Fixed area unmanned aerial vehicle countering method and system

Abstract

The invention discloses a fixed area unmanned aerial vehicle countering method and system. The method comprises the steps of obtaining a three-dimensional coordinate of a to-be-identified unmanned aerial vehicle; inputting the obtained spectrum data of the to-be-identified unmanned aerial vehicle into an invasive unmanned aerial vehicle detection model constructed by a deep learning algorithm to generate an invasive unmanned aerial vehicle signal; in response to the invading unmanned aerial vehicle signal, acquiring multispectral imaging data of the invading unmanned aerial vehicle based on the visual image acquisition base station; performing feature-level fusion on the radar wave detection data, the spectrum data and the multispectral imaging data based on a multi-modal fusion algorithm of space-time registration to obtain fusion feature data of the target area; performing feature extraction of different dimensions on the fused feature data, and establishing a threat situation three-dimensional thermodynamic diagram of the invading unmanned aerial vehicle based on the extracted features; and executing a security countering strategy generated based on the threat situation three-dimensional thermodynamic diagram. According to the method, the unmanned aerial vehicle countering requirement in a complex environment is met.

Description

Technical Field

[0001] This application relates to the technical field of unmanned aerial vehicles, and particularly to a method and system for countering unmanned aerial vehicles in a fixed area. Background Art

[0003] Unmanned aerial vehicles are small in size and strong in mobility. Traditional security means rely on single monitoring means and it is difficult to effectively supervise them. For example, conventional surveillance cameras are limited by the viewing angle and distance and cannot detect unmanned aerial vehicles flying at a long distance or at a low altitude in a timely manner; while ordinary radar systems also have certain limitations in the detection ability of small unmanned aerial vehicles, and false alarms or missed alarms are likely to occur.

[0004] Therefore, how to optimize the existing methods for countering unmanned aerial vehicles has become a technical problem that needs to be solved urgently by those skilled in the art. Summary of the Invention

[0005] This application provides a method and system for countering unmanned aerial vehicles in a fixed area to solve the technical problem of how to optimize the existing methods for countering unmanned aerial vehicles.

[0006] To solve the above technical problem, an embodiment of this application provides a method for countering unmanned aerial vehicles in a fixed area, including:

[0007] Setting a plurality of radar detection base stations, spectrum analysis base stations and visual image acquisition base stations in the target area, the method for countering unmanned aerial vehicles in the fixed area includes:

[0008] Analyzing the radar wave detection data captured by the radar detection base station, and obtaining the three-dimensional coordinates of the unmanned aerial vehicle to be identified according to the analysis result;

[0009] Based on the spectrum analysis base station and the three-dimensional coordinates, obtaining the spectrum data of the unmanned aerial vehicle to be identified, inputting the spectrum data into an intrusion unmanned aerial vehicle detection model constructed by a deep learning algorithm, and correspondingly generating an intrusion unmanned aerial vehicle signal according to the recognition result output by the intrusion unmanned aerial vehicle detection model;

[0010] In response to the intrusion unmanned aerial vehicle signal, obtaining the multi-spectral imaging data of the intrusion unmanned aerial vehicle based on the visual image acquisition base station;

[0011] Based on a multi-modal fusion algorithm of spatio-temporal registration, performing feature-level fusion on the radar wave detection data, the spectrum data and the multi-spectral imaging data to obtain the fusion feature data of the target area;

[0012] Extract features of the fused feature data in different dimensions, and establish a three-dimensional heat map of the threat situation of the intruding UAV based on the extracted features, where the dimensions at least include the coordinate dimension, the function dimension, the flight attitude dimension, and the communication dimension;

[0013] Execute the security countermeasure strategy generated based on the three-dimensional heat map of the threat situation.

[0014] In a preferred embodiment of the present application, the deployment quantity N of the radar detection base stations 雷达 is expressed as:

[0015]

[0016] where L and W are respectively the length and width of the target area, and R 雷达 is the effective detection radius of a single radar;

[0017] The deployment quantity N of the spectrum analysis base stations 频谱 and the maximum spacing D 频谱 are expressed as:

[0018]

[0019] where R 频谱 is the effective coverage radius of a single spectrum analysis base station;

[0020] The deployment quantity N of the visual image acquisition base stations 视觉 is expressed as:

[0021] N 视觉 = 2min(2×H, R 视觉 )

[0022] where H is the protection height of the target area, and R 视觉 is the effective recognition distance of the visual image acquisition base station.

[0023] In a preferred embodiment of the present application, inputting the spectrum data into an intrusion UAV detection model constructed by a deep learning algorithm, and correspondingly generating an intrusion UAV signal according to the recognition result output by the intrusion UAV detection model, includes:

[0024] Obtain the historical spectrum data of the intrusion UAV, and perform preprocessing operations on the historical spectrum data;

[0025] Construct an initial intrusion UAV detection model based on a convolutional neural network, input the preprocessed historical spectrum data into the initial intrusion UAV detection model for training, and obtain a trained intrusion UAV detection model;

[0026] Input the spectrum data obtained in real time into the intrusion UAV detection model, and determine the type of the UAV to be identified based on the output result, and generate an intrusion UAV signal according to the identification result.

[0027] In a preferred embodiment of the present application, the analysis of the radar wave detection data captured by the radar detection base station, and obtaining the three-dimensional coordinates of the UAV to be identified according to the analysis result includes:

[0028] Obtain the azimuth data and distance data of the UAV to be identified relative to each radar detection base station collected synchronously by at least three radar detection base stations;

[0029] Taking the position of any radar detection base station as the center of the sphere, establish a spherical coordinate system equation set;

[0030] Solve the spherical coordinate system equation set according to the azimuth data and the distance data to obtain the coordinates of the UAV to be identified;

[0031] Among them, the spherical coordinate system equation set is expressed as:

[0032]

[0033] Among them, (x0,y0,z0) represents the coordinates of the selected millimeter radar wave, (x1,y1,z1), (x2,y2,z2) represent the coordinates of other millimeter radar waves outside the selected millimeter radar wave, and r0, r1, r2 respectively represent the distances from the UAV to be identified to the corresponding millimeter radar waves.

[0034] In a preferred embodiment of the present application, for the multi-modal fusion algorithm based on spatio-temporal registration, perform feature-level fusion on the radar wave detection data, the spectrum data and the multi-spectral imaging data to obtain the fusion feature data of the target area, specifically including:

[0035] Perform time consistency processing on the radar wave detection data, the spectrum data and the multi-spectral imaging data, and map the radar wave detection data, the spectrum data and the multi-spectral imaging data after time consistency processing to the same spatial coordinate system to complete spatio-temporal registration;

[0036] Respectively perform feature extraction on the radar wave detection data, the spectrum data and the multi-spectral imaging data after spatio-temporal registration to obtain the pose feature data and visual feature data of the target area;

[0037] Based on the fusion algorithm, fuse the pose feature data and the visual feature data to obtain the fusion feature data of the target area.

[0038] In a preferred embodiment of the present application, the feature extraction of the fusion feature data in different dimensions and the establishment of a three-dimensional heat map of the threat situation of the intruding unmanned aerial vehicle based on the extracted features include:

[0039] Obtain the unique identifier of the intruding unmanned aerial vehicle, associate the fusion feature data with the unique identifier to obtain the status information data of the intruding unmanned aerial vehicle, and the status information data set includes coordinate dimension data, function dimension data, flight attitude dimension data, and communication dimension data;

[0040] Quantize the coordinate dimension data, the function dimension data, the flight attitude dimension data, and the communication dimension data respectively, and perform weighted calculation according to the quantized status information data to obtain the threat level of the intruding unmanned aerial vehicle;

[0041] Establish a three-dimensional space rectangular coordinate system for the target area, map the position coordinates and threat level of the intruding unmanned aerial vehicle into the three-dimensional space rectangular coordinate system to construct the three-dimensional heat map of the threat situation.

[0042] In a preferred embodiment of the present application, the execution of the security countermeasure strategy generated based on the three-dimensional heat map of the threat situation includes:

[0043] Obtain the threat level of the intruding unmanned aerial vehicle based on the three-dimensional heat map of the threat situation, where the threat level includes low threat level, medium threat level, and high threat level, and the threat level reflects the number and flight speed of the intruding unmanned aerial vehicle;

[0044] If the threat level is low threat level, use the nearby ground electromagnetic interference base station to perform directional countermeasure on the intruding unmanned aerial vehicle;

[0045] If the threat level is medium threat level, dispatch the aerial unmanned aerial vehicle countermeasure platform to perform sub-region encirclement countermeasure on the intruding unmanned aerial vehicle;

[0046] If the threat level is high threat level, perform directional countermeasure on the intruding unmanned aerial vehicle through the ground electromagnetic interference base station, and dispatch the aerial unmanned aerial vehicle countermeasure platform to perform sub-region encirclement countermeasure on the intruding unmanned aerial vehicle.

[0047] In a preferred embodiment of the present application, the dispatch of the aerial unmanned aerial vehicle countermeasure platform to perform sub-region encirclement countermeasure on the intruding unmanned aerial vehicle includes:

[0048] Determine the airborne equipment that matches the type of the intruding unmanned aerial vehicle, where the airborne equipment includes radio countermeasure equipment, directional antenna, omnidirectional antenna, strong light irradiation lamp, and capture net, and the working frequency band of the airborne equipment is compatible with the interference frequency band of the ground electromagnetic interference base station;

[0049] The airborne device is installed on the aerial UAV countermeasure platform by driving a crawler with a stepping motor, and the aerial UAV countermeasure platform is controlled to perform sub-region encirclement countermeasures against the intruding UAV.

[0050] In a preferred embodiment of the present application, the deployment method of the ground electromagnetic interference base station is expressed as:

[0051]

[0052] D 干扰 ≤1.5R 干扰

[0053] Wherein, N 干扰 represents the deployment quantity of the ground electromagnetic interference base stations, D 干扰 represents the distance between any two ground electromagnetic interference base stations, R 干扰 is the effective interference radius of a single ground electromagnetic interference base station, and L and W respectively represent the length and width of the target area.

[0054] Another embodiment of the present application provides a fixed-area UAV countermeasure system, including:

[0055] A first acquisition module, configured to analyze the radar wave detection data captured by the radar detection base station, and obtain the three-dimensional coordinates of the UAV to be identified according to the analysis result;

[0056] An identification module, configured to obtain the spectrum data of the UAV to be identified based on the spectrum analysis base station and the three-dimensional coordinates, input the spectrum data into an intrusion UAV detection model constructed by a deep learning algorithm, and generate an intrusion UAV signal corresponding to the identification result output by the intrusion UAV detection model;

[0057] A second acquisition module, configured to, in response to the intrusion UAV signal, obtain the multi-spectral imaging data of the intrusion UAV based on the visual image acquisition base station;

[0058] A fusion module, configured to perform feature-level fusion on the radar wave detection data, the spectrum data, and the multi-spectral imaging data based on a multi-modal fusion algorithm of spatio-temporal registration to obtain the fusion feature data of the target area;

[0059] A construction module, configured to perform feature extraction on the fusion feature data in different dimensions, and establish a three-dimensional heat map of the threat situation of the intrusion UAV based on the extracted features, wherein the dimensions at least include a coordinate dimension, a function dimension, a flight attitude dimension, and a communication dimension;

[0060] An execution module, configured to execute a security countermeasure strategy generated based on the three-dimensional heat map of the threat situation.

[0061] Compared with the prior art, the beneficial effects of the embodiments of the present application are at least one of the following:

[0062] (1) The multi-modal fusion algorithm based on spatio-temporal registration in the present application performs feature-level fusion on radar wave detection data, spectrum data, and multi-spectral imaging data, fully combining the advantages of radar in long-distance detection, target positioning, and speed measurement, as well as the advantages of multi-spectral imaging in target feature recognition. More comprehensive and accurate target information can be obtained, overcoming the limitations of a single data source, improving the detection, tracking, and recognition performance of unmanned aerial vehicles (UAVs), and enhancing the adaptability and reliability of the system in complex environments.

[0063] (2) The present application performs multi-dimensional feature extraction on the fused feature data and establishes a three-dimensional heat map of the threat situation, which can visually and visually present the threat situation of the UAV. From the coordinate dimension, the position information of the UAV can be accurately grasped. From the function dimension, the tasks it can perform or the equipment it carries can be understood. From the flight attitude dimension, it helps to judge its flight intention and trajectory changes. From the communication dimension, its communication status and whether there are abnormal communication behaviors can be monitored. It provides comprehensive and clear situation awareness for decision-makers, facilitating the rapid and accurate formulation of countermeasures. Description of the Drawings

[0064] Figure 1 is a schematic flowchart of a method for countering UAVs in a fixed area in one embodiment of the present application;

[0065] Figure 2 is a schematic diagram of a system for countering UAVs in a fixed area in one embodiment of the present application. Detailed Embodiments

[0066] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0067] In the description of the present application, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first", "second", "third", etc. may explicitly or implicitly include one or more of such features. In the description of the present application, unless otherwise specified, the meaning of "a plurality" is two or more.

[0068] In the description of the present application, it should be noted that unless otherwise clearly specified and defined, the terms "installed", "connected", and "coupled" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two components. The terms "vertical", "horizontal", "left", "right", "upper", "lower" and similar expressions used herein are only for the purpose of illustration, rather than indicating or implying that the device or component referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present application. The term "and / or" used herein includes any and all combinations of one or more of the related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0069] In the description of the present application, it should be noted that unless otherwise defined, all technical and scientific terms used in the present application have the same meaning as those commonly understood by those skilled in the technical field to which the present application belongs. The terms used in the specification of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0070] An embodiment of the present application provides a method for countering drones in a fixed area. Specifically, please refer to Figure 1 , Figure 1 which shows a schematic flow chart of the method for countering drones in a fixed area in one of the embodiments of the present application. It includes setting a number of radar detection base stations, spectrum analysis base stations and visual image acquisition base stations in the target area. In a preferred embodiment of the present application, the deployment quantity N of the radar detection base stations 雷达 is expressed as:

[0071]

[0072] where L and W are the length and width of the target area respectively, and R 雷达 is the effective detection radius of a single radar;

[0073] The deployment quantity N of the spectrum analysis base stations 频谱 and the maximum spacing D 频谱 are expressed as:

[0074]

[0075] where R 频谱 is the effective coverage radius of a single spectrum analysis base station;

[0076] The deployment quantity N of visual image acquisition base stations 视觉 It is expressed as:

[0077] N 视觉 = 2min(2×H, R 视觉 )

[0078] where H is the protection height of the target area, and R 视觉 is the effective recognition distance of the visual image acquisition base station.

[0079] Among them, the main function of the radar detection base station is to detect targets in the target area, and determine information such as the position and speed of the target by transmitting and receiving radar waves. The spectrum analysis base station is used to monitor and analyze the electromagnetic spectrum in the target area to detect whether there are abnormal electromagnetic signals, such as signals with specific frequencies that drones may emit. In order to ensure comprehensive monitoring of the spectrum of the target area, it is necessary to reasonably arrange the deployment quantity and spacing of the spectrum analysis base stations. The visual image acquisition base station assists in identifying targets by collecting visual images of the target area, such as identifying information such as the appearance and position of drones. Considering that the target area has a certain protection height, it is necessary to determine the deployment quantity according to the protection height and the effective recognition distance of the base station. The method for countering drones in a fixed area includes:

[0080] S1: Analyze the radar wave detection data captured by the radar detection base station, and obtain the three-dimensional coordinates of the drone to be identified according to the analysis results;

[0081] In a preferred embodiment of the present application, analyzing the radar wave detection data captured by the radar detection base station and obtaining the three-dimensional coordinates of the drone to be identified according to the analysis results includes:

[0082] Obtain the azimuth data and distance data of the drone to be identified relative to each radar detection base station synchronously collected by at least three radar detection base stations;

[0083] Taking the position of any radar detection base station as the center of the sphere, establish a spherical coordinate system equation set;

[0084] Solve the spherical coordinate system equation set according to the azimuth data and distance data to obtain the coordinates of the drone to be identified;

[0085] Among them, the spherical coordinate system equation set is expressed as:

[0086]

[0087] Among them, (x0, y0, z0) represents the coordinates of the selected millimeter radar wave, and (x1, y1, z1), (x2, y2, z2) represent the coordinates of other millimeter radar waves outside the selected millimeter radar wave, and r0, r1, r2 respectively represent the distances from the drone to be identified to the corresponding millimeter radar waves.

[0088] Reasonably arrange at least three radar detection base stations around the target area. These radar detection base stations need to work precisely synchronously to ensure that the collected data can accurately reflect the state of the intruding drone at the same moment. The synchronization method can be achieved through a hardware synchronization circuit or a software synchronization algorithm based on a high-precision clock signal. Each radar detection base station has the ability to accurately measure the azimuth angle and distance of the target drone. The radar detection base station works by transmitting millimeter-wave signals and receiving the echo signals reflected by the target drone. According to the time difference between the echo signal and the transmitted signal, the distance from the target drone (i.e., the drone to be identified) to the radar can be calculated. After detecting the signal of the drone to be identified, each radar detection base station synchronously starts the data acquisition process to obtain the azimuth angle data and distance data of the intruding drone relative to itself in real time. The collected data will be transmitted to a central processing unit or a data fusion center, where it is integrated to form a data set containing the azimuth angle and distance information of the intruding drone from all radar detection base stations for positioning calculation.

[0089] S2: Obtain the spectral data of the drone to be identified based on the spectral analysis base station and the three-dimensional coordinates, input the spectral data into the intrusion drone detection model constructed by the deep learning algorithm, and generate the intrusion drone signal corresponding to the recognition result output by the intrusion drone detection model;

[0090] In a preferred embodiment of the present application, inputting the spectral data into the intrusion drone detection model constructed by the deep learning algorithm and generating the intrusion drone signal corresponding to the recognition result output by the intrusion drone detection model includes:

[0091] Obtain the historical spectral data of the intrusion drone and perform preprocessing operations on the historical spectral data;

[0092] Construct an initial intrusion drone detection model based on a convolutional neural network, input the preprocessed historical spectral data into the initial intrusion drone detection model for training, and obtain the trained intrusion drone detection model;

[0093] Input the real-time obtained spectral data into the intrusion drone detection model, determine the type of the drone to be identified based on the output result, and generate the intrusion drone signal corresponding to the recognition result.

[0094] Specifically, check whether there are missing values, outliers, or incorrect data in the historical spectrum data. For missing values, mean filling, median filling, or other more complex imputation methods can be used to fill them according to the characteristics of the data. For outliers, it is necessary to determine whether they are caused by measurement errors or real special situations. If it is a measurement error, it can be corrected or removed according to statistical methods. In addition, considering that the UAV signal can be affected by various factors in practical applications, resulting in limited and unevenly distributed spectrum data collected. To increase the diversity of the data and improve the generalization ability of the model, data augmentation operations can be performed. For example, perform transformations such as translation, scaling, and rotation on the spectrum data to simulate the changes in the spectrum characteristics of UAVs under different flight postures, distances, and angles.

[0095] Convolutional neural networks have the ability to automatically extract data features and are particularly suitable for processing data with a grid structure, such as spectrum data. When constructing the initial intrusion UAV detection model, multiple convolutional layers, pooling layers, and fully connected layers are usually included.

[0096] Divide the preprocessed historical spectrum data into a training set, a validation set, and a test set. The training set is used to train the model so that it learns the mapping relationship between the spectrum data and the intrusion UAV; the validation set is used to adjust the hyperparameters of the model during training, such as the learning rate, the number of layers, the number of neurons, etc., to prevent the model from overfitting; the test set is used to evaluate the performance of the trained model on unknown data.

[0097] During the training process, input the training set data into the initial intrusion UAV detection model. The model performs forward propagation on the input data according to the current parameters and calculates the loss between the predicted result and the true label (common loss functions such as the cross-entropy loss function). Then, calculate the gradient of the loss function with respect to the model parameters through the backpropagation algorithm, and use an optimizer (such as Stochastic Gradient Descent SGD, Adam, etc.) to update the model parameters according to the gradient, so that the loss function gradually decreases. This process is repeated continuously until the performance of the model on the validation set no longer improves or reaches the preset number of training epochs. At this time, the trained intrusion UAV detection model is obtained.

[0098] When the system enters the actual operation stage, after processing the real-time spectrum data in the same preprocessing manner as the historical data, input it into the trained intrusion UAV detection model. The model performs feature extraction and analysis on the input spectrum data through forward propagation, and finally outputs a recognition result, which usually represents the possibility that the input data belongs to different classes in the form of a probability distribution. For example, if the probability of the UAV class in the result output by the model exceeds a preset threshold (such as 0.9), it is determined that the current spectrum data corresponds to an intrusion UAV signal.

[0099] Based on the recognition result of the model, the system will correspondingly generate an intrusion drone signal. If the recognition result determines an intrusion drone, the generated signal can be an electrical signal that triggers an alarm, and this signal can be transmitted to the alarm system to trigger an audible and visual alarm to alert relevant personnel. At the same time, this signal can also be used as a control signal to initiate subsequent countermeasures.

[0100] S3: In response to the intrusion drone signal, obtain the hyperspectral imaging data of the intrusion drone based on the visual image acquisition base station;

[0101] According to factors such as the characteristics of the target area, monitoring requirements, and budget, select a suitable hyperspectral imaging device, such as a hyperspectral camera. Parameters such as its spectral band range, resolution, and sensitivity need to be considered. For example, when monitoring vegetation, a camera with bands sensitive to vegetation characteristics, such as the near-infrared band, should be selected. Mount the hyperspectral imaging device on a suitable platform, such as a fixed-wing drone, a rotary-wing drone, or a ground monitoring station, etc. If mounted on a drone, ensure that it is firmly installed and does not affect the flight performance of the drone. At the same time, calibrate the hyperspectral imaging device, including spectral calibration, radiometric calibration, and geometric calibration, etc., to ensure the accuracy and reliability of the collected data. When the monitoring system receives the intrusion drone signal, the signal is transmitted to the data acquisition system or the control center. The system analyzes and processes the signal, confirms that it is a signal triggered by an intrusion drone, and at the same time obtains information related to the intrusion drone, such as the approximate location, flight direction, etc.

[0102] According to the information of the intrusion drone and the situation of the target area, the control center or system software automatically plans the hyperspectral imaging data acquisition task, determines parameters such as the acquisition area, acquisition time, and acquisition angle, and sends the acquisition instructions to the hyperspectral imaging device. The hyperspectral imaging device acquires hyperspectral images of the target area according to the set parameters and instructions. During the acquisition process, different sensors in the device sense and record the light of different spectral bands, convert the optical signal into an electrical signal or a digital signal, and generate hyperspectral image data. The acquisition method can be to take static images at certain time intervals, or to continuously shoot videos and then extract key frames from the videos as hyperspectral images. The acquired hyperspectral image data is first temporarily stored in the cache of the device. When the cache reaches a certain capacity or after a specific area is acquired, the data is transmitted to the data storage center or processing platform through wired or wireless communication for subsequent analysis and processing.

[0103] S4: Based on the multi-modal fusion algorithm of spatio-temporal registration, perform feature-level fusion on the radar wave detection data, spectral data, and hyperspectral imaging data to obtain the fusion feature data of the target area;

[0104] Radar wave detection data, spectral data, and hyperspectral imaging data are usually collected by different sensors. There can be differences in the acquisition frequencies, start times, etc. of these sensors, resulting in data that is not synchronized in time. Time consistency processing aims to eliminate this time difference and ensure that the two types of data reflect the state of the target area at the same moment.

[0105] In a preferred embodiment of the present application, based on a multi-modal fusion algorithm for spatio-temporal registration, the radar wave detection data, spectral data, and hyperspectral imaging data are subjected to feature-level fusion to obtain fused feature data of the target area, specifically including:

[0106] Perform time consistency processing on the radar wave detection data, spectral data, and hyperspectral imaging data, and map the radar wave detection data, spectral data, and hyperspectral imaging data after time consistency processing to the same spatial coordinate system to complete spatio-temporal registration;

[0107] Extract features from the radar wave detection data, spectral data, and hyperspectral imaging data after spatio-temporal registration respectively to obtain pose feature data and visual feature data of the target area;

[0108] Fuse the pose feature data and visual feature data based on the fusion algorithm to obtain fused feature data of the target area.

[0109] Specifically, the installation positions and viewing angles of different sensors in the target area are different, which makes the data they collect in different spatial coordinate systems. For example, a radar may be installed at a fixed position on the ground and detect at a specific angle; a hyperspectral imaging device may be installed at a high place with different field of view ranges and directions. In order to map the radar wave detection data, spectral data, and hyperspectral imaging data after time consistency processing to the same spatial coordinate system, it is necessary to determine the relative positions and pose relationships between the sensors. The relative positions and pose relationships between the sensors can be determined by measuring parameters such as the installation positions and angles of the sensors, and a transformation matrix can be established. Using these transformation matrices, the data collected by different sensors can be transformed from their respective local coordinate systems to a unified global coordinate system, thereby completing spatial registration.

[0110] First, the transformation relationship between the data coordinate systems needs to be determined. This can be achieved through known sensor position and pose information, as well as some calibration points. For example, if a radar and a hyperspectral imaging device are installed on the same platform, the relative positions and angles between the devices can be measured to establish a rotation matrix and a translation vector to describe the transformation relationship between the coordinate systems. For more complex situations, multiple calibration points with known positions can be used, and accurate coordinate transformation parameters can be solved through algorithms such as the least squares method. After determining the transformation relationship, coordinate transformation can be performed on each point in the radar wave detection data, spectral data, and hyperspectral imaging data.

[0111] S5: Extract features of the fused feature data in different dimensions, and establish a three-dimensional heat map of the threat situation of the intruding UAV based on the extracted features, where the dimensions at least include the coordinate dimension, the function dimension, the flight attitude dimension, and the communication dimension;

[0112] In a preferred embodiment of the present application, establishing a three-dimensional heat map of the threat situation of the intruding UAV according to the fused feature data and the positioning result of the intruding UAV includes:

[0113] Obtain the unique identifier of the intruding UAV, and associate the fused feature data with the unique identifier to obtain the status information data of the intruding UAV. The status information data set includes coordinate dimension data, function dimension data, flight attitude dimension data, and communication dimension data;

[0114] Quantify the coordinate dimension data, function dimension data, flight attitude dimension data, and communication dimension data respectively, and perform weighted calculation according to the quantified status information data to obtain the threat level of the intruding UAV;

[0115] Establish a three-dimensional space rectangular coordinate system of the target area, and map the position coordinates and threat level of the intruding UAV into the three-dimensional space rectangular coordinate system to construct a three-dimensional heat map of the threat situation.

[0116] Specifically, in the entire monitoring system, each intruding UAV needs to be assigned a unique identifier. This identifier can be a serial number automatically generated by the system when the UAV is first detected, or the identification information determined based on the specific identification signal emitted by the UAV itself. For example, some UAVs will send data packets containing information such as their own model and number during communication. By parsing these data packets, their unique identifier can be obtained.

[0117] Associate the obtained unique identifier of the intruding UAV with the previously fused feature data. Since the fused feature data has integrated the characteristics of radar wave detection data, spectrum data, and multi-spectral imaging data, it contains rich information about the intruding UAV. Through this association, this information can be integrated into the data set indexed by the unique identifier to form the status information data of this intruding UAV.

[0118] The status information data is further divided according to different dimensions, respectively obtaining coordinate dimension data, function dimension data, flight attitude dimension data, and communication dimension data. The coordinate dimension data reflects the spatial position information of the intruding drone in the target area, which can come from the position determined by radar positioning or the combination of multi-spectral images and geographic information system (GIS); the function dimension data is related to the tasks that the drone can perform, such as reconnaissance, delivering items, or other functions, and this information can be obtained through the analysis of the drone's shape, equipped devices, and flight behavior patterns; the flight attitude dimension data includes the pitch angle, yaw angle, roll angle, etc. of the drone, reflecting its attitude in the air, and can be obtained through the fine analysis of radar echoes or the attitude recognition of the drone in multi-spectral images; the communication dimension data involves information such as the communication frequency band, communication intensity, and communication protocol of the drone, and is obtained by monitoring and analyzing the communication signals emitted by the drone.

[0119] After that, the data of each dimension is quantized and the threat level is calculated by weighting. The coordinate dimension data is usually continuous spatial position information, and it needs to be converted into discrete quantization values for subsequent processing. For example, the target area can be divided into multiple grids, and corresponding quantization values are assigned according to the grid position where the intruding drone is located. If the target area is a rectangular area, it can be equally divided into several small intervals in the horizontal and vertical directions, and the position of the small interval where the drone is located corresponds to a quantized coordinate value. The purpose of doing this is to convert the continuous spatial position information into discrete numerical values suitable for calculation and comparison.

[0120] The different function types in the function dimension are discrete category information in themselves, and they need to be converted into numerical forms. Different quantization values can be set for different function types. For example, the reconnaissance function is set to 1, and the delivery function is set to 2, etc. This quantization method can unify different function types to a numerical scale, facilitating subsequent comprehensive calculation with data of other dimensions.

[0121] The angular values of the flight attitude are also continuous data and also need to be quantized. For example, for the pitch angle, yaw angle, and roll angle, their angular ranges can be divided into several intervals, and each interval corresponds to a quantization value. For example, the pitch angle is set to 1 when it is between 0 - 30 degrees, and set to 2 when it is between 30 - 60 degrees, etc. Through this quantization, the flight attitude information is converted into numerical values convenient for calculation.

[0122] The information such as the frequency band and intensity in the communication dimension also needs to be quantized. For the communication frequency band, different quantization values can be assigned according to the divided range of the frequency band; the communication intensity can be divided into several levels according to its intensity magnitude, and each level corresponds to a quantization value. For example, the communication intensity is divided into three levels: weak, medium, and strong, and is quantized as 1, 2, and 3 respectively.

[0123] Data in different dimensions has different importance for evaluating the threat level of an intruding drone. Therefore, corresponding weights need to be assigned to the data for each dimension. The determination of the weights can be based on expert experience, statistical analysis of historical data, or obtained through training with machine learning algorithms. For example, in certain scenarios, the dimension of the drone's coordinates near a sensitive area may have a higher weight; while in the scenario of preventing communication eavesdropping, the communication dimension may have a greater weight, and the threat level is calculated based on the weights.

[0124] Based on the target area, a three-dimensional rectangular coordinate system is established. The origin of this coordinate system can be set according to the actual situation. For example, it can be set at a landmark position in the target area or the position of the monitoring center. The x-axis, y-axis, and z-axis respectively represent different dimensions of the target area in the horizontal and vertical directions, and their units and scales are determined according to the size and accuracy requirements of the actual target area.

[0125] Map the position coordinates of the intruding drone (i.e., the position in the three-dimensional space corresponding to the quantized coordinate dimension data) and the threat level to the established three-dimensional rectangular coordinate system. For each intruding drone, at its corresponding spatial position, different color or brightness values are assigned according to the level of the threat. The higher the threat level, the more vivid the corresponding color may be (such as red) or the higher the brightness; the lower the threat level, the dimmer the color may be (such as blue) or the lower the brightness. In this way, in the three-dimensional rectangular coordinate system, drones at different positions are represented by points of different colors or brightness, and these points together form a three-dimensional heat map of the threat situation. This heat map can intuitively display the distribution of intruding drones in the target area and the threat level of each drone, providing clear and intuitive situation awareness for decision-makers so as to take corresponding countermeasures in a timely manner.

[0126] S6: Execute the security countermeasure strategy based on the three-dimensional heat map of the threat situation.

[0127] In a preferred embodiment of the present application, before executing the security countermeasure strategy based on the three-dimensional heat map of the threat situation, it further includes:

[0128] Obtain the historical trajectory data of the drone, where the historical trajectory data includes the drone's speed data, acceleration data, and position data;

[0129] Construct an initial drone path prediction model based on the LSTM network, and train the initial drone path prediction model according to the drone historical trajectory data to obtain a trained drone path prediction model;

[0130] Input the position information and speed information of the real-time obtained intruding drone into the pre-constructed drone path prediction model to obtain the path prediction result of the intruding drone.

[0131] Specifically, preprocess the obtained historical trajectory data of the drone, including operations such as data cleaning and normalization. Data cleaning is used to remove noise, outliers, missing values, etc. in the data to ensure the quality of the data. Normalization is to map the data to a certain range, such as [0, 1] or [-1, 1], to accelerate the training convergence speed of the model and prevent some features from having an adverse impact on model training due to overly large or small numerical values.

[0132] Divide the preprocessed data into a training set, a validation set, and a test set. Usually, the training set is used for updating the parameters and learning of the model, the validation set is used to monitor the performance of the model during training and adjust the hyperparameters of the model, such as the learning rate, the number of iterations, etc., to prevent the model from overfitting. The test set is used to evaluate the generalization ability and prediction accuracy of the model after the model training is completed.

[0133] Determine the parameters during the training process, such as the learning rate, batch size, number of iterations, etc. The learning rate determines the step size of parameter update of the model during training. An overly large learning rate may cause the model to fail to converge or overfit, while an overly small learning rate will slow down the training speed. The batch size refers to the number of samples input to the model each time during training. An appropriate batch size can improve the training efficiency and the stability of the model. The number of iterations indicates the number of times the model traverses the training data. Generally, an appropriate number of iterations needs to be determined according to the convergence situation of the model and the performance of the validation set.

[0134] Input the training set data into the initial LSTM model, calculate the loss function of the model through the backpropagation algorithm, and update the weights and biases of the model according to the gradient of the loss function. During the training process, the model will continuously adjust its own parameters to minimize the loss function, thereby learning the laws and patterns in the historical trajectory data of the drone. At the same time, use the validation set data to regularly evaluate the model, observe the performance indicators of the model, such as the mean squared error (MSE), mean absolute error (MAE), etc., and adjust the training parameters and model structure according to the evaluation results until the model reaches the optimal performance or meets the preset stop conditions, such as the loss function converges to a certain threshold, the performance of the validation set no longer improves, etc., to obtain a trained drone path prediction model.

[0135] Continuously obtain the position information and speed information of the invading drone through real-time monitoring devices. Input the processed real-time data into the trained drone path prediction model. The model predicts the future path of the invading drone according to the learned laws and patterns. The prediction results may include information such as the position and speed of the drone at multiple future time points, thereby providing a basis for subsequent security countermeasures. For example, it can anticipate in advance the dangerous areas that the drone may reach, so as to take corresponding defense measures in a timely manner.

[0136] In a preferred embodiment of the present application, implementing a security countermeasure strategy based on a three-dimensional heat map of threat situation includes:

[0137] Obtaining the threat level of the intruding unmanned aerial vehicle (UAV) based on the three-dimensional heat map of threat situation, where the threat level includes low threat level, medium threat level, and high threat level, and the threat level reflects the number and flight speed of the intruding UAVs;

[0138] If the threat level is low threat level, use the nearby ground electromagnetic interference base station to conduct directional countermeasures against the intruding UAV;

[0139] If the threat level is medium threat level, dispatch the aerial UAV countermeasure platform to conduct zonal encirclement countermeasures against the intruding UAV;

[0140] If the threat level is high threat level, conduct directional countermeasures against the intruding UAV through the ground electromagnetic interference base station, and dispatch the aerial UAV countermeasure platform to conduct zonal encirclement countermeasures against the intruding UAV.

[0141] Specifically, the three-dimensional heat map of threat situation integrates multi-dimensional information such as the coordinates, functions, flight postures, and communications of the intruding UAVs, and calculates the threat level of each intruding UAV through weighted calculation. According to the preset thresholds, the threat level is divided into low threat level, medium threat level, and high threat level. For example, the threat level score is in the range of 0 - 30 points (assuming a total score of 100 points) for the low threat level, 31 - 60 points for the medium threat level, and 61 - 100 points for the high threat level. The setting of these thresholds needs to comprehensively consider factors such as the security importance of the target area, the possible types of UAV threats faced, and the existing countermeasure resources.

[0142] In another embodiment of the present application, for a single intruding UAV, preferably use the nearby ground electromagnetic interference base station for directional countermeasures, and dispatch 1 - 2 aerial UAV countermeasure platforms to cooperate in interception when necessary;

[0143] For multiple intruding UAVs, dispatch 3 - 5 aerial UAV countermeasure platforms to conduct zonal encirclement countermeasures, and at the same time activate all ground electromagnetic interference base stations;

[0144] For high-speed crossing UAVs, adopt a countermeasure method combining pre-interference and interception tracking, and deploy countermeasure forces in advance on the predicted path;

[0145] For UAVs with unknown communication characteristics, activate the full-band scanning interference mode.

[0146] In a preferred embodiment of the present application, dispatching the aerial UAV countermeasure platform to conduct zonal encirclement countermeasures against the intruding UAV includes:

[0147] Determine the airborne equipment that matches the type of the invading drone, where the airborne equipment includes radio countermeasure equipment, directional antennas, omnidirectional antennas, strong light irradiation lamps, and capture nets, and the operating frequency bands of the airborne equipment are compatible with the interference frequency bands of the ground electromagnetic interference base stations;

[0148] Drive the crawler by a stepper motor to install the airborne equipment on the aerial drone countermeasure platform, and control the aerial drone countermeasure platform to conduct zonal encirclement countermeasures against the invading drone.

[0149] The mode of the drone carrying the airborne electromagnetic interference equipment is a one - select - many mode.

[0150] Before departure, the drone does not carry the airborne device. After the ground detection equipment determines the characteristics of the invading drone, the corresponding airborne device is selected according to the type of the invading drone.

[0151] The airborne device is installed on the traveling crawler, driven by a stepper motor, and each step can switch an airborne device. At the same time, the airborne device is installed directly below the drone. Through the airborne device sent by the control center, the control crawler transports the corresponding airborne device to directly below the drone and installs it. After the drone installs the countermeasure airborne equipment, it takes off. The whole process is controlled within 10 s.

[0152] The purpose of such a design is, on the one hand, to prevent the drone from carrying a lot of devices each time, which affects the mobility of the drone, and on the other hand, to ensure that after the countermeasure signal is selected, it will not interfere with the signal of the aerial drone countermeasure platform.

[0153] Among them, for any fixed rectangular area, determine the equipment deployment plan based on the following method:

[0154] Uniformly deploy radar detection base stations inside the area to achieve full - coverage area detection;

[0155] Uniformly deploy broadband radio spectrum analysis base stations inside the area to achieve full - coverage area detection;

[0156] Deploy multi - spectral visual image acquisition base stations around key facilities;

[0157] Set no less than 4 ground electromagnetic interference base stations at the area boundary.

[0158] In a preferred embodiment of the present application, the deployment method of the ground electromagnetic interference base stations is expressed as:

[0159]

[0160] D 干扰 ≤1.5R 干扰

[0161] Among them, N 干扰Denotes the deployment quantity of ground electromagnetic interference base stations, D 干扰 Denotes the distance between any two ground electromagnetic interference base stations, R 干扰 Is the effective interference radius of a single ground electromagnetic interference base station. L and W respectively denote the length and width of the target area.

[0162] Another embodiment of this application provides a fixed - area unmanned aerial vehicle counter - measure system. Specifically, please refer to Figure 2 , Figure 2 Shown is a schematic diagram of the fixed - area unmanned aerial vehicle counter - measure system in one of the embodiments of this application, which includes:

[0163] The first acquisition module 11 is used to analyze the radar wave detection data captured by the radar detection base station, and obtain the three - dimensional coordinates of the unmanned aerial vehicle to be identified according to the analysis result;

[0164] The identification module 12 is used to obtain the spectrum data of the unmanned aerial vehicle to be identified based on the spectrum analysis base station and the three - dimensional coordinates, input the spectrum data into the intrusion unmanned aerial vehicle detection model constructed by the deep - learning algorithm, and generate an intrusion unmanned aerial vehicle signal corresponding to the identification result output by the intrusion unmanned aerial vehicle detection model;

[0165] The second acquisition module 13 is used to, in response to the intrusion unmanned aerial vehicle signal, obtain the multi - spectral imaging data of the intrusion unmanned aerial vehicle based on the visual image acquisition base station;

[0166] The fusion module 14 is used to perform feature - level fusion on the radar wave detection data, spectrum data, and multi - spectral imaging data based on the multi - modal fusion algorithm of spatio - temporal registration, and obtain the fusion feature data of the target area;

[0167] The construction module 15 is used to extract features of different dimensions from the fusion feature data, and establish a three - dimensional heat map of the threat situation of the intrusion unmanned aerial vehicle based on the extracted features, where the dimensions at least include the coordinate dimension, function dimension, flight attitude dimension, and communication dimension;

[0168] The execution module 16 is used to execute the security counter - measure strategy generated based on the three - dimensional heat map of the threat situation.

[0169] Compared with the prior art, the beneficial effects of the embodiments of this application are at least one of the following:

[0170] (1) The multi-modal fusion algorithm based on spatio-temporal registration in this application performs feature-level fusion on radar wave detection data, spectral data, and multi-spectral imaging data, fully combining the advantages of radar in long-distance detection, target positioning, and speed measurement, as well as the advantages of multi-spectral imaging in target feature recognition. More comprehensive and accurate target information can be obtained, overcoming the limitations of a single data source, improving the detection, tracking, and recognition performance of unmanned aerial vehicles (UAVs), and enhancing the adaptability and reliability of the system in complex environments.

[0171] (2) This application performs multi-dimensional feature extraction on the fused feature data and establishes a three-dimensional heat map of the threat situation, which can intuitively and visually present the threat situation of UAVs. From the coordinate dimension, the position information of UAVs can be accurately grasped. From the function dimension, the tasks that can be performed or the equipment carried can be understood. From the flight attitude dimension, it helps to judge its flight intention and trajectory changes. From the communication dimension, its communication status and whether there are abnormal communication behaviors can be monitored. It provides comprehensive and clear situation awareness for decision-makers, facilitating the rapid and accurate formulation of countermeasures.

[0172] The above-described embodiments merely represent several implementation manners of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent of this application. It should be noted that for those of ordinary skill in the art, without departing from the concept of this application, several modifications and improvements can still be made, and these all belong to the protection scope of this application. Therefore, the protection scope of the patent of this application shall be subject to the appended claims.

Claims

1. A method for countering drones in a fixed area, characterized in that: A number of radar detection base stations, spectrum analysis base stations and visual image acquisition base stations are set up in the target area. The fixed area drone countermeasure method includes: Analyze the radar wave detection data captured by the radar detection base station, and obtain the three-dimensional coordinates of the drone to be identified according to the analysis results; Acquire spectrum data of the drone to be identified based on the spectrum analysis base station and the three-dimensional coordinates, input the spectrum data into an intrusion drone detection model constructed by a deep learning algorithm, and generate an intrusion drone signal according to the identification result output by the intrusion drone detection model; In response to the intrusion drone signal, acquiring multispectral imaging data of the intrusion drone based on the visual image acquisition base station; Based on a multimodal fusion algorithm of spatiotemporal registration, the radar wave detection data, the spectrum data and the multispectral imaging data are fused at a feature level to obtain fused feature data of the target area; Extracting features of different dimensions from the fused feature data, and establishing a three-dimensional heat map of the threat situation of the invading UAV based on the extracted features, wherein the dimensions include at least a coordinate dimension, a functional dimension, a flight attitude dimension, and a communication dimension; Execute a security countermeasure strategy generated based on the three-dimensional heat map of the threat situation.

2. The fixed area drone countermeasure method according to claim 1, characterized in that: The number of radar detection base stations deployed is N 雷达 It is expressed as: Among them, L and W are the length and width of the target area respectively, and R 雷达 is the effective detection radius of a single radar; The number of spectrum analysis base stations deployed is N 频谱 and the maximum spacing D 频谱 It is expressed as: Among them, R 频谱 Analyze the effective coverage radius of a base station for a single spectrum; The number of deployments of the visual image acquisition base stations N 视觉 It is expressed as: N 视觉 =2min(2×H,R 视觉 ) Among them, H is the protection height of the target area, R 视觉 It is the effective recognition distance of the visual image acquisition base station.

3. The fixed area drone countermeasure method according to claim 1, characterized in that: The step of inputting the spectrum data into an intrusion drone detection model constructed by a deep learning algorithm, and generating an intrusion drone signal according to a recognition result output by the intrusion drone detection model, includes: Acquire historical spectrum data of the intruding UAV, and perform preprocessing operations on the historical spectrum data; Constructing an initial intrusion drone detection model based on a convolutional neural network, inputting the preprocessed historical spectrum data into the initial intrusion drone detection model for training, and obtaining a trained intrusion drone detection model; The spectrum data acquired in real time is input into the intrusion drone detection model, and the type of the drone to be identified is determined based on the output result, and an intrusion drone signal is generated accordingly according to the identification result.

4. The fixed area drone countermeasure method according to claim 1, characterized in that: The step of analyzing the radar wave detection data captured by the radar detection base station and obtaining the three-dimensional coordinates of the drone to be identified according to the analysis result includes: Acquire azimuth data and distance data of the unmanned aerial vehicle to be identified relative to each radar detection base station, which are synchronously collected by at least three radar detection base stations; Taking the position of any radar detection base station as the center of the sphere, establish a spherical coordinate system equation group; Solving the spherical coordinate system equations according to the azimuth data and the distance data to obtain the coordinates of the drone to be identified; Wherein, the spherical coordinate system equations are expressed as: Among them, (x0, y0, z0) represents the coordinates of the selected millimeter radar wave, (x1, y1, z1) and (x2, y2, z2) represent the coordinates of other millimeter radar waves other than the selected millimeter radar wave, and r0, r1, and r2 respectively represent the distance from the drone to be identified to the corresponding millimeter radar wave.

5. The fixed area drone countermeasure method according to claim 1, characterized in that: The multimodal fusion algorithm based on spatiotemporal registration performs feature-level fusion on the radar wave detection data, the spectrum data and the multispectral imaging data to obtain fused feature data of the target area, specifically including: Performing time consistency processing on the radar wave detection data, the spectrum data and the multi-spectral imaging data, and mapping the radar wave detection data, the spectrum data and the multi-spectral imaging data after the time consistency processing to the same spatial coordinate system to complete time-space registration; Performing feature extraction on the radar wave detection data, the spectrum data and the multi-spectral imaging data after time-space registration respectively to obtain position feature data and visual feature data of the target area; The posture feature data and the visual feature data are fused based on a fusion algorithm to obtain fused feature data of the target area.

6. The fixed area drone countermeasure method according to claim 1, characterized in that: The extracting features of different dimensions from the fused feature data and establishing a three-dimensional heat map of the threat situation of the invading drone based on the extracted features includes: Obtain a unique identifier of the invading drone, and associate the fused feature data with the unique identifier to obtain state information data of the invading drone, wherein the state information data set includes coordinate dimension data, function dimension data, flight attitude dimension data, and communication dimension data; Quantifying the coordinate dimension data, the function dimension data, the flight attitude dimension data, and the communication dimension data respectively, and performing weighted calculation according to the state information data after the quantization process to obtain the threat level of the invading drone; A three-dimensional rectangular coordinate system is established for the target area, and the position coordinates and threat level of the invading drone are mapped into the three-dimensional rectangular coordinate system to construct a three-dimensional heat map of the threat situation.

7. The fixed area drone countermeasure method according to claim 1, characterized in that: The executing of the security countermeasure strategy generated based on the threat situation three-dimensional heat map includes: Obtaining the threat level of the invading drones based on the threat situation three-dimensional heat map, wherein the threat level includes a low threat level, a medium threat level, and a high threat level, and the threat level reflects the number and flight speed of the invading drones; If the threat level is a low threat level, use a nearby ground electromagnetic interference base station to carry out a directional countermeasure against the invading drone; If the threat level is a medium threat level, dispatch the aerial drone countermeasure platform to surround and counter the invading drone in different areas; If the threat level is a high threat level, the invading drone is directedly countered through the ground electromagnetic interference base station, and the aerial drone countermeasure platform is dispatched to surround and counter the invading drone in different areas.

8. The method for countering a UAV in a fixed area as claimed in claim 7, characterized in that: The dispatching of the aerial drone countermeasure platform to surround and counter the invading drone in different areas includes: Determine an airborne device that matches the type of the intruding drone, wherein the airborne device includes a radio countermeasure device, a directional antenna, an omnidirectional antenna, a strong light irradiation lamp, and a capture net, and the operating frequency band of the airborne device is compatible with the interference frequency band of the ground electromagnetic interference base station; The airborne equipment is installed on the aerial drone countermeasure platform by driving the crawler tracks through a stepper motor, and the aerial drone countermeasure platform is controlled to perform regional encirclement and countermeasure on the invading drone.

9. The fixed area drone countermeasure method according to claim 7, characterized in that: The ground electromagnetic interference base station deployment method is expressed as: D 干扰 ≤1.5R 干扰 Among them, N 干扰 represents the number of ground electromagnetic interference base stations deployed, D 干扰 Represents the distance between any two ground electromagnetic interference base stations, R 干扰 is the effective interference radius of a single ground electromagnetic interference base station, L and W represent the length and width of the target area respectively.

10. A fixed area drone countermeasure system, characterized in that: include: A first acquisition module is used to analyze the radar wave detection data captured by the radar detection base station, and obtain the three-dimensional coordinates of the drone to be identified according to the analysis results; an identification module, configured to obtain spectrum data of the to-be-identified drone based on the spectrum analysis base station and the three-dimensional coordinates, input the spectrum data into an intrusion drone detection model constructed by a deep learning algorithm, and generate an intrusion drone signal according to the identification result output by the intrusion drone detection model; A second acquisition module is used to respond to the intrusion drone signal and acquire multispectral imaging data of the intrusion drone based on the visual image acquisition base station; A fusion module, for performing feature-level fusion on the radar wave detection data, the spectrum data and the multi-spectral imaging data based on a multi-modal fusion algorithm of spatiotemporal registration to obtain fused feature data of the target area; A construction module is used to extract features of different dimensions from the fused feature data, and establish a three-dimensional heat map of the threat situation of the invading drone based on the extracted features, wherein the dimensions include at least a coordinate dimension, a functional dimension, a flight attitude dimension, and a communication dimension; An execution module is used to execute a security countermeasure strategy generated based on the three-dimensional heat map of the threat situation.

Images