A method and system for generating UAV countermeasure plan based on artificial intelligence technology

Abstract

The present application provides a method and system for generating a UAV countermeasure plan based on artificial intelligence technology. Specifically, the present application analyzes the dynamic monitoring data stream of the target UAV, identifies its flight trajectory characteristics and electromagnetic signal spectrum, and constructs a three-dimensional spatial threat situation model to assess the threat level. Based on the threat level, the target UAV's behavior pattern and attack intention parameters are decomposed, and a dynamic decision factor set is generated in combination with geographic fence constraints and environmental interference factors. Multiple candidate countermeasure plans are generated and a plan logic tree is constructed by matching the basic countermeasure plans in the plan knowledge graph. A multidimensional evaluation parameter set is generated based on the three-dimensional spatial threat situation model, the dynamic decision factor set, and the countermeasure effectiveness index. The plan logic tree is collaboratively optimized to generate the optimal countermeasure plan and convert it into an executable instruction set, which is synchronized to the countermeasure device for execution, thereby improving the real-time performance and scenario adaptability of the UAV countermeasure plan.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of drone technology, and in particular to a method and system for generating a drone countermeasure plan based on artificial intelligence technology. Background Art

[0002] With the rapid development of drone technology, drones are increasingly used in both civilian and military applications. However, illegal or unauthorized drone activities pose a serious threat to public safety, privacy protection, and the security of critical infrastructure. Effectively monitoring and countering these potential threats has become an urgent technical need, especially in sensitive areas such as large public events, airports, and military facilities.

[0003] Existing drone countermeasures primarily rely on a single type of sensor, such as radar and radio frequency detectors, to monitor and identify drones. Once an unusual flying object is detected, the data obtained is manually analyzed to assess the threat level and formulate appropriate countermeasures. Furthermore, some advanced systems attempt to incorporate geofencing technology to restrict drones from entering specific airspaces. However, most of these solutions lack the ability to deeply integrate multi-source heterogeneous sensor data, making it difficult to construct accurate three-dimensional spatial threat situation models. Furthermore, the basic countermeasures are typically statically pre-set and cannot be dynamically adjusted based on real-time environmental factors, resulting in limited countermeasure effectiveness. These issues collectively contribute to the sluggish response of existing countermeasure systems and the inaccurate response strategies, which may be ineffective in preventing potential threats posed by drones. Summary of the Invention

[0004] The embodiments of the present application provide a method and system for generating drone countermeasure plans based on artificial intelligence technology, which is used to solve the problem in the existing technology that the existing technology lacks the ability to deeply integrate multi-source heterogeneous sensor data and cannot dynamically adjust according to real-time changing environmental factors, resulting in limited countermeasure effects.

[0005] In a first aspect, an embodiment of the present application provides a method for generating a drone countermeasure plan based on artificial intelligence technology, comprising:

[0006] Analyze the dynamic monitoring data stream obtained from the target UAV, identify the flight trajectory characteristics and electromagnetic signal spectrum of the target UAV, and build a three-dimensional spatial threat situation model by combining multi-source heterogeneous sensor fusion technology. Use the three-dimensional spatial threat situation model to assess the threat level of the target UAV and generate a threat level assessment result;

[0007] Based on the threat level assessment results, the target drone's behavior patterns and attack intention parameters are decomposed, and a dynamic decision factor set is generated by combining geo-fence constraints and environmental interference factors.

[0008] Generate multiple candidate countermeasure plans by matching the basic countermeasure plans preset in the plan knowledge graph through a dynamic decision factor set, and construct a plan logic tree based on the multiple candidate countermeasure plans;

[0009] Generate a multi-dimensional evaluation parameter set based on the three-dimensional threat situation model, the dynamic decision factor set, and the countermeasure effectiveness indicators preset in the plan knowledge graph;

[0010] The plan logic tree is collaboratively optimized based on a multi-dimensional evaluation parameter set to generate an optimal countermeasure plan, which is then converted into an executable instruction set and synchronized to the countermeasure device in the target area.

[0011] Optionally, a collaborative optimization process is performed on the plan logic tree based on a multi-dimensional evaluation parameter set to generate an optimal countermeasure plan, and the optimal countermeasure plan is converted into an executable instruction set and synchronized to the countermeasure device in the target area, including:

[0012] Extract trajectory prediction deviation and electromagnetic suppression priority from the three-dimensional threat situation model, extract geo-fence penetration risk value and environmental interference attenuation coefficient from the dynamic decision factor set, and extract response delay threshold and spectrum interference compatibility from the emergency plan knowledge graph.

[0013] The trajectory prediction deviation and response delay threshold are combined through the time series compression algorithm to generate dynamic response margin parameters;

[0014] The peak value of the cross-correlation function between the electromagnetic suppression priority and the spectrum interference compatibility is calculated through the frequency domain overlap analysis algorithm. When the peak value exceeds the preset interference threshold, the spectrum avoidance coefficient is generated and the countermeasures of the corresponding frequency band are locked;

[0015] An attenuation compensation model is constructed using the geo-fence penetration risk value and the environmental interference attenuation coefficient to dynamically correct the transmit power weight factor of the countermeasure device.

[0016] The dynamic response margin parameters, spectrum avoidance coefficient and transmission power weight factor are input into the node activation function of the plan logic tree. The plan paths in the plan logic tree are pruned with confidence weighted by the Monte Carlo tree search algorithm. The plan paths with confidence higher than the preset dynamic threshold are used to generate the optimal countermeasure plan. The optimal countermeasure plan is converted into an executable instruction set and synchronized to the countermeasure device in the target area.

[0017] Optionally, the dynamic response margin parameter, the spectrum avoidance coefficient, and the transmit power weight factor are input into the node activation function of the plan logic tree, and the plan paths in the plan logic tree are pruned with confidence weights using a Monte Carlo tree search algorithm. The plan paths with confidence levels higher than a preset dynamic threshold are used to generate an optimal countermeasure plan, and the optimal countermeasure plan is converted into an executable instruction set and synchronized to the countermeasure device in the target area, including:

[0018] A time-sensitive search window is generated based on the dynamic response margin parameter, wherein the width of the time-sensitive search window shrinks exponentially as the trajectory prediction deviation increases. The time-sensitive search window is then spatiotemporally aligned with the node execution timestamps of the plan logic tree to screen candidate paths within the coverage of the time-sensitive search window.

[0019] Perform frequency band compliance checks on candidate paths based on the spectrum avoidance coefficient. If the transmission frequency band of the countermeasure in the candidate path overlaps with the currently locked conflicting frequency band and the dynamic spectrum migration mechanism is not configured, the candidate path is marked as a disabled candidate path and removed from the contingency plan logic tree.

[0020] The real-time energy consumption estimate of each candidate path is calculated using the transmit power weight factor. A dynamic fuse threshold is generated based on the remaining power of the countermeasure device and the environmental interference attenuation coefficient. When the cumulative energy consumption of a candidate path exceeds the fuse threshold, a backtracking mechanism is triggered and a low-power node is reselected.

[0021] By integrating electromagnetic suppression priority data from a three-dimensional threat situation model, the confidence of candidate paths containing coordinated countermeasure nodes is weighted and improved. The coordinated countermeasure nodes must simultaneously meet the timing linkage conditions of radio frequency interference and navigation deception methods.

[0022] The pruning threshold is dynamically adjusted according to the level of the geo-fence penetration risk value, and the plan paths with confidence levels higher than the preset dynamic threshold are subjected to conflict detection and redundancy elimination to generate the optimal countermeasure plan.

[0023] Optionally, the dynamic monitoring data stream obtained by the target UAV is parsed to identify the flight trajectory characteristics and electromagnetic signal spectrum of the target UAV, and a three-dimensional space threat situation model is constructed in combination with multi-source heterogeneous sensor fusion technology. The threat level of the target UAV is evaluated using the three-dimensional space threat situation model, and a threat level assessment result is generated. The threat level assessment result includes:

[0024] Analyze the dynamic monitoring data stream of the target UAV collected synchronously by multi-source sensors, and perform time alignment processing on the position coordinates, velocity vectors and electromagnetic spectrum data in the dynamic monitoring data stream to generate the original data set;

[0025] Based on the position coordinates and velocity vectors in the original data set, the target drone's heading angle change rate and acceleration fluctuation characteristics are calculated. The trajectory prediction deviation is generated by combining the preset trajectory prediction model, and the flight trajectory feature vector is constructed.

[0026] Perform frequency domain feature extraction on the electromagnetic spectrum data in the original data set to obtain the main frequency band energy distribution and signal frequency hopping pattern of the target UAV, and generate the electromagnetic signal spectrum feature vector;

[0027] The heading angle change rate, acceleration fluctuation characteristics and trajectory prediction deviation in the flight trajectory feature vector are fused with the main frequency band energy distribution and signal frequency hopping pattern in the electromagnetic signal spectrum feature vector to generate the threat confidence parameter of the target UAV.

[0028] A three-dimensional threat situation model is constructed based on the threat confidence parameters. The real-time position coordinates of the target UAV are mapped to the three-dimensional grid space under the constraints of the geographic fence. The threat level assessment result is generated by combining the trajectory prediction deviation and signal frequency hopping law.

[0029] Optionally, based on the threat level assessment results, the target drone's behavior pattern and attack intent parameters are decomposed, and a dynamic decision factor set is generated by combining geo-fence constraints and environmental interference factors, including:

[0030] Based on the trajectory prediction deviation and signal frequency hopping pattern in the threat level assessment results, the target UAV's behavior pattern category is identified, where the behavior pattern category includes: reconnaissance mode, interference mode, and attack mode;

[0031] Extract the target drone's attack intention parameters based on the behavior pattern category. Among them, the reconnaissance mode corresponds to the target area scanning frequency, the interference mode corresponds to the electromagnetic suppression intensity of the countermeasure device, and the attack mode corresponds to the target drone's dive angle and speed change rate.

[0032] Spatial matching of attack intent parameters with geofence constraints is performed, the minimum distance between the target drone’s current location and the geofence boundary is calculated, and the spatial threat weight is generated in combination with environmental interference factors.

[0033] The attack intention parameters are dynamically modified based on the spatial threat weight to obtain the modified attack intention parameters. In the reconnaissance mode, the target area scanning frequency increases as the distance decreases, the electromagnetic suppression intensity in the interference mode decreases as the environmental interference factor increases, and the dive angle and speed change rate in the attack mode increase as the distance decreases.

[0034] The modified attack intention parameters are combined with the behavior pattern categories to generate a dynamic decision factor set including spatial threat weights, modified attack intention parameters and behavior pattern categories.

[0035] Optionally, it is characterized in that a plurality of candidate countermeasure plans are generated by matching the basic countermeasure plans preset in the plan knowledge graph through a dynamic decision factor set, and a plan logic tree is constructed according to the plurality of candidate countermeasure plans, including:

[0036] Based on the behavioral pattern categories in the dynamic decision factor set, matching basic countermeasure plans are extracted from the plan knowledge graph. Among them, the reconnaissance mode corresponds to the spectrum detection and interference plan, the interference mode corresponds to the frequency band suppression and navigation deception plan, and the attack mode corresponds to the physical interception and directional strike plan;

[0037] According to the modified attack intention parameters in the dynamic decision factor set, the basic countermeasure plan extracted from the plan knowledge graph is adjusted to obtain the basic countermeasure plan after parameter adaptation. Among them, the scanning frequency of the spectrum detection plan is synchronized with the scanning frequency of the target area, the transmission power of the frequency band suppression plan is matched with the electromagnetic suppression intensity, and the triggering timing of the physical interception plan is associated with the dive angle and speed change rate.

[0038] Based on the spatial threat weights in the dynamic decision factor set, the basic countermeasure plans after parameter adaptation are prioritized to generate a candidate countermeasure plan set containing execution timing, intensity parameters, and trigger conditions;

[0039] The plan nodes in the candidate countermeasure plan set are constructed into a plan logic tree according to the execution time sequence and spatial position relationship, where the connecting edges between the plan nodes represent the switching conditions of the countermeasure measures, where the switching conditions include: electromagnetic suppression intensity threshold, navigation deception success rate and physical interception feasibility.

[0040] Optionally, based on the three-dimensional threat situation model, the dynamic decision factor set, and the countermeasure effectiveness indicators preset in the plan knowledge graph, a multi-dimensional evaluation parameter set is generated, including:

[0041] The real-time position coordinates, trajectory prediction deviation, and electromagnetic suppression priority of the target UAV are extracted from the three-dimensional spatial threat situation model. Combined with the countermeasure effectiveness indicators in the plan knowledge graph, spatial threat assessment parameters are generated. The spatial threat assessment parameters include: target distance attenuation coefficient, trajectory coverage completeness, and electromagnetic suppression effectiveness value.

[0042] The response time parameter and intensity matching parameter of the countermeasure are calculated from the modified attack intention parameter based on the dynamic decision factor set. The response time parameter is calculated based on the speed change rate of the target UAV and the deployment position of the countermeasure device, and the intensity matching parameter is calculated based on the electromagnetic suppression intensity and the transmission power range of the countermeasure.

[0043] Based on the countermeasure effectiveness indicators in the plan knowledge graph, the spectrum compatibility parameters and energy efficiency parameters of the countermeasures are extracted. The spectrum compatibility parameter is calculated by the overlapping area between the countermeasure's transmission frequency band and the protection frequency band, and the energy efficiency parameter is calculated by the countermeasure's energy consumption per unit time and the remaining power of the countermeasure device.

[0044] The spatial threat assessment parameters, response time parameters, intensity matching parameters, spectrum compatibility parameters and energy efficiency parameters are normalized to generate a multidimensional assessment parameter set containing weight coefficients, where the weight coefficients are dynamically adjusted through the spatial threat weights in the dynamic decision factor set.

[0045] In a second aspect, an embodiment of the present application provides a drone countermeasure plan generation system based on artificial intelligence technology, comprising:

[0046] The parsing module is used to parse the dynamic monitoring data stream obtained from the target UAV, identify the flight trajectory characteristics and electromagnetic signal spectrum of the target UAV, and build a three-dimensional spatial threat situation model based on multi-source heterogeneous sensor fusion technology. The three-dimensional spatial threat situation model is used to assess the threat level of the target UAV and generate a threat level assessment result.

[0047] A decomposition module is used to decompose the target drone's behavior patterns and attack intention parameters based on the threat level assessment results, and generate a dynamic decision factor set based on geo-fence constraints and environmental interference factors;

[0048] A construction module is used to match the basic countermeasure plans preset in the plan knowledge graph through a dynamic decision factor set to generate multiple candidate countermeasure plans, and to construct a plan logic tree based on the multiple candidate countermeasure plans;

[0049] A generation module is used to generate a multi-dimensional evaluation parameter set based on the three-dimensional threat situation model, the dynamic decision factor set, and the countermeasure effectiveness indicators preset in the plan knowledge graph;

[0050] The synchronization module is used to collaboratively optimize the plan logic tree based on a multi-dimensional evaluation parameter set to generate an optimal countermeasure plan, convert the optimal countermeasure plan into an executable instruction set, and synchronize it to the countermeasure device in the target area.

[0051] In a third aspect, an embodiment of the present application provides a computing device comprising a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute a method for generating a drone countermeasure plan based on artificial intelligence technology as described in any one of the first aspects.

[0052] In a fourth aspect, an embodiment of the present application provides a computer storage medium having computer program instructions stored thereon, which, when executed by a processor, implements a method for generating a drone countermeasure plan based on artificial intelligence technology as described in any one of the first aspects.

[0053] In an embodiment of the present application, a dynamic monitoring data stream obtained by a target UAV is parsed to identify the flight trajectory characteristics and electromagnetic signal spectrum of the target UAV, and a three-dimensional spatial threat situation model is constructed in combination with multi-source heterogeneous sensor fusion technology. The threat level of the target UAV is evaluated using the three-dimensional spatial threat situation model to generate a threat level assessment result; based on the threat level assessment result, the behavior pattern and attack intention parameters of the target UAV are decomposed, and a dynamic decision factor set is generated in combination with geographic fence constraints and environmental interference factors; the dynamic decision factor set is matched with the basic countermeasure plan preset in the plan knowledge graph to generate multiple candidate countermeasure plans, and a plan logic tree is constructed based on the multiple candidate countermeasure plans; based on the three-dimensional spatial threat situation model, the dynamic decision factor set, and the countermeasure effectiveness indicators preset in the plan knowledge graph, a multidimensional evaluation parameter set is generated; based on the multidimensional evaluation parameter set, the plan logic tree is collaboratively optimized to generate an optimal countermeasure plan, and the optimal countermeasure plan is converted into an executable instruction set and synchronized to the countermeasure device in the target area.

[0054] The technical solution of this application has the following beneficial effects:

[0055] The above method effectively improves the accuracy of threat assessment, dynamically adjusts the countermeasure strategy to adapt to complex and changing environmental conditions, and generates the optimal countermeasure plan through optimization processing, thereby greatly enhancing the security protection capability of the target area and reducing the risks brought by potential threats. In addition, this method also improves the automation level and response flexibility of the countermeasure system, reduces the need for human intervention, and makes the entire countermeasure process more intelligent and efficient.

[0056] Furthermore, by extracting key parameters such as trajectory prediction deviation, electromagnetic suppression priority, and geo-fence penetration risk value from the three-dimensional threat situation model, dynamic decision factor set, and plan knowledge graph, and generating dynamic response margin parameters through a time series compression algorithm, the peak value of the cross-correlation function is calculated using a frequency domain overlap analysis algorithm to determine whether to generate a spectrum avoidance coefficient to lock the corresponding frequency band of the countermeasure. At the same time, an attenuation compensation model is constructed to dynamically correct the transmission power weight factor of the countermeasure device. Finally, these parameters are input into the node activation function of the plan logic tree, and confidence-weighted pruning is performed through a Monte Carlo tree search algorithm to generate the optimal countermeasure plan.

[0057] It can not only dynamically adjust the countermeasure strategy according to real-time changing environmental conditions, but also effectively avoid unnecessary spectrum interference and optimize the transmission power of the countermeasure device, thereby ensuring the rapid and accurate generation of the optimal countermeasure plan in complex environments, greatly improving the system's response flexibility and overall effectiveness, and ensuring the safety of the target area.

[0058] These and other aspects of the present application will become more readily apparent from the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0060] Figure 1 A flowchart of a method for generating a drone countermeasure plan based on artificial intelligence technology provided in an embodiment of the present application;

[0061] Figure 2 A schematic diagram of the structure of a UAV countermeasure plan generation system based on artificial intelligence technology provided in an embodiment of the present application;

[0062] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0063] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0064] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to being different types.

[0065] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0066] Figure 1 The present invention provides a flowchart of a method for generating a UAV countermeasure plan based on artificial intelligence technology, such as Figure 1As shown, the method includes:

[0067] Step 101: parse the dynamic monitoring data stream obtained from the target UAV, identify the flight trajectory characteristics and electromagnetic signal spectrum of the target UAV, and build a three-dimensional spatial threat situation model using multi-source heterogeneous sensor fusion technology. Use the three-dimensional spatial threat situation model to assess the threat level of the target UAV and generate a threat level assessment result, including:

[0068] In this step, dynamic monitoring of data streams refers to the use of various sensors and technical means to collect real-time data generated by drones during flight, including but not limited to flight trajectory and electromagnetic signal spectrum. Flight trajectory characteristics refer to the movement path of the drone in space and the characteristics of the drone. The electromagnetic signal spectrum is the frequency range of radio signals used for communication between the drone and the drone controller. Multi-source heterogeneous sensor fusion technology is the integrated processing of data from different types of sensors (such as radar, cameras, radio receivers, etc.).

[0069] In this step, a series of multi-source heterogeneous sensor networks, including but not limited to radar systems, optical cameras, and radio spectrum analyzers, are first deployed to capture the drone's flight trajectory and electromagnetic signal information. Specialized data processing algorithms are then used to clean and pre-process this raw data, extracting the drone's flight trajectory characteristics and electromagnetic signal spectrum. Advanced sensor fusion technology is then used to integrate this information to construct an accurate three-dimensional threat situation model. Finally, based on this three-dimensional threat situation model, a threat assessment algorithm is used to determine the drone's threat level and generate a threat level assessment result.

[0070] For example, in a real-world example of airport perimeter security monitoring, a multi-source heterogeneous sensor network, including high-precision radar, infrared cameras, and radio spectrum monitoring equipment, was deployed to prevent illegal drone intrusions from disrupting normal flight operations. Upon detecting a drone entering the monitoring area, the system immediately begins collecting its flight trajectory and electromagnetic signal data, processing it through a powerful backend data analysis platform. Machine learning algorithms are used to identify drone behavior patterns, and a three-dimensional spatial threat situation model is constructed using a geographic information system (GIS) to accurately assess the drone's threat level. Based on this assessment, the airport security department can quickly activate the appropriate early warning mechanism and implement effective countermeasures to ensure airspace safety and stability, effectively enhancing its monitoring and response capabilities to drone threats.

[0071] Step 102: Decompose the target drone's behavior pattern and attack intent parameters based on the threat level assessment results, and generate a dynamic decision factor set based on geo-fence constraints and environmental interference factors, including:

[0072] In this step, geofence constraints refer to pre-set electronic boundaries that restrict drones from entering specific areas. Environmental interference factors include the impact of external factors such as weather conditions and radio interference on drone operations. The dynamic decision factor set is key data extracted from information such as the target drone's behavior patterns, attack intent parameters, geofence constraints, and environmental interference factors. This data is used to assess the specific threat situation of the drone and provide a basis for developing corresponding countermeasure plans.

[0073] In this step, based on the threat level assessment results provided by the three-dimensional spatial threat situation model, a behavioral analysis algorithm is used to analyze the target drone's behavioral patterns, such as whether it is in a hovering state or has an abnormal flight path. The target drone's potential attack intent is then inferred based on historical data. Geographical fence constraints (such as airport no-fly zones) and real-time monitored environmental interference factors (such as strong winds or electromagnetic interference) are then taken into consideration. By integrating this information, a dynamic decision factor set is established. This set not only includes the drone's behavioral patterns and attack intent, but also integrates the specific impact of geofencing and environmental interference, providing a detailed reference for the subsequent matching of countermeasures in the plan knowledge graph.

[0074] For example, when an illegal drone approaching an airport and its flight trajectory characteristics and electromagnetic signal spectrum have been identified, the threat level of the illegal drone is assessed through a three-dimensional spatial threat situation model. Next, based on the threat level assessment results, the drone's behavior pattern is deeply analyzed, and it is found that the drone is trying to cross the no-fly zone near the airport, and there are signs that the drone may be carrying some kind of jamming equipment. At the same time, strong radio interference and unfavorable weather conditions are detected. Based on this information, a dynamic decision factor set is generated, including a high geo-fence penetration risk value and a large environmental interference attenuation coefficient. The decision factor set is immediately input into the plan logic tree to quickly match the countermeasure plan that best suits the current situation, ensuring that this potential threat can be responded to in a timely and effective manner, protecting the safety of the airport airspace, and improving the airport's response efficiency to illegal drone activities.

[0075] Step 103, matching the basic countermeasure plans preset in the plan knowledge graph with the dynamic decision factor set, generates multiple candidate countermeasure plans, and constructs a plan logic tree based on the multiple candidate countermeasure plans, including:

[0076] In this step, the plan knowledge graph is a knowledge base containing a variety of basic plans for preset countermeasures, which stores response strategies designed according to different types of threat scenarios. Each basic plan in the map describes in detail the specific countermeasures and their applicable conditions, aiming to match a specific set of dynamic decision factors to generate an effective countermeasure plan. The basic countermeasure plan refers to a pre-set standard response plan for different types of drone threats. Each basic plan includes detailed execution steps and required resources, such as using radio jammers to suppress signals, deploying physical interception devices, or activating electromagnetic pulse devices, to ensure that potential threats can be effectively responded to. Candidate countermeasure plans are multiple potentially applicable countermeasure plans generated by matching the dynamic decision factor set with the plan knowledge graph. These plans are preliminarily screened based on their adaptability to the current threat scenario to form a selection set for further evaluation. The plan logic tree is a logical structure used to display the relationship and priority between candidate countermeasure plans. By constructing the plan logic tree, the multi-dimensional evaluation parameters under different plan paths can be systematically analyzed to determine the optimal countermeasure plan. This tree structure helps optimize the decision-making process and ensure that the most effective countermeasures are quickly found and implemented in complex environments.

[0077] In this step, the dynamic decision factor set is first input into the contingency plan knowledge graph system. The contingency plan knowledge graph system contains a series of preset basic countermeasure plans for different threat scenarios. Based on the specific parameters in the dynamic decision factor set (such as the geo-fence penetration risk value and the environmental interference attenuation coefficient), the most relevant basic plans are automatically screened as candidate countermeasure plans. Then, according to the logical relationship and priority between the candidate plans, a contingency plan logic tree is constructed. The contingency plan logic tree not only shows the hierarchical relationship between the candidate plans, but also provides a framework for the subsequent multi-dimensional evaluation of each plan path. Finally, by analyzing the contingency plan logic tree, the most suitable countermeasure plan for the current situation can be more accurately determined.

[0078] For example, in the previous step, an illegal drone that attempted to cross the airport's no-fly zone and may have carried jamming equipment was identified, and a dynamic decision factor set containing a high geo-fence penetration risk value and a large-environment interference attenuation coefficient was generated. Next, this set is applied to the plan knowledge graph to match multiple basic countermeasure plans suitable for the current scenario, such as using a radio jammer to suppress its signal, deploying a capture net to intercept, or activating an electromagnetic pulse device to disable it. Based on these candidate plans, a plan logic tree is constructed, which lists in detail the execution order and conditions of each plan. For example, when the radio interference effect is not good, the physical capture plan is given priority, ensuring that even in a complex and changeable actual environment, the most effective countermeasures can be quickly found and implemented, maximizing the safe operation of the airport and improving the ability to respond quickly to drone threats.

[0079] Step 104: Generate a multi-dimensional evaluation parameter set based on the three-dimensional threat situation model, the dynamic decision factor set, and the countermeasure effectiveness indicators preset in the plan knowledge graph, including:

[0080] In this step, the countermeasure effectiveness index is a pre-set evaluation standard in the plan knowledge graph, covering the effectiveness evaluation data of various countermeasure measures, such as response speed, success rate, resource consumption, etc., used to evaluate the actual execution effect of different countermeasure plans. The multi-dimensional evaluation parameter set is a data set generated based on the three-dimensional spatial threat situation model and the dynamic decision factor set, including but not limited to trajectory prediction deviation, electromagnetic suppression priority, geo-fence penetration risk value, etc., designed to comprehensively evaluate the effectiveness and applicability of each candidate countermeasure plan.

[0081] In this step, the latest data provided by the three-dimensional threat situation model is first combined with the key evaluation data in the dynamic decision factor set. Then, the preset countermeasure effectiveness indicators are extracted from the plan knowledge graph to ensure that all relevant evaluation criteria are covered. Then, through comprehensive analysis of this information, a multidimensional evaluation parameter set is generated. This multidimensional evaluation parameter set not only includes various parameters that affect the effectiveness of countermeasures, but also takes into account various limiting factors in actual operations. Finally, the multidimensional evaluation parameter set is used to comprehensively evaluate each candidate countermeasure plan to determine the optimal countermeasure plan.

[0082] For example, in the previous step, multiple candidate countermeasure plans for illegal drones have been matched based on the dynamic decision factor set, and a plan logic tree has been constructed. Next, the real-time flight trajectory data and electromagnetic signal information in the three-dimensional space threat situation model are integrated, combined with the geographic fence penetration risk value and environmental interference attenuation coefficient in the dynamic decision factor set, and the preset countermeasure effectiveness indicators in the plan knowledge graph, such as response speed and success rate, to generate a detailed multi-dimensional evaluation parameter set; for example, for the plan of using radio jammers to suppress drone signals, the system evaluates the response time, success rate and possible spectrum interference of the plan. At the same time, for the physical interception plan, the deployment time and the probability of successful capture are analyzed. Based on these detailed evaluation results, the most effective countermeasure plan can be selected more accurately to ensure a quick and accurate response to the threat of illegal drones and protect the safety of sensitive areas.

[0083] Step 105, based on the multi-dimensional evaluation parameter set, performs collaborative optimization processing on the plan logic tree to generate an optimal countermeasure plan, converts the optimal countermeasure plan into an executable instruction set, and synchronizes it to the countermeasure device in the target area, including:

[0084] In this step, collaborative optimization processing refers to the process of analyzing and optimizing the plan logic tree using algorithms (such as the Monte Carlo tree search algorithm), selecting the plan path with the highest confidence as the optimal countermeasure plan through techniques such as weighted pruning; the executable instruction set refers to the conversion of the optimal countermeasure plan into specific action guidelines or operating commands, including but not limited to activating specific equipment and adjusting setting parameters, so that they can be directly applied to the countermeasure device; the countermeasure device refers to multiple countermeasure devices deployed in the target area, which can receive and execute the instruction set from the system and implement specific countermeasures against illegal drones;

[0085] In this step, each plan path in the plan logic tree is first evaluated based on a multi-dimensional evaluation parameter set, and the plan paths are confidence-weighted pruned using a collaborative optimization algorithm (such as the Monte Carlo tree search algorithm). Then, based on the optimization results, one or more plan paths with the highest confidence are selected as the optimal countermeasure plan. Next, this optimal countermeasure plan is converted into a detailed executable instruction set to ensure that each step is clear and specific. Finally, these instruction sets are synchronized to the distributed countermeasure devices in the target area via a communication network, enabling them to immediately execute the corresponding countermeasures.

[0086] For example, in the previous step, a multi-dimensional evaluation parameter set has been generated and the plan logic tree has been preliminarily screened, identifying several possible countermeasure plans. Now, these countermeasure plan paths are further analyzed using collaborative optimization processing technology, and finally a plan with the highest confidence is determined as the optimal countermeasure plan. For example, when facing illegal drones attempting to cross the no-fly zone, a comprehensive solution of using radio interference combined with a physical interception network is chosen as the optimal countermeasure plan. This plan is then converted into a series of executable instruction sets, including activating radio jammers in specific frequency bands, adjusting the interference power, and the specific location and time points for deploying the physical interception network. These executable instruction sets are quickly synchronized to the countermeasure devices deployed around the airport to ensure that they can respond immediately and effectively to the threat of illegal drones, thereby ensuring the safety and stability of the airspace.

[0087] To address the complexity and uncertainty issues encountered in the drone threat assessment process and significantly improve the accuracy and response efficiency of countermeasures, in some embodiments, according to step 105, a collaborative optimization process is performed on the plan logic tree based on a multi-dimensional evaluation parameter set to generate an optimal countermeasure plan. The optimal countermeasure plan is converted into an executable instruction set and synchronized to the countermeasure device in the target area, specifically including:

[0088] The trajectory prediction deviation and electromagnetic suppression priority are extracted from the three-dimensional threat situation model, the geofence penetration risk value and environmental interference attenuation coefficient are extracted from the dynamic decision factor set, and the response delay threshold and spectrum interference compatibility are extracted from the plan knowledge graph. The trajectory prediction deviation and response delay threshold are converted into dynamic response margin parameters through a time series compression algorithm. The peak value of the cross-correlation function between the electromagnetic suppression priority and the spectrum interference compatibility is calculated through a frequency domain overlap analysis algorithm. When the peak value exceeds the preset interference threshold, a spectrum avoidance coefficient is generated and the countermeasure measures in the corresponding frequency band are locked. The geofence penetration risk value and the environmental interference attenuation coefficient are used to construct an attenuation compensation model to dynamically correct the transmission power weight factor of the countermeasure device. The dynamic response margin parameters, spectrum avoidance coefficient, and transmission power weight factor are input into the node activation function of the plan logic tree. The plan paths in the plan logic tree are pruned based on confidence using the Monte Carlo tree search algorithm. The plan paths with confidence values higher than the preset dynamic threshold are used to generate the optimal countermeasure plan. The optimal countermeasure plan is converted into an executable instruction set and synchronized to the countermeasure device in the target area.

[0089] In this embodiment, the dynamic response margin parameter is a comprehensive indicator calculated using a time series compression algorithm by combining the trajectory prediction deviation and the response delay threshold. It is used to measure the system's real-time response capability when facing drone threats. The spectrum avoidance coefficient is calculated by calculating the peak value of the cross-correlation function between the electromagnetic suppression priority and the spectrum interference compatibility. When the peak value exceeds the preset interference threshold, a parameter is generated to guide the selection of appropriate frequency bands to avoid unnecessary spectrum interference. The attenuation compensation model is constructed based on the geo-fence penetration risk value and the environmental interference attenuation coefficient. It is used to dynamically adjust the transmit power weighting factor of the countermeasure device to ensure its effectiveness under various environmental conditions.

[0090] In an embodiment of the present application, first, necessary parameters are extracted from the three-dimensional space threat situation model, the dynamic decision factor set and the plan knowledge graph, such as trajectory prediction deviation, electromagnetic suppression priority, etc. Next, the dynamic response margin parameters are generated using the time series compression algorithm, and the spectrum avoidance coefficient is calculated by the frequency domain overlap analysis algorithm to lock the frequency band that may be interfered with. Then, an attenuation compensation model is constructed based on the geographic fence penetration risk value and the environmental interference attenuation coefficient, and the transmission power weight factor of the countermeasure device is dynamically corrected. Finally, the dynamic response margin parameters, spectrum avoidance coefficient and transmission power weight factor generated above are input into the node activation function of the plan logic tree, and the Monte Carlo tree search algorithm is used to perform confidence-weighted pruning on the plan path, and the plan with the highest confidence is selected as the optimal countermeasure plan, and it is converted into a specific executable instruction set and synchronized to the countermeasure device;

[0091] For example, in a security monitoring scenario around an airport, the system detected an illegal drone attempting to cross a no-fly zone and engaging in potential radio interference. It first extracted relevant parameters such as trajectory prediction deviation and electromagnetic suppression priority, determined the optimal response strategy after analysis, and used a time series compression algorithm to calculate the dynamic response margin parameters to ensure a rapid response. At the same time, it used a frequency domain overlap analysis algorithm to identify the frequency bands that needed to be avoided and adjusted the corresponding countermeasures. In addition, considering the impact of current meteorological conditions (such as strong winds) on drone operations, the system also constructed an attenuation compensation model to optimize the transmission power of the countermeasures. Finally, it optimized the plan logic tree through the Monte Carlo tree search algorithm, selected the most effective countermeasure plan, and quickly synchronized the instructions to the countermeasures around the airport, successfully preventing the intrusion of illegal drones and ensuring airspace safety.

[0092] To further improve the adaptability and reliability of countermeasures, according to the previous embodiment, the dynamic response margin parameter, spectrum avoidance coefficient, and transmit power weight factor are input into the node activation function of the plan logic tree. The plan paths in the plan logic tree are pruned with confidence weights using the Monte Carlo tree search algorithm. The plan paths with confidence levels higher than a preset dynamic threshold are used to generate the optimal countermeasure plan. The optimal countermeasure plan is converted into an executable instruction set and synchronized to the countermeasure device in the target area, specifically including:

[0093] A time-sensitive search window is generated based on the dynamic response margin parameter, wherein the width of the time-sensitive search window shrinks exponentially with the increase of the trajectory prediction deviation, and the time-sensitive search window is time-space aligned with the node execution timestamp of the plan logic tree to screen out candidate paths within the coverage of the time-sensitive search window; the frequency band compliance of the candidate path is checked according to the spectrum avoidance coefficient, and when the transmission frequency band of the countermeasure in the candidate path overlaps with the currently locked conflict frequency band and the dynamic spectrum migration mechanism is not configured, the candidate path is marked as a disabled candidate path and removed from the plan logic tree; the transmission power weight factor is used to calculate the compliance of each candidate path. Real-time energy consumption estimation of the path, combined with the remaining power of the countermeasure device and the environmental interference attenuation coefficient, generates a dynamic fuse threshold. When the cumulative energy consumption of the candidate path exceeds the fuse threshold, the backtracking mechanism is triggered and a low-power node is reselected. The electromagnetic suppression priority data of the three-dimensional space threat situation model is integrated to perform a confidence weighted increase on the candidate paths containing the coordinated countermeasure nodes. The coordinated countermeasure nodes must simultaneously meet the timing linkage conditions of radio frequency interference and navigation deception methods. The pruning threshold is dynamically adjusted according to the level of the geo-fence penetration risk value. The plan paths with confidence levels higher than the preset dynamic threshold are subjected to conflict detection and redundancy elimination to generate the optimal countermeasure plan.

[0094] In this embodiment, the time-sensitive search window refers to a time range generated based on the dynamic response margin parameter, the width of which shrinks exponentially with the increase of the trajectory prediction deviation, and is used to perform spatiotemporal alignment with the node execution timestamps in the plan logic tree, thereby screening out feasible candidate paths within a specific time. Frequency band compliance verification is a process that uses the spectrum avoidance coefficient to check whether the transmission frequency band of the countermeasures in the candidate path overlaps with the currently locked conflicting frequency band to ensure that unnecessary spectrum interference is not generated. The dynamic fuse threshold is a threshold calculated based on the transmission power weight factor, the remaining power and the environmental interference attenuation coefficient, which is used to trigger the backtracking mechanism to avoid equipment failure due to excessive energy consumption. A collaborative countermeasure node refers to a node that meets the timing linkage conditions of radio frequency interference and navigation deception means at the same time, which can effectively enhance the countermeasure effect;

[0095] In an embodiment of the present application, a time-sensitive search window is first generated based on a dynamic response margin parameter, and it is aligned in time and space with the node execution timestamp of the plan logic tree to screen out candidate paths within the coverage of the time-sensitive search window. Then, the frequency band compliance of these candidate paths is checked according to the spectrum avoidance coefficient, and paths that do not meet the conditions are removed. Then, the real-time energy consumption estimation of each candidate path is calculated using the transmission power weight factor, and a dynamic fuse threshold is generated in combination with the remaining power of the countermeasure device and the environmental interference attenuation coefficient. When the cumulative energy consumption of the candidate path exceeds this threshold, the backtracking mechanism is triggered to reselect the low-power node. In addition, the electromagnetic suppression priority data in the three-dimensional space threat situation model is integrated to perform confidence weighting on the path containing the collaborative countermeasure node to ensure efficient countermeasure. Finally, the pruning threshold is dynamically adjusted according to the geographic fence penetration risk value, and conflict detection and redundancy elimination are performed on the candidate paths. The path with a confidence higher than the preset dynamic threshold is screened out as the optimal countermeasure plan.

[0096] For example, in a security monitoring scenario around an airport, an illegal drone was detected attempting to cross a no-fly zone and engaging in potential radio interference. First, a time-sensitive search window was generated based on dynamic response margin parameters to ensure that only planned paths that could be executed within a reasonable time were considered. The spectrum avoidance coefficient was then used to verify the frequency band compliance of candidate paths, eliminating those that could cause spectrum conflicts. Considering that current environmental conditions (such as strong winds) may affect device energy consumption, the system used transmit power weighting factors to calculate real-time energy consumption estimates and set dynamic fuse thresholds to prevent device failure due to excessive energy consumption. For paths that include coordinated countermeasure nodes (such as those using a combination of RF interference and navigation deception), the system performed a confidence-weighted boost to ensure that these high-efficiency paths were prioritized. Finally, the pruning threshold was dynamically adjusted based on the geofence penetration risk value, selecting the optimal countermeasure plan and quickly synchronizing the instructions to distributed countermeasure devices around the airport, successfully preventing the intrusion of illegal drones, ensuring airspace security, and improving the effectiveness and adaptability of countermeasures.

[0097] To further improve the monitoring accuracy and response speed of illegal or abnormal drone activities, as another embodiment, according to step 101, the dynamic monitoring data stream obtained by the target drone is parsed to identify the flight trajectory characteristics and electromagnetic signal spectrum of the target drone. A three-dimensional spatial threat situation model is constructed by combining multi-source heterogeneous sensor fusion technology. The threat level of the target drone is assessed using this model, and a threat level assessment result is generated, which specifically includes:

[0098] The dynamic monitoring data stream of the target UAV is collected synchronously by multi-source sensors, and the position coordinates, velocity vectors and electromagnetic spectrum data in the dynamic monitoring data stream are time-aligned to generate an original data set. The heading angle change rate and acceleration fluctuation characteristics of the target UAV are calculated based on the position coordinates and velocity vectors in the original data set, and the trajectory prediction deviation is generated by combining with the preset trajectory prediction model, and a flight trajectory feature vector is constructed. The electromagnetic spectrum data in the original data set is subjected to frequency domain feature extraction to obtain the main frequency band energy distribution and signal frequency hopping pattern of the target UAV, and an electromagnetic signal spectrum feature vector is generated. The heading angle change rate, acceleration fluctuation characteristics and trajectory prediction deviation in the flight trajectory feature vector are combined with the main frequency band energy distribution and signal frequency hopping pattern in the electromagnetic signal spectrum feature vector for multi-source data fusion to generate the threat confidence parameter of the target UAV. A three-dimensional threat situation model is constructed based on the threat confidence parameter, and the real-time position coordinates of the target UAV are mapped to the three-dimensional grid space under the constraints of the geographic fence. The threat level assessment result is generated by combining the trajectory prediction deviation and signal frequency hopping pattern.

[0099] In this embodiment, the raw data set includes dynamic monitoring data of the target UAV, such as position coordinates, velocity vectors, and electromagnetic spectrum data, collected synchronously from multiple source sensors. After time alignment, the basic data set is formed for subsequent analysis and modeling. The flight trajectory feature vector consists of the heading angle change rate, acceleration fluctuation characteristics, and trajectory prediction deviation, which is used to describe the flight behavior characteristics of the target UAV. The electromagnetic signal spectrum feature vector includes the main frequency band energy distribution and signal frequency hopping pattern, reflecting the main characteristics of the UAV communication signal and its dynamic changes. The threat confidence parameter is an evaluation indicator derived by comprehensively analyzing the flight trajectory feature vector and the electromagnetic signal spectrum feature vector, which is used to quantify the potential threat level of the target UAV.

[0100] In an embodiment of the present application, first, dynamic monitoring data streams of the target UAV are synchronously collected by multi-source sensors (such as radar, optical cameras, radio receivers, etc.) deployed in the monitoring area, and time alignment processing is performed on these data to ensure that data from different sources can be analyzed on the same time basis. Then, the heading angle change rate and acceleration fluctuation characteristics of the target UAV are calculated based on the position coordinates and velocity vector in the original data set, and the trajectory prediction deviation is generated by combining with a preset trajectory prediction model to construct a flight trajectory feature vector. At the same time, frequency domain feature extraction is performed on the electromagnetic spectrum data in the original data set to obtain the main frequency band energy distribution and signal frequency hopping pattern of the target UAV and generate an electromagnetic signal spectrum feature vector. Then, multi-source data fusion is performed on the flight trajectory feature vector and the electromagnetic signal spectrum feature vector to generate a threat confidence parameter for the target UAV. Finally, a three-dimensional threat situation model is constructed based on this parameter, and the real-time position coordinates of the target UAV are mapped to a three-dimensional grid space under the constraints of the geo-fence. The threat level assessment result is generated by combining the trajectory prediction deviation and the signal frequency hopping pattern.

[0101] For example, when a drone suspected of illegal intrusion is detected, the drone's position coordinates, velocity vector and electromagnetic spectrum data are synchronously collected through various sensors deployed around the airport, and time alignment processing is performed. The system then calculates the drone's heading angle change rate and acceleration fluctuation characteristics, and finds that its flight path has abnormal fluctuations, which may be intended to evade monitoring. At the same time, through the analysis of electromagnetic spectrum data, it is identified that the drone uses a specific frequency band for communication and has a certain frequency hopping pattern. This information is fused through multi-source data to generate threat confidence parameters, and a detailed three-dimensional spatial threat situation model is constructed based on this. The model not only displays the real-time position of the drone, but also predicts its possible flight trajectory. The threat level of the drone is evaluated using this model. According to the threat level assessment results, the corresponding early warning mechanism is quickly activated, and necessary countermeasures are prepared to be taken, thereby effectively ensuring the safety of the airspace.

[0102] To address the complexity and uncertainty in drone threat identification and further improve targeted countermeasures against drones with different behavior patterns, as another embodiment, as described in step 102, the target drone's behavior pattern and attack intent parameters are decomposed based on the threat level assessment results, and a dynamic decision factor set is generated in combination with geo-fence constraints and environmental interference factors. Specifically, the dynamic decision factor set includes:

[0103] Based on the trajectory prediction deviation and signal frequency hopping pattern in the threat level assessment results, the behavior pattern category of the target UAV is identified, wherein the behavior pattern categories include: reconnaissance mode, interference mode, and attack mode; the attack intention parameters of the target UAV are extracted according to the behavior pattern category, wherein the reconnaissance mode corresponds to the target area scanning frequency, the interference mode corresponds to the electromagnetic suppression intensity of the countermeasure device, and the attack mode corresponds to the dive angle and speed change rate of the target UAV; the attack intention parameters are spatially matched with the geo-fence constraint conditions, the minimum distance between the current position of the target UAV and the geo-fence boundary is calculated, and the spatial threat weight is generated in combination with the environmental interference factor; the attack intention parameters are dynamically corrected based on the spatial threat weight to obtain corrected attack intention parameters, wherein the target area scanning frequency in the reconnaissance mode increases with decreasing distance, the electromagnetic suppression intensity in the interference mode decreases with increasing environmental interference factors, and the dive angle and speed change rate in the attack mode increase with decreasing distance; the corrected attack intention parameters are combined with the behavior pattern category to generate a dynamic decision factor set including the spatial threat weight, the corrected attack intention parameters, and the behavior pattern category;

[0104] In this embodiment, behavioral pattern categories refer to different types of drone behavior based on flight characteristics (such as trajectory prediction deviation and signal frequency hopping patterns), primarily including reconnaissance mode, jamming mode, and attack mode. Attack intent parameters are specific parameters extracted based on different behavioral pattern categories, such as scanning frequency in reconnaissance mode, electromagnetic suppression intensity in jamming mode, and dive angle and velocity change rate in attack mode. The spatial threat weight is a weighted value that comprehensively considers the minimum distance between the drone's current location and the geofence boundary and environmental interference factors, and is used to quantify the actual threat level of the drone to a specific area.

[0105] In an embodiment of the present application, the behavior pattern category of the target UAV is first identified based on the trajectory prediction deviation and signal frequency hopping pattern in the threat level assessment result. Then, the corresponding attack intention parameters are extracted according to the identified behavior pattern category. Then, these attack intention parameters are spatially matched with the geo-fence constraint conditions, and the minimum distance between the current position of the target UAV and the geo-fence boundary is calculated. The spatial threat weight is generated in combination with the environmental interference factor. Based on this weight, the attack intention parameter is dynamically modified so that the target area scanning frequency in the reconnaissance mode increases with decreasing distance, the electromagnetic suppression intensity in the interference mode decreases with increasing environmental interference factors, and the dive angle and speed change rate in the attack mode increase with decreasing distance. Finally, the modified attack intention parameter is combined with the behavior pattern category to generate a dynamic decision factor set including the spatial threat weight, the modified attack intention parameter, and the behavior pattern category.

[0106] For example, when the system detects a drone suspected of illegally invading, it identifies that the drone is in reconnaissance mode based on the trajectory prediction deviation and signal frequency hopping pattern of the drone. The system extracts the target area scanning frequency of the drone as the attack intention parameter, and spatially matches it with the geographic fence constraint conditions to calculate the minimum distance between the drone's current position and the boundary of the airport's no-fly zone. Taking into account the existence of certain radio interference in the current environment, the system generates a spatial threat weight based on the environmental interference factor. Based on this weight, the system dynamically corrects the drone's scanning frequency so that it gradually increases as it approaches the boundary of the no-fly zone. In addition, the system also considers other possible behavior patterns, such as interference mode or attack mode, and calculates the corresponding corrected attack intention parameters through similar methods. Finally, a detailed set of dynamic decision factors is generated, which not only includes the drone's current behavior pattern and corrected attack intention parameters, but also reflects its actual threat level to airport security, thereby providing key information for formulating accurate and effective countermeasure plans.

[0107] In order to solve the problem of accuracy and pertinence of plan matching during the drone countermeasure process and further improve the ability to quickly respond to drones with different behavior patterns, as another embodiment, according to step 103, a dynamic decision factor set is matched with the basic countermeasure plan preset in the plan knowledge graph to generate multiple candidate countermeasure plans, and a plan logic tree is constructed based on these plans. Specifically, it includes:

[0108] Based on the behavioral pattern categories in the dynamic decision factor set, the matching basic countermeasure plan is extracted from the plan knowledge graph, where the reconnaissance mode corresponds to the spectrum detection and interference plan, the interference mode corresponds to the frequency band suppression and navigation deception plan, and the attack mode corresponds to the physical interception and directional strike plan; according to the modified attack intention parameters in the dynamic decision factor set, the basic countermeasure plan extracted from the plan knowledge graph is adapted and adjusted to obtain the basic countermeasure plan after parameter adaptation, where the scanning frequency of the spectrum detection plan is synchronized with the scanning frequency of the target area, and the transmission power of the frequency band suppression plan is synchronized with the power of the electric field. Matching the intensity of magnetic suppression, the triggering timing of the physical interception plan is associated with the dive angle and speed change rate; based on the spatial threat weight in the dynamic decision factor set, the basic countermeasure plan after parameter adaptation is prioritized to generate a candidate countermeasure plan set containing execution timing, intensity parameters and triggering conditions; the plan nodes in the candidate countermeasure plan set are constructed into a plan logic tree according to the execution timing and spatial position relationship, where the connecting edges between the plan nodes represent the switching conditions of the countermeasure, where the switching conditions include: electromagnetic suppression intensity threshold, navigation deception success rate and physical interception feasibility;

[0109] In this embodiment, the basic countermeasure plan after parameter adaptation refers to a specific plan obtained by adjusting the basic plan extracted from the plan knowledge graph according to the modified attack intent parameters (such as scanning frequency, electromagnetic suppression intensity, etc.) in the dynamic decision factor set; priority sorting is the process of sorting the adapted basic countermeasure plans based on spatial threat weights to determine the priority order of each plan in actual operation; the plan logic tree is a structured tree model that contains multiple plan nodes and the connecting edges between them, which is used to show the execution order of different countermeasures and their switching conditions;

[0110] In an embodiment of the present application, first, based on the behavior pattern category (reconnaissance mode, interference mode or attack mode) in the dynamic decision factor set, the corresponding basic countermeasure plan is extracted from the plan knowledge graph. For example, for the reconnaissance mode, spectrum detection and jamming plans are extracted; for the jamming mode, frequency band suppression and navigation deception plans are extracted; for the attack mode, physical interception and directional strike plans are extracted. Then, according to the modified attack intention parameters in the dynamic decision factor set, the parameters of the extracted basic plans are adapted and adjusted to ensure that the scanning frequency of the spectrum detection plan is synchronized with the scanning frequency of the target area, the transmission power of the frequency band suppression plan matches the electromagnetic suppression intensity, and the triggering timing of the physical interception plan is associated with the dive angle and speed change rate. Then, based on the spatial threat weight in the dynamic decision factor set, the basic countermeasure plans after parameter adaptation are prioritized to generate a set of candidate countermeasure plans containing execution timing, intensity parameters and triggering conditions. Finally, these candidate countermeasure plans are constructed into a plan logic tree according to the execution timing and spatial position relationship, where the connecting edges between the plan nodes represent the switching conditions of the countermeasures, such as the electromagnetic suppression intensity threshold, the navigation deception success rate and the feasibility of physical interception.

[0111] For example, in a security monitoring system around an airport, the system identified a drone in reconnaissance mode and extracted a spectrum detection and interference plan as the basic plan. Based on the modified attack intention parameters in the dynamic decision factor set, the system adjusted the scanning frequency of the spectrum detection plan to keep it consistent with the scanning frequency of the drone's target area. At the same time, considering the radio interference in the current environment, the system also adjusted the transmission power of the frequency band suppression plan to match the current electromagnetic suppression intensity requirements. Next, based on the spatial threat weight, the system prioritized the adjusted plans to ensure that the most effective countermeasures could be taken quickly in high-threat situations. Finally, these adjusted plans were constructed into a detailed plan logic tree according to the execution sequence and spatial position relationship, clarifying the switching conditions between different countermeasures. For example, when the electromagnetic suppression intensity reaches a certain threshold, the system will automatically switch to the navigation deception plan; if the navigation deception fails, the physical interception plan will be further initiated.

[0112] In order to solve the problem of accuracy and comprehensiveness of multi-dimensional evaluation parameters in the drone countermeasure process and further improve the adaptability and response efficiency to different threat scenarios, as another embodiment, according to step 104, based on the three-dimensional threat situation model, the dynamic decision factor set, and the countermeasure effectiveness indicators preset in the plan knowledge graph, a multi-dimensional evaluation parameter set is generated, specifically including:

[0113] The real-time position coordinates, trajectory prediction deviation and electromagnetic suppression priority of the target UAV are extracted from the three-dimensional space threat situation model, and the space threat assessment parameters are generated by combining the countermeasure effectiveness indicators in the plan knowledge graph. The space threat assessment parameters include: target distance attenuation coefficient, trajectory coverage completeness and electromagnetic suppression effectiveness value; the response time parameter and intensity matching parameter of the countermeasure are calculated from the modified attack intention parameters based on the dynamic decision factor set. The response time parameter is calculated by the speed change rate of the target UAV and the deployment position of the countermeasure device, and the intensity matching parameter is calculated by the electromagnetic suppression intensity and the countermeasure. Calculation of the transmission power range; based on the countermeasure effectiveness indicators in the plan knowledge graph, extract the spectrum compatibility parameters and energy efficiency parameters of the countermeasures, where the spectrum compatibility parameters are calculated by the overlapping area between the transmission frequency band of the countermeasures and the protection frequency band, and the energy efficiency parameters are calculated by the energy consumption per unit time of the countermeasures and the remaining power of the countermeasure device; normalize the spatial threat assessment parameters, response time parameters, intensity matching parameters, spectrum compatibility parameters and energy efficiency parameters to generate a multidimensional evaluation parameter set containing weight coefficients, where the weight coefficients are dynamically adjusted through the spatial threat weights in the dynamic decision factor set.

[0114] In this embodiment, the spatial threat assessment parameters are a set of parameters derived by combining the real-time position coordinates of the target UAV in the three-dimensional spatial threat situation model, the trajectory prediction deviation, and the electromagnetic suppression priority with the countermeasure effectiveness index in the plan knowledge graph, and are used to quantify the actual threat level of the UAV to a specific area; the response time parameter refers to the time delay calculated based on the speed change rate of the target UAV and the deployment position of the countermeasure device, reflecting the timeliness of the initiation of the countermeasure; the intensity matching parameter is a parameter derived by comparing the electromagnetic suppression intensity with the transmission power range of the countermeasure to ensure the effectiveness of the countermeasure; the spectrum compatibility parameter is a parameter derived by calculating the overlapping area between the countermeasure transmission frequency band and the protection frequency band to evaluate whether the countermeasure will interfere with the critical communication frequency band; the energy efficiency parameter is a parameter that evaluates the sustainable operation capability of the countermeasure by the relationship between the energy consumption per unit time and the remaining power of the countermeasure device;

[0115] In an embodiment of the present application, the real-time position coordinates, trajectory prediction deviation, and electromagnetic suppression priority of the target UAV are first extracted from the three-dimensional space threat situation model, and the countermeasure effectiveness indicators in the plan knowledge graph are combined to generate space threat assessment parameters, such as the target distance attenuation coefficient, trajectory coverage completeness, and electromagnetic suppression effectiveness value. Then, based on the modified attack intention parameters in the dynamic decision factor set, the response time parameter and intensity matching parameter of the countermeasure are calculated. For example, the response time parameter is determined by the speed change rate of the target UAV and the deployment position of the countermeasure device, and the intensity matching parameter is calculated by the electromagnetic suppression intensity and the transmission power range of the countermeasure. Then, based on the countermeasure effectiveness indicators in the plan knowledge graph, the spectrum compatibility parameter and energy efficiency parameter of the countermeasure are extracted to ensure that the countermeasure does not interfere with key communication frequency bands and has good energy utilization efficiency. Finally, the space threat assessment parameters, response time parameter, intensity matching parameter, spectrum compatibility parameter, and energy efficiency parameter are normalized to generate a multidimensional evaluation parameter set including weight coefficients, wherein the weight coefficient is dynamically adjusted according to the space threat weight in the dynamic decision factor set to ensure the accuracy and adaptability of the evaluation results.

[0116] For example, a security monitoring system around an airport identified a suspected intruder drone and extracted its real-time location coordinates, trajectory prediction deviation, and electromagnetic suppression priority from a three-dimensional spatial threat situation model. The system then combined the countermeasure effectiveness indicators in the emergency plan knowledge graph to generate spatial threat assessment parameters, such as the target distance attenuation coefficient and the electromagnetic suppression effectiveness value. The system then calculated the response time parameters of the countermeasures, taking into account the drone's speed change rate and the deployment location of the countermeasures to ensure a rapid response. The system also calculated the intensity matching parameters, comparing the electromagnetic suppression intensity with the countermeasure's transmission power range to ensure the effectiveness of the countermeasures. Furthermore, the system extracted the countermeasures' spectrum compatibility parameters and energy efficiency parameters to ensure that they would not interfere with the airport's critical communication bands and ensure the countermeasures' energy efficiency. Finally, the system normalized these parameters to generate a multidimensional evaluation parameter set containing weight coefficients, which were dynamically adjusted based on the spatial threat weights in the dynamic decision factor set.

[0117] Figure 2 The present invention provides a schematic diagram of a UAV countermeasure plan generation system based on artificial intelligence technology, as shown in FIG. Figure 2 As shown, the system includes:

[0118] The parsing module 21 is used to parse the dynamic monitoring data stream obtained from the target UAV, identify the flight trajectory characteristics and electromagnetic signal spectrum of the target UAV, and build a three-dimensional spatial threat situation model in combination with multi-source heterogeneous sensor fusion technology. The three-dimensional spatial threat situation model is used to assess the threat level of the target UAV and generate a threat level assessment result.

[0119] A decomposition module 22 is configured to decompose the target UAV's behavior pattern and attack intention parameters based on the threat level assessment result, and generate a dynamic decision factor set in combination with geo-fence constraints and environmental interference factors;

[0120] A construction module 23 is configured to match the basic countermeasure plans preset in the plan knowledge graph through a dynamic decision factor set to generate multiple candidate countermeasure plans, and construct a plan logic tree based on the multiple candidate countermeasure plans;

[0121] A generation module 24 is configured to generate a multi-dimensional evaluation parameter set based on the three-dimensional threat situation model, the dynamic decision factor set, and the countermeasure effectiveness indicators preset in the plan knowledge graph;

[0122] The synchronization module 25 is used to perform collaborative optimization processing on the plan logic tree based on the multi-dimensional evaluation parameter set to generate an optimal countermeasure plan, convert the optimal countermeasure plan into an executable instruction set and synchronize it to the countermeasure device in the target area.

[0123] Figure 2 The UAV countermeasure plan generation system based on artificial intelligence technology can execute Figure 1 The implementation principles and technical effects of the AI-based drone countermeasure plan generation method described in the illustrated embodiment are not further elaborated. The specific manner in which each module and unit performs operations in the AI-based drone countermeasure plan generation system described in the above embodiment has been described in detail in the embodiments of the method and will not be further elaborated here.

[0124] In one possible design, Figure 2 The UAV countermeasure plan generation system based on artificial intelligence technology in the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0125] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0126] The processing component 32 is configured to parse the dynamic monitoring data stream acquired from the target UAV, identify the flight trajectory characteristics and electromagnetic signal spectrum of the target UAV, and construct a three-dimensional spatial threat situation model in combination with multi-source heterogeneous sensor fusion technology. The three-dimensional spatial threat situation model is used to assess the threat level of the target UAV and generate a threat level assessment result. Based on the threat level assessment result, the target UAV's behavior pattern and attack intent parameters are decomposed, and a dynamic decision factor set is generated in combination with geographic fence constraints and environmental interference factors. The dynamic decision factor set is matched with the basic countermeasure plan preset in the plan knowledge graph to generate multiple candidate countermeasure plans, and a plan logic tree is constructed based on the multiple candidate countermeasure plans. A multi-dimensional evaluation parameter set is generated based on the three-dimensional spatial threat situation model, the dynamic decision factor set, and the countermeasure effectiveness indicators preset in the plan knowledge graph. The plan logic tree is collaboratively optimized based on the multi-dimensional evaluation parameter set to generate an optimal countermeasure plan, and the optimal countermeasure plan is converted into an executable instruction set and synchronized to the countermeasure device in the target area.

[0127] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0128] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile memory 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.

[0129] Of course, a computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0130] The input / output interface provides an interface between the processing component and the peripheral interface module, which can be an output device, an input device, etc.

[0131] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0132] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0133] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The embodiment shown is a method for generating a UAV countermeasure plan based on artificial intelligence technology.

[0134] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0135] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0136] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0137] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for generating a UAV countermeasure plan based on artificial intelligence technology, characterized in that: include: The dynamic monitoring data stream of the target UAV is collected synchronously by multi-source sensors, and the position coordinates, velocity vectors and electromagnetic spectrum data in the dynamic monitoring data stream are time-aligned to generate an original data set. The heading angle change rate and acceleration fluctuation characteristics of the target UAV are calculated based on the position coordinates and velocity vectors in the original data set, and the trajectory prediction deviation is generated by combining with the preset trajectory prediction model, and a flight trajectory feature vector is constructed. The electromagnetic spectrum data in the original data set is subjected to frequency domain feature extraction to obtain the main frequency band energy distribution and signal frequency hopping pattern of the target UAV, and an electromagnetic signal spectrum feature vector is generated. The heading angle change rate, acceleration fluctuation characteristics and trajectory prediction deviation in the flight trajectory feature vector are combined with the main frequency band energy distribution and signal frequency hopping pattern in the electromagnetic signal spectrum feature vector for multi-source data fusion to generate the threat confidence parameter of the target UAV. A three-dimensional threat situation model is constructed based on the threat confidence parameter, and the real-time position coordinates of the target UAV are mapped to the three-dimensional grid space under the constraints of the geographic fence. The threat level assessment result is generated by combining the trajectory prediction deviation and signal frequency hopping pattern. Based on the trajectory prediction deviation and signal frequency hopping pattern in the threat level assessment results, the behavior pattern category of the target UAV is identified, wherein the behavior pattern categories include: reconnaissance mode, interference mode, and attack mode; the attack intention parameters of the target UAV are extracted according to the behavior pattern category, wherein the reconnaissance mode corresponds to the target area scanning frequency, the interference mode corresponds to the electromagnetic suppression intensity of the countermeasure device, and the attack mode corresponds to the dive angle and speed change rate of the target UAV; the attack intention parameters are spatially matched with the geo-fence constraint conditions, the minimum distance between the current position of the target UAV and the geo-fence boundary is calculated, and the spatial threat weight is generated in combination with the environmental interference factor; the attack intention parameters are dynamically corrected based on the spatial threat weight to obtain corrected attack intention parameters, wherein the target area scanning frequency in the reconnaissance mode increases with decreasing distance, the electromagnetic suppression intensity in the interference mode decreases with increasing environmental interference factors, and the dive angle and speed change rate in the attack mode increase with decreasing distance; the corrected attack intention parameters are combined with the behavior pattern category to generate a dynamic decision factor set including the spatial threat weight, the corrected attack intention parameters, and the behavior pattern category; Generate multiple candidate countermeasure plans by matching the basic countermeasure plans preset in the plan knowledge graph through a dynamic decision factor set, and construct a plan logic tree based on the multiple candidate countermeasure plans; Generate a multi-dimensional evaluation parameter set based on the three-dimensional threat situation model, the dynamic decision factor set, and the countermeasure effectiveness indicators preset in the plan knowledge graph; The plan logic tree is collaboratively optimized based on a multi-dimensional evaluation parameter set to generate an optimal countermeasure plan, which is then converted into an executable instruction set and synchronized to the countermeasure device in the target area.

2. The method according to claim 1, characterized in that Based on a multi-dimensional evaluation parameter set, the plan logic tree is collaboratively optimized to generate the optimal countermeasure plan. The optimal countermeasure plan is converted into an executable instruction set and synchronized to the countermeasure device in the target area, including: Extract trajectory prediction deviation and electromagnetic suppression priority from the three-dimensional threat situation model, extract geo-fence penetration risk value and environmental interference attenuation coefficient from the dynamic decision factor set, and extract response delay threshold and spectrum interference compatibility from the emergency plan knowledge graph. The trajectory prediction deviation and response delay threshold are combined through the time series compression algorithm to generate dynamic response margin parameters; The peak value of the cross-correlation function between the electromagnetic suppression priority and the spectrum interference compatibility is calculated through the frequency domain overlap analysis algorithm. When the peak value exceeds the preset interference threshold, the spectrum avoidance coefficient is generated and the countermeasures of the corresponding frequency band are locked; An attenuation compensation model is constructed using the geo-fence penetration risk value and the environmental interference attenuation coefficient to dynamically correct the transmit power weight factor of the countermeasure device. The dynamic response margin parameters, spectrum avoidance coefficient and transmission power weight factor are input into the node activation function of the plan logic tree. The plan paths in the plan logic tree are pruned with confidence weighted by the Monte Carlo tree search algorithm. The plan paths with confidence higher than the preset dynamic threshold are used to generate the optimal countermeasure plan. The optimal countermeasure plan is converted into an executable instruction set and synchronized to the countermeasure device in the target area.

3. The method according to claim 2, characterized in that The dynamic response margin parameter, spectrum avoidance coefficient, and transmit power weight factor are input into the node activation function of the plan logic tree. The plan paths in the plan logic tree are pruned based on their confidence level using the Monte Carlo tree search algorithm. The optimal countermeasure plan is generated from the plan paths with confidence levels higher than a preset dynamic threshold. The optimal countermeasure plan is converted into an executable instruction set and synchronized to the countermeasure device in the target area, including: A time-sensitive search window is generated based on the dynamic response margin parameter, wherein the width of the time-sensitive search window shrinks exponentially as the trajectory prediction deviation increases. The time-sensitive search window is then spatiotemporally aligned with the node execution timestamps of the plan logic tree to screen candidate paths within the coverage of the time-sensitive search window. Perform frequency band compliance checks on candidate paths based on the spectrum avoidance coefficient. If the transmission frequency band of the countermeasure in the candidate path overlaps with the currently locked conflicting frequency band and the dynamic spectrum migration mechanism is not configured, the candidate path is marked as a disabled candidate path and removed from the contingency plan logic tree. The real-time energy consumption estimate of each candidate path is calculated using the transmit power weight factor. A dynamic fuse threshold is generated based on the remaining power of the countermeasure device and the environmental interference attenuation coefficient. When the cumulative energy consumption of a candidate path exceeds the fuse threshold, a backtracking mechanism is triggered and a low-power node is reselected. By integrating electromagnetic suppression priority data from a three-dimensional threat situation model, the confidence of candidate paths containing coordinated countermeasure nodes is weighted and improved. The coordinated countermeasure nodes must simultaneously meet the timing linkage conditions of radio frequency interference and navigation deception methods. The pruning threshold is dynamically adjusted according to the level of the geo-fence penetration risk value, and the plan paths with confidence levels higher than the preset dynamic threshold are subjected to conflict detection and redundancy elimination to generate the optimal countermeasure plan.

4. The method according to claim 1, wherein By matching the dynamic decision factor set with the basic countermeasures plan preset in the plan knowledge graph, multiple candidate countermeasure plans are generated, and a plan logic tree is constructed based on the multiple candidate countermeasure plans, including: Based on the behavioral pattern categories in the dynamic decision factor set, matching basic countermeasure plans are extracted from the plan knowledge graph. Among them, the reconnaissance mode corresponds to the spectrum detection and interference plan, the interference mode corresponds to the frequency band suppression and navigation deception plan, and the attack mode corresponds to the physical interception and directional strike plan; According to the modified attack intention parameters in the dynamic decision factor set, the basic countermeasure plan extracted from the plan knowledge graph is adjusted to obtain the basic countermeasure plan after parameter adaptation. Among them, the scanning frequency of the spectrum detection plan is synchronized with the scanning frequency of the target area, the transmission power of the frequency band suppression plan is matched with the electromagnetic suppression intensity, and the triggering timing of the physical interception plan is associated with the dive angle and speed change rate. Based on the spatial threat weights in the dynamic decision factor set, the basic countermeasure plans after parameter adaptation are prioritized to generate a candidate countermeasure plan set containing execution timing, intensity parameters, and trigger conditions; The plan nodes in the candidate countermeasure plan set are constructed into a plan logic tree according to the execution time sequence and spatial position relationship, where the connecting edges between the plan nodes represent the switching conditions of the countermeasure measures, where the switching conditions include: electromagnetic suppression intensity threshold, navigation deception success rate and physical interception feasibility.

5. The method according to claim 1, wherein Based on the three-dimensional threat situation model, the dynamic decision factor set, and the preset countermeasure effectiveness indicators in the plan knowledge graph, a multi-dimensional evaluation parameter set is generated, including: The real-time position coordinates, trajectory prediction deviation, and electromagnetic suppression priority of the target UAV are extracted from the three-dimensional spatial threat situation model. Combined with the countermeasure effectiveness indicators in the plan knowledge graph, spatial threat assessment parameters are generated. The spatial threat assessment parameters include: target distance attenuation coefficient, trajectory coverage completeness, and electromagnetic suppression effectiveness value. The response time parameter and intensity matching parameter of the countermeasure are calculated from the modified attack intention parameter based on the dynamic decision factor set. The response time parameter is calculated based on the speed change rate of the target UAV and the deployment position of the countermeasure device, and the intensity matching parameter is calculated based on the electromagnetic suppression intensity and the transmission power range of the countermeasure. Based on the countermeasure effectiveness indicators in the plan knowledge graph, the spectrum compatibility parameters and energy efficiency parameters of the countermeasures are extracted. The spectrum compatibility parameter is calculated by the overlapping area between the countermeasure's transmission frequency band and the protection frequency band, and the energy efficiency parameter is calculated by the countermeasure's energy consumption per unit time and the remaining power of the countermeasure device. The spatial threat assessment parameters, response time parameters, intensity matching parameters, spectrum compatibility parameters and energy efficiency parameters are normalized to generate a multidimensional assessment parameter set containing weight coefficients, where the weight coefficients are dynamically adjusted through the spatial threat weights in the dynamic decision factor set.

6. A UAV countermeasure plan generation system based on artificial intelligence technology, characterized by: include: The parsing module parses the dynamic monitoring data stream of the target UAV collected synchronously by multi-source sensors, and performs time alignment processing on the position coordinates, velocity vectors and electromagnetic spectrum data in the dynamic monitoring data stream to generate an original data set; calculates the heading angle change rate and acceleration fluctuation characteristics of the target UAV based on the position coordinates and velocity vectors in the original data set, generates a trajectory prediction deviation degree in combination with a preset trajectory prediction model, and constructs a flight trajectory feature vector; extracts frequency domain features from the electromagnetic spectrum data in the original data set to obtain the main frequency band energy distribution and signal frequency hopping pattern of the target UAV, and generates an electromagnetic signal spectrum feature vector; performs multi-source data fusion on the heading angle change rate, acceleration fluctuation characteristics and trajectory prediction deviation degree in the flight trajectory feature vector with the main frequency band energy distribution and signal frequency hopping pattern in the electromagnetic signal spectrum feature vector to generate a threat confidence parameter for the target UAV; constructs a three-dimensional threat situation model based on the threat confidence parameters, maps the real-time position coordinates of the target UAV to a three-dimensional grid space under the constraints of the geographic fence, and generates a threat level assessment result in combination with the trajectory prediction deviation degree and signal frequency hopping pattern; A decomposition module identifies the target UAV's behavior pattern category based on the trajectory prediction deviation and signal frequency hopping pattern in the threat level assessment result, wherein the behavior pattern categories include reconnaissance mode, interference mode, and attack mode; extracts the target UAV's attack intention parameters based on the behavior pattern category, wherein the reconnaissance mode corresponds to the target area scanning frequency, the interference mode corresponds to the electromagnetic suppression intensity of the countermeasure device, and the attack mode corresponds to the target UAV's dive angle and speed change rate; spatially matches the attack intention parameters with the geo-fence constraint conditions, calculates the minimum distance between the target UAV's current position and the geo-fence boundary, and generates a spatial threat weight based on the environmental interference factor; dynamically modifies the attack intention parameters based on the spatial threat weight to obtain modified attack intention parameters, wherein the target area scanning frequency in the reconnaissance mode increases with decreasing distance, the electromagnetic suppression intensity in the interference mode decreases with increasing environmental interference factors, and the dive angle and speed change rate in the attack mode increase with decreasing distance; and combines the modified attack intention parameters with the behavior pattern category to generate a dynamic decision factor set including the spatial threat weight, the modified attack intention parameters, and the behavior pattern category; A construction module is used to match the basic countermeasure plans preset in the plan knowledge graph through a dynamic decision factor set to generate multiple candidate countermeasure plans, and to construct a plan logic tree based on the multiple candidate countermeasure plans; A generation module is used to generate a multi-dimensional evaluation parameter set based on the three-dimensional threat situation model, the dynamic decision factor set, and the countermeasure effectiveness indicators preset in the plan knowledge graph; The synchronization module is used to collaboratively optimize the plan logic tree based on a multi-dimensional evaluation parameter set to generate an optimal countermeasure plan, convert the optimal countermeasure plan into an executable instruction set, and synchronize it to the countermeasure device in the target area.

7. A computing device, characterized in that It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a method for generating a drone countermeasure plan based on artificial intelligence technology as described in any one of claims 1 to 5.

8. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, the method for generating a UAV countermeasure plan based on artificial intelligence technology as described in any one of claims 1 to 5 is implemented.

Images