Aerial target comprehensive threat assessment method and system

Abstract

The invention provides a comprehensive threat assessment method for an air target, and the method comprises the steps: S1, building a threat assessment model through setting a threat assessment factor; s2, determining parameters and calculation rules of the threat evaluation model; s3, calculating probability distribution of threat evaluation factor states according to the data acquired in real time; s4, inputting the probability distribution of the threat evaluation factor state into the threat evaluation model, evaluating a target threat, and obtaining a single threat value for a security object; s5, repeating the steps S3 to S4 to obtain single threat values of all surrounding objects, and integrating the single threat values into a target comprehensive threat value; and S6, judging whether a target to be evaluated exists or not, and if not, outputting a target comprehensive threat evaluation result sequence. The method is based on dynamic Bayesian network reasoning, a self-adaptive discrete method is adopted, comprehensive threat assessment of the aerial target is achieved, and the method can effectively adapt to application scenes with incomplete data.

Description

Technical Field

[0001] The present invention relates to the field of computer networks, and in particular, to a method and system for comprehensive threat assessment of aerial targets, and in particular, to a method and system for comprehensive threat assessment of aerial targets based on an adaptive dynamic Bayesian network. Background Art

[0002] Modern complex air defense operations are full of unknown and uncertain factors, which place high demands on commanders to grasp battlefield opportunities. The traditional method of relying mainly on the personal qualities and combat experience of commanders to judge the battlefield situation is no longer applicable to high-intensity and high-confrontation combat scenarios. The currently used target threat assessment methods can solve the target threat assessment in transparent battlefields with full information, but most of them cannot adapt to combat scenarios in semi-transparent and non-transparent situations, and cannot make comprehensive use of a large amount of historical experience and accumulated equipment data.

[0003] The publication number is CN114358558A, which discloses a method for assessing the threat of air strike targets based on commander experience learning, including: obtaining factors that affect the threat assessment of incoming targets and quantifying them; using the ring-comparison scoring method to obtain the subjective weight vector of each influencing factor, using the entropy method to obtain the objective weight vector, and then obtaining the comprehensive weight vector; using the ideal solution approximation method to obtain the relative approximation of the target to the ideal solution, and using the approximation to sort the target threats from large to small; by adjusting the weights that affect the threat degree of the incoming target, the threat degree of different targets is adjusted; when the weight adjustment still cannot meet the requirements of manual intervention, the threat degree of different targets is adjusted by adding a membership factor to the threat factor. Compared with the prior art, the present invention dynamically introduces the manual intervention factor into the threat assessment model, so that the air strike target threat assessment method has the ability to learn from the commander's experience. This patent does not integrate historical threat degrees and other comprehensive parameters, and the results obtained are relatively simple.

[0004] Therefore, a comprehensive threat assessment method for aerial targets based on an adaptive dynamic Bayesian network is needed. Summary of the invention

[0005] In view of the defects in the prior art, the object of the present invention is to provide a method and system for comprehensive threat assessment of aerial targets.

[0006] A method for comprehensive threat assessment of aerial targets provided by the present invention comprises:

[0007] Step S1: Establish a threat assessment model by setting threat assessment factors;

[0008] Step S2: Determine the parameters and calculation rules of the threat assessment model;

[0009] Step S3: Calculate the probability distribution of the threat assessment factor state based on the data acquired in real time;

[0010] Step S4: inputting the probability distribution of the threat assessment factor state into the threat assessment model, assessing the target threat, and obtaining a single threat value for a specific protection object;

[0011] Step S5: Repeat steps S3 to S4 to obtain the single threat values ​​of all the surrounded objects and integrate them into the target comprehensive threat value;

[0012] Step S6: Determine whether there is a target to be evaluated. If yes, update the target information to be evaluated and jump to step S3; if no, output the target comprehensive threat assessment result sequence.

[0013] Preferably, step S1 comprises:

[0014] Step S1.1: Determine the threat assessment factor from three perspectives: target, protected object, and historical threat assessment result; specifically, the target attributes include: target type L, target firepower strike capability E, target number N, route shortcut P, ​​radial arrival time T, and distance R; the protected object attributes include the protected object importance level S, and the historical threat assessment result attributes include the threat assessment result at the previous moment;

[0015] Step S1.2: Divide the threat assessment factor status; specifically, the target firepower strike capability, route shortcut, radial arrival time, distance, and threat assessment result of the previous moment are continuous threat assessment factors; target type, target number, and importance level of the protected object are discrete threat assessment factors;

[0016] Step S1.3: Based on the Bayesian network reasoning method, a threat assessment model is established, which adopts a one-layer reasoning structure and can generate target basic threat, target single threat value and target comprehensive threat value.

[0017] Preferably, step S2 comprises:

[0018] Step S2.1: Determine the prior probability of the continuous threat assessment factor node, including the prior probability of the target firepower strike capability, the prior probability of the route shortcut, the prior probability of the radial arrival time, the prior probability of the distance, and the prior probability of the threat assessment result at the previous moment;

[0019] Step S2.2: Set the transfer probability between the node and the parent node, including: the transfer probability of target threat and target firepower strike capability condition, the transfer probability of target threat and route shortcut condition, the transfer probability of target threat and radial arrival time condition, and the transfer probability of target threat and distance condition;

[0020] Step S2.3: Determine the conversion method between the target threat level probability distribution and the target basic threat value, and use the linear weighting method to convert the threat level probability distribution value into the target basic threat value; wherein the target threat level probability is divided into high, medium and low threat probabilities, and the weight values ​​are [0.6, 0.3, 0.1] respectively;

[0021] Step S2.4: Determine the weight values ​​corresponding to the discrete evaluation factors through the hierarchical analysis method, and use the linear weighting method for transformation.

[0022] Preferably, step S3 includes:

[0023] Step S3.1: Based on the data acquired in real time, for the discrete threat assessment factors, determine the single state to which they belong based on the observed value or inherent attribute, that is, obtain the corresponding weight value, including the weight value of the target type, the number of targets, and the importance level of the protected object;

[0024] Step S3.2: Based on the real-time acquired data, for the continuous threat assessment factors, the probability distribution of different states is calculated based on the adaptive normal cloud model method, including the probability distribution of target firepower strike capability, route shortcut, radial arrival time, and distance;

[0025] In step S3.3, when the continuous threat assessment factor parameter cannot be identified, the prior probability is used as a weight value instead.

[0026] Preferably, step S4 comprises:

[0027] Step S4.1: sequentially input the probability distribution of the continuous threat assessment factor into the threat assessment model, calculate the posterior probability distribution of the target threat at the current moment according to the Bayesian formula and the total probability formula, and use the linear weighted method to calculate the target basic threat value;

[0028] Step S4.2: The weight values ​​of the discrete threat assessment factors are evaluated in sequence, so that the target basic threat value is calculated to obtain a single threat value for a specific protection object.

[0029] Preferably, step S5 comprises:

[0030] Step S5.1: Repeat steps S3 to S4 according to the protected object sequence to calculate the single threat value of the target to all protected objects;

[0031] Step S5.2: Count the threat values ​​of the top three in the ranking, take the average value, and obtain the comprehensive threat value of the target;

[0032] The target comprehensive threat value calculation method is as follows:

[0033]

[0034] Among them, Te syn is the comprehensive threat value of the aerial target, Te' is the single target threat value of the target relative to a specific protected object, and i is sorted from large to small, with values ​​of 1, 2, and 3. When the number of protected objects is less than 3, the actual number is used for calculation.

[0035] Preferably, step S1.2 comprises:

[0036] The state division includes: target types include [L1, L2, L3, L4], representing missiles, attack aircraft, support aircraft and others respectively; target firepower strike capabilities include [E1, E2, E3], representing strong, medium and weak strike capabilities respectively; target numbers include [N1, N2, N3], representing single, double and multiple targets respectively; route shortcuts include [P1, P2, P3], representing small, medium and large route shortcuts respectively; radial arrival time includes [T1, T2, T3], representing short, medium and long time respectively; distance includes [R1, R2, R3], representing short, medium and long distance respectively; protection object importance level includes [S1, S2, S3], representing I, II and III importance levels respectively; target threats include [Te1, Te2, Te3], representing high, medium and low threat probability values ​​respectively.

[0037] Preferably, step S3.2 comprises:

[0038] The adaptive normal cloud model discretization method is specifically as follows:

[0039]

[0040] Among them, μ(x) is the membership of the observed value to the state, x is the observed value or indirect calculated value of the threat assessment factor, E x is the state expectation, E n is the state entropy, E x 、E n The value of can be set according to each state interval; specifically, the threat assessment factor is divided into three states, and the observation interval of the threat assessment factor is divided into four equal sections, E x The value is the middle node of each segment, E n The length of each segment is taken as the value to determine the state model of the three states; the observed values ​​or calculated values ​​are input into the three state models to obtain the membership degrees of the three states, which are μ1(x), μ2(x), and μ3(x) respectively. After normalization, the corresponding state probability distribution can be obtained to realize data discrete application.

[0041] Preferably, step S4 comprises:

[0042] The step S4.1 comprises:

[0043] The target basic threat value calculation method adopts a linear weighted method, specifically:

[0044] Te=Te1×0.6+Te2×0.3+Te3×0.1

[0045] Among them, Te is the basic threat value of the aerial target to the designated defense object, Te1, Te2, and Te3 are the probability values ​​of high threat, medium threat, and low threat respectively;

[0046] The step S4.2 comprises:

[0047] The target single threat value calculation method adopts a linear weighted method, specifically:

[0048] Te'=Te×k1×k2×k3

[0049] Among them, Te' is the single target threat value of the target relative to a specific protected object, k1, k2 and k3 are the weight values ​​of the discrete assessment factors of target type, target number and protected object importance level respectively.

[0050] An aerial target comprehensive threat assessment system provided by the present invention comprises:

[0051] Module M1: Establish a threat assessment model by setting threat assessment factors;

[0052] Module M2: Determine the parameters and calculation rules of the threat assessment model;

[0053] Module M3: Calculate the probability distribution of threat assessment factor status based on real-time acquired data;

[0054] Module M4: Input the probability distribution of the threat assessment factor state into the threat assessment model, assess the target threat, and obtain a single threat value for a specific protection object;

[0055] Module M5: Repeatedly trigger modules M3 to M4 to obtain the single threat value of all surrounded objects and integrate them into the target comprehensive threat value;

[0056] Module M6: Determine whether there is a target to be evaluated. If so, update the target information to be evaluated and jump to module M3; if not, output the target comprehensive threat assessment result sequence.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] 1. The present invention is based on dynamic Bayesian network reasoning, adopts an adaptive discrete method, integrates the attributes of targets, protected objects, and historical threat assessment results, covers target and protected object factors, realizes comprehensive threat assessment of aerial targets, and can effectively adapt to application scenarios with incomplete data;

[0059] 2. The present invention uses the normal cloud model method to achieve fuzzification of continuous observation variables to meet the input of the dynamic Bayesian network model;

[0060] 3. The present invention realizes the rational use and uniform application of observation data by distinguishing between discrete and continuous types of influence on evaluation factors based on variable characteristics. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0062] Figure 1 A flowchart of a method for comprehensive threat assessment of aerial targets according to the present invention;

[0063] Figure 2 A schematic diagram of a dynamic Bayesian network for comprehensive target threat assessment according to the present invention;

[0064] Figure 3 is a scene graph of embodiment 1 of the present invention;

[0065] Figure 4 The target threat variation curve for multiple protected objects over time in Example 1 of the present invention;

[0066] Figure 5 This is a curve showing how the target comprehensive threat changes over time in Example 1 of the present invention. DETAILED DESCRIPTION

[0067] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0068] Example 1

[0069] The present embodiment is described below in conjunction with the accompanying drawings. A method for comprehensive threat assessment of aerial targets includes the following steps:

[0070] Step 1: Establish a threat assessment model;

[0071] Step 1.1: Determine the target type, target firepower strike capability, target quantity, route shortcut, radial arrival time, distance, importance level of the protected object, and threat assessment result at the previous moment as threat assessment factors.

[0072] Step 1.2, according to the range of the evaluation factor, the status is rationally divided, including: target type includes [L1, L2, L3, L4], representing missiles, attack aircraft, support aircraft and others; target firepower strike capability includes [E1, E2, E3], representing strong, medium and weak strike capability; target number includes [N1, N2, N3], representing single, double and multiple targets; route shortcuts include [P1, P2, P3], representing small, medium and large route shortcuts; radial arrival time includes [T1, T2, T3], representing short, medium and long time; distance includes [R1, R2, R3], representing short, medium and long distance; the importance level of the protected object includes [S1, S2, S3], representing I, II and III importance levels; target threat includes [Te1, Te2, Te3], representing the probability value of high, medium and low threat;

[0073] The target firepower strike capability, route shortcut, radial arrival time, and distance are all continuous values. The value range corresponding to their specific states is determined by the equal-width discrete method. Taking the route shortcut as an example, assuming that the lower limit is 0km and the upper limit is 120km, the value ranges corresponding to the three states of small, medium, and large route shortcuts are [0,60], [30,90], and [60,120] respectively.

[0074] Step 1.3: Based on the Bayesian network reasoning method, a target threat assessment model based on a dynamic Bayesian network is established, such as Figure 2 As shown in the figure, a one-layer reasoning structure is used to generate the target basic threat through comprehensive reasoning based on the observation data of continuous factors such as target firepower strike capability, route shortcut, radial arrival time, distance, and threat assessment results at the previous moment;

[0075] Step 1.4: Based on the observed data of discrete evaluation factors such as target type, target quantity, and importance level of protected objects, as well as the corresponding weight distribution, linear weighted processing is performed to obtain a single threat value of the target against a single protected object;

[0076] Step 1.5: Superimpose the single threat values ​​of the target to multiple protected objects and take the average value to obtain the target comprehensive threat value.

[0077] Step 2: Establish basic model parameters

[0078] Step 2.1, determine the prior probability of the node, including the prior probability of the target firepower strike capability, the prior probability of the route shortcut, the prior probability of the radial arrival time, the prior probability of the distance, and the prior probability of the threat assessment result at the previous moment;

[0079] Step 2.2, determine the conditional transfer probability between the node and the parent node, including: the conditional transfer probability between target threat and target firepower strike capability, the conditional transfer probability between target threat and route shortcut, the conditional transfer probability between target threat and radial arrival time, and the conditional transfer probability between target threat and distance;

[0080] Step 2.3, determine the conversion method between the target threat probability distribution and the target threat value, and use the linear weighted method to convert the probability distribution value into the basic threat value. The weight value of the target threat level [high threat, medium threat, low threat] is [0.6, 0.3, 0.1]. Specifically, determine the conversion method between the target threat probability distribution and the target basic threat value. To ensure the comprehensiveness of the threat assessment, the probability distribution of high, medium and low threats is comprehensively considered, and the linear weighted method is used to convert the threat level probability distribution value into the target basic threat value. For the determination of the weight, the present invention adopts the expert evaluation method to comprehensively consider the influence of high, medium and low threat probabilities on the target basic threat value, that is: high threat> medium threat> low threat;

[0081] Step 2.4, determine the weight values ​​corresponding to discrete assessment factors such as target type, target quantity and importance level of protected objects, and use linear weighting method to convert the basic threat value into a single threat value; for determining the weights, the present invention uses hierarchical analysis method.

[0082] Step 2.5, determine the statistical strategy of a single threat value in the case of multiple protected objects. According to the threat sequence, the three largest threat values ​​can be selected as the reference of the comprehensive threat value;

[0083] Step 3: confirm the probability distribution of the evaluation factor status based on the data obtained in real time;

[0084] Step 3.1: For discrete evaluation factors, determine the single state to which they belong based on their observed values ​​or inherent attributes, that is, obtain the corresponding weight value, including: target type, target number, and importance level of the protected object;

[0085] Step 3.2: For continuous evaluation factors, based on the adaptive normal cloud model method, the probability distribution of different states is calculated, including: target firepower strike capability, route shortcut, radial arrival time, and distance;

[0086] In the adaptive normal cloud model discretization method described in step 3.2, specifically:

[0087]

[0088] Among them, μ(x) is the membership of the observed value to the state, x is the observed value or indirect calculated value of the evaluation factor, E x is the state expectation, E n is the state entropy, and the latter two values ​​are set according to each state interval. Taking the parameters in the above route shortcut as an example, E x The values ​​are 30, 60, 90, E n Specifically, the threat assessment factor is divided into three states according to the threat level, and the observation interval of the threat assessment factor is divided into four equal sections. x The values ​​are respectively taken as the intermediate nodes, E n The length of each segment is taken as the value to determine the state model of the three states. The observed values ​​or calculated values ​​are input into the three state models to obtain the membership of the three states, namely μ1(x), μ2(x), and μ3(x). After normalization, the corresponding state probability distribution can be obtained to realize data discrete application.

[0089] Step 3.3: When the target type or motion parameter cannot be identified or measured, the prior probability is used instead.

[0090] Step 4: Assess target threats;

[0091] Step 4.1, input the state probability distribution of the continuous evaluation factor into the dynamic Bayesian network inference model in turn, and after conversion, calculate the posterior probability distribution of the target threat at the current moment according to the Bayesian formula and the total probability formula, and use the linear weighted method to calculate the target basic threat value;

[0092] The target basic threat value calculation method adopts a linear weighted method, specifically:

[0093] Te=Te1×0.6+Te2×0.3+Te3×0.1

[0094] Among them, Te is the basic threat value of the aerial target to the designated defense object, Te1, Te2, and Te3 are the probability values ​​of high threat, medium threat, and low threat respectively;

[0095] Step 4.2, consider the weights of discrete evaluation factors in turn, so that the target basic threat value is calculated, and a single threat value of the target for a certain protected object is obtained by using a linear weighted method;

[0096] The target single threat value calculation method adopts a linear weighted method, specifically:

[0097] Te'=Te×k1×k2×k3

[0098] Among them, Te' is the single target threat value of the target relative to a specific protected object, k1, k2 and k3 are the weight values ​​of the discrete assessment factors of target type, target number and protected object importance level respectively.

[0099] Step 5, calculate the target comprehensive threat;

[0100] Step 5.1, repeat steps 3 and 4 according to the protected object sequence, and calculate the single threat value of the target to all protected objects;

[0101] Step 5.2: Count and sort the top three threat values, take the average value, and get the comprehensive threat value of the target.

[0102] The target comprehensive threat value calculation method is as follows:

[0103]

[0104] Among them, Te syn is the comprehensive threat value of the aerial target, Te' is the single target threat value of the target relative to a specific protected object, and they are sorted from large to small, with i taking values ​​of 1, 2, and 3; Te i ′ is the threat value of the ith single target. When the number of protected objects is less than 3, the actual number is used for calculation.

[0105] Step 6: Output the comprehensive threat assessment result of the target and check whether there are any targets to be assessed in this round. If so, update the target information to be assessed and jump to step 3. Otherwise, output the comprehensive threat sequence of the target.

[0106] Example 1

[0107] Set up three defense targets: the airport, the position and the bridge. The deployment positions are [121.4212,31.3208,0], [121.0683,31.0877,0] and [121.3435,30.7729,0], and the importance levels are I, II and III respectively;

[0108] Three targets, a cruise missile, a fighter jet and an electronic jammer, are set. Their initial positions are [125.2915, 31.4742, 3000], [125.3356, 31.4783, 3000] and [125.3751, 31.4816, 3000] respectively. They maneuver in the same direction at a speed of 300 m / s and an altitude of 3000 m. The detailed information is shown in Table 1. The scenario situation is as follows Figure 3 shown.

[0109] Table 1

[0110] Target Firepower strike capability / km Speed ​​m / s type quantity 1 600 300 missile 1 2 300 300 Attack aircraft 1 3 160 300 Support aircraft 1

[0111] The method of the present invention is used to carry out threat assessment on the target. The change of the single threat value of the cruise missile to all the protected objects over time is as follows: Figure 4 shown.

[0112] The method of the present invention is used to conduct comprehensive threat assessment on multiple targets. The comprehensive threat value of each target changes over time as shown in the following figure: Figure 5 shown.

[0113] In summary, it can be concluded that the present invention can achieve comprehensive threat assessment of different targets in different situations, realize threat assessment of battlefield situations, and significantly improve battlefield analysis and cognitive capabilities.

[0114] The present invention also provides an aerial target comprehensive threat assessment system, which can be implemented by executing the process steps of the aerial target comprehensive threat assessment method, that is, those skilled in the art can understand the aerial target comprehensive threat assessment method as a preferred implementation of the aerial target comprehensive threat assessment system.

[0115] A comprehensive threat assessment system for aerial targets comprises: a module M1: establishing a threat assessment model by setting threat assessment factors; a module M2: determining parameters and calculation rules of the threat assessment model; a module M3: calculating the probability distribution of threat assessment factor states according to data acquired in real time; a module M4: inputting the probability distribution of threat assessment factor states into the threat assessment model, assessing target threats, and obtaining a single threat value for a specific protected object; a module M5: repeatedly triggering modules M3 to M4, obtaining single threat values ​​for all surrounded objects, and integrating them into a comprehensive threat value for the target; a module M6: judging whether there is a target to be assessed, and if so, updating the target information to be assessed and jumping to execute module M3; if not, outputting a sequence of comprehensive threat assessment results for the target.

[0116] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A method for comprehensive threat assessment of aerial targets, characterized in that: include: Step S1: Establish a threat assessment model by setting threat assessment factors; Step S2: Determine the parameters and calculation rules of the threat assessment model; Step S3: Calculate the probability distribution of the threat assessment factor state based on the data acquired in real time; Step S4: input the probability distribution of the threat assessment factor state into the threat assessment model, assess the target threat, and obtain a single threat value for a specific protection object; Step S5: Repeat steps S3 to S4 to obtain the single threat values ​​of all the surrounded objects and integrate them into the target comprehensive threat value; Step S6: Determine whether there is a target to be evaluated. If yes, update the target information to be evaluated and jump to step S3; If not, output the target comprehensive threat assessment result sequence.

2. The method for comprehensive threat assessment of aerial targets according to claim 1, characterized in that: The step S1 comprises: Step S1.1: Determine the threat assessment factor from three perspectives: target, protected object, and historical threat assessment result; specifically, the target attributes include: target type L, target firepower strike capability E, target number N, route shortcut P, ​​radial arrival time T, and distance R; the protected object attributes include the protected object importance level S, and the historical threat assessment result attributes include the threat assessment result at the previous moment; Step S1.2: Divide the threat assessment factor status; specifically, the target firepower strike capability, route shortcut, radial arrival time, distance, and threat assessment result of the previous moment are continuous threat assessment factors; target type, target number, and importance level of the protected object are discrete threat assessment factors; Step S1.3: Based on the Bayesian network reasoning method, a threat assessment model is established, which adopts a one-layer reasoning structure and can generate target basic threat, target single threat value and target comprehensive threat value.

3. The method for comprehensive threat assessment of aerial targets according to claim 2, characterized in that: The step S2 comprises: Step S2.1: Determine the prior probability of the continuous threat assessment factor node, including the prior probability of the target firepower strike capability, the prior probability of the route shortcut, the prior probability of the radial arrival time, the prior probability of the distance, and the prior probability of the threat assessment result at the previous moment; Step S2.2: Set the transfer probability between the node and the parent node, including: the transfer probability of target threat and target firepower strike capability condition, the transfer probability of target threat and route shortcut condition, the transfer probability of target threat and radial arrival time condition, and the transfer probability of target threat and distance condition; Step S2.3: Determine the conversion method between the target threat level probability distribution and the target basic threat value, and use the linear weighting method to convert the threat level probability distribution value into the target basic threat value; wherein the target threat level probability is divided into high, medium and low threat probabilities, and the weight values ​​are [0.6, 0.3, 0.1] respectively; Step S2.4: Determine the weight values ​​corresponding to the discrete evaluation factors through the hierarchical analysis method, and use the linear weighting method for transformation.

4. The method for comprehensive threat assessment of aerial targets according to claim 3, characterized in that: The step S3 comprises: Step S3.1: Based on the data acquired in real time, for the discrete threat assessment factors, determine the single state to which they belong based on the observed value or inherent attribute, that is, obtain the corresponding weight value, including the weight value of the target type, the number of targets, and the importance level of the protected object; Step S3.2: Based on the real-time acquired data, for the continuous threat assessment factors, the probability distribution of different states is calculated based on the adaptive normal cloud model method, including the probability distribution of target firepower strike capability, route shortcut, radial arrival time, and distance; In step S3.3, when the continuous threat assessment factor parameter cannot be identified, the prior probability is used as a weight value instead.

5. The method for comprehensive threat assessment of aerial targets according to claim 3, characterized in that: The step S4 comprises: Step S4.1: Input the probability distribution of the continuous threat assessment factor into the threat assessment model in turn, and after conversion, calculate the posterior probability distribution of the target threat at the current moment according to the Bayesian formula and the total probability formula, and use the linear weighted method to calculate the target basic threat value; Step S4.2: The weight values ​​of the discrete threat assessment factors are evaluated in sequence, so that the target basic threat value is calculated to obtain a single threat value for a specific protection object.

6. The method for comprehensive threat assessment of aerial targets according to claim 5, characterized in that: The step S5 comprises: Step S5.1: Repeat steps S3 to S4 according to the protected object sequence to calculate the single threat value of the target to all protected objects; Step S5.2: Count the threat values ​​of the top three in the ranking, take the average value, and obtain the comprehensive threat value of the target; The target comprehensive threat value calculation method is as follows: Among them, Te syn is the comprehensive threat value of the aerial target, Te′ is the single target threat value of the target relative to a specific protected object, and they are sorted from large to small, with i taking values ​​of 1, 2, and 3; Te i ′ is the threat value of the ith single target. When the number of protected objects is less than 3, the actual number is used for calculation.

7. The method for comprehensive threat assessment of aerial targets according to claim 2, characterized in that: The step S1.2 comprises: The state division includes: target types include [L1, L2, L3, L4], representing missiles, attack aircraft, support aircraft and others respectively; target firepower strike capabilities include [E1, E2, E3], representing strong, medium and weak strike capabilities respectively; target numbers include [N1, N2, N3], representing single, double and multiple targets respectively; route shortcuts include [P1, P2, P3], representing small, medium and large route shortcuts respectively; radial arrival time includes [T1, T2, T3], representing short, medium and long time respectively; distance includes [R1, R2, R3], representing close, medium and long distance respectively; the importance level of the protected object includes [S1, S2, S3], representing I, II and III importance levels respectively; target threats include [Te1, Te2, Te3], representing the probability values ​​of high, medium and low threats respectively.

8. The method for comprehensive threat assessment of aerial targets according to claim 4, characterized in that: The step S3.2 comprises: The adaptive normal cloud model discretization method is specifically as follows: Among them, μ(x) is the membership of the observed value to the state, x is the observed value or indirect calculated value of the threat assessment factor, E x is the state expectation, E n is the state entropy, E x 、E n The value of can be set according to each state interval; specifically, the threat assessment factor is divided into three states, and the observation interval of the threat assessment factor is divided into four equal sections, E x The value is the middle node of each segment, E n The length of each segment is taken as the value to determine the state model of the three states; the observed values ​​or calculated values ​​are input into the three state models to obtain the membership degrees of the three states, which are μ1(x), μ2(x), and μ3(x) respectively. After normalization, the corresponding state probability distribution can be obtained to realize data discrete application.

9. The method for comprehensive threat assessment of aerial targets according to claim 5, characterized in that: The step S4 comprises: The step S4.1 comprises: The target basic threat value calculation method adopts a linear weighted method, specifically: Te=Te1×0.6+Te2×0.3+Te3×0.1 Among them, Te is the basic threat value of the aerial target to the designated defense object, Te1, Te2, and Te3 are the probability values ​​of high threat, medium threat, and low threat respectively; The step S4.2 comprises: The target single threat value calculation method adopts a linear weighted method, specifically: Te'=Te×k1×k2×k3 Among them, Te' is the single target threat value of the target relative to a specific protected object, k1, k2 and k3 are the weight values ​​of the discrete assessment factors of target type, target number and protected object importance level respectively.

10. An aerial target integrated threat assessment system, used to execute the aerial target integrated threat assessment method according to claim 1, characterized in that: include: Module M1: Establish a threat assessment model by setting threat assessment factors; Module M2: Determine the parameters and calculation rules of the threat assessment model; Module M3: Calculate the probability distribution of threat assessment factor status based on real-time acquired data; Module M4: Input the probability distribution of the threat assessment factor state into the threat assessment model, assess the target threat, and obtain a single threat value for a specific protection object; Module M5: Repeatedly trigger modules M3 to M4 to obtain the single threat value of all surrounded objects and integrate them into the target comprehensive threat value; Module M6: Determine whether there is a target to be evaluated. If so, update the target information to be evaluated and jump to module M3; if not, output the target comprehensive threat assessment result sequence.

Images