Unmanned aerial vehicle monitoring and countering integrated system

Abstract

The invention discloses an unmanned aerial vehicle monitoring and countering integrated system, and relates to the technical field of unmanned aerial vehicle monitoring, and the system comprises a sensing module which is used for carrying out the monitoring of an environment through a plurality of sensors, and obtaining original data; the fusion module is used for processing the original data by adopting a data fusion algorithm to obtain target information; the identification module is used for outputting the type and behavior mode of the unmanned aerial vehicle; the analysis module is used for evaluating the type and behavior mode of the unmanned aerial vehicle by adopting a dynamic threat evaluation method and generating early warning information; the strategy module is used for formulating a dynamic countering strategy; the execution module is used for implementing corresponding countering measures and monitoring the countering effect in real time; and the recording module is used for recording all results in the whole process. Through the technical means of multi-sensor cooperative monitoring, data fusion processing, classification identification, dynamic threat assessment, game decision making, intelligent countering and the like, omnibearing perception, accurate identification, intelligent assessment and efficient disposal of the unmanned aerial vehicle are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) monitoring, and in particular to an integrated UAV monitoring and countermeasure system. Background Art

[0002] With the rapid development and widespread application of drone technology, the security threats posed by illegal or malicious drone use are becoming increasingly prominent. Unauthorized drone intrusions are frequent around military facilities, critical infrastructure, large-scale public events, and sensitive areas, posing serious challenges to national security, public safety, and personal privacy. Therefore, efficient and reliable drone monitoring and countermeasures have become a key research area in the current security defense field.

[0003] Existing drone monitoring technologies primarily rely on single or simple combinations of sensor devices, such as radar systems, radio frequency detectors, and optoelectronic devices. These systems have significant limitations in practical applications: single radar systems have limited detection capabilities for small, low-altitude targets and are susceptible to terrain interference; radio frequency detection devices can only identify drones using specific communication protocols and are ineffective against autonomous drones; and optoelectronic devices are severely limited by weather and lighting conditions. While some research has attempted to combine multiple sensors, the lack of effective data fusion algorithms results in each sensor operating independently, preventing them from fully leveraging their synergy.

[0004] When it comes to drone identification and threat assessment, existing technologies generally employ simple rule-matching or fixed threshold judgment methods. Regarding countermeasures, existing systems typically employ pre-defined, fixed countermeasures, such as electromagnetic interference and GPS spoofing. Furthermore, existing drone monitoring and countermeasure systems are often designed separately, making it difficult to seamlessly transmit target information acquired by the monitoring system to the countermeasure system, resulting in information delays and decision-making lags. Furthermore, most systems lack effective effect evaluation and experience-based learning mechanisms, lacking the ability to self-optimize and thus struggling to cope with the increasingly complex drone threat.

[0005] Therefore, there is an urgent need to develop an intelligent system that integrates monitoring, identification, evaluation, decision-making and countermeasures. Summary of the Invention

[0006] In view of the above-mentioned existing problems, the present invention provides an integrated drone monitoring and countermeasure system. Through multi-sensor collaborative monitoring and efficient data fusion, it can accurately identify the type and behavior pattern of drones, and introduce a threat assessment method based on dynamic weight coefficients. It solves the problems that the existing early warning mechanism relies only on simple threshold judgment, lacks adaptive adjustment to environmental conditions and historical data, and cannot accurately assess diverse drone threats. It also realizes intelligent countermeasure strategy selection based on game theory to effectively deal with various drone intrusion scenarios.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, the present invention provides an integrated drone monitoring and countermeasure system, comprising:

[0009] The sensing module is used to monitor the environment using multiple sensors and obtain raw data;

[0010] The fusion module is used to process the original data using the data fusion algorithm to obtain the target information;

[0011] The recognition module is used to process target information using a pre-trained classification model to identify and output the drone type and behavior pattern;

[0012] An analysis module, which uses dynamic threat assessment methods to evaluate drone types and behavior patterns, predict potential threats, and generate early warning information;

[0013] Strategy module, used to process early warning information using game theory and formulate dynamic countermeasure strategies;

[0014] The execution module is used to implement corresponding countermeasures according to the dynamic countermeasure strategy and monitor the countermeasure effects in real time;

[0015] The recording module is used to record all results of the entire process, including original data, processing results, decision basis and final countermeasure effect.

[0016] Preferably, the dynamic threat assessment method comprises the following steps:

[0017] Based on the drone type and behavior pattern, a multi-dimensional threat feature vector is constructed, wherein the multi-dimensional threat feature vector includes spatial position, motion state, behavior pattern and historical trajectory characteristics;

[0018] The threat level of the drone is calculated using the threat assessment function, and its mathematical expression is:

[0019]

[0020] Where, T i (t) represents the threat level of the i-th UAV target at time t, V i (t) represents the multi-dimensional threat feature vector of the target at time t, E(t) represents the environmental condition vector at time t, and f k represents the kth threat assessment subfunction, α k (t) represents the dynamic weight coefficient of the kth threat assessment sub-function, and n represents the number of threat assessment sub-functions.

[0021] Preferably, the dynamic weight coefficient α k (t) is calculated as follows:

[0022]

[0023] Where, β k is the basic weight coefficient, g k is the weight adjustment function, H(t) is the historical threat data sequence;

[0024] Among them, the weight adjustment function g k The applicability of the kth threat assessment sub-function is dynamically adjusted based on historical threat assessment accuracy and current environmental conditions.

[0025] Preferably, the threat assessment subfunction includes:

[0026] Regional sensitivity function, expression is:

[0027]

[0028] Where, d i (t) is the distance from the UAV to the nearest sensitive area, d max is the maximum reference distance, p i (t) is the current position of the UAV, S(p) is the regional sensitivity function of position p, and σ is the sigmoid function;

[0029] Dynamic approach speed function, the expression is:

[0030]

[0031] Where, is the UAV velocity vector, is the direction vector from the UAV to the nearest sensitive target, v ref is the reference speed.

[0032] Preferably, the threat assessment subfunction further includes:

[0033] The behavior abnormality function is expressed as:

[0034]

[0035] Where B i (t) is the behavior characteristic distribution of the UAV at time t, B norm (T d ) is of type T d The normal behavior characteristic distribution of UAV, D KL is the KL divergence, λ is the scaling factor;

[0036] The historical trajectory consistency function is expressed as:

[0037]

[0038] Where H i (t) is the historical trajectory sequence of the UAV, P i (t) is the predicted trajectory based on historical data, Δ represents the deviation measure between the actual trajectory and the predicted trajectory, and γ is a non-negative adjustment parameter.

[0039] Preferably, the dynamic threat assessment method further comprises the following steps:

[0040] According to the calculated threat level T i (t) and preset threat thresholds to provide graded warnings for drone targets;

[0041] Compare the threat assessment results with the actual countermeasure effects and continuously optimize the dynamic weight coefficient α k (t) calculation;

[0042] In a multi-target scenario, the threat correlation between targets is calculated to form a group threat assessment result.

[0043] Preferably, the multiple sensors used in the sensing module include a lidar, an infrared thermal imager, a radio frequency signal detector, an acoustic microphone array, and a visible light camera, and synchronously collect real-time information in the target area under unified clock control.

[0044] Preferably, the steps of the data fusion algorithm include:

[0045] Apply Kalman filtering to the raw data for time series state estimation;

[0046] Extracting a multimodal feature set from the filtered data;

[0047] Apply principal component analysis to reduce the dimensionality of multimodal feature sets;

[0048] The physical feature parameter set of the target is extracted from the feature matrix after dimensionality reduction as the target information.

[0049] Preferably, the game theory method adopted by the strategy module includes the following steps:

[0050] Extract the warning information into threat state vector and input it into the game theory decision model;

[0051] Based on the current threat state, a game model between the defender and the invader is constructed, and the Nash equilibrium algorithm is applied to find the optimal strategy combination.

[0052] The optimal strategy combination obtained is comprehensively analyzed with historical confrontation data and real-time environmental factors, and the most suitable countermeasure strategy for the current situation is selected through the artificial intelligence-assisted decision-making module;

[0053] The finalized countermeasure strategy is converted into execution instructions and passed to the execution module.

[0054] Preferably, the countermeasures implemented by the execution module include:

[0055] Electromagnetic interference, which interferes with the drone's communication link by emitting electromagnetic signals in a specific frequency band;

[0056] GPS spoofing, which misleads a drone's navigation system by sending fake GPS signals;

[0057] Radio frequency detection and suppression, accurately identifying and blocking the communication signals between the drone and the remote controller;

[0058] Laser blinding, which uses a directed laser beam to interfere with the drone's electro-optical sensors;

[0059] Net capture interception: using a physical net capture device to physically intercept the drone;

[0060] Trick the drone into returning home by sending specific instructions to trigger the drone's loss of connection protection mechanism to make it return home automatically.

[0061] The beneficial effects of the present invention are as follows: through multi-level technical means such as multi-sensor collaborative monitoring, data fusion processing, classification and identification, dynamic threat assessment, game decision-making and intelligent countermeasures, all-round perception, accurate identification, intelligent assessment and efficient disposal of drones are achieved. The system uses a heterogeneous sensor array to obtain high-quality raw data, and uses Kalman filtering, PCA dimensionality reduction and other algorithms for efficient processing, which significantly improves the target recognition accuracy. The threat assessment method based on dynamic weight coefficients realizes the accurate quantification of threat levels, and comprehensively analyzes drone threats through evaluation sub-functions such as regional sensitivity and dynamic approach speed, and can adaptively adjust evaluation weights according to environmental conditions and historical data, thereby enhancing the system's ability to cope with diverse threats. The game theory method is used to formulate dynamic countermeasures, combined with the coordinated implementation of multiple countermeasures such as electromagnetic interference and GPS spoofing, to achieve accurate and efficient strikes against invading drones. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0063] Figure 1 Schematic diagram of the integrated drone monitoring and countermeasure system in Example 1. DETAILED DESCRIPTION

[0064] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0065] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0066] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0067] Reference Figure 1 The present invention provides an integrated drone monitoring and countermeasure system, comprising:

[0068] The sensing module is used to monitor the environment using multiple sensors and obtain raw data;

[0069] The fusion module is used to process the original data using the data fusion algorithm to obtain the target information;

[0070] The recognition module is used to process target information using a pre-trained classification model to identify and output the drone type and behavior pattern;

[0071] An analysis module, which uses dynamic threat assessment methods to evaluate drone types and behavior patterns, predict potential threats, and generate early warning information;

[0072] Strategy module, used to process early warning information using game theory and formulate dynamic countermeasure strategies;

[0073] The execution module is used to implement corresponding countermeasures according to the dynamic countermeasure strategy and monitor the countermeasure effects in real time;

[0074] The recording module is used to record all results of the entire process, including original data, processing results, decision basis and final countermeasure effect.

[0075] Specifically, in one embodiment of the present invention, a heterogeneous sensor array is deployed, which includes a lidar, an infrared thermal imager, a radio frequency signal detector, an acoustic microphone array, and a visible light camera; the sensors are synchronously started under unified clock control, collect real-time information within the target area, and output a raw data set; the standardized sensor data is integrated using a weighted summation method to generate a unified raw data vector.

[0076] It should be noted that the heterogeneous sensor array achieves full coverage of the target environment through multimodal perception means. LiDAR provides high-precision spatial position information, infrared thermal imagers can still effectively detect in low light or night conditions, radio frequency signal detectors can capture drone remote control and image transmission signals, acoustic microphone arrays are used to identify propeller noise characteristics, and visible light cameras provide the basis for image recognition. All sensors work synchronously under unified clock control to ensure the time consistency of data collection and avoid data fusion errors caused by time deviation. The standardized sensor data is integrated using weighted summation, and the weights can be dynamically adjusted according to the performance of different sensors to improve the overall data quality.

[0077] Specifically, in one embodiment of the present invention, the data fusion algorithm includes:

[0078] The Kalman filter is used to perform preliminary fusion on the unified original data vector to obtain preliminary fusion data;

[0079] Perform feature extraction on the preliminary fusion data to extract the multimodal feature set M;

[0080] Principal component analysis (PCA) is used to reduce the dimensionality of the multimodal feature set M and output the reduced dimensionality feature matrix M reduced , the expression is:

[0081] M reduced = PCA(M);

[0082] Among them, PCA(.) represents the principal component analysis function, M is the multimodal feature set, M reduced is the feature matrix after dimensionality reduction;

[0083] Based on physical feature matching and rule judgment, the feature matrix M after dimensionality reduction is reduced Perform classification and recognition to obtain preliminary target category information;

[0084] From the feature matrix M after dimensionality reduction reduced Extract the target's physical feature parameter set F phy , the expression is:

[0085] F phy ={v,a,s,h,p};

[0086] Where v is the target's velocity, a is the target's acceleration, s is the target's size, h is the target's height, and p is the target's power signal strength.

[0087] The extracted physical feature parameter set F phy Compare with the preset drone feature template library, match the most similar template, and obtain a set of candidate target categories;

[0088] Select the most likely target category based on the minimum distance criterion and generate the final preliminary target category information;

[0089] It should be noted that the initial fusion process of Kalman filtering can effectively reduce the noise interference in the original data and improve data stability; the multimodal feature set subsequently extracted covers information in multiple dimensions such as space, motion, and electromagnetic, providing rich discrimination basis for subsequent classification and recognition. The principal component analysis (PCA) method is used to reduce the dimensionality of the features, which not only reduces the computational burden brought by redundant information, but also retains the most representative key features, thereby improving recognition efficiency and accuracy. Based on the physical feature matching and rule judgment mechanism, the system can achieve rapid target classification in the absence of a large number of training samples, which is especially suitable for drone recognition scenarios under complex backgrounds.

[0090] Specifically, in one embodiment of the present invention, the processing of the identification module includes:

[0091] Process the preliminary fusion data and the final target information, classify objects into different categories, and obtain the category classification results;

[0092] Based on the classification results and original data, a classification model is constructed and pre-trained. The expression is:

[0093] M class =BUILD(G,DB);

[0094] Among them, BUULD(.) represents the classification model building function, G is the category division result, D is the original data, M class The classification model constructed;

[0095] Input the target information into the classification model, perform the classification operation, and output the drone type. The expression is:

[0096] T d =CLASSIFY(O,M class );

[0097] Among them, CLASSIFY(.) represents the classification function, O is the target information, M class is the classification model, T d is the type of drone identified;

[0098] Based on drone type T d And historical trajectory data, analyze the target's behavior pattern and obtain behavior pattern B m ;

[0099] Set drone type to T d and behavior pattern B mCombine to form complete drone type and behavior pattern information;

[0100] It should be noted that by combining clustering division with classification models, a target recognition process from coarse to fine is realized. When constructing the classification model, the original data and fusion results are fully combined to improve the generalization ability of the model. Historical trajectory data is used to assist in analyzing target behavior patterns, so that the system can not only identify the type of drone, but also further understand its flight intentions, such as patrolling, hovering, approaching and other behaviors, which significantly enhances the system's intelligent recognition level. The final output of complete drone type and behavior pattern information provides a reliable basis for subsequent threat assessment and strategy formulation.

[0101] Specifically, in one embodiment of the present invention, the analysis module uses a dynamic threat assessment method to evaluate drone types and behavior patterns, predict potential threats, and generate early warning information. This method combines multiple threat assessment dimensions to accurately quantify the potential risks posed by drone targets. It also dynamically adjusts the assessment strategy based on environmental conditions and historical data, providing highly accurate threat assessment results. The specific implementation process is as follows:

[0102] First, the analysis module receives the drone type and behavior pattern information from the recognition module, and combines it with the target's historical trajectory data to construct a multi-dimensional threat feature vector V i (t). This vector contains the following four key features:

[0103] Spatial location features: including the three-dimensional coordinates of the drone (x i (t),y i (t),z i (t)), the shortest distance d to the boundary of the sensitive area i (t) and the distance matrix D to the sensitive target point i (t);

[0104] Motion state characteristics: including the velocity vector of the drone acceleration vector and the angular velocity ω i (t);

[0105] Behavior mode characteristics: including the current flight mode M i (t) (such as hovering, straight flight, circling flight, etc.), the typical behavior parameters P corresponding to the UAV model i (T d ) and the frequency of behavioral changes F i (t);

[0106] Historical trajectory features: including the position sequence of the past N time points {p i (t-nΔt)|n=1,2,...,N} and the corresponding velocity sequence

[0107] In addition, the system also maintains an environmental condition vector E(t), which includes environmental factors such as current weather conditions, lighting conditions, electromagnetic environment, and the importance of the protected area.

[0108] Based on the drone type and behavior pattern, a multi-dimensional threat feature vector is constructed, wherein the multi-dimensional threat feature vector includes spatial position, motion state, behavior pattern and historical trajectory characteristics;

[0109] Based on the constructed multi-dimensional threat feature vector, the analysis module uses the following threat assessment function to calculate the threat level of the drone target:

[0110]

[0111] Where, T i (t) represents the threat level of the i-th UAV target at time t, V i (t) represents the multi-dimensional threat feature vector of the target at time t, E(t) represents the environmental condition vector at time t, and f k represents the kth threat assessment subfunction, α k (t) represents the dynamic weight coefficient of the kth threat assessment sub-function, satisfying n represents the number of threat assessment sub-functions. In this embodiment, n=4.

[0112] Dynamic weight coefficient α k (t) is calculated as follows:

[0113]

[0114] Where, β k is the basic weight coefficient, g k is the weight adjustment function, H(t) is the historical threat data sequence;

[0115] Among them, the weight adjustment function g k The applicability of the kth threat assessment sub-function is dynamically adjusted based on historical threat assessment accuracy and current environmental conditions.

[0116] In this embodiment, the weight adjustment function g k The specific implementation form is:

[0117] g k (H(t),E(t))=w1·Acc k (H(t))+w2·Rel k (E(t))

[0118] Among them, Acc k(H(t)) represents the accuracy score of the k-th evaluation subfunction in the historical evaluation, and the calculation formula is:

[0119]

[0120] V l and E l Represent the feature vector and environmental conditions of the lth sample in the historical records, A l Indicates the actual threat level of the sample (obtained by inferring the countermeasure effect), and L represents the total number of historical samples;

[0121] Rel k (E(t)) represents the applicability score of the k-th evaluation sub-function to the current environmental conditions, and the calculation formula is:

[0122]

[0123] Among them, e m (t) represents the mth component of the environmental condition vector E(t), M represents the dimension of the environmental condition, c m represents the importance coefficient of the mth environmental factor, φ k Represents the fitness function of the kth evaluation sub-function to a specific environmental factor.

[0124] w1 and w2 are weight coefficients for balancing historical accuracy and current applicability, satisfying w1+w2=1.

[0125] This embodiment uses four threat assessment sub-functions to assess the threat level of drones from different perspectives:

[0126] The regional sensitivity function evaluates the spatial relationship between the drone and the sensitive area, and its expression is:

[0127]

[0128] Where, d i (t) is the distance from the UAV to the nearest sensitive area, d max is the maximum reference distance, p i (t) is the current position of the UAV, S(p) is the regional sensitivity function of position p, and σ is the sigmoid function; the regional sensitivity function S(p) is defined as:

[0129]

[0130] In the formula, Q represents the set of all sensitive target points, q represents the position of a sensitive target point, and I q Represents the importance coefficient of the sensitive target, σ qIt represents the influence range parameter of the sensitive target, and ||pq|| represents the Euclidean distance from position p to target point q.

[0131] The dynamic approach speed function evaluates the speed and direction of the UAV approaching the sensitive target, and the expression is:

[0132]

[0133] Where, is the UAV velocity vector, is the direction vector from the UAV to the nearest sensitive target, v ref The function outputs 0 when the drone is far away from the sensitive target, and the output value is proportional to the approach speed when the drone is flying towards the sensitive target. ref When , the function value can be truncated to ensure that the output does not exceed 1.

[0134] The behavior abnormality function evaluates the degree of abnormal behavior by comparing the current behavior of the drone with the normal behavior pattern of the drone of this type:

[0135]

[0136] Where B i (t) is the behavior characteristic distribution of the UAV at time t, B norm (T d ) is of type T d The normal behavior characteristic distribution of UAV, D KL is the KL divergence, and λ is the scaling factor.

[0137] The historical trajectory consistency function evaluates the degree of abnormality of the flight trajectory by comparing the deviation between the actual trajectory of the drone and the predicted trajectory:

[0138]

[0139] Where H i (t) is the historical trajectory sequence of the UAV, P i (t) is the trajectory predicted based on historical data, Δ represents the deviation measure between the actual trajectory and the predicted trajectory, and γ is a non-negative adjustment parameter. The trajectory deviation measure Δ is calculated as:

[0140]

[0141] Where p i (t-jΔt) represents the position at time t-jΔt in the actual trajectory, represents the position at the corresponding moment in the predicted trajectory, w jis the time weighting coefficient, and N is the number of historical time points considered. Trajectory prediction uses methods such as Kalman filtering or long short-term memory (LSTM) networks to predict future trajectories based on past position and velocity sequences. A significant deviation between the actual trajectory and the predicted trajectory indicates that the drone may have performed unusual maneuvers, increasing the threat level.

[0142] According to the calculated threat level T i (t), the system divides drone targets into different alert levels:

[0143] (1) Security level: T i (t) <T safe , indicating that the target poses no obvious threat and the system continues to monitor;

[0144] (2) Attention level: T safe ≤T i (t) <T warn , indicating that the target needs to be focused on and the monitoring frequency should be increased;

[0145] (3) Warning level: T warn ≤T i (t) <T alert , indicating that the target is a potential threat, the system enters the early warning state and prepares countermeasures;

[0146] (4) Danger level: T i (t)≥T alert , indicating that the target poses an obvious threat, and the system immediately initiates the countermeasure process.

[0147] Among them, T safe 、T warn and T alert Threat thresholds are preset and can be dynamically adjusted based on the security requirements of the protected area. The system configures visual and audible alarms for different threat levels and automatically adjusts monitoring resource allocation based on the threat level.

[0148] The system continuously optimizes the calculation model of dynamic weight coefficients by comparing threat assessment results with actual countermeasure effects. The specific steps are as follows:

[0149] (1) Record the target information, evaluation results, and actual effects of each countermeasure action; (2) Analyze the contribution of each evaluation sub-function to the final threat assessment; (3) Calculate the accuracy score of each evaluation sub-function; (4) Update the parameters of the weight adjustment function based on new historical data; (5) Adjust the calculation model of the applicability score according to changes in environmental conditions.

[0150] The optimization process uses methods such as gradient descent or genetic algorithms to minimize the error between the assessment results and the actual threat level, and continuously improve the system's threat assessment accuracy.

[0151] In multi-target scenarios, the system not only assesses the threat of a single drone, but also considers the group threat posed by the coordinated behavior of multiple drones. Group threat assessment includes the following steps:

[0152] (1) Calculate the threat correlation matrix C between drone targets, where the element C_{ij} represents the correlation between the i-th and j-th drones:

[0153]

[0154] Among them, σ d and σ v are the influencing factors of distance and speed, S behav (M i ,M j ) represents the similarity of behavior patterns;

[0155] (2) Identify potential cooperative drone groups based on the correlation matrix;

[0156] (3) Calculate the group threat level for the identified group:

[0157]

[0158] Among them, G represents the drone group, δ is the group effect coefficient, and F corr (G) is the correlation strength function of the drones in the group;

[0159] (4) Adjust the warning level and countermeasure strategy priority based on the group threat level.

[0160] Through the above-mentioned group threat assessment mechanism, the system can effectively respond to multi-UAV collaborative invasion scenarios and identify and deal with potential group threats in advance.

[0161] Strategy module, used to process early warning information using game theory combined with artificial intelligence methods and formulate dynamic countermeasure strategies;

[0162] Furthermore, we define the players in the game theory, where the defender is defined as the drone monitoring and countermeasure system, and the attacker is defined as the invading drone;

[0163] Set the state space S, which includes the position, speed and behavior pattern of the drone, and set the action set A for each party;

[0164] Define a utility function to reflect the preferences of each party for different outcomes;

[0165] Based on the above settings, we construct a bounded rational game model G, which is expressed as:

[0166] G=(P,S,A,U);

[0167] Among them, P represents the set of participants, S represents the state space, A represents the action set, and U represents the utility function set;

[0168] The utility function U includes:

[0169] The defender's utility function represents the benefits gained by the defender from taking actions in the state and facing the attacker's actions;

[0170] The attacker's utility function represents the benefits gained by the attacker from taking actions in the state and facing the defender's actions;

[0171] Input the warning information W into the game theory decision model G, extract the current threat state parameter set, and obtain the threat state vector;

[0172] The Nash equilibrium solution algorithm is used to solve the game model G, find the optimal strategy combination, and obtain the optimal counter-strategy candidate set;

[0173] The optimal countermeasure candidate set is input into the AI-assisted decision-making module, which combines historical confrontation data and real-time environmental factors to select the most suitable countermeasure strategy.

[0174] The artificial intelligence-assisted decision-making module includes a strategy evaluator, a historical matching engine, a real-time factor fusion device and a decision optimizer;

[0175] The strategy evaluator is used to score the candidate counter-strategies output by game theory using XGBoost;

[0176] The historical matching engine is used to retrieve similar cases based on historical adversarial data to improve decision robustness;

[0177] The real-time factor aggregator is used to dynamically adjust the strategy weights in combination with the current environment parameters;

[0178] The decision optimizer uses a weighted scoring mechanism to comprehensively consider the strategy score, historical matching degree, and environmental adaptability, and outputs the final countermeasure strategy, specifically:

[0179] Input the current threat state vector and the candidate strategy set output by game theory into the strategy evaluator, and output the comprehensive score corresponding to each strategy;

[0180] Based on the current status, similar confrontation cases are searched in the historical database to extract past successful / failed strategic experiences to provide reference for current decision-making;

[0181] Integrate dynamic environmental factors into strategy scoring, dynamically weight the applicability of different countermeasures, calculate the weighted total score, and select the optimal strategy;

[0182] Output the final countermeasure strategy to the execution module, triggering the corresponding countermeasure device to implement countermeasures;

[0183] It should be noted that the design of the bounded rationality game model fully reflects the essence of the strategic interaction between the two sides in a confrontational environment. It takes the drone monitoring and countermeasure system as the defender and the invading drone as the attacker, constructs the state space and action set, and defines the utility function to reflect the interest preferences of all parties. It has strong theoretical support and practical adaptability. The Nash equilibrium solution algorithm ensures that the system can find the optimal response plan among multiple possible strategies, and the artificial intelligence-assisted decision-making module combines historical confrontation data and real-time environmental factors to dynamically optimize strategy selection, greatly improving the flexibility and actual combat effectiveness of the countermeasure strategy.

[0184] The execution module is used to implement corresponding countermeasures according to the dynamic countermeasure strategy and monitor the countermeasure effects in real time;

[0185] Furthermore, the final selected countermeasure strategy is converted into an actual operation instruction set O;

[0186] Send the operation instruction set O to the corresponding countermeasure device, execute the countermeasures, and obtain the actual execution result;

[0187] Countermeasures include electromagnetic jamming, GPS spoofing, radio frequency detection and suppression, laser blinding, net capture and interception, acoustic jamming, and decoy return;

[0188] It should be noted that the execution module has highly integrated command conversion and execution capabilities, which can quickly convert the abstract strategies output by game decisions into specific operational instructions, and support the coordinated implementation of multiple countermeasures. Electromagnetic interference, GPS deception, radio frequency suppression, laser blinding, net interception, acoustic interference and decoy return measures each have their own focus and are suitable for different types of drones and different combat environments. The system has a real-time monitoring and feedback mechanism, which can dynamically adjust the strategy according to the execution effect to achieve closed-loop control. This module not only improves the automation level of the countermeasure system, but also enhances the ability to respond to sudden threats and ensures the safety and effectiveness of the countermeasure process.

[0189] Recording module, used to record all results of the whole process in detail;

[0190] All results include raw data, processing results, decision basis and final countermeasure effects.

[0191] This embodiment also provides a computer device suitable for the integrated drone monitoring and countermeasure system, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the integrated drone monitoring and countermeasure system proposed in the above embodiment.

[0192] The computer device may be a terminal, comprising a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0193] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the integrated drone monitoring and countermeasure system proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0194] In summary, the present invention realizes all-round perception, precise identification, intelligent evaluation and efficient disposal of aerial targets through multi-level technical means such as multi-sensor collaborative monitoring, data fusion processing, classification and identification, behavior analysis, game decision-making and dynamic countermeasures. The system uses a heterogeneous sensor array to obtain high-quality raw data, and performs data fusion and feature extraction through advanced algorithms such as Kalman filtering and PCA dimensionality reduction, which significantly improves the target recognition accuracy and robustness. It realizes quantitative threat assessment based on the calculation of behavioral risk index, forms a hierarchical early warning mechanism, and enhances the real-time response capability of the system. It adopts game theory combined with artificial intelligence methods to formulate dynamic countermeasure strategies, ensuring the flexibility and effectiveness of the response measures. Through the coordinated implementation of multiple countermeasures, it achieves rapid and precise strikes on invading drones.

[0195] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. The integrated drone monitoring and countermeasure system is characterized by: include: The sensing module is used to monitor the environment using multiple sensors and obtain raw data; The fusion module is used to process the original data using the data fusion algorithm to obtain the target information; The recognition module is used to process target information using a pre-trained classification model to identify and output the drone type and behavior pattern; An analysis module, which uses dynamic threat assessment methods to evaluate drone types and behavior patterns, predict potential threats, and generate early warning information; Strategy module, used to process early warning information using game theory and formulate dynamic countermeasure strategies; The execution module is used to implement corresponding countermeasures according to the dynamic countermeasure strategy and monitor the countermeasure effects in real time; The recording module is used to record all results of the entire process, including original data, processing results, decision basis and final countermeasure effect.

2. The integrated drone monitoring and countermeasure system according to claim 1, characterized in that: The dynamic threat assessment method includes the following steps: Based on the drone type and behavior pattern, a multi-dimensional threat feature vector is constructed, wherein the multi-dimensional threat feature vector includes spatial position, motion state, behavior pattern and historical trajectory characteristics; The threat level of the drone is calculated using the threat assessment function, and its mathematical expression is: Where, T i (t) represents the threat level of the i-th UAV target at time t, V i (t) represents the multi-dimensional threat feature vector of the target at time t, E(t) represents the environmental condition vector at time t, and f k represents the kth threat assessment subfunction, α k (t) represents the dynamic weight coefficient of the kth threat assessment sub-function, and n represents the number of threat assessment sub-functions.

3. The integrated drone monitoring and countermeasure system according to claim 2, characterized in that: Dynamic weight coefficient α k (t) is calculated as follows: Where, β k is the basic weight coefficient, g k is the weight adjustment function, H(t) is the historical threat data sequence; Among them, the weight adjustment function g k The applicability of the kth threat assessment sub-function is dynamically adjusted based on historical threat assessment accuracy and current environmental conditions.

4. The integrated drone monitoring and countermeasure system according to claim 2, wherein: Threat assessment sub-functions include: Regional sensitivity function, expression is: Where, d i (t) is the distance from the UAV to the nearest sensitive area, d max is the maximum reference distance, p i (t) is the current position of the UAV, S(p) is the regional sensitivity function of position p, and σ is the sigmoid function; Dynamic approach speed function, the expression is: Where, is the UAV velocity vector, is the direction vector from the UAV to the nearest sensitive target, v ref is the reference speed.

5. The integrated drone monitoring and countermeasure system according to claim 4, characterized in that: The threat assessment sub-function also includes: The behavior abnormality function is expressed as: Where B i (y) is the distribution of the UAV’s behavior characteristics at time t, B norm (T d ) is of type T d The normal behavior characteristic distribution of UAV, D KL is the KL divergence, λ is the scaling factor; The historical trajectory consistency function is expressed as: Where H i (t) is the historical trajectory sequence of the UAV, P i (t) is the predicted trajectory based on historical data, Δ represents the deviation measure between the actual trajectory and the predicted trajectory, and γ is a non-negative adjustment parameter.

6. The integrated drone monitoring and countermeasure system according to claim 2, characterized in that: The dynamic threat assessment method also includes the following steps: According to the calculated threat level T i (t) and preset threat thresholds to provide graded warnings for drone targets; Compare the threat assessment results with the actual countermeasure effects and continuously optimize the dynamic weight coefficient α k (t) calculation; In a multi-target scenario, the threat correlation between targets is calculated to form a group threat assessment result.

7. The integrated drone monitoring and countermeasure system according to claim 1, wherein: The sensing module uses multiple sensors including lidar, infrared thermal imager, radio frequency signal detector, acoustic microphone array and visible light camera, and synchronously collects real-time information in the target area under unified clock control.

8. The integrated drone monitoring and countermeasure system according to claim 1, wherein: The steps of the data fusion algorithm include: Apply Kalman filtering to the raw data for time series state estimation; Extracting a multimodal feature set from the filtered data; Apply principal component analysis to the multimodal feature set for dimensionality reduction; The physical feature parameter set of the target is extracted from the feature matrix after dimensionality reduction as the target information.

9. The integrated drone monitoring and countermeasure system according to claim 1, wherein: The game theory approach employed by the strategy module involves the following steps: Extract the warning information into threat state vectors and input them into the game theory decision model; Based on the current threat state, a game model between the defender and the invader is constructed, and the Nash equilibrium algorithm is applied to find the optimal strategy combination. The optimal strategy combination obtained is comprehensively analyzed with historical confrontation data and real-time environmental factors, and the most suitable countermeasure strategy for the current situation is selected through the artificial intelligence-assisted decision-making module; The finalized countermeasure strategy is converted into execution instructions and passed to the execution module.

10. The integrated drone monitoring and countermeasure system according to claim 1, wherein: The countermeasures implemented by the execution module include: Electromagnetic interference, which interferes with the drone's communication link by emitting electromagnetic signals in a specific frequency band; GPS spoofing, which misleads a drone's navigation system by sending fake GPS signals; Radio frequency detection and suppression, accurately identifying and blocking the communication signals between the drone and the remote controller; Laser blinding, which uses a directed laser beam to interfere with the drone's electro-optical sensors; Net capture interception: using a physical net capture device to physically intercept the drone; Trick the drone into returning home by sending specific instructions to trigger the drone's loss of connection protection mechanism to make it return home automatically.

Images