Unmanned aerial vehicle threat assessment method and system

Abstract

The invention discloses an unmanned aerial vehicle threat assessment method and system, and the method specifically comprises the steps: obtaining each static capability index value and an environment meteorological index value of a to-be-assessed unmanned aerial vehicle, and calculating a static capability threat score and a meteorological threat score of the to-be-assessed unmanned aerial vehicle based on the reference values of the static capability index and the environment meteorological index; obtaining a dynamic capability index value of the to-be-evaluated unmanned aerial vehicle, and calculating a dynamic capability threat score of the to-be-evaluated unmanned aerial vehicle based on a pre-trained neural network model; distribution of threat scores of the to-be-evaluated unmanned aerial vehicle is acquired, and weight values corresponding to the threat scores are calculated based on information entropies corresponding to the threat scores; and calculating a comprehensive threat score of each to-be-evaluated unmanned aerial vehicle based on the weight value corresponding to each threat score. Compared with the prior art, a more accurate and effective unmanned aerial vehicle threat assessment strategy is realized; the problem of model over-fitting or insufficient generalization ability is solved; and the method has a stronger capturing capability for a complex environment and a non-linear relationship.

Description

Technical Field

[0001] The present invention belongs to the technical field of threat assessment, and in particular relates to a method and system for assessing the threat of an unmanned aerial vehicle. Background Art

[0002] In recent years, the rapid development of drone technology and its widespread application have significantly increased the influence of drones in military and civilian fields. However, at the same time, the threat of drones has become increasingly prominent, especially in the military field. Drones can be used for reconnaissance, attack, interference and other tasks, posing a serious threat to targets. Therefore, how to effectively assess the threat level of drones has become an important issue that needs to be solved urgently. Drone threat assessment aims to provide a comprehensive assessment of the threat posed by drones to targets by quantitatively analyzing the capabilities, behaviors and combat environment of drones, and to provide decision makers with a reliable basis for judgment.

[0003] Existing drone threat assessment methods mainly rely on expert experience and superimpose threat indicators in a fixed weighted manner. Although intuitive and easy to use, they lack the ability to capture complex environments and nonlinear relationships. Machine learning models (such as support vector machines and decision trees) have been used to a certain extent in recent years. They have strong processing capabilities for multidimensional data, but in dynamic environments, the models may have problems of overfitting or insufficient generalization. In addition, in the assessment process, existing methods fail to fully consider the combined effects of drones' static capabilities (such as inherent attributes such as strike and maneuverability), dynamic capabilities (such as dynamic changes in speed, altitude, and distance targets), and combat environments (such as rainfall, wind speed, etc.), which limits the comprehensiveness and accuracy of the assessment results.

[0004] Therefore, in response to the above technical issues, it is necessary to provide a strategy to evaluate the threat level of drones.

[0005] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the invention and should not be regarded as an acknowledgment or any form of suggestion that the information constitutes the prior art already known to a person skilled in the art. Summary of the invention

[0006] The purpose of the present invention is to provide a drone threat assessment method and system, which can analyze drone threats more comprehensively and scientifically and provide effective support for defense strategies in different scenarios.

[0007] In order to achieve the above object, a technical solution provided by a specific embodiment of the present invention is as follows:

[0008] In a first aspect, the present invention provides a drone threat assessment method, which comprises:

[0009] Obtaining various static capability index values ​​and environmental meteorological index values ​​of the UAV to be evaluated, and calculating the static capability threat score and meteorological threat score of the UAV to be evaluated based on the reference values ​​of the static capability index and the environmental meteorological index;

[0010] Obtaining a dynamic capability index value of the UAV to be evaluated, and calculating a dynamic capability threat score of the UAV to be evaluated based on a pre-trained neural network model;

[0011] The distribution of each threat score of the drone to be evaluated is obtained respectively, and the weight value corresponding to the threat score is calculated based on the information entropy corresponding to each threat score;

[0012] Based on the weight values ​​corresponding to each of the threat scores, a comprehensive threat score of each of the drones to be evaluated is calculated.

[0013] In one or more embodiments of the present invention, calculating the static capability threat score of the drone to be evaluated includes:

[0014] Dividing the static capability of the drone to be evaluated into multiple static sub-capabilities, each static sub-capability including multiple static capability indicators;

[0015] Calculate the ratio of each static capability indicator value corresponding to the static sub-capability to the reference value corresponding to the static capability indicator, and obtain the static sub-capability score of the corresponding static sub-capability based on the average of the ratios;

[0016] Comparing the static sub-capabilities in pairs, and generating a first threat factor indicator comparison matrix based on the importance of each static sub-capability;

[0017] Normalizing the first threat factor indicator comparison matrix, and calculating the average of the elements in each row after normalization, where the average of the elements in each row is the weight value of the static sub-capability corresponding to the row in the matrix;

[0018] The product of the weight value of each static sub-capability and the corresponding static sub-capability score is calculated, and the average of the products is the static capability threat score of the unmanned aerial vehicle to be evaluated.

[0019] In one or more embodiments of the present invention, calculating the meteorological threat score of the unmanned aerial vehicle to be evaluated includes:

[0020] Obtaining a meteorological index score of a corresponding meteorological index based on a ratio of a meteorological index value of an environment in which the UAV to be evaluated is located to a reference value corresponding to the meteorological index;

[0021] Comparing the meteorological indicators in pairs, and generating a second threat factor indicator comparison matrix based on the importance of each meteorological indicator;

[0022] Normalizing the second threat factor index comparison matrix, and calculating the mean of the elements in each row after normalization, wherein the mean of the elements in each row is the weight value of the meteorological index corresponding to the row in the matrix;

[0023] The product of the weight value of each meteorological indicator and the corresponding meteorological indicator score is calculated, and the average of the products is the meteorological indicator threat score of the unmanned aerial vehicle to be evaluated.

[0024] In one or more embodiments of the present invention, the pre-trained neural network model includes:

[0025] Constructing an LSTM network structure model and setting relevant parameters of the LSTM network structure model;

[0026] Acquire a training sample set, wherein the training sample set includes various dynamic capability indicator values ​​of the drone and dynamic capability threat scores corresponding to the various dynamic capability indicator values ​​of the drone;

[0027] The LSTM network structure model is trained based on the training sample set, and the parameters of the LSTM network structure model are corrected until the error of the dynamic capability threat score of the drone to be evaluated by the LSTM network structure model is less than or equal to a preset threshold.

[0028] In one or more embodiments of the present invention, the formula for calculating the weight value corresponding to the threat score is:

[0029]

[0030]

[0031] Among them, p ij represents the proportion of the standardized value of the j-th category threat score of the i-th drone to be evaluated to the total threat score, z ij is the normalized value of the j-th category threat score of the i-th drone to be evaluated, and n is the number of drones to be evaluated; e j The information entropy for scoring the j-th threat category; w j is the first weight value of the j-th category threat score; m is the total number of indicators.

[0032] In one or more embodiments of the present invention, the index value is standardized, including:

[0033]

[0034] Among them, x ij is the j-th threat score corresponding to the i-th drone to be evaluated, max(x j ) is the maximum value of the j-th threat score among all the drones to be evaluated, min(xj ) is the minimum value of the j-th threat score among all the drones to be evaluated.

[0035] In one or more embodiments of the present invention, calculating the weight value corresponding to the threat score further includes:

[0036] Compare each threat score pairwise, and generate a third threat factor indicator comparison matrix based on the importance of each threat score;

[0037] Normalizing the third threat factor indicator comparison matrix, and calculating the average of the elements in each row after normalization, wherein the average of the elements in each row is a second weight value of the threat score corresponding to the row in the matrix;

[0038] In one or more embodiments of the present invention, the weight value corresponding to the threat score is calculated as follows:

[0039] w″ j =αw j +(1-α)w′ j

[0040] Among them, w″ j is the third weight value of the j-th category threat score, w j The first weight value of the j-th category threat score, w' j is the second weight value of the j-th category threat score, and α is the adjustment coefficient.

[0041] In one or more embodiments of the present invention, the step of calculating the comprehensive threat score of each drone to be evaluated based on the weight values ​​corresponding to each threat score includes:

[0042]

[0043] Among them, w″ j is the third weight value of the j-th category threat score, z ij is the standardized value of the j-th threat score of the i-th drone to be evaluated, m is the number of threat score types, S i is the comprehensive threat score of the i-th drone to be evaluated.

[0044] In a second aspect, the present invention provides a drone threat assessment system, comprising:

[0045] The first calculation module is used to obtain the static capability index values ​​and environmental meteorological index values ​​of the UAV to be evaluated, and calculate the static capability threat score and meteorological threat score of the UAV to be evaluated based on the reference values ​​of the static capability index and the environmental meteorological index;

[0046] A deduction module, used to obtain the dynamic capability index value of the UAV to be evaluated, and to infer the dynamic capability threat score of the UAV to be evaluated based on a pre-trained neural network model;

[0047] The second calculation module is used to obtain the distribution of each threat score of the drone to be evaluated, and calculate the weight value corresponding to the threat score based on the information entropy corresponding to each threat score;

[0048] The third calculation module is used to calculate the comprehensive threat score of each of the drones to be evaluated based on the weight values ​​corresponding to each of the threat scores.

[0049] Compared with the prior art, the drone threat assessment method provided by the present invention fully considers the various properties of drones and improves the comprehensiveness and accuracy of the assessment results. At the same time, based on the impact on drone threats, the present invention classifies the indicators that affect the degree of drone threats at multiple levels, refines the action mechanism of each indicator, and classifies the threat scores of each indicator to achieve a more accurate and effective drone threat assessment strategy.

[0050] Secondly, for different types of threat indicators, the present invention adopts different targeted scoring methods. For static indicators with small changes and clear impact on the threat level, as well as the meteorological indicators of the environment where the drone is located, the weights are calculated by fitting the threat factor indicator comparison matrix; for dynamic indicators with large changes and difficult to define the importance of the threat level to the drone, the dynamic capability threat score is evaluated by introducing a neural network model. This method further improves the accuracy of threat assessment and solves the problem of overfitting or insufficient generalization of the neural network model.

[0051] Finally, the entropy values ​​of various threat scores are integrated to generate dynamic weights to aggregate the various threat scores into a comprehensive threat score for the drone to be evaluated. This abandons the fixed weight aggregation method of transmission and has a stronger ability to capture complex environments and nonlinear relationships. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0053] Figure 1 is a schematic diagram of a drone threat assessment scenario in one embodiment of the present invention;

[0054] Figure 2is a flow chart of a drone threat assessment method in one embodiment of the present invention;

[0055] Figure 3 is a structural block diagram of a drone threat assessment system in one embodiment of the present invention;

[0056] Figure 4 It is a structural block diagram of an electronic device in one embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0058] Unless explicitly stated otherwise, throughout the specification and claims, the term “comprise” or variations such as “include” or “comprising”, etc., will be understood to include the stated elements or components but not to exclude other elements or components.

[0059] In the existing technical solutions, there are methods for conducting threat assessment on drones, specifically including:

[0060] Obtain the maneuverability attribute information of enemy combat aircraft at the preset time, including speed, altitude, and distance change from our target. Set corresponding weight values ​​for each threat indicator based on expert experience. Normalize the above-obtained local fighter information data, multiply and add the corresponding weight values, and obtain the threat assessment value of the corresponding enemy fighter.

[0061] However, under this technical solution, the threat assessment index of UAV is limited to the dynamic performance at the current moment, and the comprehensive influence of the static capabilities of UAV (such as inherent attributes such as strike and maneuverability) and the combat environment (such as terrain, climate, enemy and friendly situation) is not fully considered, which limits the comprehensiveness and accuracy of the assessment results. In addition, the aggregation weight of each threat score depends on expert experience, so the solution itself is extremely subjective. At the same time, the above dynamic performance parameters are actually difficult to distinguish the importance of threat scores, so the processing strategy with strong subjectivity is prone to cause large errors. At the same time, it is difficult to make dynamic adaptive changes in different scenarios, and the weight value can only be readjusted based on experts. Therefore, this technical solution will inevitably have problems such as poor adaptability, strong subjectivity, and lack of ability to capture complex environments and nonlinear relationships.

[0062] The inventor of the present invention discovered the main shortcomings of the prior art and proposed a new technical implementation idea based on the shortcomings of the prior art: expand the selection of influencing indicators for drone threat assessment, take into account the static performance parameters of drones and the external environment in which drones are located, and comprehensively analyze the threat score of drones. On the other hand, as the number of indicators involved in threat assessment increases, different types of indicators are evaluated based on their respective properties, which further improves the accuracy of threat assessment. Finally, based on the entropy value of each indicator type, the corresponding dynamic weight is solved, and further aggregated into a comprehensive threat score corresponding to the drone to be evaluated.

[0063] Please refer to Figure 1 , which is a schematic diagram of an application scenario of the drone threat assessment method provided by the present invention under one implementation mode, and the scenario specifically includes: a database 101, a scanning and identification module 102, and a calculation module 103.

[0064] It should be noted that there are communication connections between the database 101, the scanning and identification module 102, and the computing module 103. The communication network derived from the communication connection may include various connection types, including but not limited to: wired connection, wireless connection, or optical fiber cable connection, etc. At the same time, the communication network may be a local area network, a metropolitan area network, a wide area network, or any combination of the three.

[0065] The database 101 stores the appearance, models and static capability index values ​​of multiple drones. The scanning and identification module 102 is used to identify the appearance of the drone to be evaluated, and transmit the image information back to the database, and obtain the static capability index values ​​of the drone to be evaluated by comparison index. At the same time, the scanning and identification module 102 is also used to obtain the dynamic capability index values ​​of the drone to be evaluated and the current environmental meteorological index values, and send them to the calculation module 103 together with the static capability index values ​​matched by the database 101 index. The calculation module is configured with the formulas and models required for calculating each stage of the comprehensive threat score of the drone to be evaluated, and finally outputs the comprehensive threat score of the drone to be evaluated through the above formulas and models.

[0066] It should also be noted that the drone threat assessment method of the embodiment of the present invention can be applied to the drone threat assessment system of the embodiment of the present invention. The drone threat assessment system can be configured in a terminal. The terminal can include but is not limited to a PC (Personal Computer), a PDA (tablet computer), a smart phone, a smart wearable device, and the like.

[0067] Please refer to Figure 2 FIG. 1 is a flow chart of a drone threat assessment method according to an embodiment of the present invention. The drone threat assessment method specifically comprises the following steps:

[0068] S201: Obtaining various static capability index values ​​and environmental meteorological index values ​​of the drone to be evaluated, and calculating the static capability threat score and meteorological threat score of the drone to be evaluated based on the reference values ​​of the static capability index and environmental meteorological index in the drone cluster;

[0069] It should be noted that the static capability indicators refer to the inherent properties of the drone when it leaves the factory due to its model, structure, material, etc., and have nothing to do with the actual combat environment and combat situation. The static capability indicators include but are not limited to the maximum speed, endurance, climb rate, maximum weapon load, maximum attack distance, anti-interference capability, communication interference intensity, reconnaissance range, reconnaissance resolution, etc. of the drone.

[0070] The acquisition of static capability index values ​​can be based on scanning sensor devices such as radar to obtain the appearance, model, etc. of the corresponding drone to be evaluated, and index acquisition is performed based on a preset database. The embodiment of the present invention does not limit the method for obtaining static capability index values.

[0071] Environmental meteorological indicators are indicators used to evaluate the meteorological conditions of the current battlefield environment that have an impact on the UAV threat assessment. Environmental meteorological indicators may include, but are not limited to, wind speed, precipitation, snowfall, cloud coverage per unit area, visibility, etc. Similarly, based on the type of UAV to be evaluated, different UAV operating environments / UAV simulation environments, the selection of environmental meteorological indicators can be adaptively adjusted. For example, for UAVs equipped with microwave radars, infrared radars, or ultrasonic radars, visibility has no effect or little effect on their threat assessment. Therefore, visibility indicators may not be considered when selecting environmental meteorological indicators. In summary, the embodiments of the present invention do not limit the specific selection of UAV static capability indicators and environmental meteorological indicators.

[0072] It should be noted that different static capability indicators of the UAV correspond to its different static capabilities. Since the number of static capability indicators involved in the consideration is huge, the static capabilities can be classified in one implementation so that the static capabilities of the UAV to be evaluated are divided into multiple static sub-capabilities, and each static sub-capability contains multiple static capability indicators.

[0073] For example, the static capability of a drone can be divided into maneuverability, strike capability, interference capability, and reconnaissance capability. The above-mentioned maneuverability, strike capability, interference capability, and reconnaissance capability are the static sub-capabilities corresponding to the static capability in this embodiment. Furthermore, the maneuverability is specifically divided into: maximum speed, endurance time, and climb rate; the strike capability is specifically divided into: maximum weapon load, maximum attack distance; the interference capability is specifically divided into: anti-interference capability, communication interference intensity; the reconnaissance capability is specifically divided into: reconnaissance range, reconnaissance resolution. The maximum speed, endurance time, climb rate, etc. are the static capability indicators corresponding to the static sub-capabilities.

[0074] In an exemplary embodiment, the static capability threat score of the unmanned aerial vehicle to be evaluated is calculated, including: calculating the ratio of each static capability index value corresponding to the static sub-capability to the reference value corresponding to the static capability index, and obtaining the static sub-capability score of the corresponding static sub-capability based on the average of the ratios; comparing each static sub-capability pairwise, and generating a first threat factor index comparison matrix based on the importance of each static sub-capability; normalizing the first threat factor index comparison matrix, and calculating the average of the elements in each row after normalization, the average of the elements in each row is the weight value of the static sub-capability corresponding to the row in the matrix; calculating the product of the weight value of each static sub-capability and the corresponding static sub-capability score, the average of the product is the static capability threat score of the unmanned aerial vehicle to be evaluated.

[0075] It is understandable that, since different index values ​​correspond to different intervals, in order to standardize the various index values, by setting a reference value, the meteorological index value of the environment in which the drone to be evaluated is compared with the reference value corresponding to the meteorological index, and the "index values" with different ranges are unified into a "proportion of the reference value" with a consistent interval, and then the corresponding score can be obtained through the preset coefficient. The coefficient can be used to adjust the score interval of the final index score, and then the average of the various index scores can be taken to obtain the corresponding meteorological index threat score. That is, let

[0076]

[0077] Among them, T endurance is a meteorological index value of the current UAV to be evaluated, T ref is the meteorological index value T of the UAV to be evaluated endurance The corresponding meteorological index reference value, k is the preset constant coefficient.

[0078] The value of the preset constant coefficient k can be dynamically adjusted based on the expected interval of the score value. The reference value can be the average value of the corresponding indicator of the drone cluster or the maximum value of the corresponding indicator in the cluster. The embodiment of the present invention does not limit the reference value and the value of the preset constant coefficient.

[0079] It should also be noted that item a in the threat factor index comparison matrix ij Indicates the importance of indicator i compared to indicator j. In addition, a ij ×a ji = 1 and when i = j, a ij =1.

[0080] For example, in a specific embodiment, by comparing the static sub-capabilities of the drones to be evaluated in pairs, the first threat factor index comparison of maneuverability, strike capability, interference capability, and reconnaissance capability is constructed as shown in Table 1:

[0081] Table 1 Comparison matrix of indicators of the first threat factor

[0082] Mobility Strike capability Interference ability Reconnaissance capabilities Mobility 1 3 5 7 Strike capability 1 / 3 1 3 5 Interference ability 1 / 5 1 / 3 1 3 Reconnaissance capabilities 1 / 7 1 / 5 1 / 3 1

[0083] Furthermore, the embodiment of the present invention does not limit the specific method of normalizing the threat factor index comparison matrix. Optionally, each item in the matrix can be divided by the sum of its column; or the value of each column of the matrix can be divided by the absolute value of the sum of the squares of all elements in each column.

[0084] The mean of the normalized row elements is the weight value of the static sub-capability corresponding to the row in the matrix; the product of the weight value of each static sub-capability and the corresponding static sub-capability score is calculated, and the mean of the product is the static capability threat score of the drone to be evaluated.

[0085] Similarly, in an exemplary embodiment, the meteorological threat score of the unmanned aerial vehicle to be evaluated is calculated, including: calculating the average of the ratios of the meteorological index values ​​of the environment in which the unmanned aerial vehicle to be evaluated is located to the reference values ​​corresponding to the meteorological index as the meteorological index score; comparing the meteorological indexes in pairs, and generating a second threat factor index comparison matrix based on the importance of each meteorological index; normalizing the second threat factor index comparison matrix, and calculating the average of the elements in each row after normalization, the average being the weight value of the meteorological index corresponding to the row in the matrix; and calculating the average of the product of the weight values ​​of each meteorological index and the corresponding meteorological index score as the meteorological index threat score of the unmanned aerial vehicle to be evaluated.

[0086] S202: Obtaining a dynamic capability index value of the drone to be evaluated, and calculating a dynamic threat score of the drone to be evaluated based on a pre-trained neural network model;

[0087] It should be noted that dynamic capabilities refer to the threats generated by drones during mission execution, which are usually the real-time variable properties of drones in specific combat environments, such as the drone’s speed, altitude, and distance from the target.

[0088] It is understandable that the dynamic attributes of the drone may change at any time during the execution of the mission. If the same weight calculation method as the static capability index is used, it will lead to excessive calculation amount, and the huge calculation delay makes it difficult to ensure the timeliness of the drone threat assessment. On the other hand, it is difficult to measure the importance of the two dynamic capability indicators. Therefore, even under the premise of meeting the calculation delay, the generation of the threat factor index comparison matrix will produce large errors, which seriously affects the accuracy of the comprehensive threat score. Therefore, the present invention adopts a pre-trained neural network model to obtain the dynamic capability threat score. At the same time, different types of data use different threat score acquisition methods, which can also avoid the problem of overfitting or insufficient generalization ability of the model.

[0089] It should be noted that the present invention does not limit the type of neural network model used, but because the dynamic data of the drone is time-series and the input sequence is long, it is preferred to use an LSTM network structure model to evaluate the dynamic capability threat score. The LSTM (Long short-term memory) network structure model can solve the gradient vanishing and gradient exploding problems in the long sequence training process and has better performance in longer sequences. At the same time, the flight of the drone is continuous, and the data of a single time segment is easily disturbed. Therefore, using an LSTM structure that can process time-series data can effectively utilize the historical data during the drone flight.

[0090] In an exemplary embodiment of the present invention, a pre-trained neural network model includes: constructing an LSTM network structure model and setting relevant parameters of the LSTM network structure model; obtaining a training sample set, wherein the training sample set includes various dynamic capability indicator values ​​of the drone and dynamic capability threat scores corresponding to the various dynamic capability indicator values ​​of the drone; training the LSTM network structure model based on the training sample set, and correcting the parameters of the LSTM network structure model until the error of the dynamic capability threat score of the drone to be evaluated by the LSTM network structure model is less than or equal to a preset threshold.

[0091] During the training process, the dynamic capability index values ​​of the drone are usually formed into a vector input model, and the loss function is continuously reduced by correcting the relevant parameters of the model, thereby improving the reasoning accuracy of the model. In one embodiment of the present invention, the input at time t can be expressed as:

[0092] x t ={μ velocity,t ,μ height,t ,μ angle,t ,μ distance,t ,μ load,t μ task,t}

[0093] The loss function can be:

[0094]

[0095] Among them, x t is the dynamic capability index value vector input at time t, μ velocity,t is the speed of the drone at time t, μ height,t is the height of the drone at time t, μ angle,t is the azimuth of the UAV, μ distance,t is the distance between the UAV and the target, μ load,t is the payload of the UAV, μ task,tIt is the UAV mission phase; is the output, which represents the dynamic capability threat assessment value of the drone in the current time segment, y i is the threat experience value determined based on expert experience and historical data, and δ is a hyperparameter.

[0096] After the LSTM network structure model is trained, in an actual evaluation environment, the dynamic capability index value corresponding to the UAV to be evaluated is obtained based on detection equipment such as sensors or radars, and a dynamic capability index value vector is generated. The vector is used as the model input, and the output is the dynamic capability threat score of the UAV to be evaluated at the current moment.

[0097] S203: Obtain the distribution of each threat score of the drone to be evaluated, and calculate the weight value corresponding to the threat score based on the information entropy corresponding to each threat score;

[0098] It should be noted that information entropy is a measure of uncertainty or randomness of data. The more random and uncertain the data is, the greater its information entropy is; and the more certain and regular the data is, the smaller its information entropy is. As information entropy increases, the discreteness of the corresponding indicator information decreases, uncertainty decreases, the amount of information increases, and the comprehensive evaluation ability of the indicator information becomes stronger. Therefore, the indicator weight corresponding to the indicator should also be smaller in the end.

[0099] Since different threat scores come from the aggregation of different types of indicator data, the data range of each threat score may be different. Therefore, before calculating the weight value, in order to ensure that the final result is not affected by the dimension of each threat score, each threat score should be standardized.

[0100] Specifically, in an exemplary embodiment of the present invention, the normalization processing formula may be:

[0101]

[0102] Among them, x ij is the j-th threat score corresponding to the i-th drone to be evaluated, max(x j ) is the maximum value of the j-th threat score among all the drones to be evaluated, min(x j ) is the minimum value of the j-th threat score among all the drones to be evaluated.

[0103] It should be noted that, since different types of threat scores have different effects on the comprehensive threat score of the drone to be evaluated, the embodiment of the present invention adopts a classification and standardization strategy for different types of threat scores. Specifically, the various types of threat scores used to aggregate and generate the comprehensive threat score of the drone to be evaluated are divided into positive indicators and negative indicators. Based on the above formula, further standardization processing is performed. This processing method can keep the original characteristics of the load data to the greatest extent, and the obtained standardized results are also the most accurate.

[0104] The present invention also provides another set of implementation methods. After classifying various threat scores, the reverse indicators in the selected threat scores can be converted into positive indicators based on a preset indicator positive conversion method, such as the inverse reciprocal conversion method, and then all indicators can be brought into the same set of standardized formulas. The purpose of data standardization required by the present invention can also be achieved. It is understandable that the positive indicators can also be uniformly converted into reverse indicators, and the present invention will not repeat this implementation method.

[0105] After each threat score is standardized, the weight value can be calculated using a preset formula. The formula for calculating the weight value corresponding to the threat score is:

[0106]

[0107] Among them, p ij represents the proportion of the standardized value of the j-th category threat score of the i-th drone to be evaluated to the total threat score, z ij is the normalized value of the j-th category threat score of the i-th drone to be evaluated, and n is the number of drones to be evaluated; e j The information entropy for scoring the j-th threat category; w j is the first weight value of the j-th category threat score; m is the total number of indicators. In particular, in order to avoid the meaninglessness of the logarithmic function, it should also be stipulated that if p ij =0, then define p ij ln(p ij )=0.

[0108] In a specific embodiment, the threat parameters of three drones to be evaluated are shown in Table 2:

[0109] Table 2

[0110] Sample No. Static ability score Dynamic capability score Combat Environment Score 1 80 70 90 2 60 85 75 3 75 65 80

[0111] Taking the static capability score as an example, the corresponding standardized static capability scores of the three drones to be evaluated are: Sample 1 is (80-60) / (80-60)=1.00; Sample 2 is (60-60) / (80-60)=0.00; Sample 3 is (75-60) / (80-60)=0.75.

[0112] The standardized data are shown in Table 3:

[0113] Table 3

[0114]

[0115] Taking the static capability score as an example, the standardized static capability score ratio p of the three drones to be evaluated is i1 They are: Sample 1: 1.00 / (1.00+0.00+0.75)=0.5714; Sample 2 numerator is 0, so p 21 is 0; Sample 3: 0.75 / (1.00+0.00+0.75)=0.4286.

[0116] The corresponding static capability information entropy value is:

[0117] e1=-0.9102×[(0.5714ln(0.5714))+(0·ln(0.0000))+(0.4286ln(0.4286))]=0.6826

[0118] Similarly, in the above embodiment, the dynamic capability score entropy value is: j =0.5004, Combat environment score entropy value: e j =0.5623.

[0119] The final calculation of the weights corresponding to each threat score is:

[0120] Static capability weight:

[0121] Dynamic capability weight:

[0122] Combat environment weight:

[0123] It is understandable that the first weight value can be directly used as the weight of each corresponding threat score to calculate the comprehensive threat score of the drone to be evaluated. The first weight value is obtained only based on the distribution of the acquired threat scores, eliminating the result bias caused by human subjective assignment, avoiding the influence of subjective factors, improving the objectivity and accuracy of the evaluation results, and is suitable for experimental scenarios with large data volumes.

[0124] In another embodiment of the present invention, the threat scores of each item may be compared in pairs, and a third threat factor indicator comparison matrix may be generated based on the importance of each threat score; the third threat factor indicator comparison matrix may be normalized, and the mean of each row element after normalization may be calculated, wherein the mean of each row element is a second weight value of the threat score corresponding to the row in the matrix;

[0125] Since the establishment of the threat indicator matrix of this scheme needs to rely on subjective experience judgment, in this embodiment, subjective judgment of the importance of the indicators can be made to assist in generating the weights of each threat score, thereby improving the performance of the threat assessment of this application when facing new types of drones or data scarcity.

[0126] In another embodiment of the present invention, the advantages of the above two embodiments are combined to form a third weight value, which is used to aggregate and generate a comprehensive threat score for the drone to be evaluated. The formula of the third weight value is:

[0127] w″ j =αw j +(1-α)w′ j

[0128] Among them, w″ j is the third weight value of the j-th category threat score, w j The first weight value of the j-th category threat score, w' j is the second weight value of the j-th category threat score, and α is the adjustment coefficient. It can be understood that the adjustment coefficient value can be dynamically adjusted according to the actual application scenario of the present invention, and the embodiment of the present invention does not limit the setting of the adjustment coefficient.

[0129] S204: Calculate the comprehensive threat score of each of the drones to be evaluated based on the weight values ​​corresponding to each of the threat scores.

[0130] It is understandable that the product of each threat score and its corresponding weight value is calculated, and the average value of the product is the comprehensive threat score of the corresponding drone to be evaluated. The weight value may be the corresponding first weight value, second weight value, or third weight value.

[0131] Preferably, the third weight value is used to calculate the comprehensive threat score of the drone to be evaluated, and the formula is as follows:

[0132]

[0133] Among them, w″ j is the third weight value of the j-th category threat score, z ij is the standardized value of the j-th threat score of the i-th drone to be evaluated, m is the number of threat score types, S i is the comprehensive threat score of the i-th drone to be evaluated.

[0134] Please refer to Figure 3 As shown, based on the same inventive concept as the aforementioned drone threat assessment method, an embodiment of the present invention provides a drone threat assessment system 300, which includes: a first computing module 301, a deduction module 302, a second computing module 303 and a third computing module 304.

[0135] Specifically, the first calculation module 301 is used to obtain the static capability index values ​​and environmental meteorological index values ​​of the UAV to be evaluated, and calculate the static capability threat score and meteorological threat score of the UAV to be evaluated based on the reference values ​​of the static capability index and the environmental meteorological index; the deduction module 302 is used to obtain the dynamic capability index value of the UAV to be evaluated, and infer the dynamic capability threat score of the UAV to be evaluated based on the pre-trained neural network model; the second calculation module 303 is used to respectively obtain the distribution of each threat score of the UAV to be evaluated, and calculate the weight value corresponding to the threat score based on the information entropy corresponding to each threat score; the third calculation module 304 is used to calculate the comprehensive threat score of each UAV to be evaluated based on the weight value corresponding to each threat score.

[0136] Please refer to Figure 4 As shown, an embodiment of the present invention further provides an electronic device 400, which includes at least one processor 401, a memory 402 (such as a non-volatile memory), a memory 403, and a communication interface 404, and the at least one processor 401, the memory 402, the memory 403, and the communication interface 404 are connected together via an internal bus 405. At least one processor 401 is used to call at least one program instruction stored or encoded in the memory 402, so that the at least one processor 401 performs various operations and functions of the drone threat assessment method described in various embodiments of this specification.

[0137] In an embodiment of the present specification, the electronic device 400 may include, but is not limited to, a personal computer, a server computer, a workstation, a desktop computer, a laptop computer, a notebook computer, a mobile electronic device, a smart phone, a tablet computer, a cellular phone, a personal digital assistant (PDA), a handheld device, a messaging device, a wearable electronic device, a consumer electronic device, and the like.

[0138] An embodiment of the present invention also provides a computer-readable medium, which carries computer-executable instructions. When the computer-executable instructions are executed by a processor, they can be used to implement various operations and functions of the drone threat assessment method described in various embodiments of this specification.

[0139] The computer-readable medium in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device.

[0140] In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0141] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0142] The present invention is described with reference to flowcharts and / or block diagrams of methods, apparatuses, systems, and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as a combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the process in the flowchart. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0143] The foregoing description of specific exemplary embodiments of the present invention is for the purpose of illustration and demonstration. These descriptions are not intended to limit the present invention to the precise form disclosed, and it is clear that many changes and variations can be made based on the above teachings. The purpose of selecting and describing the exemplary embodiments is to explain the specific principles of the present invention and its practical application, so that those skilled in the art can realize and utilize various different exemplary embodiments of the present invention and various different selections and changes. The scope of the present invention is intended to be limited by the claims and their equivalents.

[0144] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

[0145] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A drone threat assessment method, characterized in that: include: Obtaining various static capability index values ​​and environmental meteorological index values ​​of the UAV to be evaluated, and calculating the static capability threat score and meteorological threat score of the UAV to be evaluated based on the reference values ​​of the static capability index and the environmental meteorological index; Obtaining a dynamic capability index value of the UAV to be evaluated, and calculating a dynamic capability threat score of the UAV to be evaluated based on a pre-trained neural network model; The distribution of each threat score of the drone to be evaluated is obtained respectively, and the weight value corresponding to the threat score is calculated based on the information entropy corresponding to each threat score; Based on the weight values ​​corresponding to each of the threat scores, a comprehensive threat score of each of the drones to be evaluated is calculated.

2. The drone threat assessment method according to claim 1, characterized in that: Calculating the static capability threat score of the drone to be evaluated, including: Dividing the static capability of the drone to be evaluated into multiple static sub-capabilities, each static sub-capability including multiple static capability indicators; Calculate the ratio of each static capability indicator value corresponding to the static sub-capability to the reference value corresponding to the static capability indicator, and obtain the static sub-capability score of the corresponding static sub-capability based on the average of the ratios; Comparing the static sub-capabilities in pairs, and generating a first threat factor indicator comparison matrix based on the importance of each static sub-capability; Normalizing the first threat factor indicator comparison matrix, and calculating the average of the elements in each row after normalization, where the average of the elements in each row is the weight value of the static sub-capability corresponding to the row in the matrix; The product of the weight value of each static sub-capability and the corresponding static sub-capability score is calculated, and the average of the products is the static capability threat score of the unmanned aerial vehicle to be evaluated.

3. The drone threat assessment method according to claim 1, characterized in that: Calculate the meteorological threat score of the drone to be evaluated, including: Obtaining a meteorological index score of a corresponding meteorological index based on a ratio of a meteorological index value of an environment in which the UAV to be evaluated is located to a reference value corresponding to the meteorological index; Comparing the meteorological indicators in pairs, and generating a second threat factor indicator comparison matrix based on the importance of each meteorological indicator; Normalizing the second threat factor index comparison matrix, and calculating the mean of the elements in each row after normalization, wherein the mean of the elements in each row is the weight value of the meteorological index corresponding to the row in the matrix; The product of the weight value of each meteorological indicator and the corresponding meteorological indicator score is calculated, and the average of the products is the meteorological indicator threat score of the unmanned aerial vehicle to be evaluated.

4. The drone threat assessment method according to claim 1, characterized in that: The pre-trained neural network model comprises: Constructing an LSTM network structure model and setting relevant parameters of the LSTM network structure model; Acquire a training sample set, wherein the training sample set includes various dynamic capability indicator values ​​of the drone and dynamic capability threat scores corresponding to the various dynamic capability indicator values ​​of the drone; The LSTM network structure model is trained based on the training sample set, and the parameters of the LSTM network structure model are corrected until the error of the dynamic capability threat score of the drone to be evaluated by the LSTM network structure model is less than or equal to a preset threshold.

5. The drone threat assessment method according to claim 1, characterized in that: The formula for calculating the weight value corresponding to the threat score is: Among them, p ij represents the proportion of the standardized value of the j-th category threat score of the i-th drone to be evaluated to the total threat score, z ij is the normalized value of the j-th category threat score of the i-th drone to be evaluated, and n is the number of drones to be evaluated; e j The information entropy for scoring the j-th threat category; w j is the first weight value of the j-th category threat score; m is the total number of indicators.

6. The drone threat assessment method according to claim 5, characterized in that: Standardize indicator values, including: Among them, x ij is the j-th threat score corresponding to the i-th drone to be evaluated, max(x j ) is the maximum value of the j-th threat score among all the drones to be evaluated, min(x j ) is the minimum value of the j-th threat score among all the drones to be evaluated.

7. The drone threat assessment method according to claim 5, characterized in that: Calculating the weight value corresponding to the threat score also includes: Compare each threat score pairwise, and generate a third threat factor indicator comparison matrix based on the importance of each threat score; The third threat factor indicator comparison matrix is ​​normalized, and the average of the normalized row elements is calculated, where the average of the row elements is the second weight value of the threat score corresponding to the row in the matrix.

8. The drone threat assessment method according to claim 7, characterized in that: The weight value corresponding to the threat score is calculated as: w″ j =αw j +(1-α)w′ j Among them, w″ j is the third weight value of the j-th category threat score, w j is the first weight value of the j-th category threat score, w′ j is the second weight value of the j-th category threat score, and α is the adjustment coefficient.

9. The drone threat assessment method according to claim 8, characterized in that: The calculating, based on the weight values ​​corresponding to the threat scores, of the comprehensive threat scores of the drones to be evaluated comprises: Among them, w″ j is the third weight value of the j-th category threat score, z ij is the standardized value of the j-th threat score of the i-th drone to be evaluated, m is the number of threat score types, S i is the comprehensive threat score of the i-th drone to be evaluated.

10. A drone threat assessment system, characterized in that: include: The first calculation module is used to obtain the static capability index values ​​and environmental meteorological index values ​​of the UAV to be evaluated, and calculate the static capability threat score and meteorological threat score of the UAV to be evaluated based on the reference values ​​of the static capability index and the environmental meteorological index; A deduction module, used to obtain the dynamic capability index value of the UAV to be evaluated, and to infer the dynamic capability threat score of the UAV to be evaluated based on a pre-trained neural network model; The second calculation module is used to obtain the distribution of each threat score of the drone to be evaluated, and calculate the weight value corresponding to the threat score based on the information entropy corresponding to each threat score; The third calculation module is used to calculate the comprehensive threat score of each of the drones to be evaluated based on the weight values ​​corresponding to each of the threat scores.

Images