployed radar-electro-optical monitoring equipment to generate multi-source monitoring data of airspace UAVs, wherein the multi-source monitoring data of airspace UAVs includes radar monitoring data and electro-optical monitoring data of airspace UAVs. In this embodiment of the invention, radar-optoelectronic integrated monitoring equipment is deployed in a grid pattern within the protected airspace, with a grid spacing of 500 meters. Each monitoring device integrates both a radar detection module and an optoelectronic detection module. The radar detection module operates in the X-band (8-12GHz), with a detection angle covering azimuth 0-360° and elevation angle -10° to 90°. The optoelectronic detection module includes a visible light camera and an infrared thermal imager. The visible light camera has a resolution of 1920×1080 and a frame rate of 30 frames / second. The infrared thermal imager has a detection band of 8-14μm and a resolution of 640×512. The monitoring equipment continuously scans and monitors the airspace. The radar detection module emits X-band electromagnetic waves and receives reflected echoes to collect radar monitoring data of UAVs in the airspace, including echo signal strength, Doppler frequency shift, range, azimuth angle, and pitch angle. The photoelectric detection module captures visible light images of UAVs through a visible light camera and thermal radiation images of UAVs through an infrared thermal imager, collecting photoelectric monitoring data of UAVs in the airspace. The two types of data are collected and summarized simultaneously to form multi-source monitoring data of UAVs in the airspace. The multi-source monitoring data of UAVs in the airspace includes only radar monitoring data and photoelectric monitoring data of UAVs in the airspace.

[0025] Step S12: Analyze radar monitoring and perception features based on UAV radar monitoring data in the airspace to generate radar monitoring and perception feature data; In this embodiment of the invention, echo signal related data is extracted from the radar monitoring data of UAVs in the airspace. First, the echo signal is filtered using an adaptive Kalman filter algorithm with a filter coefficient set to 0.05 to remove environmental electromagnetic interference signals. Then, feature extraction is performed on the filtered echo signal to extract the peak value, pulse width, rise time, and fall time, generating radar monitoring echo signal feature data. The radar monitoring echo signal feature data is converted into radar monitoring electromagnetic wave sensing spectrum data according to a preset parsing rule. The parsing rule is based on the electromagnetic wave frequency range corresponding to the echo signal pulse width: a pulse width of 0.1 μs corresponds to a frequency of 10 MHz, a pulse width of 0.5 μs corresponds to a frequency of 2 MHz, and so on, completing the parsing from echo signal feature data to electromagnetic wave sensing spectrum data. Based on the radar monitoring electromagnetic wave sensing spectrum data, three core features are extracted: spectrum peak value, spectrum bandwidth, and spectrum center frequency. The spectrum peak value threshold is set to -30 dBm, the spectrum bandwidth is the frequency difference between the peak value and the peak value at 3 dB, and the spectrum center frequency is the center frequency of the concentrated energy region of the spectrum. Through the above feature extraction, radar monitoring sensing feature analysis is completed, generating radar monitoring sensing feature data.

[0026] Step S13: Perform photoelectric observation feature analysis based on airspace UAV photoelectric monitoring data to generate photoelectric observation feature data; In this embodiment of the invention, the photoelectric monitoring data of the UAV in the airspace is classified and extracted. An image segmentation algorithm is used to split the photoelectric monitoring data into airspace photoelectric visual data and airspace photoelectric thermal radiation data. The image segmentation algorithm uses a threshold segmentation method, with a threshold set to 128 (8-bit grayscale value). Regions with grayscale values ​​greater than 128 are considered photoelectric visual data, and regions with grayscale values ​​less than or equal to 128 are considered photoelectric thermal radiation data. Visual feature analysis of the contour and texture of the airspace photoelectric visual data is performed. The Canny edge detection algorithm is used to extract the contour features of the UAV. The low threshold for edge detection is set to 50, and the high threshold is set to 150, obtaining the coordinate point sequence of the UAV contour. Then, a grayscale co-occurrence matrix is ​​used to extract texture features. The distance of the grayscale co-occurrence matrix is ​​set to 1, and the angle is set to 0°, 45°, and 90°. At 135°, four texture feature parameters—contrast, correlation, energy, and entropy—are extracted to obtain photoelectric visual feature data. Photoelectric thermal radiation distribution feature analysis is performed on the spatial photoelectric thermal radiation data, extracting the mean temperature, temperature variance, area of ​​high-temperature regions, and center coordinates of high-temperature regions from the thermal radiation image. The mean temperature is the average temperature of all pixels in the thermal radiation image, and the temperature variance is the average of the squares of the temperature and the mean of all pixels. High-temperature regions are defined as areas with temperatures above 35°C. The mean temperature and center coordinates of high-temperature regions from the photoelectric thermal radiation distribution feature data are mapped to the contour coordinate system of the photoelectric visual feature data, ensuring a precise correspondence between thermal radiation features and visual contour features. Combining these two features completes the photoelectric observation feature analysis, generating photoelectric observation feature data.

[0027] Step S14: Perform radar monitoring and sensing feature analysis on the UAV spatial trajectory features of the radar monitoring and sensing feature data to generate radar monitoring and sensing spatial trajectory feature data; In this embodiment of the invention, distance, azimuth, and elevation angle data from radar monitoring and sensing feature data are acquired. Based on these data, the real-time coordinates of the UAV in three-dimensional space are calculated. The coordinate calculation uses the formula for converting spherical coordinates to rectangular coordinates, with the radar monitoring device as the coordinate origin, the X-axis pointing due east, the Y-axis pointing due north, and the Z-axis perpendicular to the ground and upward. The azimuth, elevation, and distance are converted into X, Y, and Z three-dimensional coordinates. Based on 10 consecutive frames of three-dimensional coordinate data, a polynomial fitting algorithm is used for trajectory fitting, with the fitting order set to 3rd order. The spatial trajectory curve of the UAV is obtained through fitting, and three feature parameters are extracted from the trajectory curve: curvature, tangent direction, and rate of change of displacement. The curvature is calculated by solving for the reciprocal of the circumference of a circle using the coordinates of three adjacent points. The rate of change of displacement is calculated by the ratio of the Euclidean distance between two adjacent frames to the time interval, with the time interval set to 0.1 seconds. Through the above analysis, the spatial trajectory feature analysis of the UAV monitored and sensed by radar is completed, generating radar monitoring and sensing spatial trajectory feature data.

[0028] Step S15: Perform photoelectric observation feature analysis on the UAV target features of photoelectric observation to generate photoelectric observation target feature data; In this embodiment of the invention, photoelectric visual feature data and photoelectric thermal radiation distribution feature data are acquired from photoelectric observation feature data. The contour coordinate sequence in the photoelectric visual feature data is analyzed to calculate the length, width, and height of the UAV contour. The size calculation uses pixel conversion, with a pixel-to-actual-distance conversion ratio of 1 pixel to 0.1 meters, to obtain the actual size parameters of the UAV. The contrast and energy in the texture feature parameters are analyzed. Contrast reflects the clarity of the UAV surface texture, and energy reflects the uniformity of the texture. The two are combined to determine the surface material type of the UAV. The area and average temperature of the high-temperature region in the photoelectric thermal radiation distribution feature data are analyzed to determine the location and heat intensity of the UAV engine. The area where the high-temperature region accounts for more than 10% of the UAV contour area is determined to be the engine location. The UAV size parameters, surface material type, engine location, and heat intensity are integrated to complete the UAV target feature analysis for photoelectric observation and generate photoelectric observation target feature data.

[0029] Step S16: Based on radar monitoring and sensing spatial trajectory feature data and photoelectric observation target feature data, perform cross-modal coupled sensing feature analysis of UAV target monitoring to generate UAV target monitoring coupled sensing feature data.

[0030] In this embodiment of the invention, a three-dimensional coordinate sequence and trajectory fitting curve are obtained from the radar monitoring and sensing spatial trajectory feature data. A Kalman prediction algorithm is used to predict the spatial coordinates of the UAV for the next five frames, with a prediction step size of 0.1 seconds, a process noise covariance of 0.01, and an observation noise covariance of 0.05, generating radar monitoring and sensing spatial trajectory prediction data. Each predicted coordinate in the radar monitoring and sensing spatial trajectory prediction data is mapped to the image coordinate system of the photoelectric observation target feature data. A 50×50 pixel neighborhood is defined as a candidate matching region, centered on the image position corresponding to the predicted coordinate. The similarity between the photoelectric observation target features and the predicted trajectory features within the candidate matching region is calculated using the Euclidean distance method. A match is considered successful if the distance is less than 5 pixels, generating spatial trajectory prediction-observation neighborhood candidate matching data. Based on the spatial trajectory prediction-observation neighborhood candidate matching data, correlation analysis is performed on each set of successfully matched data to extract the trajectory from the matched data. The association parameters between the features and target features include the deviation between the trajectory coordinates and the target contour center coordinates, and the angle between the trajectory tangent direction and the target orientation. Matching data with a deviation value less than 0.5 meters and an angle less than 10° are considered valid association data, generating candidate association data for UAV cross-modal monitoring. Dynamic target association optimization is performed on the candidate association data of UAV cross-modal monitoring. The Hungarian algorithm is used to match and optimize the valid association data. The optimization objective is to minimize the sum of association deviation values. The number of iterations is set to 10 to obtain the optimal association result, generating dynamic target association data for UAV cross-modal monitoring. Coupled analysis is performed on the dynamic target association data of UAV cross-modal monitoring. The trajectory features monitored by radar are fused with the target features monitored by electro-optical devices. The weight of the trajectory features is set to 0.6, and the weight of the target features is set to 0.4. The comprehensive feature parameters of the UAV target are obtained through weighted fusion, completing the cross-modal coupled sensing feature analysis of UAV target monitoring and generating coupled sensing feature data for UAV target monitoring.

[0031] Furthermore, step S12 includes the following steps: Step S121: Analyze the characteristics of radar monitoring echo signals based on the UAV radar monitoring data in the airspace, and generate radar monitoring echo signal characteristic data; In this embodiment of the invention, radar echo signal feature analysis is performed based on airspace UAV radar monitoring data. The radar echo signal is then filtered using an adaptive Kalman filter algorithm with a filter coefficient set to 0.05. This algorithm filters out electromagnetic interference signals in the airspace environment, ensuring the purity of the echo signal and preventing interference signals from affecting subsequent feature extraction. After filtering, feature extraction is performed on the clean echo signal. Four core features are extracted: peak value, pulse width, rise time, and fall time. The peak value extraction threshold is set to -35dBm, and only peak data exceeding this threshold are extracted. Pulse width extraction uses a fixed time window detection method with a window size of 0.01μs. The duration of the echo signal is determined by scanning the window. Rise time is defined as the time required for the echo signal to rise from 10% to 90% of its peak value, and fall time is defined as the time required for the echo signal to fall from 90% to 10% of its peak value. By accurately detecting these four feature parameters, radar monitoring echo signal feature data is integrated, providing a foundation for subsequent electromagnetic wave sensing spectrum data analysis.

[0032] Step S122: Parse the radar monitoring echo signal characteristic data into radar monitoring electromagnetic wave sensing spectrum data; In this embodiment of the invention, radar monitoring echo signal characteristic data is parsed into radar monitoring electromagnetic wave sensing spectrum data. The parsing logic revolves around the correspondence between echo signal pulse width and electromagnetic wave frequency, explicitly setting the corresponding parameters for pulse width and frequency. Specifically, a pulse width of 0.1 μs corresponds to a frequency of 10 MHz, a pulse width of 0.2 μs corresponds to a frequency of 5 MHz, a pulse width of 0.3 μs corresponds to a frequency of 3.33 MHz, a pulse width of 0.4 μs corresponds to a frequency of 2.5 MHz, and a pulse width of 0.5 μs corresponds to a frequency of 2 MHz. According to this correspondence, each set of pulse width data in the radar monitoring echo signal characteristic data is accurately converted into the corresponding electromagnetic wave frequency data. At the same time, combined with the echo signal peak data, the signal strength corresponding to each set of frequency data is determined, forming complete radar monitoring electromagnetic wave sensing spectrum data. This data can accurately reflect the electromagnetic wave characteristics of the UAV detected by the radar, providing a reliable spectrum basis for subsequent radar monitoring sensing feature analysis.

[0033] Step S123: Analyze the radar monitoring and sensing characteristics based on the radar monitoring electromagnetic wave sensing spectrum data to generate radar monitoring and sensing characteristic data.

[0034] In this embodiment of the invention, radar monitoring and sensing feature analysis is performed based on radar monitoring electromagnetic wave sensing spectrum data. Three core sensing features are extracted from the data: peak frequency, bandwidth, and center frequency. Peak frequency extraction employs a peak detection algorithm with a detection threshold of -30dBm, extracting only peaks above this threshold as valid peak data. Bandwidth calculation uses the 3dB bandwidth method, taking the frequency difference where the signal strength drops by 3dB on either side of the peak frequency; this difference reflects the distribution range of the electromagnetic wave spectrum. Center frequency calculation uses an energy-weighted average method, weighting all frequency points in the spectrum by their signal strength to obtain the center frequency, which reflects the core frequency characteristics of the UAV's electromagnetic wave signal. By extracting and integrating these three core features, radar monitoring and sensing feature analysis is completed, generating radar monitoring and sensing feature data that accurately reflects the spatial presence of the UAV.

[0035] Furthermore, step S13 includes the following steps: Step S131: Extract airspace photoelectric visual data and airspace photoelectric thermal radiation data based on airspace UAV photoelectric monitoring data; In this embodiment of the invention, airspace photoelectric visual data and airspace photoelectric thermal radiation data are extracted from airspace UAV photoelectric monitoring data. A threshold segmentation method is used to segment the airspace UAV photoelectric monitoring data, setting a grayscale threshold of 128 (8-bit grayscale value). Pixel regions with grayscale values ​​greater than 128 correspond to airspace photoelectric visual data, containing visual information such as the visible light outline and surface texture of the UAV. Pixel regions with grayscale values ​​less than or equal to 128 correspond to airspace photoelectric thermal radiation data, containing thermal characteristic information such as the thermal radiation intensity and distribution range of various parts of the UAV. A neighboring pixel verification mechanism is used during segmentation. A 3×3 pixel neighborhood unit is used; if more than 70% of the pixels within a unit meet the corresponding grayscale threshold condition, the unit is determined to be the corresponding data type. This avoids data extraction deviations caused by single-pixel errors. Through the above segmentation and verification operations, airspace photoelectric visual data and airspace photoelectric thermal radiation data are accurately extracted, ensuring the independence and integrity of the two types of data.

[0036] Step S132: Perform visual feature analysis on the contour and texture of the spatial optoelectronic visual data to obtain optoelectronic visual feature data; perform optoelectronic thermal radiation distribution feature analysis on the spatial optoelectronic thermal radiation data to generate optoelectronic thermal radiation distribution feature data. In this embodiment of the invention, the extracted spatial optoelectronic visual data and spatial optoelectronic thermal radiation data are used as objects to conduct visual feature and thermal radiation feature analysis respectively. Visual feature analysis of contour and texture is performed on the spatial optoelectronic visual data. The Canny edge detection algorithm is used to extract the UAV contour features, with a low threshold of 50 and a high threshold of 150. The algorithm scans the optoelectronic visual data to capture the coordinate point sequence of the UAV edge contour, forming a complete UAV contour feature. Texture features are extracted using a gray-level co-occurrence matrix, with the distance of the gray-level co-occurrence matrix set to 1 and the angles to 0°, 45°, 90°, and 135°. Four texture parameters—contrast, correlation, energy, and entropy—are extracted through matrix calculation. Contrast reflects the clarity of the UAV surface texture, correlation reflects the correlation of texture pixels, energy reflects the uniformity of the texture, and entropy reflects the complexity of the texture. The contour coordinate sequence and texture parameters are integrated to obtain optoelectronic visual feature data. Photoelectric and thermal radiation distribution characteristics are analyzed on spatial photoelectric and thermal radiation data. The mean temperature, temperature variance, area of ​​high-temperature region, and center coordinates of high-temperature region are extracted from the thermal radiation image. The mean temperature is the average temperature of all pixels in the thermal radiation image, and the temperature variance is the average of the squares of the temperature of all pixels and the mean temperature. The criterion for judging a high-temperature region is a pixel region with a temperature higher than 35℃. The center coordinates of the high-temperature region are calculated by the arithmetic mean of the coordinates of all pixels in the high-temperature region. The above four thermal radiation parameters are integrated to generate photoelectric and thermal radiation distribution characteristic data.

[0037] Step S133: Map the photoelectric thermal radiation distribution characteristic data to the photoelectric visual characteristic data to perform photoelectric observation characteristic analysis and generate photoelectric observation characteristic data.

[0038] In this embodiment of the invention, photoelectric thermal radiation distribution feature data is mapped to photoelectric visual feature data for photoelectric observation feature analysis. A photoelectric visual feature coordinate system is established, with the upper left corner of the photoelectric visual data image as the origin, the X-axis along the horizontal direction, and the Y-axis along the vertical direction. The coordinates of the UAV outline center in the photoelectric visual feature data are set as the reference point, with coordinates (480, 320) (corresponding to a 640×512 resolution image). The average temperature and the center coordinates of the high-temperature area in the photoelectric thermal radiation distribution feature data are mapped to this photoelectric visual feature coordinate system according to the conversion ratio between image pixels and actual coordinates (1 pixel corresponds to 0.1 meters), so that the center coordinates of the high-temperature area accurately correspond to the UAV outline coordinates, ensuring the spatial consistency between thermal radiation features and visual features. After mapping, the two types of feature data are fused and analyzed. The contour size and texture parameters in the optoelectronic visual features are correlated with the average temperature and high-temperature area in the optoelectronic thermal radiation features. The thermal radiation intensity corresponding to different regions of the UAV contour is clarified. For example, the UAV engine area (high-temperature area) corresponds to the contour area with high texture contrast. Through this correlation analysis, complete optoelectronic observation feature data is integrated to provide accurate optoelectronic feature support for subsequent cross-modal coupled sensing feature analysis, ensuring that the optoelectronic observation features can fully reflect the target characteristics of the UAV.

[0039] Furthermore, step S16 includes the following steps: Step S161: Perform spatial trajectory prediction processing on the radar monitoring and sensing spatial trajectory feature data to generate radar monitoring and sensing spatial trajectory prediction data; In this embodiment of the invention, spatial trajectory prediction processing is performed on the radar monitoring and sensing spatial trajectory feature data. This data includes parameters such as the three-dimensional spatial coordinates, trajectory curvature, and displacement change rate of the UAV for 10 consecutive frames. The three-dimensional coordinates are centered on the radar monitoring device, with the X-axis pointing due east, the Y-axis pointing due north, and the Z-axis perpendicular to the ground and upwards. A Kalman prediction algorithm is used to predict the subsequent trajectory of the UAV, with a prediction step size of 0.1 seconds, 5 prediction frames, a process noise covariance of 0.01, and an observation noise covariance of 0.05. The algorithm fits and extrapolates the historical trajectory coordinates, calculating the X, Y, and Z three-dimensional values ​​of the predicted coordinates for each frame. Simultaneously, combined with the trajectory curvature and displacement change rate parameters, the subsequent flight direction and speed changes of the UAV are predicted. The predicted coordinates and related trajectory parameters are integrated to form radar monitoring and sensing spatial trajectory prediction data, ensuring that the predicted trajectory is consistent with the actual flight trend of the UAV.

[0040] Step S162: Map the radar monitoring and sensing spatial trajectory prediction data to the photoelectric observation target feature data to perform observation neighborhood candidate matching analysis for spatial trajectory prediction, and generate spatial trajectory prediction-observation neighborhood candidate matching data; In this embodiment of the invention, radar monitoring and sensing spatial trajectory prediction data is mapped to photoelectric observation target feature data for spatial trajectory prediction observation neighborhood candidate matching analysis. First, a unified coordinate mapping system is established. The three-dimensional spatial coordinates in the radar monitoring and sensing spatial trajectory prediction data are converted into two-dimensional coordinates of the photoelectric observation image according to a preset conversion rule. The conversion rule is that the ratio of the X-axis spatial coordinate to the image X-axis coordinate is 1 meter to 10 pixels, the ratio of the Y-axis spatial coordinate to the image Y-axis coordinate is 1 meter to 10 pixels, and the Z-axis coordinate corresponds to the image grayscale value (the higher the Z-axis, the larger the grayscale value). A 50×50 pixel neighborhood is defined as the candidate matching region centered on each converted two-dimensional prediction coordinate. The similarity between the photoelectric observation target features and the predicted trajectory features within the candidate matching region is calculated using the Euclidean distance method. A distance threshold of 5 pixels is set. When the calculated Euclidean distance is less than 5 pixels, the photoelectric target within the candidate region is determined to match the predicted trajectory. The predicted trajectory parameters and photoelectric target feature parameters corresponding to the match are recorded. All successfully matched parameter pairs are integrated to generate spatial trajectory prediction-observation neighborhood candidate matching data, laying the foundation for subsequent candidate association analysis.

[0041] Step S163: Perform candidate association analysis for UAV cross-modal monitoring based on spatial trajectory prediction-observation neighborhood candidate matching data to generate UAV cross-modal monitoring candidate association data; In this embodiment of the invention, candidate association analysis for UAVs in cross-modal monitoring is performed based on candidate matching data from spatial trajectory prediction and observation neighborhoods. Each pair of matching parameters in the candidate matching data is extracted, with a focus on analyzing two core association parameters: the deviation between the trajectory coordinates and the target contour center coordinates, and the angle between the trajectory tangent direction and the target orientation. A deviation threshold of 0.5 meters and an angle threshold of 10° are set. Each pair of matching parameters is validated. When the deviation is less than 0.5 meters and the angle is less than 10°, the pair is considered a valid association, and the corresponding radar trajectory prediction features and photoelectric target features are retained. When the deviation is greater than or equal to 0.5 meters or the angle is greater than or equal to 10°, the pair is considered an invalid association and is discarded. Simultaneously, the valid association data is deduplicated. If multiple sets of matching data correspond to the same UAV target, only the set with the smallest deviation and angle is retained. Through the above validation and deduplication operations, all valid association data are integrated to generate candidate association data for UAV cross-modal monitoring, providing a high-quality association foundation for subsequent dynamic target association optimization.

[0042] Step S164: Optimize the dynamic target association of UAVs in cross-modal monitoring based on the cross-modal candidate association data of UAV monitoring, and generate UAV cross-modal dynamic target association data; In this embodiment of the invention, dynamic target association optimization for UAV monitoring across modalities is performed based on cross-modal candidate association data. The Hungarian algorithm is used to match and optimize the effective association data. The optimization objective is to minimize the sum of deviations of all association data. The number of iterations is set to 10. In each iteration, the matching correspondence of the association data is adjusted, and the sum of deviations is calculated until the optimal matching result with the minimum sum of deviations is obtained after 10 iterations. During the optimization process, the angle between the trajectory tangent direction and the target orientation is simultaneously verified to ensure that the optimized association data still meets the requirement of an angle less than 10°. If the angle of a certain set of data is greater than or equal to 10° after optimization, the matching relationship of that set of data is readjusted until all association data meet the preset standard. Through the above optimization operations, accurate dynamic target association results are obtained. The optimized association parameter pairs are integrated to generate dynamic target association data for UAV cross-modal monitoring, ensuring accurate correspondence between radar trajectory features and photoelectric target features.

[0043] Step S165: Perform cross-modal coupled sensing feature analysis on the dynamic target association data of UAV cross-modal monitoring to generate UAV target monitoring coupled sensing feature data.

[0044] In this embodiment of the invention, radar trajectory features and optoelectronic target features in dynamic target association data are coupled and analyzed. The radar trajectory feature weight is set to 0.6, and the optoelectronic target feature weight is set to 0.4. A weighted fusion algorithm is used to fuse the two types of features. For radar trajectory features, three core parameters are selected: trajectory curvature, displacement rate of change, and predicted coordinates. For optoelectronic target features, three core parameters are selected: UAV size, surface texture parameters, engine position, and heat intensity. After standardizing the parameters of each type of feature, they are weighted and summed according to the set weights to obtain the comprehensive feature parameters of the UAV target. Simultaneously, the spatial location information of the radar trajectory is correlated with the detailed feature information of the optoelectronic target to clarify the detailed features such as surface texture and heat intensity of the UAV at different spatial locations. By integrating the comprehensive feature parameters and the associated information, cross-modal coupled sensing feature analysis is completed, generating UAV target monitoring coupled sensing feature data. This ensures that the data can comprehensively and accurately reflect the spatial state and target characteristics of the UAV.

[0045] Furthermore, step S2 includes the following steps: Step S21: Based on the UAV target monitoring coupled sensing feature data, perform UAV spatial energy peak migration feature analysis to generate UAV spatial energy peak migration feature data; In this embodiment of the invention, UAV spatial energy peak migration characteristic analysis is performed based on UAV target monitoring coupled sensing feature data. The UAV target monitoring coupled sensing feature data includes comprehensive UAV feature parameters, spatial location information, and detailed feature information. Spatial energy-related parameters of the UAV are extracted from this data, with a focus on two core energy parameters: UAV engine thermal radiation energy and radar reflection energy. A time-series feature analysis method is used to process these two types of energy parameters, setting the time-series analysis time window to 1 second and collecting energy parameters every 0.1 seconds to form 10 consecutive sets of time-series energy data. Spatial dynamic energy peak identification is performed using the time-series energy data. An energy peak identification threshold of 500W is set, and the spatial coordinates corresponding to energy data exceeding this threshold are extracted to determine the location of the UAV's spatial dynamic energy peak. Then, the spatial distance and time difference between adjacent energy peaks are calculated to obtain the energy peak migration speed and migration direction. Integrating parameters such as energy peak location, migration speed, and migration direction generates UAV spatial energy peak migration characteristic data, clearly reflecting the dynamic changes in the UAV's energy distribution.

[0046] Step S22: Analyze the dynamic behavior characteristics of the UAV based on the spatial energy peak migration characteristic data, and generate UAV dynamic behavior characteristic data; In this embodiment of the invention, UAV dynamic behavior characteristics are analyzed based on UAV spatial energy peak migration feature data to accurately capture dynamic changes during UAV flight and support subsequent flight behavior pattern analysis. Parameters such as energy peak migration speed, migration direction, and energy peak change rate are extracted from the UAV spatial energy peak migration feature data. These parameters are then combined with spatial trajectory parameters from UAV target monitoring coupled sensing feature data to construct a dynamic behavior analysis model. The model calculates three core dynamic parameters of the UAV: ​​acceleration, angular velocity, and turning radius. Acceleration is calculated using the ratio of the difference in energy peak migration speed between two adjacent frames to the time interval, which is set to 0.1 seconds. Angular velocity is calculated using the ratio of the angle between migration directions between two adjacent frames to the time interval, with the angle calculated using the vector dot product method. The turning radius is calculated based on the trajectory curvature and migration speed; the turning radius is equal to the ratio of the square of the migration speed to the trajectory curvature. Simultaneously, the power output status of the UAV is determined by combining the peak energy change rate. When the peak energy change rate is greater than 100W / s, it is determined to be enhanced power; when it is less than -100W / s, it is determined to be weakened power; and when it is equal to 0, it is determined to be stable power. By integrating dynamic parameters and power output status, dynamic behavior characteristic data of the UAV is generated, which comprehensively reflects the dynamic characteristics of the UAV during flight.

[0047] Step S23: Perform UAV visual structure state feature analysis based on UAV target monitoring coupled perception feature data to generate UAV visual structure state feature data; In this embodiment of the invention, UAV visual structural state feature analysis is performed based on UAV target monitoring coupled sensing feature data. Detailed features related to photoelectric observation are extracted from the UAV target monitoring coupled sensing feature data, with a focus on visually relevant parameters such as UAV outline dimensions, surface texture parameters, engine position, and heat intensity. Real-time analysis of the UAV outline dimensions is performed, calculating the real-time values ​​of the UAV's length, width, and height using outline coordinate sequences. A pixel-to-actual-distance conversion ratio of 1 pixel to 0.1 meters is used to ensure the accuracy of the dimensional data. Contrast and energy parameters in the surface texture are analyzed to determine the integrity of the UAV's surface structure. A contrast ratio below 20 indicates surface damage, while an energy ratio above 0.8 indicates uniform surface texture. Engine position and heat intensity are monitored, recording the positional changes and temperature fluctuations of the engine's high-temperature areas. A positional shift of more than 5 pixels in the high-temperature area indicates abnormal engine operation, and a temperature fluctuation exceeding 5°C indicates unstable power output. Integrating the above visual structure-related analysis results and parameters generates UAV visual structural state feature data, comprehensively reflecting the UAV's structural integrity and operational status.

[0048] Step S24: Analyze the UAV flight behavior pattern using UAV dynamic behavior feature data and UAV visual structural state feature data to generate UAV flight behavior pattern data. In this embodiment of the invention, UAV flight behavior pattern analysis is performed using UAV dynamic behavior characteristic data and UAV visual structural state characteristic data to establish a flight behavior pattern analysis system. Three core flight modes are defined: uniform straight-line flight, uniform turning flight, and acceleration / deceleration flight, with clear criteria for each mode. The criteria for uniform straight-line flight are: absolute acceleration less than 0.5 m / s², angular velocity less than 5° / s, and stable visual structural state; the criteria for uniform turning flight are: absolute acceleration less than 0.5 m / s², angular velocity greater than or equal to 5° / s and less than 15° / s, and stable turning radius; the criteria for acceleration / deceleration flight are: absolute acceleration greater than or equal to 0.5 m / s², and absolute rate of change of peak energy greater than 100 W / s. The dynamic behavior characteristic data and visual structure state characteristic data are jointly verified. Dynamic parameters such as acceleration, angular velocity, and turning radius are substituted into the judgment criteria along with visual parameters such as visual structure stability and power output state to determine the current flight behavior mode of the UAV. At the same time, parameters such as the duration and switching time of the flight mode are recorded and integrated to form UAV flight behavior mode data, ensuring that the judgment of flight behavior mode is accurate and reliable and closely matches the actual flight state of the UAV.

[0049] Step S25: Analyze the propagation characteristics of UAV flight behavior based on the UAV flight behavior pattern data to generate UAV flight behavior propagation characteristic data.

[0050] In this embodiment of the invention, UAV flight behavior propagation characteristics are analyzed based on UAV flight behavior pattern data. Core parameters such as flight mode type, duration, flight speed, and flight direction are extracted from the UAV flight behavior pattern data. Combined with UAV spatial location information, a flight behavior propagation analysis model is constructed. A propagation analysis spatial grid is set with a grid spacing of 10 meters, dividing the UAV flight trajectory into several trajectory nodes. Each node corresponds to a set of flight behavior pattern parameters and spatial coordinates. The flight behavior similarity between adjacent trajectory nodes is calculated using the cosine similarity method, with a similarity threshold of 0.8. When the similarity is greater than 0.8, it is determined to be continuous propagation of the same flight behavior; when it is less than or equal to 0.8, it is determined to be a change in flight behavior. Simultaneously, the propagation speed and range of flight behavior at different spatial locations are analyzed. Combined with the UAV's visual structural state and dynamic characteristics, the subsequent propagation trend of flight behavior is predicted. By integrating trajectory node parameters, behavior similarity, propagation speed, propagation range, and predicted trends, UAV flight behavior propagation characteristic data is generated, comprehensively reflecting the spatial propagation laws of UAV flight behavior and providing reliable behavioral dimension support for subsequent threat assessment and countermeasure strategy formulation.

[0051] Furthermore, step S21 includes the following steps: Step S211: Perform time-series feature analysis on the UAV target monitoring coupled sensing feature data to generate UAV target monitoring coupled sensing time-series feature data; In this embodiment of the invention, temporal feature analysis is performed on the coupled sensing feature data for UAV target monitoring. This data includes comprehensive feature parameters resulting from the weighted fusion of radar and optoelectronic data, UAV spatial location information, and detailed feature information. Two types of core energy-related parameters are extracted: radar reflection energy parameters and UAV engine thermal radiation energy parameters. Both types of parameters originate from the comprehensive feature data after cross-modal coupling and are consistent with radar monitoring and sensing features and optoelectronic observation target features. A sliding time window method is used for temporal feature analysis, with a time window length of 1 second and a sliding step of 0.1 seconds. Ten sets of energy parameter data are collected within each time window, synchronously correlated with parameters such as the UAV's three-dimensional spatial coordinates and trajectory curvature at the corresponding time points. Through time window scanning, the continuously collected energy and spatial parameters are arranged chronologically to form a continuous temporal data sequence. Each temporal data point corresponds to a set of energy parameters, spatial coordinates, and trajectory parameters. Integrating all temporal data sequences generates UAV target monitoring coupled sensing temporal feature data, clarifying the changing pattern of energy parameters over time.

[0052] Step S212: Based on the UAV target monitoring coupled sensing time series feature data, perform UAV spatial dynamic energy peak identification processing to generate UAV spatial dynamic energy peak data; In this embodiment of the invention, UAV spatial dynamic energy peak identification processing is performed based on the UAV target monitoring coupled sensing time-series feature data to accurately capture the peak position and changes of UAV energy distribution, supporting subsequent energy peak migration feature analysis. Time-series sequences of two core energy parameters—radar reflection energy and engine thermal radiation energy—are extracted from the time-series feature data. A unified energy peak identification threshold is set, with the radar reflection energy peak identification threshold set at 450W and the engine thermal radiation energy peak identification threshold set at 500W. Only energy data exceeding the corresponding thresholds in the time-series are extracted. Peak values ​​are determined for the extracted high-energy data. The determination criterion is that a certain energy data is greater than the values ​​of the three adjacent energy data sets before and after it; that is, the energy data is determined to be an energy peak. The energy value, occurrence time, and corresponding UAV three-dimensional spatial coordinates (X, Y, Z) of the energy peak are recorded. The three-dimensional coordinates are consistent with the coordinate system set in step S14, with the radar monitoring equipment as the origin, the X-axis pointing due east, the Y-axis pointing due north, and the Z-axis perpendicular to the ground and upward. At the same time, the radar reflection energy peak and the engine thermal radiation energy peak are distinguished, and the types of the two types of energy peaks are clearly marked in the data to avoid confusion. The energy values, occurrence time, spatial coordinates and type information of all energy peaks are integrated to generate UAV spatial dynamic energy peak data to ensure the accuracy of energy peak identification.

[0053] Step S213: Perform UAV spatial energy peak migration characteristic analysis on the UAV spatial dynamic energy peak data to generate UAV spatial energy peak migration characteristic data.

[0054] In this embodiment of the invention, the spatial energy peak migration characteristics of UAVs are analyzed using spatial dynamic energy peak data. Continuous peak information of the same type of energy peaks (radar reflection energy peaks or engine thermal radiation energy peaks) is extracted from the spatial dynamic energy peak data and arranged in chronological order of appearance. The spatial migration distance and time difference between two adjacent energy peaks are calculated. The spatial migration distance is calculated using the Euclidean distance formula for two-point three-dimensional coordinates, and the time difference is the difference in the appearance times of the two energy peaks, uniformly retained to 0.1 seconds. Based on the spatial migration distance and time difference, the energy peak migration speed is calculated, which is equal to the ratio of the spatial migration distance to the time difference. The energy peak migration direction is determined using a vector calculation method. A migration vector is constructed, with the spatial coordinates of the previous energy peak as the starting point and the spatial coordinates of the next energy peak as the ending point. The direction of the vector is the energy peak migration direction, expressed as an azimuth angle, ranging from 0-360°, consistent with the azimuth angle setting for radar detection. Simultaneously, the energy difference between adjacent energy peaks is calculated to analyze the changing trend of energy peak intensity. The energy difference is the difference between the energy values ​​of the subsequent energy peak and the preceding energy peak; a positive value indicates energy enhancement, and a negative value indicates energy weakening. By integrating the energy peak migration speed, migration direction, spatial migration distance, time difference, and energy difference, spatial energy peak migration characteristic data of the UAV is generated, comprehensively reflecting the spatial migration law of the UAV's energy peaks and providing reliable energy dimension support for subsequent dynamic behavior analysis.

[0055] Furthermore, as an embodiment of the present invention, reference is made to... Figure 2 As shown, Figure 1 A detailed flowchart illustrating the implementation steps of step S3 is provided in this embodiment. Step S3 includes: Step S31: Perform spatial threat multi-track grid cell analysis on the UAV flight behavior propagation characteristic data to generate spatial threat multi-track grid cell data; In this embodiment of the invention, spatial threat multi-track grid cell analysis of UAV flight behavior propagation characteristic data within a protected area is performed to clarify the specific range and boundaries of the protected area. The protected area is defined as a square region with sides of 1000 meters. A unified two-dimensional plane coordinate system is established with the center point of the protected area as the origin, where the X-axis points due east and the Y-axis points due north. The units of the coordinate system are consistent with the actual spatial distance to ensure accurate spatial positioning. Subsequently, according to standardized grid division rules, the entire protected area is divided into uniform grid cells, with the side length of each grid cell strictly set to 10 meters. Based on this, the entire protected area is calculated to be divided into 100×100 grid cells. Each grid cell is assigned a unique grid number, arranged sequentially from left to right and top to bottom, facilitating accurate marking and management of the threat status of each grid cell. Next, core parameters such as drone flight trajectory nodes, flight speed, flight direction, and propagation range are comprehensively extracted from the drone flight behavior propagation characteristic data. Flight trajectory nodes contain the spatial coordinates of the drone at different points in time; flight speed and flight direction reflect the drone's real-time motion state; and propagation range reflects the spatial coverage of the drone's flight behavior. The three-dimensional spatial coordinates of all trajectory nodes are transformed to a preset two-dimensional plane coordinate system of the protection area using coordinate transformation rules. Ignoring the influence of Z-axis height and retaining only the X and Y axis coordinates, the grid cell containing each trajectory node is accurately determined. Grid cells containing trajectory nodes are directly marked as threat grid cells, ensuring that existing drone threats can be accurately located. Simultaneously, combining the propagation speed and flight direction from the drone flight behavior propagation characteristics, a trajectory prediction method is used to calculate the grid cells that the drone's subsequent flight trajectory may cover. These predicted covered grid cells are also marked as threat grid cells, achieving early prediction and marking of potential threats. Finally, the core information of each threat grid cell is recorded in detail, including the grid number, parameters of all trajectory nodes, flight behavior patterns of the corresponding nodes, and predicted coverage time. The relevant information of all threat grid cells is systematically integrated to form complete spatial threat multi-track grid cell data, realizing the grid-based division of UAV threats within the protected area and providing a precise spatial carrier for subsequent threat analysis.

[0056] Step S32: Analyze the space threat diffusion trend based on the space threat multi-track grid cell data to generate space threat diffusion trend data; In this embodiment of the invention, spatial threat diffusion trend analysis is performed based on multi-track grid cell data of spatial threats. From the multi-track grid cell data, the grid number, trajectory node parameters, flight behavior pattern, and predicted coverage time of each threat grid cell are extracted one by one to ensure completeness and accuracy of the extracted information. All threat grid cells are sorted according to their chronological order of appearance, with a fixed time interval of 0.1 seconds. Using this time interval as the unit, the spatial distribution range of each threat grid cell at each time point is characterized, forming a time-space corresponding threat distribution sequence, clearly presenting the coverage of threats at different time points. A neighborhood diffusion analysis method is used, taking each current threat grid cell as the center, comprehensively analyzing its eight adjacent grid cells, namely the grid cells in the four positive directions (up, down, left, right) and four diagonal directions, to clarify the possibility of each adjacent grid cell being covered by threat diffusion. Through a fixed calculation logic, the time for the threat to diffuse to each adjacent grid cell is calculated. The calculation logic for the diffusion time is the ratio of the grid cell side length to the actual flight speed of the UAV, where the UAV flight speed is directly taken from the actual flight speed in the UAV flight behavior propagation characteristic data, ensuring the accuracy of the calculation results. Simultaneously, by combining the flight direction from the propagation characteristics of UAV flight behavior, the main direction of threat diffusion is identified. Adjacent grid cells along the flight direction are designated as priority diffusion grids, with their diffusion time reduced by 20% compared to the original calculation results. Adjacent grid cells in non-flight directions retain their original diffusion time, thus reflecting the directional differences in threat diffusion. Through the above series of detailed analysis operations, the expansion patterns of threat grid cells, the diffusion speed in different directions, the core diffusion direction, and the threat coverage at different time points are clarified. This dynamic diffusion information is systematically integrated to form spatial threat diffusion trend data, clearly reflecting the dynamic diffusion process of UAV threats within the protected area.

[0057] Step S33: Perform threat indicator feature analysis on the spatial threat diffusion trend data to generate spatial diffusion trend threat indicator feature data; In this embodiment of the invention, threat indicator feature analysis is performed on spatial threat diffusion trend data to identify three core threat indicators: threat diffusion speed, threat coverage density, and threat duration. Clear and fixed calculation rules and anomaly judgment criteria are established for each indicator to ensure the standardization and consistency of indicator analysis. Specifically, the threat diffusion speed is calculated as the ratio of the maximum distance between two adjacent time points to the time interval, which is strictly set to 0.1 seconds. This indicator directly reflects the speed of threat diffusion; a larger value indicates a faster diffusion speed. Threat coverage density is calculated as the ratio of the number of threat grid cells at a certain time point to the total number of grid cells in the protected area; a larger ratio indicates a wider threat coverage and a greater spatial impact. Threat duration is determined by recording the time from when each threat grid cell is marked as a threat until the threat completely disappears. The criterion for threat disappearance is that the UAV's flight trajectory completely leaves the grid cell, and trajectory prediction determines that subsequent flight trajectories will no longer cover the grid cell, ensuring the accuracy of the threat duration recording. Subsequently, for each time point and each threat grid cell in the spatial threat diffusion trend data, the specific values ​​of the three core threat indicators were calculated one by one according to the aforementioned calculation rules. The specific values ​​of each indicator were recorded in detail. Simultaneously, anomalies in each indicator were marked against the anomaly judgment criteria. Threat diffusion speeds greater than 10 m / s, threat coverage densities greater than 0.3, and threat durations greater than 10 seconds were all judged as anomalies. The corresponding time point, grid cell, and anomalous value for each anomalous indicator were clearly marked. Finally, all indicator values, anomaly markers, and corresponding time point and grid cell information were comprehensively integrated to form complete spatial diffusion trend threat indicator characteristic data, enabling quantitative analysis of threat diffusion trends.

[0058] Step S34: Analyze the associated threat characteristics of the spatial diffusion trend based on the spatial diffusion trend threat index characteristic data, and generate spatial diffusion trend associated threat characteristic data; In this embodiment of the invention, threat characteristic analysis of the diffusion trend is performed based on the spatial diffusion trend threat indicator feature data. Time-series data of three core indicators—threat diffusion speed, threat coverage density, and threat duration—are extracted from the spatial diffusion trend threat indicator feature data. The time-series data retains the indicator values ​​at each time point in chronological order, ensuring data continuity and integrity. The time-series data of the three types of indicators are precisely aligned in chronological order to ensure that the values ​​of the three types of indicators at each time point correspond, laying the foundation for subsequent correlation analysis. A correlation analysis method is used to systematically analyze the correlation between the three types of indicators. A fixed correlation judgment threshold of 0.7 is set. When the correlation coefficient between two indicators is greater than 0.7, a strong correlation is determined; when it is less than or equal to 0.7, a weak correlation is determined. The specific analysis logic is as follows: First, analyze the correlation between threat spread speed and threat coverage density. When the threat spread speed increases, observe the changing trend of threat coverage density simultaneously. If both show a synchronous increase, it indicates that the accelerated threat spread speed will directly lead to a rapid expansion of the threat coverage area, and there is a strong positive correlation between the two. Second, analyze the correlation between threat spread speed and threat duration. If the threat spread speed increases and the threat duration shortens, it indicates that the threat is in a rapid transit state, and the impact time on the protected area is short. If the threat spread speed decreases and the threat duration lengthens, it indicates that the threat is in a lingering spread state, and the impact time on the protected area is longer, and there is an inverse correlation between the two. Third, analyze the correlation between threat coverage density and threat duration. If the threat coverage density increases and the threat duration lengthens, it indicates that the threat is in a continuous spread state and the range is constantly expanding, and the threat level to the protected area is constantly increasing, and there is a strong positive correlation between the two. Simultaneously, by combining flight patterns from UAV flight behavior propagation characteristic data, the correlation patterns of the three types of indicators under different flight modes are analyzed in depth. For example, when the UAV is in a uniform straight-line flight mode, the threat diffusion speed remains stable, the threat coverage density shows a uniform distribution, and the indicator correlation is relatively stable; when in a uniform turning flight mode, the threat diffusion speed fluctuates slightly, the threat coverage density shows a non-uniform distribution, and the indicator correlation changes with the turning direction. By comprehensively integrating the above indicator correlations, correlation patterns under different flight modes, and corresponding flight mode information, spatial diffusion trend-related threat characteristic data is generated, comprehensively reflecting the inherent correlation between threat indicators.

[0059] Step S35: Analyze the threat intensity distribution characteristics of the diffusion trend based on the spatial diffusion trend threat index feature data and the spatial diffusion trend related threat feature data, and generate spatial diffusion trend threat intensity distribution feature data; In this embodiment of the invention, threat intensity distribution characteristics of the diffusion trend are analyzed based on spatial diffusion trend threat index feature data and spatial diffusion trend-related threat feature data. This quantifies the threat intensity at different locations within the protected area, providing a basis for subsequent threat propagation field analysis. A standardized threat intensity calculation model is established, defining threat intensity as a weighted sum of three core indicators: threat diffusion speed, threat coverage density, and threat duration. The weight of threat diffusion speed is set to 0.4, the weight of threat coverage density is set to 0.3, and the weight of threat duration is set to 0.3. The weighted sum is calculated by direct multiplication and summation: Threat Intensity = (Threat Diffusion Speed ​​× 0.4) + (Threat Coverage Density × 0.3) + (Threat Duration × 0.3). This weight allocation highlights the core impact of threat diffusion speed on overall threat intensity while also considering the influence of threat coverage and duration, ensuring the rationality of the threat intensity calculation. Subsequently, for each grid cell within the protected area, the values ​​of the three core indicators corresponding to that grid cell in the spatial diffusion trend threat indicator feature data were individually substituted into the threat intensity calculation model to accurately calculate the threat intensity value of each grid cell, ensuring the accuracy and relevance of the calculation results. Simultaneously, the calculated threat intensity values ​​were precisely corrected by incorporating the indicator correlation relationships in the spatial diffusion trend related threat indicator feature data. If a grid cell's three indicators showed a strong correlation and were all in an abnormal state, it indicated that the threat level of that grid cell was significantly higher than other grid cells, and its threat intensity value was increased by 30% to reflect the impact of abnormal indicator correlations on threat intensity. After calculation and correction, all grid cells were divided into three distinct threat levels based on their threat intensity values: grid cells with a threat intensity value less than 2 were classified as low threat level, grid cells with a threat intensity value between 2 and 5 were classified as medium threat level, and grid cells with a threat intensity value greater than 5 were classified as high threat level. The threat intensity level, specific value, and corresponding time point of each grid cell were clearly marked to ensure that the threat intensity information of each grid cell was clearly traceable. Finally, the threat intensity information of all grid cells is systematically integrated to generate spatial diffusion trend threat intensity distribution characteristic data, clearly presenting the spatial distribution pattern of threat intensity within the protected area.

[0060] Step S36: Analyze the UAV threat propagation field characteristics of the spatial threat diffusion trend data using the spatial diffusion trend threat intensity distribution characteristic data, and generate UAV threat propagation field characteristic data.

[0061] In this embodiment of the invention, the spatial threat diffusion trend data is analyzed using spatial diffusion trend threat intensity distribution characteristic data to perform UAV threat propagation field characteristic analysis. Using the grid cells of the protected area as the basic carrier, the threat intensity level and specific value of each grid cell in the spatial diffusion trend threat intensity distribution characteristic data are precisely correlated with dynamic information such as diffusion speed, diffusion direction, and coverage area in the spatial threat diffusion trend data. This ensures a one-to-one correspondence between threat intensity and dynamic diffusion information, constructing a complete UAV threat propagation field model. Based on a defined two-dimensional coordinate system of the protected area, the model clearly marks the threat intensity, diffusion speed, and diffusion direction of each grid cell, intuitively presenting the distribution and diffusion status of threats within the protected area. Subsequently, the threat propagation field model is dynamically analyzed, depicting the overall state of the threat propagation field at each time point in chronological order. The center position and diffusion radius of the threat propagation field are accurately calculated. The center position of the threat propagation field is the arithmetic mean of the coordinates of all high-threat-level grid cells, accurately locating the core area of ​​the threat; the diffusion radius of the threat propagation field is the maximum distance from the center position of the threat propagation field to the outermost threat grid cell, clearly reflecting the spatial coverage range of the threat. Simultaneously, by combining flight patterns from UAV flight behavior propagation characteristic data, the dynamic changes of the threat propagation field under different flight modes are analyzed in depth. For example, when the UAV is in a uniform turning flight mode, the center position of the threat propagation field shifts synchronously with the turning direction, while the diffusion radius remains relatively stable. When in an accelerating flight mode, the diffusion radius of the threat propagation field increases rapidly, and the center position moves continuously with the flight direction. When in a uniform straight-line flight mode, the center position, diffusion radius, and diffusion speed of the threat propagation field all remain stable. Finally, the core parameters, dynamic changes, center position, diffusion radius, and threat information of each grid cell of the threat propagation field model are comprehensively integrated to generate UAV threat propagation field characteristic data. This data comprehensively and accurately reflects the propagation patterns and intensity distribution of UAV threats within the protected area, providing reliable support for countermeasure strategy analysis.

[0062] Furthermore, step S31 includes the following steps: Step S311: Perform UAV multi-track node feature analysis based on UAV flight behavior propagation feature data to generate UAV multi-track node feature data; In this embodiment of the invention, UAV multi-track node feature analysis is performed based on UAV flight behavior propagation feature data. The UAV flight behavior propagation feature data includes parameters such as flight mode type, duration, flight speed, flight direction, and trajectory nodes. All node information of the UAV flight trajectory is extracted from this data, and the trajectory nodes are sorted chronologically. A node acquisition interval of 0.1 seconds is set, with one trajectory node every 0.1 seconds, and each node is associated with a unique timestamp. Feature extraction is performed on each trajectory node, focusing on three types of parameters: first, the node's spatial coordinates, with the radar monitoring device as the origin, the X-axis pointing due east, the Y-axis pointing due north, and the Z-axis perpendicular to the ground upwards, clearly defining the X, Y, and Z three-dimensional coordinates of each node; second, the node's corresponding flight behavior parameters, including the flight speed, flight direction, and flight mode at that node. The flight speed is rounded to the nearest integer, the flight direction is represented by an azimuth angle (0-360°), and the flight mode corresponds to one of three types: uniform straight-line, uniform turning, and acceleration / deceleration; third, the node's energy association parameters, taken from spatial energy peak migration feature data, recording the energy peak type and energy value corresponding to each node. By integrating the spatial coordinates, flight behavior parameters, energy correlation parameters, and timestamps of all trajectory nodes and arranging them in chronological order to form an ordered node dataset, multi-track node feature data of UAVs is generated.

[0063] Step S312: Perform threat multi-track node feature analysis based on UAV multi-track node feature data to generate UAV threat multi-track node feature data; In this embodiment of the invention, threat multi-track node feature analysis is performed based on UAV multi-track node feature data to screen out threatening track nodes, achieving accurate differentiation between threatening and non-threatening nodes and supporting subsequent spatial threat grid cell analysis. First, the criteria for determining threat nodes are clarified. Combining the UAV countermeasure analysis requirements with fused radar-electro-optical data, two types of judgment conditions are set. A node is determined to be a threat node if either condition is met: 1) the node's flight speed is greater than 8 m / s, and its flight direction points towards the core area of ​​the protected zone (the core area of ​​the protected zone is defined as a circular area with a radius of 100 meters centered on the center point of the protected zone); 2) the node's corresponding energy peak value is greater than 480W (a comprehensive threshold integrating radar reflection energy and engine thermal radiation energy), and the energy peak change rate is greater than 80W / s. For each node in the UAV multi-track node feature data, the above judgment conditions are checked one by one. Nodes that meet the conditions are marked as threatening multi-track nodes, and non-threatening nodes that do not meet the conditions are eliminated. Further feature enhancements were made to the marked multi-track threat nodes. The straight-line distance between each threat node and the core area of ​​the protected zone was calculated using the three-dimensional Euclidean distance formula. The duration of each threat node (the time difference between its appearance and the appearance of the next non-threat node) was also recorded, with durations accumulated in 0.1-second increments. The spatial coordinates, flight behavior parameters, energy correlation parameters, timestamps, distances to the core area, and durations of all multi-track threat nodes were integrated to generate UAV threat multi-track node feature data, ensuring comprehensive feature information for each threat node.

[0064] Step S313: Perform spatial threat multi-track grid cell analysis on the protected area based on the multi-track node feature data of UAV threats, and generate spatial threat multi-track grid cell data.

[0065] In this embodiment of the invention, spatial threat multi-track grid cell analysis of the protected area is performed based on the multi-track node feature data of UAV threats. First, the scope of the protected area and the grid division rules are defined. The protected area is set as a square area with a side length of 1000 meters. A two-dimensional plane coordinate system is established with the center point of the protected area as the origin (consistent with the grid system in subsequent steps S3). The X-axis points due east, and the Y-axis points due north. The protected area is divided into uniform grid cells, with a side length of 10 meters. The entire protected area is divided into 100×100 grid cells, each assigned a unique grid number, arranged sequentially from left to right and top to bottom. The three-dimensional coordinates of each threat node in the multi-track node feature data of UAV threats are transformed to this two-dimensional plane coordinate system (ignoring the influence of Z-axis height, only retaining X and Y coordinates). The grid cell containing each threat node is determined, and the grid cell containing the threat node is marked as the threat grid cell. Simultaneously, by combining the flight speed, flight direction, and duration of each threat node, the potential trajectory nodes that the threat node may subsequently extend to are predicted. The coordinates of the predicted trajectory nodes are calculated, and their respective grid cells are determined and marked as threat grid cells. The number of each threat grid cell, the parameters of the included threat nodes (spatial coordinates, flight mode, energy value), the predicted coverage time, and the duration of the threat nodes are recorded. The relevant information of all threat grid cells is integrated to generate multi-track grid cell data of space threats, ensuring accurate grid cell division and complete threat information.

[0066] Furthermore, step S4 includes the following steps: Step S41: Perform drone countermeasure state characteristic analysis on the drone threat propagation field characteristic data to generate drone countermeasure state characteristic data; In this embodiment of the invention, UAV threat propagation field characteristic data is analyzed for UAV countermeasure status characteristics. The UAV threat propagation field characteristic data includes parameters such as threat propagation field model parameters, dynamic change patterns, center position, diffusion radius, and threat intensity level of each grid cell. Three types of core countermeasure-related characteristic parameters are extracted from this data: First, core threat characteristics, including the center coordinates of the threat propagation field, diffusion radius, and distribution range of high-threat-level grid cells. The center coordinates of the threat propagation field follow the established two-dimensional coordinate system of the protection area, and the diffusion radius is recorded according to the actual calculated value. Second, dynamic threat characteristics, including threat diffusion speed and threat intensity change rate. The threat intensity change rate is calculated as the ratio of the difference in average threat intensity of high-threat-level grid cells at two adjacent time points to the time interval, which is set to 0.1 seconds. Third, UAV-related characteristics, taken from flight behavior propagation characteristic data and cross-modal coupled sensing characteristic data, including UAV flight speed, flight mode, energy peak, and target size. Correlation analysis is performed on three types of characteristic parameters to clarify the correspondence between threat status and the drone's own status. For example, when the threat level is high and the spread speed is greater than 10 m / s, the countermeasure priority is determined to be the highest. All characteristic parameters and correlation analysis results are integrated to generate drone countermeasure status characteristic data to ensure that the countermeasure status analysis is comprehensive and accurate.

[0067] Step S42: Utilize the preset UAV prior countermeasure strategy library to perform UAV countermeasure state feature analysis on UAV countermeasure strategy matching feature data, and generate UAV countermeasure strategy matching feature data. In this embodiment of the invention, a pre-defined UAV prior countermeasure strategy library is used to perform UAV countermeasure state feature analysis on UAV countermeasure strategy matching, achieving a preliminary correspondence between countermeasure strategies and threat states, supporting the design of subsequent countermeasure action space. The pre-defined UAV prior countermeasure strategy library includes three core countermeasure strategies: electromagnetic interference strategy, physical interception strategy, and navigation trapping strategy. Each strategy corresponds to specific applicable conditions and characteristic parameter thresholds. The electromagnetic interference strategy is applicable when the UAV threat intensity level is high and the flight speed is less than 15 m / s, with corresponding characteristic parameter thresholds of threat intensity greater than 5 and flight speed less than 15 m / s. The physical interception strategy is applicable when the UAV threat intensity level is medium-high and the diffusion radius is less than 50 meters, with corresponding characteristic parameter thresholds of threat intensity greater than 3 and diffusion radius less than 50 meters. The navigation trapping strategy is applicable when the UAV threat intensity level is medium-low and the flight mode is uniform linear, with corresponding characteristic parameter thresholds of threat intensity less than 5 and flight mode marked as uniform linear. The feature parameters in the UAV countermeasure status feature data are compared one by one with the applicable conditions in the prior countermeasure strategy library. The matching degree of each strategy is calculated. The matching degree is calculated as the ratio of the number of matching feature parameters to the total number of feature parameters. A matching degree greater than 0.8 is considered a high match, 0.6 to 0.8 is considered a moderate match, and less than 0.6 is considered a mismatch. The matching degree, applicable condition matching status, and corresponding feature parameters of each strategy are recorded and integrated to form UAV countermeasure strategy matching feature data, which clarifies the optimal matching countermeasure strategy and alternative strategies.

[0068] Step S43: Design the action space of the UAV countermeasure strategy based on the matching feature data of the UAV countermeasure strategy, and generate the action space data of the UAV countermeasure strategy. In this embodiment of the invention, the action space design for UAV countermeasure strategies is carried out based on the matching feature data of UAV countermeasure strategies. Taking the countermeasure strategy with the highest matching degree as the core, a corresponding countermeasure action space is designed, clarifying the type, execution parameters, triggering conditions, and execution order of the countermeasure actions. If the optimal matching strategy is an electromagnetic interference strategy, the action space design includes three core countermeasure actions: First, electromagnetic interference frequency adjustment, setting the interference frequency range to 2.4-5.8GHz, corresponding to the commonly used communication frequency band of UAVs. The interference frequency is adjusted according to the peak energy of the UAV; when the peak energy is greater than 500W, the interference frequency is set to 5.8GHz, and when it is less than or equal to 500W, it is set to 2.4GHz. Second, interference power control, setting interference power levels: 50W for high threat status, 30W for medium threat status, and 10W for low threat status. Third, interference direction alignment, using the center coordinates of the UAV threat propagation field as the target, adjusting the direction of the interference antenna, controlling the alignment error within 5°. If the optimal matching strategy is physical interception or navigation-based decoy, corresponding action parameters such as interception trajectory and decoy signal parameters should be designed. The execution time, sequence, and associated conditions of each action should be clearly defined. For example, a physical interception action requires first adjusting the position of the interception device to 10 meters ahead of the threat diffusion trajectory before initiating the interception action. All countermeasure action types, parameters, triggering conditions, and execution sequences should be integrated to generate UAV countermeasure strategy action space data, ensuring that the countermeasure action design is accurate and executable.

[0069] Step S44: Analyze the intelligent control strategy for UAV countermeasures using the UAV countermeasures action space data to generate intelligent control strategy data for UAV countermeasures.

[0070] In this embodiment of the invention, intelligent control strategy analysis for UAV countermeasures is performed using UAV countermeasure action space data. An intelligent control analysis model for countermeasures is established. Based on the countermeasure action space data and combined with the dynamic change patterns of threats in the UAV countermeasure state characteristic data, the execution effect of countermeasure actions is predicted. The prediction logic is as follows: The effective time of the countermeasure action is calculated based on the threat propagation speed and the countermeasure action response speed. The effective time is equal to the difference between the countermeasure action preparation time and the remaining time for the threat to propagate to the core of the protected area. The preparation time is set to 0.5 seconds. Each action in the countermeasure action space is prioritized. The priority determination criterion is the weighted sum of the countermeasure effect and the response speed. The weight of the countermeasure effect is set to 0.6, and the weight of the response speed is set to 0.4. The higher the weighted sum, the higher the priority. Simultaneously, a countermeasure action adjustment mechanism is set. When the threat intensity change rate is greater than 1W / second, the countermeasure action parameters are adjusted in real time, such as increasing the electromagnetic interference power by 20% and shifting the physical interception trajectory by 5 meters. Through the above intelligent control analysis, the execution sequence, effective time, parameter adjustment rules, and emergency backup actions of countermeasures are clarified. All control parameters and execution rules are integrated to generate intelligent control strategy data for UAV countermeasures. This ensures that the countermeasure strategy can dynamically adapt to the real-time changes in UAV threats, improve countermeasure efficiency and effectiveness, and provide a complete and accurate intelligent control basis for UAV countermeasures.

[0071] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, it is intended that all variations falling within the meaning and scope of the equivalents of the application be incorporated into the invention.

[0072] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for analyzing UAV countermeasures by fusing radar and optoelectronic data, characterized in that, Includes the following steps: Step S1: Perform multi-source monitoring and processing of airspace UAVs using deployed radar-electro-optical monitoring equipment to generate multi-source monitoring data for airspace UAVs; Cross-modal coupled sensing feature analysis of UAV target monitoring is performed on multi-source monitoring data of UAVs in the airspace to generate coupled sensing feature data of UAV target monitoring; Step S2: Analyze the propagation characteristics of UAV flight behavior based on the coupled sensing feature data of UAV target monitoring, and generate UAV flight behavior propagation feature data; Step S3: Analyze the characteristics of drone threat propagation field based on drone flight behavior propagation feature data, and generate drone threat propagation field feature data; Step S4: Analyze the intelligent control strategy for drone countermeasures based on the drone threat propagation field characteristic data, and generate drone countermeasure intelligent control strategy data.

2. The UAV countermeasure analysis method fusion radar-electro-optical data according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Perform multi-source monitoring and processing of airspace UAV data through deployed radar-electro-optical monitoring equipment to generate multi-source monitoring data of airspace UAVs, wherein the multi-source monitoring data of airspace UAVs includes radar monitoring data and electro-optical monitoring data of airspace UAVs. Step S12: Analyze radar monitoring and perception features based on UAV radar monitoring data in the airspace to generate radar monitoring and perception feature data; Step S13: Perform photoelectric observation feature analysis based on airspace UAV photoelectric monitoring data to generate photoelectric observation feature data; Step S14: Perform radar monitoring and sensing feature analysis on the UAV spatial trajectory features of the radar monitoring and sensing feature data to generate radar monitoring and sensing spatial trajectory feature data; Step S15: Perform photoelectric observation feature analysis on the UAV target features of photoelectric observation to generate photoelectric observation target feature data; Step S16: Based on radar monitoring and sensing spatial trajectory feature data and photoelectric observation target feature data, perform cross-modal coupled sensing feature analysis of UAV target monitoring to generate UAV target monitoring coupled sensing feature data.

3. The UAV countermeasure analysis method fusion radar-electro-optical data according to claim 2, characterized in that, Step S12 includes the following steps: Step S121: Analyze the characteristics of radar monitoring echo signals based on the UAV radar monitoring data in the airspace, and generate radar monitoring echo signal characteristic data; Step S122: Parse the radar monitoring echo signal characteristic data into radar monitoring electromagnetic wave sensing spectrum data; Step S123: Analyze the radar monitoring and sensing characteristics based on the radar monitoring electromagnetic wave sensing spectrum data to generate radar monitoring and sensing characteristic data.

4. The UAV countermeasure analysis method fusion radar-electro-optical data according to claim 2, characterized in that, Step S13 includes the following steps: Step S131: Extract airspace photoelectric visual data and airspace photoelectric thermal radiation data based on airspace UAV photoelectric monitoring data; Step S132: Perform visual feature analysis on the contour and texture of the spatial optoelectronic visual data to obtain optoelectronic visual feature data; perform optoelectronic thermal radiation distribution feature analysis on the spatial optoelectronic thermal radiation data to generate optoelectronic thermal radiation distribution feature data. Step S133: Map the photoelectric thermal radiation distribution characteristic data to the photoelectric visual characteristic data to perform photoelectric observation characteristic analysis and generate photoelectric observation characteristic data.

5. The UAV countermeasure analysis method fusion radar-electro-optical data according to claim 3, characterized in that, Step S16 includes the following steps: Step S161: Perform spatial trajectory prediction processing on the radar monitoring and sensing spatial trajectory feature data to generate radar monitoring and sensing spatial trajectory prediction data; Step S162: Map the radar monitoring and sensing spatial trajectory prediction data to the photoelectric observation target feature data to perform observation neighborhood candidate matching analysis for spatial trajectory prediction, and generate spatial trajectory prediction-observation neighborhood candidate matching data; Step S163: Perform candidate association analysis for UAV cross-modal monitoring based on spatial trajectory prediction-observation neighborhood candidate matching data to generate UAV cross-modal monitoring candidate association data; Step S164: Optimize the dynamic target association of UAVs in cross-modal monitoring based on the cross-modal candidate association data of UAV monitoring, and generate UAV cross-modal dynamic target association data; Step S165: Perform cross-modal coupled sensing feature analysis on the dynamic target association data of UAV cross-modal monitoring to generate UAV target monitoring coupled sensing feature data.

6. The UAV countermeasure analysis method fusion radar-electro-optical data according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Based on the UAV target monitoring coupled sensing feature data, perform UAV spatial energy peak migration feature analysis to generate UAV spatial energy peak migration feature data; Step S22: Analyze the dynamic behavior characteristics of the UAV based on the spatial energy peak migration characteristic data, and generate UAV dynamic behavior characteristic data; Step S23: Perform UAV visual structure state feature analysis based on UAV target monitoring coupled perception feature data to generate UAV visual structure state feature data; Step S24: Analyze the UAV flight behavior pattern using UAV dynamic behavior feature data and UAV visual structural state feature data to generate UAV flight behavior pattern data. Step S25: Analyze the propagation characteristics of UAV flight behavior based on the UAV flight behavior pattern data to generate UAV flight behavior propagation characteristic data.

7. The UAV countermeasure analysis method fusion radar-electro-optical data according to claim 6, characterized in that, Step S21 includes the following steps: Step S211: Perform time-series feature analysis on the UAV target monitoring coupled sensing feature data to generate UAV target monitoring coupled sensing time-series feature data; Step S212: Based on the UAV target monitoring coupled sensing time series feature data, perform UAV spatial dynamic energy peak identification processing to generate UAV spatial dynamic energy peak data; Step S213: Perform UAV spatial energy peak migration characteristic analysis on the UAV spatial dynamic energy peak data to generate UAV spatial energy peak migration characteristic data.

8. The UAV countermeasure analysis method fusion radar-electro-optical data according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Perform spatial threat multi-track grid cell analysis on the UAV flight behavior propagation characteristic data to generate spatial threat multi-track grid cell data; Step S32: Analyze the space threat diffusion trend based on the space threat multi-track grid cell data to generate space threat diffusion trend data; Step S33: Perform threat indicator feature analysis on the spatial threat diffusion trend data to generate spatial diffusion trend threat indicator feature data; Step S34: Analyze the associated threat characteristics of the spatial diffusion trend based on the spatial diffusion trend threat index characteristic data, and generate spatial diffusion trend associated threat characteristic data; Step S35: Analyze the threat intensity distribution characteristics of the diffusion trend based on the spatial diffusion trend threat index feature data and the spatial diffusion trend related threat feature data, and generate spatial diffusion trend threat intensity distribution feature data; Step S36: Analyze the UAV threat propagation field characteristics of the spatial threat diffusion trend data using the spatial diffusion trend threat intensity distribution characteristic data, and generate UAV threat propagation field characteristic data.

9. The UAV countermeasure analysis method fusion radar-electro-optical data according to claim 8, characterized in that, Step S31 includes the following steps: Step S311: Perform UAV multi-track node feature analysis based on UAV flight behavior propagation feature data to generate UAV multi-track node feature data; Step S312: Perform threat multi-track node feature analysis based on UAV multi-track node feature data to generate UAV threat multi-track node feature data; Step S313: Perform spatial threat multi-track grid cell analysis on the protected area based on the multi-track node feature data of UAV threats, and generate spatial threat multi-track grid cell data.

10. The UAV countermeasure analysis method fusion radar-electro-optical data according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Perform drone countermeasure state characteristic analysis on the drone threat propagation field characteristic data to generate drone countermeasure state characteristic data; Step S42: Utilize the preset UAV prior countermeasure strategy library to perform UAV countermeasure state feature analysis on UAV countermeasure strategy matching feature data, and generate UAV countermeasure strategy matching feature data. Step S43: Design the action space of the UAV countermeasure strategy based on the matching feature data of the UAV countermeasure strategy, and generate the action space data of the UAV countermeasure strategy. Step S44: Analyze the intelligent control strategy for UAV countermeasures using the UAV countermeasures action space data to generate intelligent control strategy data for UAV countermeasures.

Images