Damage calculation method and device based on Bayesian network

Abstract

The invention relates to the technical field of machine learning, and provides a damage calculation method and device based on a Bayesian network. The method comprises the following steps: determining a network structure by taking various factors of our asset damage caused by an opposite target as event nodes; determining a prior probability of each event node according to prior knowledge of experts; and establishing a Bayesian network model according to the prior probability of each event node and the network structure, and determining the damage probability of our assets according to the Bayesian network model. According to the method, the Bayesian estimation network based on the target capability and the target intention is constructed, accurate estimation of our asset damage in a high dynamic environment is achieved, and the problem that our asset damage is difficult to calculate is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of machine learning, and in particular to a damage calculation method and device based on a Bayesian network. Background Art

[0002] Damage estimation refers to the process of estimating the damage that a target may cause to our assets. It is a must-do task before making subsequent plans. Its purpose is to estimate the loss effectiveness based on the target threat and target intention, and to know in advance the remaining effectiveness of the target after being damaged, so as to provide a reference for subsequent plans. For example, in the field of drone applications, such as drone simulated confrontation and drone firefighting, pre-estimating the damage caused by the target can effectively make subsequent plans. In various battle games, damage estimation can also provide effective reference for game users, thereby improving the gaming experience.

[0003] However, in actual use, it is difficult to obtain complete information about the target, so damage estimation based on observed situation information alone is often not accurate enough; in addition, the damage caused by the target may include many very complex aspects such as the coupling relationship between the main structure, main power equipment and electronic equipment. These seriously affect the damage estimation and the smooth implementation of subsequent plans.

[0004] In view of this, how to overcome the defects of the existing technology and quickly and accurately perform damage estimation in a high dynamic environment is a problem to be solved in this technical field. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a damage calculation method and device based on a Bayesian network.

[0006] The present invention adopts the following technical solution:

[0007] In a first aspect, the present invention provides a damage calculation method based on a Bayesian network, comprising:

[0008] Using various factors that may cause damage to our assets as event nodes, we can determine the network structure.

[0009] Determine the prior probability of each event node based on the prior knowledge of experts;

[0010] According to the prior probability of each event node and the network structure, a Bayesian network model is established to determine the damage probability of our assets according to the Bayesian network model.

[0011] Preferably, the event node includes one or more of the opponent target type, opponent target damage range, opponent target position, opponent target speed, opponent target distance, distance between the opponent target and the nearest approach point, opponent target estimated arrival time, opponent target threat level, opponent target intention and damage probability.

[0012] Preferably, the network structure is as follows: the opponent target type is used as a prerequisite for determining the opponent target damage range and the opponent target speed, the opponent target position is used as a prerequisite for calculating the opponent target distance and the distance between the opponent target and the nearest approach point, the opponent target distance and the opponent target damage range are used as prerequisites for inferring the opponent target threat level, the distance between the opponent target and the nearest approach point and the opponent target speed are used as prerequisites for calculating the opponent target estimated arrival time, the opponent target estimated arrival time is used as a prerequisite for judging the opponent target intention, and the opponent target threat level and the opponent target intention are used as prerequisites for determining the damage probability.

[0013] Preferably, the determining the prior probability of each event node based on the prior knowledge of the expert specifically includes:

[0014] Determine the prior probability P(EffectRange|TargetType) of the damage range of the opponent target corresponding to different opponent target types;

[0015] Determine the prior probability P (Threat|EffectRange,Distance) of the enemy target threat level corresponding to different enemy target damage ranges and different enemy target distances;

[0016] Determine the prior probability P(Intent|TCPA) of the other party's intention corresponding to different expected arrival times of the other party;

[0017] Determine the prior probability P(Damage|Threat,Intent) of the damage probability corresponding to different threat levels of the opponent target and different intentions of the opponent target.

[0018] Preferably, determining the damage probability of our assets according to the Bayesian network model specifically includes:

[0019] Determine the enemy target's location, enemy target type, enemy target speed, and our asset's location;

[0020] Calculate the distance to the other party's target based on the other party's target location and our asset location;

[0021] Determine the location of the nearest approach point based on the location of the other party's target, the type of the other party's target, and the location of our assets; calculate the estimated arrival time of the other party's target based on the location of the nearest approach point;

[0022] The opponent's target type, opponent's target speed, opponent's target distance, distance between the opponent's target and the nearest approach point, and various prior probabilities are substituted into the Bayesian network model to calculate the damage probability of our assets.

[0023] Preferably, the calculating the distance to the other party's target according to the other party's target position and the position of our asset specifically includes:

[0024] Calculate the distance to the other target

[0025] Among them, (x zz ,y zz , z zz ) is the coordinate of our asset location, (x target ,y target , z target ) are the coordinates of the other party’s target position.

[0026] Preferably, determining the position of the nearest approach point according to the other party's target position, the other party's target type and the position of our asset specifically includes:

[0027] The position of the closest approach point is calculated using a combined equation; the combined equation includes a first equation, a second equation, a third equation, a fourth equation, a fifth equation and a sixth equation;

[0028] The first equation is (x cpa -x zz )·(x cpa -x target )=0;

[0029] The second equation is (y cpa -y zz )·(y cpa -y target )=0;

[0030] The third equation is (z cpa -z zz )·(z cpa -z target )=0;

[0031] The fourth equation is

[0032] The fifth equation is

[0033] The sixth equation is

[0034] Among them, (x cpa ,y cpa , zcpa ) is the coordinate of the nearest approach point, (x zz ,y zz , z zz ) is the coordinate of our asset location, (x target ,y target , z target ) is the coordinate of the other party’s target position, It is the vector representation of the opponent's target speed.

[0035] Preferably, the calculating the estimated arrival time of the other target according to the position of the closest approach point specifically includes:

[0036] The distance between the target and the closest approach point is calculated according to the position of the closest approach point.

[0037] Calculate the target's estimated arrival time

[0038] Among them, (x cpa ,y cpa , z cpa ) is the coordinate of the nearest approach point, (x target ,y target , z target ) is the coordinate of the other party’s target position,

[0039] In a second aspect, the present invention further provides a damage calculation device based on a Bayesian network, which is used to implement the damage calculation method based on a Bayesian network described in the first aspect, and the device comprises:

[0040] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the processor to execute the Bayesian network-based damage calculation method described in the first aspect.

[0041] In a third aspect, the present invention further provides a non-volatile computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are executed by one or more processors to complete the method described in the first aspect.

[0042] In a fourth aspect, a chip is provided, comprising: a processor and an interface, for calling and running a computer program stored in the memory from a memory, and executing the method of the first aspect.

[0043] In a fifth aspect, a computer program product comprising instructions is provided, which, when executed on a computer or a processor, causes the computer or the processor to execute the method of the first aspect.

[0044] The present invention realizes accurate estimation of damage to our assets in a highly dynamic environment based on a Bayesian estimation network of target capability and target intention, and solves the problem of difficulty in calculating damage to our assets. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0046] Figure 1 It is a flowchart of a damage calculation method based on a Bayesian network provided by an embodiment of the present invention;

[0047] Figure 2 It is a schematic diagram of the cause-effect relationship of corresponding event nodes in a damage calculation method based on a Bayesian network provided by an embodiment of the present invention;

[0048] Figure 3 It is a schematic diagram of the cause-effect relationship of corresponding event nodes in a damage calculation method based on a Bayesian network provided by an embodiment of the present invention;

[0049] Figure 4 It is a schematic diagram of the cause-effect relationship of corresponding event nodes in a damage calculation method based on a Bayesian network provided by an embodiment of the present invention;

[0050] Figure 5 It is a schematic diagram of the cause-effect relationship of corresponding event nodes in a damage calculation method based on a Bayesian network provided by an embodiment of the present invention;

[0051] Figure 6 Schematic diagram of the network structure of a Bayesian network in a damage calculation method based on a Bayesian network provided in an embodiment of the present invention;

[0052] Figure 7 is a schematic diagram of a priori probability in a damage calculation method based on a Bayesian network provided in an embodiment of the present invention;

[0053] Figure 8 is a schematic diagram of a priori probability in a damage calculation method based on a Bayesian network provided in an embodiment of the present invention;

[0054] Fig. 9is a schematic diagram of a priori probability in a damage calculation method based on a Bayesian network provided in an embodiment of the present invention;

[0055] Fig.10 is a schematic diagram of a priori probability in a damage calculation method based on a Bayesian network provided in an embodiment of the present invention;

[0056] Fig.11 is a schematic diagram of a Bayesian network in a damage calculation method based on a Bayesian network provided in an embodiment of the present invention;

[0057] Fig.12 It is a flowchart of a damage calculation method based on a Bayesian network provided by an embodiment of the present invention;

[0058] Fig.13 is a schematic diagram of the closest approach point in a damage calculation method based on a Bayesian network provided by an embodiment of the present invention;

[0059] Fig.14 It is a flowchart of a damage calculation method based on a Bayesian network provided by an embodiment of the present invention;

[0060] Fig.15 1 is a schematic diagram of a network structure of a comprehensive Bayesian network in a damage calculation method based on a Bayesian network provided in an embodiment of the present invention;

[0061] Fig.16 It is a flowchart of a damage calculation method based on a Bayesian network provided by an embodiment of the present invention;

[0062] Fig.17 It is a schematic diagram of the architecture of a damage calculation device based on a Bayesian network provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0064] Unless the context requires otherwise, throughout the specification and claims, the term "including" is to be interpreted as open inclusion, that is, "including, but not limited to". In the description of the specification, the terms "one embodiment", "some embodiments", "exemplary embodiments", "examples", "specific examples" or "some examples" and the like are intended to indicate that specific features, structures, materials or characteristics associated with the embodiment or example are included in at least one embodiment or example of the present disclosure. The schematic representation of the above terms does not necessarily refer to the same embodiment or example. In addition, the specific features, structures, materials or characteristics may be included in any one or more embodiments or examples in any appropriate manner, that is, although they may be carried in the embodiments or examples of the above terms due to reasons such as the order and position of appearance, it is not limited to that they can be carried in combination by one embodiment or example.

[0065] In the description of the present invention, the terms "first" and "second" are used only for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present disclosure, unless otherwise specified, the meaning of "multiple" is two or more. In addition, for example, the same type of nouns may be described as two independent individuals by adding "A" and "B" at the end. In this case, the corresponding features defined as "A" and "B" are only used to distinguish the same type of individuals for description purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features.

[0066] In the description of the present invention, the expression "A and / or B" (where A and B are used to formally represent specific characteristic contents) will be involved, and the corresponding expressions include the following three combinations: only A, only B, and a combination of A and B.

[0067] As used herein, "about," "substantially," or "approximately" includes the stated value and an average value that is within an acceptable range of deviation from the particular value as determined by one of ordinary skill in the art taking into account the measurements in question and the errors associated with the measurement of the particular quantity (i.e., the limitations of the measurement system).

[0068] In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0069] Embodiment 1:

[0070] Damage estimation has always been a research hotspot, and many methods have been proposed, including numerical simulation methods based on physical models and prediction methods based on machine learning.

[0071] The numerical simulation method based on physical models requires the establishment of complex physical models to describe the relationship between the target structure and the load response. This involves a large number of parameters and assumptions, and the calculation complexity is high. In addition, the establishment and verification of physical models requires a large amount of experimental data, which is often difficult to obtain. Therefore, the method based on physical models has certain limitations in practical applications.

[0072] Prediction methods based on machine learning usually require a large amount of training data to build models. However, it is usually very difficult to obtain target damage data. Therefore, insufficient training data may lead to unstable performance or poor generalization of the model. In addition, machine learning methods usually have difficulty in interpreting the prediction results of the model, which may be limited in areas such as target design and maintenance. In order to solve these problems, Example 1 of the present invention provides a damage calculation method based on a Bayesian network, such as Figure 1 As shown, including:

[0073] In step 101, the network structure is determined by taking various factors that cause damage to our assets caused by the opponent's target as event nodes; the network structure can be understood as a network used to express the causal relationship between various event nodes. In an actual application scenario, our assets can be objects with defensive capabilities, and the opponent's targets can be objects with confrontation capabilities or the ability to affect our assets. It should be noted here that our assets and opponent's targets are only a description used to distinguish the two opposing parties, and do not represent that the executor of this embodiment in actual use is our party. In actual use, taking Party A as the defender and Party B as the opponent as an example, Party A is our asset and Party B is the opponent's target.

[0074] In step 102, the prior probability of each event node is determined based on the prior knowledge of the expert.

[0075] In step 103, a Bayesian network model is established according to the prior probability of each event node and the network structure, so as to determine the damage probability of our assets according to the Bayesian network model.

[0076] Wherein, the damage probability of our assets determined according to the Bayesian network model can be calculated using Netica analysis software.

[0077] This embodiment is based on the Bayesian estimation network of target capability and target intention, which realizes the accurate estimation of our asset damage in a highly dynamic environment and solves the problem of difficulty in calculating our asset damage. In addition, the Bayesian network method has significant advantages in processing uncertain information and can evaluate the damage effects of complex targets. The characteristics of this method include the ability to process a large amount of uncertain and fuzzy information, and it is suitable for the evaluation of complex systems. This embodiment is constructed based on the Bayesian network of target threat and target intention and the simulation calculation model based on Netica, and the target damage can be quickly and accurately estimated based on the prior probability and the input target information.

[0078] In an actual application scenario, the event node includes one or more of the opponent target type, opponent target damage range, opponent target position, opponent target speed, opponent target distance, distance between the opponent target and the nearest approach point, opponent target estimated arrival time, opponent target threat level, opponent target intention and damage probability.

[0079] Among them, after analysis, it is found that there are Figure 2 , Figure 3 , Figure 4 and Figure 5 The causal relationship shown, such as Figure 2 As shown, TargetType represents the target type of the other party, and the target position L of the other party can be obtained according to the target type of the other party. target (exist Figure 2 Location in the middle), the opponent's target speed V target (exist Figure 2 Speed) and the opponent's target damage range EffectRange, that is, TargetType→{L target ,V target ,EffectRange}.

[0080] like Figure 3 As shown, L zz represents the position of our asset, then the distance to the other party's target (i.e. the distance between the other party's target and our asset) can be obtained from the position of our asset and the position of the other party's target. target ,L zz ), then the threat degree Threat of the opponent's target can be determined by the opponent's target damage range EffectRange and the opponent's target distance Distance, that is, (EffectRange, Distance)→Threat.

[0081] like Figure 4As shown in the figure, CPA represents the closest approach point, and the distance between the opponent's target and the closest approach point is expressed as range from CPA. The distance between the opponent's target and the closest approach point can be obtained by analyzing the opponent's target position and our asset position. The opponent's target estimated arrival time TCPA can be expressed as (CPA, V target )→TBH, the other party’s target estimated arrival time can be understood as the time when the other party’s target is expected to arrive at the nearest approach point, and the target intent Intent can be expressed as TCPA→Intent.

[0082] like Figure 5 As shown in the figure, the damage probability Damage of our assets can be expressed as (Threat, Intent)→Damage.

[0083] according to Figure 2 , Figure 3 , Figure 4 and Figure 5 The relationship between the event nodes shown in the figure is determined as follows: Figure 6 The network structure shown in the figure is as follows: the opponent target type is used as a prerequisite for determining the opponent target damage range and the opponent target speed, the opponent target position is used as a prerequisite for calculating the opponent target distance and the distance between the opponent target and the nearest approach point, the opponent target distance and the opponent target damage range are used as prerequisites for inferring the opponent target threat level, the distance between the opponent target and the nearest approach point and the opponent target speed are used as prerequisites for calculating the opponent target estimated arrival time, the opponent target estimated arrival time is used as a prerequisite for judging the opponent target intention, and the opponent target threat level and the opponent target intention are used as prerequisites for determining the damage probability.

[0084] In an optional implementation, determining the prior probability of each event node based on the prior knowledge of the expert specifically includes:

[0085] Determine the prior probability P(EffectRange|TargetType) of the opponent target damage range corresponding to different opponent target types; in an actual application scenario, the opponent target damage range is divided into three types: short, medium, and long. The prior probability P(EffectRange|TargetType) of the opponent target damage range corresponding to each type of opponent target is as follows: Figure 7 As shown, Figure 7 For example, there are three types of opponent targets, namely, type one targets, type two targets, and type three targets.

[0086] Determine the prior probability P(Threat|EffectRange,Distance) of the threat level of the opponent target corresponding to different opponent target damage ranges at different opponent target distances; in an actual application scenario, the opponent target distance is divided into three types: close, medium, and far, the opponent target threat level is divided into low, medium, and high, and the opponent target damage range is divided into three types: short, medium, and long, and the prior probability P(Threat|EffectRange,Distance) of the opponent target threat level of each opponent target damage range at the corresponding target distance is obtained as follows: Figure 8 shown.

[0087] Determine the prior probability P(Intent|TCPA) of the other party's intention corresponding to different expected arrival times of the other party; in an actual application scenario, the other party's intention is divided into three types: low, medium, and high, and the prior probability P(Intent|TCPA) of the other party's intention corresponding to the expected arrival time of the other party within the corresponding range is obtained as follows: Fig. 9 shown.

[0088] Determine the prior probability P(Damage|Threat,Intent) of the damage probability corresponding to different threat levels of the other party when the other party has different intentions. In a practical application scenario, the damage probability is divided into two categories: damage (i.e. true) and no damage (i.e. false). The prior probability P(Damage|Threat,Intent) of the damage probability corresponding to each target threat level when the other party has different intentions is as follows: Fig.10 shown.

[0089] According to the above-mentioned prior probabilities and network structure, a Bayesian network model is established as follows: Fig.11 shown.

[0090] In an optional embodiment, the damage probability of our assets is determined according to the Bayesian network model, such as Fig.12 As shown, specifically including:

[0091] In step 201, the position of the opponent target, the type of the opponent target, the speed of the opponent target and the position of our asset are determined.

[0092] In step 202, the distance to the other party's target is calculated based on the other party's target position and the position of our asset.

[0093] In step 203, the position of the nearest approach point is determined according to the position of the other party's target, the type of the other party's target and the position of our asset; and the estimated arrival time of the other party's target is calculated according to the position of the nearest approach point.

[0094] In step 204, the opponent target type, opponent target speed, opponent target distance, distance between the opponent target and the nearest approach point, and various prior probabilities are substituted into the Bayesian network model to calculate the damage probability of our assets.

[0095] The above steps 201 to 204 can be understood as Figure 6 The process of solving the problem in the network structure shown in the figure is to determine the target type, target speed, target location and asset location of the other party, and then calculate the target distance and estimated arrival time of the other party. After obtaining the values ​​of the above event nodes, according to Figure 7 , Figure 8 , Fig. 9 and Fig.10 The prior probability shown can determine the threat level and intention of the opponent's target, and further determine the probability of damage to our assets.

[0096] The calculation of the distance to the other party's target based on the other party's target position and the position of our asset specifically includes: calculating the distance to the other party's target Among them, (x zz ,y zz , z zz ) is the coordinate of our asset location, (x target ,y target , z target ) are the coordinates of the other party’s target position.

[0097] In actual use, the distance to the other party is converted into three values: close, medium, and far. In a specific application, P (Distance = close) is D (L target ,L zz )≤2; P(Distance=medium) is 2 <D(L target ,L zz )≤10; P(Distance=far) is 10 <D(L target ,L zz )≤40.

[0098] The closest approach point is the point where the other party's target is closest to our asset in the direction of travel, such as Fig.13 As shown, the position of the nearest approach point is determined according to the other party's target position, the other party's target type and the position of our asset, specifically including:

[0099] The position of the closest approach point is calculated using a joint equation; the joint equation includes a first equation, a second equation, a third equation, a fourth equation, a fifth equation and a sixth equation.

[0100] The first equation is (x cpa -x zz )·(x cpa -x target )=0; the second equation is (y cpa -y zz )·(y cpa -y target )=0;the third equation is (z cpa -z zz )·(z cpa -z target )=0.

[0101] The fourth equation is The fifth equation is The sixth equation is

[0102] Among them, (x cpa ,y cpa , z cpa ) is the coordinate of the nearest approach point, (x zz ,y zz , z zz ) is the coordinate of our asset location, (x target ,y target , z target ) is the coordinate of the other party’s target position, is the vector representation of the opponent's target speed. Solving the joint equation obtains the position (i.e., coordinates) of the nearest approach point.

[0103] The calculating the estimated arrival time of the other target according to the position of the closest approach point specifically includes:

[0104] The distance between the target and the closest approach point is calculated according to the position of the closest approach point. Calculate the target's estimated arrival time Among them, (x cpa ,y cpa , z cpa ) is the coordinate of the nearest approach point, (x target ,y target , z target ) is the coordinate of the other party’s target position,

[0105] Considering that in actual use, our side may also have air assets for confrontation, which are mainly used to escort our ground assets. In this case, the calculation of the damage probability of our ground assets will be more complicated. In order to cope with this situation, this embodiment further provides a preferred implementation method based on the above implementation method, such as Fig.14 As shown, specifically including:

[0106] In step 301, taking our ground assets as the defender and the opponent's target as the adversary, a first Bayesian network model between the opponent's target and our ground assets is established.

[0107] In step 302, a second Bayesian network model between our aerial assets and the opponent's targets is established, with our aerial assets as the adversary and the opponent's targets as the defender.

[0108] In step 303, a comprehensive Bayesian network model is established with the final result event node of the first Bayesian network model and the final result event node of the second Bayesian network model as the premise and the damage probability of our ground assets as the result. The structure of the comprehensive Bayesian network model is as follows: Fig.15 shown.

[0109] In step 304, based on the prior knowledge of the expert, the prior probability of each event node in the comprehensive Bayesian network model is determined.

[0110] In step 305, the first probability of damage to our ground assets is determined according to the first Bayesian network model; the second probability of damage to the opponent's target is determined according to the second Bayesian network model; and the final probability of damage to our ground assets is determined according to the prior probability of each event node in the comprehensive Bayesian network model, the first probability and the second probability. The determination of the prior probability of each event node in the comprehensive Bayesian network model is to determine the probability of damage to our ground assets corresponding to the first probability and the second probability.

[0111] In an actual application scenario, the damage probability of our assets corresponding to the first probability and the second probability conditions is specifically expressed as follows:

[0112] Damage=Damage_1×(1-k×Damage_2)

[0113] Wherein, Damage is the damage probability of our ground assets, Damage_1 is the first probability, Damage_2 is the second probability, k is a preset parameter obtained by technical personnel in this field based on empirical analysis, and k is greater than 0 and k is less than or equal to 1.

[0114] The establishment of the first Bayesian network model and the second Bayesian network model is based on the same concept as the above-mentioned implementation modes, and will not be repeated here.

[0115] Taking the first Bayesian network model as an example, the establishment of the first Bayesian network model specifically includes: taking the various factors of the opponent's target causing damage to our ground assets as event nodes to determine the network structure; determining the prior probability of each event node based on the prior knowledge of experts; and establishing the first Bayesian network model based on the prior probability of each event node and the network structure. Similarly, the second Bayesian network model is established based on the network structure determined by taking the various factors of our air assets causing damage to the opponent's target as event nodes.

[0116] Embodiment 2:

[0117] The present invention is based on the method described in Example 1, combined with specific application scenarios, and uses technical descriptions in related scenarios to illustrate the implementation process of the present invention in characteristic scenarios.

[0118] In this embodiment, our assets are taken as water targets, and air targets are taken as the other party's targets. The Netica analysis software is used to calculate the damage of water targets as an example of a practical application scenario. The embodiment of the present invention provides a damage calculation method based on a Bayesian network, such as Fig.16 As shown, specifically including:

[0119] In step 401, event nodes in the Bayesian network are determined. Factors that affect the damage estimation of the water target (ie, event nodes) are analyzed and determined.

[0120] In order to make an accurate estimate of the threat level of the opponent's target, it is necessary to carefully analyze the highly dynamic environment full of incomplete information and determine the relevant factors.

[0121] In actual scenarios, there are many factors that affect the estimation of water target damage. Combined with actual scenarios, the factors that affect the estimation of water target damage mainly include the opponent target type, opponent target position, opponent target speed, opponent target damage range, opponent target threat level and opponent target intention. According to the actual simulation environment, the above factors are divided into composite factors such as opponent target type, opponent target threat level and opponent target intention, and atomic factors such as opponent target position, opponent target speed and opponent target damage range.

[0122] In step 402, the Bayesian network structure is determined, and a hierarchical relationship between nodes is established according to the causal relationship between events.

[0123] In an actual implementation scenario, starting from atomic factors, reasoning layer by layer, and finally obtaining a process of composite factors representing the degree of damage. According to the actual situation of the simulation environment, the position of the other party's target, the speed of the other party's target, the damage range of the other party's target, and other information can be obtained from the type of the other party's target. The threat level of the other party's target is determined by the damage range of the other party's target and the distance of the target from our assets; the intention of the other party's target is determined by the estimated arrival time of the other party's target; and the estimated arrival time of the other party's target is determined by the speed of the other party's target and the nearest approach point.

[0124] Accordingly, establish Figure 2 , Figure 3 , Figure 4 and Figure 5 The hierarchical structure between the nodes shown in the figure can be integrated to obtain the following Figure 6 The Bayesian network structure shown (i.e., the network structure in Example 1).

[0125] TargetType represents the target type of the other party. According to the target type of the other party, the target position L can be obtained. target 、Opposite target speed V target and the opponent's target damage range EffectRange, that is, TargetType→{L target ,L zz ,EffectRange}.

[0126] L zz Indicates the location of key assets, D(L target ,L zz ) represents the distance Distance between the opponent target and the water target, then the threat degree Threat of the opponent target can be expressed as (EffectRange, Distance)→Threat.

[0127] CPA is the closest approach point. Fig.13 As shown in the figure, the distance from the other party's target position to CPA is expressed as range from CPA, and the other party's target estimated arrival time TBH can be expressed as (CPA, V target )→TBH, the other party’s target intent Intent can be expressed as TBH→Intent.

[0128] Damage of water targets can be expressed as (Threat, Intent)→Damage.

[0129] In step 403, the prior probability of the relevant nodes is determined: some parameters in the Bayesian network, ie, conditional probability, are determined based on expert experience.

[0130] In an actual implementation scenario, based on expert experience, it can be determined that the probability of the opponent's target damage range being short, medium, and long is determined by the opponent's target type. Figure 7 shown.

[0131] The damage range of the opponent target EffectRange and the distance D (L target ,L zz ) together get the probability of the other party's target threat being low, medium, and high. Figure 8 As shown; the probability of the other party's target intention being low, medium, and high is obtained from the other party's target estimated arrival time TBH, as shown Fig. 9 As shown in the figure, the probability of water target damage is obtained by combining the threat degree of the other party's target and the intention of the other party's target. Fig.10 shown.

[0132] The value of the distance between the other target and the water target is determined according to the specific value of the distance between the other target and the water target, which needs to be calculated; according to prior knowledge: when the distance between the other target and the water target is less than or equal to 2 kilometers, that is, D(L target ,L zz )≤2, the opponent's target is considered to be close, and Distance takes the value close. When the opponent's target is 2-10 kilometers away from the water target, that is, 2 <D(L target ,L zz )≤10, the distance between the other target and the water target is considered medium, and Distance takes the value medium. When the distance between the other target and the water target is 10-40 kilometers, that is, 10<(L target ,L zz )≤40, the opponent’s target is considered to be far away, and Distance takes the value far.

[0133] The estimated arrival time of the other party's target also needs to be calculated in real time.

[0134] In step 404, a simulation calculation model is built. Netica analysis software is used to establish a Bayesian network model for water target damage estimation.

[0135] In actual implementation, Netica analysis software is used to establish a Bayesian network model for water target damage estimation, such as Fig.11 As shown; input a determined partial conditional probability value; calculate the probability of water target damage based on the input opponent target information.

[0136] The method of determining some parameters in the Bayesian network based on expert experience specifically includes: according to the simulation environment, the types of aerial targets are mainly type I targets, type II targets and type III targets, that is, TargetType = {type I targets, type II targets, type III targets}; the values ​​of the opponent's target damage range EffectRange are short, medium and long, that is, EffectRange = {short, medium, Long}; the threat level Threat of the opponent's target is high, medium and low, that is, Threat = {high, medium, low}; the values ​​of the opponent's target intention Intent are high, medium and low, that is, Intent = {high, medium, long}; the values ​​of the water target damage Damage are true and false, that is, Damage = {true, false}.

[0137] The opponent target type is determined by the input opponent target information; the probability of obtaining the opponent target damage range from the opponent target type is P(EffectRange|TargetType) as follows Figure 7 shown.

[0138] The probability of determining the threat level of the opponent's target by the opponent's target damage range and real-time distance is P (Threat | Effect Range, Distance). Figure 8 shown.

[0139] The probability of obtaining the other party's target intention from the estimated arrival time is P(Intent|TBH) as follows: Fig. 9 As shown;

[0140] The probability of water target damage P(Damage|Threat,Intent) is obtained from the threat degree and capability of the other party's target, as follows: Fig.10 shown.

[0141] After the Bayesian network model is established, the damage probability of our assets is calculated, including: obtaining the probability of the opponent's target threat level from the specific value of the opponent's target damage range and the real-time distance; calculating the closest approach point; calculating the distance from the opponent's target to the CPA; calculating the expected arrival time TBH of the opponent's target; determining the opponent's target intention based on the expected arrival time of the opponent's target, substituting the target intention and the opponent's target threat level into the Bayesian network model to obtain the damage probability of our assets.

[0142] The probability of obtaining the threat level of the opponent's target from the specific value of the opponent's target damage range and the real-time distance specifically includes:

[0143] The probability P(EffectRange|TargetType) of obtaining the damage range of the target of the other party from the target type of the other party is as follows Figure 7 as shown

[0144] The real-time distance D(L target ,L zz ) of the target of the other party from the key asset (i.e., the water target) is calculated as follows

[0145] In actual implementation, the value of Distance is determined according to the distance D(L target ,L zz ) of the target of the other party from the water target, as follows

[0146] P(Distance = close) is D(L target ,L zz ) ≤ 2

[0147] P(Distance = medium) is 2 < D(L target ,L zz ) ≤ 10

[0148] P(Distance = far) is 10 < D(L target ,L zz ) ≤ 40

[0149] The probability of obtaining the threat level of the target of the other party from the specific value and real-time distance of the damage range of the target of the other party is shown in Figure 8 .

[0150] The calculation of the closest point of approach (CPA) is as follows

[0151] The schematic diagram of the closest point of approach (CPA) is as Fig.13 . CPA is the point closest to the ZZ point in the direction of the target speed of the other party, and the solution is as follows

[0152] Let the coordinates of CPA be (x cpa ,y cpa ,z cpa ).

[0153] First, the direction vector from the position of the target of the other party to CPA is perpendicular to the direction vector from CPA to the ZZ point, and there are the following equations: (x cpa -x zz )·(x cpa -x target ) = 0, (y cpa -y zz )·(y cpa -y target ) = 0 and (zcpa -z z2 )·(z cpa -z target )=0.

[0154] In addition, to ensure that the obtained CPA is in the direction of velocity, it is necessary to ensure that the direction vector from the opponent's target position to CPA is in the same direction as the velocity direction of the incoming weapon, as shown in the following equation: and

[0155] Solving the above equations, we can obtain the coordinates of CPA.

[0156] The calculating of the distance from the other party's target to the CPA specifically includes:

[0157] According to the coordinates of CPA, calculate the distance from the opponent's target to CPA:

[0158] The calculation of the other party's target estimated arrival time TBH specifically includes:

[0159] Define the time from the opponent's target position to CPA as T CPA .

[0160] Then, T CPA =range from CPA / |V target |

[0161] TBH=T CPA

[0162] Determining the other party's target intention according to the other party's target's estimated arrival time specifically includes:

[0163] In actual implementation, the probability of obtaining the other party's target intention from the estimated arrival time is P(Intent|TBH) as follows: Fig. 9 shown.

[0164] This embodiment analyzes and determines the factors that affect the ship damage estimation according to the actual simulation environment, establishes a hierarchical relationship between nodes according to the causal relationship between the factors, integrates the hierarchical relationship to obtain the Bayesian network structure, obtains some parameters in the Bayesian estimation network according to the expert prior knowledge, builds a simulation calculation model based on the Netica software, and inputs the target information in the simulation environment to quickly and accurately estimate the ship damage.

[0165] This embodiment achieves accurate estimation of water target damage in a high dynamic environment based on the Bayesian estimation network of target capability and target intention. Based on Netica's simulation calculation model, the problem of difficult calculation of water target damage is solved through scientific calculation methods.

[0166] Embodiment 3:

[0167] like Fig.17 , which is a schematic diagram of the architecture of a Bayesian network-based damage calculation device according to an embodiment of the present invention. The Bayesian network-based damage calculation device according to this embodiment includes one or more processors 21 and a memory 22. Fig.17 A processor 21 is taken as an example.

[0168] The processor 21 and the memory 22 may be connected via a bus or other means. Fig.17 The example of connecting through bus is taken in the following.

[0169] The memory 22 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs and non-volatile computer executable programs, such as the damage calculation method based on the Bayesian network in Example 1. The processor 21 executes the damage calculation method based on the Bayesian network by running the non-volatile software programs and instructions stored in the memory 22.

[0170] The memory 22 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some embodiments, the memory 22 may optionally include a memory remotely arranged relative to the processor 21, and these remote memories may be connected to the processor 21 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0171] The program instructions / modules are stored in the memory 22, and when executed by the one or more processors 21, the damage calculation method based on the Bayesian network in the above-mentioned embodiment 1 is executed.

[0172] It is worth noting that the information interaction, execution process, etc. between the modules and units within the above-mentioned devices and systems are based on the same concept as the processing method embodiment of the present invention. The specific contents can be found in the description of the method embodiment of the present invention and will not be repeated here.

[0173] A person skilled in the art may understand that all or part of the steps in the various methods of the embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the storage medium may include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0174] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A damage calculation method based on Bayesian network, characterized in that: include: Using the various factors that cause damage to our assets caused by the other party's goals as event nodes, determine the network structure; Determine the prior probability of each event node based on the prior knowledge of experts; According to the prior probability of each event node and the network structure, a Bayesian network model is established, and according to the Bayesian network model, the damage probability of our assets is determined.

2. The damage calculation method based on Bayesian network according to claim 1 is characterized in that: The event node includes one or more of the opponent target type, opponent target damage range, opponent target position, opponent target speed, opponent target distance, distance between the opponent target and the nearest approach point, opponent target estimated arrival time, opponent target threat level, opponent target intention and damage probability.

3. The damage calculation method based on Bayesian network according to claim 2 is characterized in that: The network structure is as follows: the opponent target type is used as a premise for determining the opponent target damage range and the opponent target speed, the opponent target position is used as a premise for calculating the opponent target distance and the distance between the opponent target and the nearest approach point, the opponent target distance and the opponent target damage range are used as premises for inferring the opponent target threat level, the distance between the opponent target and the nearest approach point and the opponent target speed are used as premises for calculating the opponent target estimated arrival time, the opponent target estimated arrival time is used as a premise for judging the opponent target intention, and the opponent target threat level and the opponent target intention are used as premises for determining the damage probability.

4. The damage calculation method based on Bayesian network according to claim 3 is characterized in that: The prior probability of each event node is determined based on the prior knowledge of the expert, specifically including: Determine the prior probability P(EffectRange|TargetType) of the damage range of the opponent target corresponding to different opponent target types; Determine the prior probability P (Threat|EffectRange,Distance) of the enemy target threat level corresponding to different enemy target damage ranges and different enemy target distances; Determine the prior probability P(Intent|TCPA) of the other party's intention corresponding to different expected arrival times of the other party; Determine the prior probability P(Damage|Threat,Intent) of the damage probability corresponding to different threat levels of the opponent target and different intentions of the opponent target.

5. The damage calculation method based on Bayesian network according to claim 3 is characterized in that: Determining the damage probability of our assets according to the Bayesian network model specifically includes: Determine the enemy target's location, enemy target type, enemy target speed, and our asset's location; Calculate the distance to the other party's target based on the other party's target location and our asset location; Determine the location of the nearest approach point based on the location of the other party's target, the type of the other party's target, and the location of our assets; calculate the estimated arrival time of the other party's target based on the location of the nearest approach point; The opponent's target type, opponent's target speed, opponent's target distance, distance between the opponent's target and the nearest approach point, and various prior probabilities are substituted into the Bayesian network model to calculate the damage probability of our assets.

6. The damage calculation method based on Bayesian network according to claim 5 is characterized in that: The calculation of the distance to the other party's target based on the other party's target position and the position of our asset specifically includes: Calculate the distance to the other target Among them, (x zz ,y zz , z zz ) is the coordinate of our asset location, (x target ,y target , z target ) are the coordinates of the other party’s target position.

7. The damage calculation method based on Bayesian network according to claim 5 is characterized in that: Determining the location of the nearest approach point based on the other party's target location, the other party's target type and the location of our assets specifically includes: The position of the closest approach point is calculated using a combined equation; the combined equation includes a first equation, a second equation, a third equation, a fourth equation, a fifth equation and a sixth equation; The first equation is (x cpa -x zz )·(x cpa -x target )=0; The second equation is (y cpa -y zz )·(y cpa -y target )=0; The third equation is (z cpa -z zz )·(z cpa -z target )=0; The fourth equation is The fifth equation is The sixth equation is Among them, (x cpa ,y cpa , z cpa ) is the coordinate of the nearest approach point, (x zz ,y zz , z zz ) is the coordinate of our asset location, (x target ,y target , z target ) is the coordinate of the other party’s target position, It is the vector representation of the opponent's target speed.

8. The damage calculation method based on Bayesian network according to claim 5 is characterized in that: The calculating the estimated arrival time of the other target according to the position of the closest approach point specifically includes: The distance between the target and the closest approach point is calculated according to the position of the closest approach point. Calculate the target's estimated arrival time Among them, (x cpa ,y cpa , z cpa ) is the coordinate of the nearest approach point, (x target ,y target , z target ) is the coordinate of the other party’s target position, 9. A non-volatile computer storage medium, characterized in that: The computer storage medium stores computer executable instructions, which are executed by one or more processors to complete the damage calculation method based on the Bayesian network described in any one of claims 1-8.

10. A damage calculation device based on Bayesian network, characterized in that: include: at least one processor; And, a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the processor to execute the Bayesian network-based damage calculation method described in any one of claims 1-8.

Images