Method and System for Selecting Strategies in the Development Plan of Weaponry and Equipment in the Game among Multiple Parties

Abstract

The present invention provides a method and system for selecting a development planning strategy of weaponry and equipment in a multi-party camp game. The method includes the following steps: obtaining the camp parameters, initial prior probabilities, and a set of strategies of weaponry and equipment development strategies of two multi-party camps participating in the camp game; calculating the camp payoffs at the initial game stage based on the camp parameters; using the clan solution method to calculate the optimal allocation results of all participants in the multi-party camps for the camp payoffs; constructing a perfect Bayesian equilibrium according to the optimal allocation results, in combination with the set of strategies and the initial prior probabilities; obtaining the optimal strategy selection and posterior probabilities at the initial game stage by solving the perfect Bayesian equilibrium; taking the posterior probabilities of each current game stage as the prior probabilities of the next game stage, and obtaining the optimal strategy selection of each game stage based on the solution of the perfect Bayesian equilibrium for each game stage. The present invention has the effect of accurately selecting the development planning strategy of weaponry and equipment.

Description

Technical Field

[0001] The present invention belongs to the field of weapon and equipment development planning, and particularly relates to a method and system for selecting strategies for weapon and equipment development planning in a multi-party camp game. Background Art

[0002] When studying the multi-party game of weapon and equipment development planning, the number of participants is usually multiple. The participants can be divided into two opposing camps according to their positions. In weapon and equipment development planning, a camp refers to a group formed by a combination of multiple participants with common interest goals or the same potential opponents. Compared with the "1-on-1" two-party game scenario, the multi-party game scenario divides the game parties into two camps and conducts game confrontations between the camps. The two camps have a sequential order in actions. According to the sequential order of making strategy decisions and taking actions, the camps can be divided into the leading camp and the following camp. When making strategy decisions, a camp can obtain information about the opponent camp during the previous game process.

[0003] In addition to the three general game characteristics of strategy dependence, incomplete information, and dynamic evolution, the multi-party game scenario also has the game characteristic of coexistence of non-cooperative and cooperative relationships. A camp is an adversarial decision-making entity in the multi-party game scenario, and a non-cooperative relationship is reflected between the camps. And each participant within the same camp can build an alliance to cooperate in weapon and equipment development planning to improve the overall benefit of the camp and further enhance the combat capability of its own weapon and equipment system, reflecting a cooperative relationship. Also, in the multi-party game scenario, the goals within a camp are relatively unified, and the goals between the camps are mutually opposed, with distinct confrontation characteristics.

[0004] In the game scenario, the strategies for weapon and equipment development planning can be analyzed and appropriate planning strategies can be selected to avoid the situation where the development planning of weapon and equipment blindly pursues the performance of a single type of weapon and equipment. In the prior art, a single non-cooperative or cooperative game theory is usually used to analyze and select the strategies for weapon and equipment development planning. However, in the multi-party game scenario, non-cooperative and cooperative relationships coexist, and it is difficult to accurately analyze the essence and process of the multi-party game scenario and thus accurately select the strategies for weapon and equipment development planning by only using a single non-cooperative or cooperative game theory. Summary of the Invention

[0005] The present invention provides a method and system for selecting strategies for weapon and equipment development planning in a multi-party camp game to solve the problem of low accuracy in selecting strategies for weapon and equipment development planning in the multi-party game scenario.

[0006] In a first aspect, the present invention provides a method for selecting strategies for weapon and equipment development planning in a multi-party camp game, and the method includes the following steps:

[0007] Obtain the camp parameters of two multi-party camps participating in the camp game, as well as the initial prior probabilities and the strategy sets of the weapon and equipment development strategies of the multi-party camps in the camp game. The initial prior probability is the probability obtained by inferring the types of participants in the opposing camp when one of the multi-party camps is in the initial game stage of the camp game;

[0008] Calculate the camp payoff at the initial game stage based on the camp parameters;

[0009] Use the pedigree solution method to calculate the optimal distribution results of all participants in the multi-party camps for the camp payoff;

[0010] Construct a perfect Bayesian equilibrium according to the optimal distribution results, in combination with the strategy set and the initial prior probability;

[0011] By solving the perfect Bayesian equilibrium, obtain the optimal strategy selection and posterior probability at the initial game stage. The posterior probability is the probability obtained by inferring the types of the participants in the opposing camp based on the strategy selection by one of the multi-party camps in the camp game;

[0012] Take the posterior probability of each current game stage as the prior probability of the next game stage, and based on the solution of the perfect Bayesian equilibrium for each game stage, obtain the optimal strategy selection for each game stage.

[0013] Optionally, the calculating the camp payoff at the initial game stage based on the camp parameters includes the following steps:

[0014] Generate the weapon and equipment system network of the multi-party camps based on the camp parameters;

[0015] Combine the weapon and equipment system networks of the two multi-party camps to generate a weapon and equipment system confrontation network between the two multi-party camps;

[0016] Evaluate and calculate the camp payoff of the multi-party camps at the initial game stage based on the weapon and equipment system confrontation network.

[0017] Optionally, the camp parameters include a camp strategy set, weapon and equipment parameters, and participant parameters in the multi-party camps. The generating the weapon and equipment system network of the multi-party camps based on the camp parameters includes the following steps:

[0018] Generate a plurality of equipment function nodes based on the weapon and equipment parameters;

[0019] Generate equipment function edges between the plurality of equipment function nodes according to the participant parameters;

[0020] Generate the weapon and equipment system network of multiple camps by combining the equipment function nodes and the equipment function edges of the said equipment.

[0021] Optionally, the steps for evaluating and calculating the camp benefits of multiple camps in the initial game stage based on the weapon and equipment system confrontation network are as follows:

[0022] Construct an association matrix between multiple equipment function nodes in multiple camps based on the weapon and equipment system network;

[0023] Calculate multiple combat ability loops of multiple camps by combining the association matrix and the weapon and equipment system confrontation network;

[0024] Aggregate multiple combat ability loops to calculate the camp benefits of multiple camps.

[0025] Optionally, the combat ability loop includes a generalized combat ability loop and a standard combat ability loop. The steps for aggregating multiple combat ability loops to calculate the camp benefits of multiple camps are as follows:

[0026] Calculate the number of ability loops of the generalized combat ability loop and the standard combat ability loop respectively;

[0027] Statistically calculate the first duration of the threat caused by the generalized combat ability loop to the enemy;

[0028] Statistically calculate the second duration of the threat caused by the standard combat ability loop to the enemy;

[0029] Calculate the camp benefits of multiple camps by combining the number of ability loops, the first duration, and the second duration.

[0030] Optionally, the steps for calculating the optimal allocation result of the camp benefits for all participants in multiple camps using the clan solution method are as follows:

[0031] Calculate the average value of the alliance benefits of all non-empty alliances in multiple camps under the camp benefits;

[0032] Solve to obtain the clan alliances of all participants based on the average value of the alliance benefits;

[0033] Determine the quasi-clan alliance of each participant;

[0034] Solve the clan core based on Pareto improvement and by combining the clan alliance and the quasi-clan alliance. The clan core is the optimal allocation result of the camp benefits for all participants.

[0035] Optionally, constructing a perfect Bayesian equilibrium based on the optimal allocation result, in combination with the strategy set and the initial prior probability, includes the following steps:

[0036] Screen out the initial strategy set of the initial game stage from the strategy set according to the optimal allocation result;

[0037] Construct a perfect Bayesian equilibrium in combination with the initial strategy set and the initial prior probability.

[0038] Optionally, obtaining the optimal strategy selection and posterior probability at the initial game stage by solving the perfect Bayesian equilibrium includes the following steps:

[0039] Calculate and infer the posterior probability based on the initial strategy set;

[0040] Calculate and analyze the optimal strategy selections of the two multi-party camps at the initial game stage according to the inferred posterior probability;

[0041] Combine the optimal strategy selection and the initial prior probability, and calculate the posterior probability at the initial game stage through Bayes' rule;

[0042] If the inferred posterior probability does not conflict with the posterior probability, the solution of the perfect Bayesian equilibrium is completed.

[0043] Optionally, the two multi-party camps are the first camp and the second camp respectively, and the formula for the perfect Bayesian equilibrium is:

[0044]

[0045] where A * (B) is the initial strategy set of the first camp; B * (Θ B ) is the initial strategy set of the second camp; is the posterior probability of the type of the participant in the second camp calculated by the first camp according to B * (Θ B ) and A * (B) and using Bayes' rule.

[0046] Second, the present invention also provides a weapon equipment development planning strategy selection system for multi-party camp games, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the method described in the first aspect is implemented.

[0047] The beneficial effects of the present invention are as follows: As can be seen from the above, the present invention provides a method for selecting a weapon and equipment development planning strategy for multi-party camp games, including the following steps: obtaining the camp parameters, initial prior probabilities, and the strategy set of weapon and equipment development strategies of two multi-party camps participating in the camp game; calculating the camp benefits at the initial game stage based on the camp parameters; using the sectarian solution method to calculate the optimal distribution results of the camp benefits for all participants in the multi-party camp; constructing a perfect Bayesian equilibrium according to the optimal distribution results, combined with the strategy set and the initial prior probabilities; obtaining the optimal strategy selection and posterior probabilities at the initial game stage by solving the perfect Bayesian equilibrium; taking the posterior probability of each current game stage as the prior probability of the next game stage, and obtaining the optimal strategy selection of each game stage based on the solution of the perfect Bayesian equilibrium for each game stage. Compared with the theoretical analysis of single non-cooperative or cooperative games, it can more accurately analyze the essence and process of multi-party game scenarios, so as to more accurately select the weapon and equipment development planning strategy in the game scenario. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a schematic flowchart of a method for selecting a weapon and equipment development planning strategy for multi-party camp games in one embodiment of the present invention.

[0049] Figure 2 It is a schematic flowchart of a method for selecting a weapon and equipment development planning strategy for multi-party camp games in one embodiment of the present invention.

[0050] Figure 3 It is a schematic flowchart of a method for selecting a weapon and equipment development planning strategy for multi-party camp games in one embodiment of the present invention.

[0051] Figure 4 It is a schematic diagram of a weapon and equipment system confrontation network for multi-party two-camp games in one embodiment of the present invention.

[0052] Figure 5 It is a schematic diagram of the dynamic game process for multi-party two-camp games in one embodiment of the present invention.

[0053] Figure 6 It is a schematic flowchart of a method for selecting a weapon and equipment development planning strategy for multi-party camp games in one embodiment of the present invention.

[0054] Figure 7 It is a schematic flowchart of a method for selecting a weapon and equipment development planning strategy for multi-party camp games in one embodiment of the present invention.

[0055] Figure 8 It is a schematic flowchart of a method for selecting a weapon and equipment development planning strategy for multi-party camp games in one embodiment of the present invention.

[0056] Figure 9 This is a schematic flowchart of a method for selecting a development planning strategy for weapon equipment in the game among multiple camps in one embodiment of the present invention.

[0057] Figure 10 This is a schematic flowchart of a method for selecting a development planning strategy for weapon equipment in the game among multiple camps in one embodiment of the present invention. Detailed implementation manner

[0058] The embodiment of the present invention specifically discloses a method for selecting a development planning strategy for weapon equipment in the game among multiple camps.

[0059] Refer to Figure 1 , the method for selecting a development planning strategy for weapon equipment in the game among multiple camps specifically includes the following steps:

[0060] S101. Obtain the camp parameters of two multiple camps participating in the camp game, as well as the initial prior probability in the camp game and the strategy set of the weapon equipment development strategy of the multiple camps.

[0061] Among them, in the camp game, there are usually two multiple camps in an opposing relationship. A multiple camp refers to a group formed by a combination of multiple participants with common interest goals or the same potential opponents. The camp game includes multiple game stages. In each game stage, the two camps have a sequential order in actions. According to the sequential order of strategy decision-making and taking actions, the camps can be divided into the leading camp and the following camp. The camp can obtain information about the opponent camp during the previous game process when making strategy decisions.

[0062] The initial prior probability is the probability obtained by one of the multiple camps when inferring the type of participants in the opponent camp during the initial game stage of the camp game. The type of participant refers to multiple participants with different levels of development intensity divided according to the amount of resources such as time and funds invested.

[0063] In this embodiment, the camp parameters mainly include the camp strategy set, weapon equipment parameters, and participant parameters in the multiple camps. The participant parameters mainly include the number of participants and the type of participants. The weapon equipment parameters mainly include the weapon equipment information of all participants and the equipment interaction relationships. The equipment interaction relationships include the interaction relationships among the equipment of each party itself, the interaction relationships among the equipment of different participants within the camp, and the equipment interaction relationships between different camps.

[0064] S102. Calculate the camp income in the initial game stage based on the camp parameters.

[0065] Among them, a multi-party two-camp weapon and equipment system confrontation network is constructed according to the camp parameters of two multi-party camps. By aggregating the capabilities of all combat ability loops of each camp in the camp weapon and equipment system confrontation network, the overall combat ability of the camp weapon and equipment system can be obtained, which is the camp benefit.

[0066] S103. Use the clan solution method to calculate the optimal allocation result of all participants in the multi-party camp for the camp benefit.

[0067] Among them, all participants in the same multi-party camp can form alliances arbitrarily within the camp. First, calculate the average alliance benefit within the multi-party camp according to the camp benefit, then solve the clan alliance in the multi-party camp according to the definition of the clan alliance, and finally solve the clan core of the multi-party camp. The clan core is the optimal allocation result of all participants for the camp benefit.

[0068] S104. Construct a perfect Bayesian equilibrium according to the optimal allocation result, combined with the strategy set and the initial prior probability.

[0069] S105. By solving the perfect Bayesian equilibrium, obtain the optimal strategy selection and posterior probability at the initial game stage.

[0070] Among them, the posterior probability is the probability obtained by inferring the types of participants in the other camp based on the strategy selection in the camp game of one multi-party camp.

[0071] S106. Take the posterior probability of each current game stage as the prior probability of the next game stage, and based on the solution of the perfect Bayesian equilibrium at each game stage, obtain the optimal strategy selection at each game stage.

[0072] The implementation principle of one implementation method of this embodiment is:

[0073] In the game among multiple parties, there not only exists a non - cooperative confrontation relationship between two multi - party camps, but also an alliance and cooperation relationship among multiple participants within the same multi - party camp. Based on the confrontation relationship between two multi - party camps, the camp benefits of the two multi - party camps in the initial game stage can be calculated according to the camp parameters. Then, based on the alliance and cooperation relationship within the multi - party camp and using the pedigree solution method, the optimal distribution result of the camp benefits for all participants can be calculated. Thus, a perfect Bayesian equilibrium is constructed and solved to obtain the optimal strategy selection and posterior probability in the initial game stage. Taking the posterior probability of each current game stage as the prior probability of the next game stage, it further affects the strategy selection in the next game stage, forming a dynamic iterative process of incomplete information update - strategy selection. Finally, according to the solution of the perfect Bayesian equilibrium for each game stage, the optimal strategy selection for each game stage is obtained. Compared with the theoretical analysis of a single non - cooperative or cooperative game, it can analyze the multi - party game scenario with co - existing non - cooperation and cooperation more accurately, and thus more accurately select the development planning strategy of weapons and equipment in the game scenario.

[0074] In one implementation manner of this embodiment, referring to Figure 2 , step S102, that is, calculating the camp benefits in the initial game stage based on the camp parameters, specifically includes the following steps:

[0075] S201. Generate a weapons and equipment system network for the multi - party camp based on the camp parameters.

[0076] Among them, in this embodiment, the camp parameters mainly include the camp strategy set, weapons and equipment parameters, and participant parameters in the multi - party camp. The participant parameters mainly include the number of participants and the participant types. The weapons and equipment parameters mainly include the weapons and equipment information of all participants and the equipment interaction relationships. The equipment interaction relationships include the interaction relationships among the parties' own equipment, the interaction relationships among the equipment of different participants within the camp, and the equipment interaction relationships between different camps. The weapons and equipment parameters in the camp parameters can be abstracted as equipment function nodes, and the command and control relationships and information transmission relationships in the equipment interaction relationships can be abstracted as equipment function edges, thereby constructing a weapons and equipment system network.

[0077] S202. Combine the weapons and equipment system networks of the two multi - party camps to generate a weapons and equipment system confrontation network between the two multi - party camps.

[0078] Among them, based on the reconnaissance relationship and the strike relationship in the equipment interaction relationships of the two multi - party camps, the connection function edges connecting the two weapons and equipment system networks are abstracted. The two weapons and equipment system networks are connected through the connection function edges to form a confrontation relationship between the two multi - party camps, thereby generating a weapons and equipment system confrontation network between the two multi - party camps.

[0079] S203. Calculate the camp benefits of multiple camps in the initial game stage based on the weapon equipment system confrontation network assessment.

[0080] Among them, in the weapon equipment system confrontation network, since alliances can be formed among the participants within the same multiple camps, enhancing the collaborative relationship between the weapon equipment of the participants in the alliance, new equipment function edges are formed in the multiple camps. The camp weapon equipment system confrontation network is jointly composed of the weapon equipment system networks of two camps. By aggregating the capabilities of all combat capability loops of each camp in the camp weapon equipment system confrontation network, the overall combat capability of the camp weapon equipment system, that is, the camp benefit, can be obtained.

[0081] The implementation principle of one implementation method of this embodiment is as follows:

[0082] First, based on the camp parameters, the weapon equipment system networks of two multiple camps can be generated respectively. The collaborative relationship between the weapon equipment of multiple participants within the same multiple camp can be reflected in the weapon equipment system network. Then, according to the reconnaissance and strike relationships between the weapon equipment of different camps, the weapon equipment system networks of the two multiple camps are combined to generate the weapon equipment system confrontation network between the two multiple camps, so that the combat capability loops of the multiple camps in the weapon equipment system confrontation network can be constructed, and the camp benefits are evaluated and calculated according to the combat capability loops.

[0083] In one implementation method of this embodiment, the camp parameters include the camp strategy set, weapon equipment parameters, and participant parameters in the multiple camps. Refer to Figure 3 , step S201, that is, generating the weapon equipment system network of the multiple camps based on the camp parameters, specifically includes the following steps:

[0084] S301. Generate multiple equipment function nodes based on the weapon equipment parameters.

[0085] Among them, in this embodiment, the weapon equipment parameters mainly include the weapon equipment information of all participants and the equipment action relationships. The equipment action relationships include the action relationships between the equipment of each party itself, the action relationships between the equipment of different participants within the camp, and the equipment action relationships between different camps. Based on the analysis of the OODA combat theory, the weapon equipment functions are divided into three categories: reconnaissance function, command function, and strike function. In this embodiment, the weapon equipment entities can be abstracted into four types of function nodes: reconnaissance node S, command node D, strike node I, and target node T.

[0086] S302. Generate the equipment function edges between multiple equipment function nodes according to the participant parameters.

[0087] Among them, in this embodiment, the participant parameters mainly include the number of participants and the participant types. The relationships between the various functional nodes, which are the equipment function relationships of the corresponding weaponry and equipment, can be abstracted as functional edges. The function relationships between the equipment function nodes can be divided into 4 categories, as shown in Table 1.

[0088] Relationship type Equipment system network function edge Constraint Reconnaissance relationship T→S Functional nodes between different camps Command and control relationship D→S, D→I, D→D Functional nodes within the same camp Strike relationship I→T Functional nodes between different camps Information transfer relationship S→D, S→S, D→D Functional nodes within the same camp

[0089] Table 1 Equipment function edge relationships in two camps

[0090] S303. Generate a weaponry and equipment system network for multiple-party camps by combining equipment function nodes and equipment function edges.

[0091] Among them, within the same camp, the participants can form alliances to enhance the overall combat capabilities of the camp. When the participants form alliances, they can use the weaponry and equipment of multiple participants to exert an impact on the opposing camp, reducing the combat capabilities of the weaponry and equipment system of the opposing camp. At the same time, they can enhance the cooperative combat capabilities among the equipment of different participants within the camp, thereby improving the combat capabilities of the camp's equipment system. Forming an alliance can correspondingly form functional edges between the function nodes of different participants within the same camp in the weaponry and equipment system network. The newly added types of functional edge relationships of the alliance include the command and control relationship and the information transmission relationship shown in Table 1.

[0092] Before forming an alliance, different participants within the same camp fought independently, and there were no connections of functional edges between the weaponry and equipment function nodes of different participants. After forming an alliance, new equipment functional edges can be formed between the weaponry and equipment function nodes of different participants, which helps to improve the cooperative combat capabilities between the function nodes within the weaponry and equipment system network of the camp, form more combat capability loops in terms of quantity and form in the weaponry and equipment system confrontation network, and enhance the overall combat effectiveness of the camp. Combine Figure 4 to specifically illustrate the impact of participants forming an alliance on the weaponry and equipment system network.

[0093] Refer to Figure 4 , Camp A and Camp B are two multi-party camps in this embodiment, and the dotted functional edges represent the newly added functional edges that can be formed by forming an alliance. Before participants i and j formed an alliance, there was no combat capability loop in Camp A targeting the reconnaissance node S6 of participant k in Camp B. Although the command node D2 in participant i could issue an attack command to the strike node I2 to strike the reconnaissance node S6 in Camp B, the command node D2 could not obtain intelligence information about the reconnaissance node S6. Although the reconnaissance node S3 in participant j could obtain intelligence information about the reconnaissance node S6, it could not strike S6.

[0094] After the formation of an alliance between Participant i and Participant j, an information transfer functional edge from S3 to D2 can be formed. S3 transfers the intelligence information about S6 obtained to D2, thereby adding a standard combat capability loop for S6: S6→S3→D2→I2→S6. Similarly, analyze the changes in the weapon equipment system network before and after the formation of an alliance in Camp B. Before the formation of an alliance between Participant k and Participant l, there was no combat capability loop in Camp B targeting the strike node I4 of Participant j in Camp A. Although the reconnaissance node S6 in Participant k can transfer information to the command node D6, and D6 can issue an attack command to I6 to strike the strike node I4 of Participant j. However, the reconnaissance node S6 cannot obtain intelligence information about the reconnaissance node I4, while the reconnaissance node S7 in Participant l can obtain intelligence information about the reconnaissance node S6. After the formation of an alliance between Participant k and Participant l, an information transfer functional edge from S7 to S6 can be formed. S7 transfers the intelligence information about I4 obtained to S6, thereby adding a generalized combat capability loop for I4: I4→S7→S6→D6→I6→I4.

[0095] After constructing the weapon equipment system confrontation network, for the convenience of analyzing the multi-party game problem of the subsequent weapon equipment development plan, the multi-party two-camp game model of the weapon equipment development plan can be assumed. The specific assumptions are as follows:

[0096] (1) In the weapon equipment development plan, only consider the research and development cost, research and development cycle, and acquisition cost of the weapon equipment. After the weapon equipment is purchased, the retirement situation is not considered, and the maintenance cost of the equipment is ignored.

[0097] (2) Multiple participants can be divided into two multi-party camps. There is no cooperation relationship between the participants in different camps, only competitive confrontation relationships. There is a cooperation relationship between the participants in the same camp. A camp contains multiple participants, and the participants in the same camp can obtain greater benefits by forming an alliance.

[0098] (3) The strategy set of the participants in each stage and the participant type set are public known information. The incomplete information is manifested as the probability inference of the participant type, and the posterior probability of the previous stage enters the next new stage of the game as the prior probability of the participant type of the next stage.

[0099] Based on the above model assumptions, a multi-party two-camp game model for weapon equipment development planning (Two-camp game model for weapon equipment development planning, TCGMFWEDP) is constructed. Assume that the two multi-party camps are Camp A and Camp B respectively. All the symbols and definition explanations included in the game model are shown in Table 2.

[0100]

[0101]

[0102] Table 2 Symbol Definitions and Explanations of the Multi-Party Two-Camp Game Model The multi-party two-camp game model for the weapon and equipment development plan can be expressed as an eight-tuple:

[0103]

[0104] where: N = (N A , N B ) is the set of participants; Ω = (A, B) is the set of camps of the participants; Θ = (Θ A , Θ B ) is the space of participant types. Among them, Θ A = (σ1, σ2,..., σ l ) represents the set of participant types in camp A, and σ i represents the i-th participant type in camp A. σ i ∈ {α1} means that the participants in camp A have only one type. Θ B = (δ1, δ2,..., δ m ) represents the set of participant types in camp B. δ i ∈ {β1, β2, β3} represents each participant type in camp B. The type is the private information of the participant. Camp A does not know the specific types of the participants in camp B, but has a prior probability judgment on the type distribution of the participants in camp B.

[0105] T is the total number of stages of the multi-stage game; S = (A, B) is the strategy set of the game players. A = A 1 × A 2 ×... × A T is the set composed of the weapon and equipment development strategies that camp A can choose at each stage in the development planning process. B = B 1 × B 2 ×... × B T is the set composed of the weapon and equipment development strategies that camp B can choose at each stage in the development planning process. Among them, represents the strategy set that camp A can choose in the k-th game stage, represents the j-th weapon and equipment development plan that camp A can choose in the k-th game stage; represents the strategy set that camp B can choose in the k-th game stage, represents the j-th weapon and equipment development plan that camp B can choose in the k-th game stage.

[0106] P = (P 1 , P 2 ,..., PT ) is the set of prior probabilities of the participant types of Camp A against Camp B. Among them, P k is the prior probability inference of the participant types of Camp A against Camp B in the k-th stage; is the set of posterior probabilities of the participant types of Camp A against Camp B. Among them, is the posterior probability judgment of the participant types of Camp A against Camp B in the k-th stage. The prior probability of each stage comes from the posterior probability of the previous game stage, that is

[0107] U = (U 1 , U 2 ,..., U T ) is the set of payoff functions of the players in each stage of the game, is the set of payoff functions of the participants in the k-th game stage. Among them, is the payoff function of the i-th participant in Camp A in the k-th game stage, is the payoff function of the i-th participant in Camp B in the k-th game stage.

[0108] Based on the multi-party two-camp game model of weapon equipment development planning, dynamic game analysis can be carried out on the multi-party equipment system confrontation network, so as to analyze the weapon equipment planning process of the two multi-party camps. In the process of development planning, the information that the two camps have about the enemy camp is incomplete, which is reflected in the uncertainty of the participant types in the enemy camp, and probability is used to represent the distribution of the decision-maker types. In each stage of the development planning, the strategy selections of the two camps have a sequential order. After obtaining the information of the strategy selections of the other camp, the two camps update the posterior probabilities of the types of the participants in the enemy camp.

[0109] Referring to Figure 5 , in each stage of the development planning, the participants in the two camps select their own strategies in this stage, and the weapon equipment in the strategies is added to the weapon equipment system. The newly added weapon equipment function nodes in the weapon equipment system network are exactly converted from the newly added weapon equipment. The newly added equipment function edges in the weapon equipment system network mainly reflect that the relationship between the equipment in the weapon equipment system network is becoming more and more complex. Part of the newly added equipment function edges comes from the influence and association relationship between the newly added weapon equipment and the original weapon equipment in the system, and the other part comes from the alliance construction between the participants in the same camp. The participants enhance the cooperative relationship between the weapon equipment by building alliances. Through the continuously updated strategies in the game stage process, the weapon equipment system confrontation network in each game stage is iteratively updated to show the dynamic game process of the multi-party two-camp.

[0110] In one implementation manner of this embodiment, referring toFigure 6 , step S203, that is, calculating the camp benefits of multiple camps in the initial game stage based on the weapon equipment system confrontation network, specifically includes the following steps:

[0111] S401. Construct an association matrix between multiple equipment function nodes in multiple camps based on the weapon equipment system network.

[0112] Among them, in this embodiment, the weapon equipment entity can be abstracted into four types of function nodes: reconnaissance node S, command node D, strike node I, and target node T. According to the equipment function edge relationship in the weapon equipment system network, an association matrix between different equipment function nodes within the camp is constructed. Among them, the value of the association matrix is [0, 1]. The association matrix value of 0 represents that there is no function edge relationship between function nodes, and the association matrix value of 1 represents that there is a function edge relationship between function nodes. The specific association matrix within the camp is shown in Table 3.

[0113] Incidence matrix Meaning Dimension <![CDATA[M T-S > Incidence matrix of target nodes - reconnaissance nodes in the camp m×n <![CDATA[M S-D > Incidence matrix of reconnaissance nodes - command nodes in the camp n×o <![CDATA[M D-I > Incidence matrix of command nodes - strike nodes in the camp o×p <![CDATA[M I-T > Incidence matrix of strike nodes - target nodes in the camp p×m <![CDATA[M S-S > Incidence matrix between different reconnaissance nodes in the camp n×n <![CDATA[M D-D > Incidence matrix between different command nodes in the camp o×o <![CDATA[M D-S > Incidence matrix of command nodes - reconnaissance nodes in the camp o×n

[0114] Table 3 Association Matrix of Equipment Function Nodes in the Camp

[0115] In Table 2, m represents the number of target nodes in the camp, n represents the number of reconnaissance nodes in the camp, o represents the number of command nodes in the camp, and p represents the number of strike nodes in the camp.

[0116] S402. Combine the association matrix and the weapon equipment system confrontation network to calculate multiple combat ability loops of multiple camps.

[0117] Among them, by combining the association matrix and the function edges between each function node in the weapon equipment system confrontation network, the combat ability loops of multiple camps and the quantity calculation formula of the combat ability loops can be constructed. The form and calculation formula of the combat ability loops are shown in Table 4.

[0118]

[0119]

[0120] Table 4 Forms and Calculation Formulas of Combat Ability Loops

[0121] S403. Aggregate multiple combat ability loops to calculate the camp benefits of multiple camps.

[0122] Among them, aggregate all combat ability loops, count the number of different types of combat ability loops, and the duration of the threat posed by different types of combat ability loops to the enemy, so as to calculate the camp benefits of multiple camps.

[0123] The implementation principle of one implementation method of this embodiment is:

[0124] The camp benefit mainly refers to aggregating the capabilities of all combat capability loops of multiple camps to obtain the overall combat capability of the camp's weapon and equipment system. Therefore, an association matrix between multiple equipment function nodes in multiple camps can be constructed according to the weapon and equipment system network, and then multiple combat capability loops of multiple camps can be calculated based on the association matrix, so as to aggregate all combat capability loops and calculate the camp benefit of multiple camps.

[0125] In one implementation manner of this embodiment, the combat capability loop includes a generalized combat capability loop and a standard combat capability loop. Referring to Figure 7 , step S403, that is, aggregating multiple combat capability loops to calculate the camp benefit of multiple camps, specifically includes the following steps:

[0126] S501. Calculate the number of capability loops of the generalized combat capability loop and the standard combat capability loop respectively.

[0127] Among them, the number of capability loops of the generalized combat capability loop and the standard combat capability loop can be calculated and counted according to formulas (1) to (6) in Table 4.

[0128] S502. Statistically calculate the first duration of the threat caused by the generalized combat capability loop to the enemy.

[0129] Among them, the first duration is the accumulation of the threat of the generalized combat capability loop to the enemy in terms of time.

[0130] S503. Statistically calculate the second duration of the threat caused by the standard combat capability loop to the enemy.

[0131] Among them, the second duration is the accumulation of the threat of the standard combat capability loop to the enemy in terms of time.

[0132] S504. Calculate the camp benefit of multiple camps by combining the number of capability loops, the first duration, and the second duration.

[0133] Among them, for example, assume that the two multiple camps in this embodiment are Camp A and Camp B respectively. Assume that in the kth game stage, the strategy composed of all participants in Camp A is The set of selection strategies composed of all participants in Camp B is The type of participants in Camp B is Θ b , and a weapon and equipment system confrontation network is generated from the weapon and equipment development plans of all participants. According to formulas (1) to (6) in Table 4, calculate that the number of standard combat capability loops of Camp A against Camp B in the weapon and equipment system confrontation network of the camp is m1, the number of generalized combat capability loops is m2, the number of standard combat capability loops of Camp B against Camp A is m3, and the number of generalized combat capability loops is m4. Then, in the kth stage, the camp benefit formula of Camp A is as follows:

[0134]

[0135] Wherein: is the camp benefit of Camp A, is the duration of the threat caused by the standard combat capability ring to the enemy, is the duration of the threat caused by the generalized combat capability ring to the enemy.

[0136] The camp benefit formula of Camp B is as follows:

[0137]

[0138] Wherein: is the camp benefit of Camp B, is the duration of the threat caused by the standard combat capability ring to the enemy, is the duration of the threat caused by the generalized combat capability ring to the enemy.

[0139] In the camp benefit formulas of Camp A and B, represents the threat ability of the k-th standard combat capability ring to the enemy, represents the threat ability of the g-th generalized combat capability ring to the enemy. The threat ability of the combat capability ring mainly depends on the functional evaluation values of three types of nodes: reconnaissance, command and control, and strike. Since the standard combat capability ring only contains 4 nodes, the reconnaissance, command and control, and strike capabilities of the standard combat capability ring correspond to the functional evaluation values of the reconnaissance, command and control, and strike nodes.

[0140] Set the functional evaluation values of the reconnaissance node, command and control node, and strike node to d s , d c , d a . Calculate the threat ability of the standard combat capability ring in a product form, and at the same time consider the influence of the enemy combat capability ring coverage number on our functional nodes. Set the threat ability D so of the standard combat capability ring, and its calculation formula is:

[0141]

[0142] Wherein: u s , u c , u a respectively represent the threat ability coefficients of the three types of nodes (reconnaissance, command and control, strike) of the standard combat capability ring.

[0143] Since the number of nodes in the same functional category (reconnaissance, command and control, strike) in the generalized combat capability ring is not unique, it affects the reconnaissance, command and control, and strike capabilities of the generalized combat capability ring. Therefore, before calculating the threat ability of the generalized combat capability ring, it is necessary to calculate the reconnaissance ability d s of the generalized combat capability ring, the command and control ability dc and the strike ability d a Suppose there are x reconnaissance nodes with information transfer relationships in a generalized combat ability loop, and there are information transfer and collaborative control relationships among the reconnaissance nodes, then the reconnaissance ability d of the generalized combat ability loop s The calculation formula is:

[0144]

[0145] where, is the function value of each reconnaissance node in this generalized combat ability loop, is the threat ability coefficient of each reconnaissance node in this generalized combat ability loop. Similarly, calculate the command and control ability d of the generalized combat ability loop c and the strike ability d a are respectively:

[0146]

[0147]

[0148] where, is the function value of the command and control node in the generalized combat ability loop, is the function value of the strike node in the generalized combat ability loop, represents the threat ability index of each command and control node and strike node in the generalized combat ability loop.

[0149] Combined with the calculated reconnaissance, command and control, and strike capabilities, the threat ability D of the generalized combat ability loop is calculated in a product form ol as:

[0150]

[0151] The implementation principle of one implementation method in this embodiment is:

[0152] Calculate the number of ability loops of the two types of combat ability loops respectively, and count the duration of the threat caused by the two types of combat ability loops to the local area. Finally, the camp income of the multi-party camp can be calculated by combining the number of ability loops and the duration.

[0153] After calculating the camp income of the multi-party camp, for the problem of the income distribution plan within the multi-party camp in this embodiment, the cooperative game theory needs to be adopted. For example: in the cooperative game G=(N, V), N represents the number of participants, V represents the income. If the N-dimensional vector x satisfies both: x i ≥V({i}) and then the vector x is called an allocation of the cooperative game. In this embodiment, the pedigree solution is used to solve the cooperative game, and the pedigree solution is proposed based on "dominance" and "egalitarianism".

[0154] The related definitions of the clan system solution include the clan system, pure clan system, mixed clan system, quasi-clan system, and clan system core. Among them, the definition of the clan system is as follows: In the cooperative game G = (N, V), for the coalition T containing the participant i, if T = argmax i∈T',T'∈N V(T') / |T'|, then the coalition T is called the clan system of the participant i. All the clan system coalitions in the cooperative game G = (N, V) are also called effective coalitions, and the coalitions dominated by the effective coalitions are called the invalid coalitions of this cooperative game.

[0155] The definitions of the pure clan system and the mixed clan system are as follows: Let the coalition T be an effective coalition of the cooperative game G = (N, V). If for each participant i ∈ T, there is then the coalition T is called the pure clan system of the participant i; if there exists a participant j ∈ T such that then the coalition T is called the mixed clan system.

[0156] The definition of the quasi-clan system is as follows: Let T be an invalid coalition of the cooperative game G = (N, V). If for each participant i ∈ T, there is then the coalition T is called the quasi-clan system of the participant i. The quasi-clan system must exist in the cooperative game G = (N, V). The definition of the clan system core is as follows: In the cooperative game G = (N, V), the solution set obtained according to the clan system is called the clan system core.

[0157] In one implementation manner of this embodiment, referring to Figure 8 , step S103, that is, using the clan system solution method to calculate the optimal distribution result of the camp income for all participants in the multi-party camp, specifically includes the following steps:

[0158] S601. Calculate the average coalition income of all non-empty coalitions in the multi-party camp under the camp income.

[0159] Among them, the non-empty coalition refers to the non-empty set coalition formed by all participants in the multi-party camp in any way and any quantity. In the cooperative game G = (N, V), for any coalition T that meets then calculate the average coalition income a(T) of the coalition T, and the calculation formula is: a(T) = V(T) / |T|.

[0160] S602. Solve to obtain the clan system coalitions of all participants based on the average coalition income.

[0161] Among them, according to the formula Solve the clan systems of all participants in the multi-party camp, and then use the set composed of all clan systems as the clan system coalition.

[0162] S603. Determine the quasi-clan system coalition of each participant.

[0163] Among them, the quasi-lineage alliance of each participant is determined from the lineage alliance.

[0164] S604. Solve the lineage core based on Pareto improvement and in combination with the lineage alliance and the quasi-lineage alliance.

[0165] Among them, when the lineage alliance collapses, the participants therein will cooperate with other participants based on their own pure lineage or quasi-lineage and in accordance with the Pareto improvement theory to form an alliance that can obtain the maximum Pareto improvement, and all participants in the alliance can divide the surplus brought by the cooperation, thereby obtaining the lineage core of the cooperative game. The lineage core in this embodiment is the optimal distribution result of the camp income for all participants.

[0166] For example: Suppose there are three participants in a multi-party camp, and the strategies of the participants are given. Calculate the incomes of different alliances respectively: U(1)=4, U(2)=1, U(3)=1, U(1,2)=6, U(1,3)=7, U(2,3)=4, U(1,2,3)=10. From this, the camp incomes under different alliance methods of the participants can be obtained, as shown in Table 5. Among them, the lineages of participants 1, 2, and 3 are {1}, {1, 2, 3}, and {1, 3} respectively. The pure lineage is {1}, and the mixed lineages are {1, 2, 3} and {1, 3}. The quasi-lineages of participants 1, 2, and 3 are {1}, {1, 2}, and {2, 3} respectively. The mixed lineages {1, 2, 3} and {1, 3} are prone to collapse due to participant 1.

[0167] Alliance method Camp benefit {1}，{2}，{3} 6 {1，2}，{3} 7 {1，3}，{2} 8 {2，3}，{1} 8 {1，2，3} 10

[0168] Table 5 Example of the solution result of the lineage solution

[0169] To achieve the alliance {1, 2, 3} that obtains the maximum Pareto improvement, participants 2 and 3 first ensure that the income of participant 1 in the alliance is at least 4. At the same time, participants 2 and 3 use the alliance U(2,3)=4 as a chip to cooperate with participant 1, cooperate to form a grand alliance on the basis of ensuring the existing interests of the participants, and equally divide the additional income brought by forming the alliance. The additional income is 2. Finally, the income of camp A is 10. Thus, the lineage core can be solved: the income distribution of the participants is (4.67, 2.67, 2.67), and the solved income distribution satisfies both the individual rationality of the participants and the alliance optimality in the multi-party camp.

[0170] The implementation principle of one implementation method in this embodiment is as follows:

[0171] Solve the lineage core in the multi-party camp to obtain the optimal distribution result of the camp income for all participants. Analyze the cooperative game by using the lineage solution method. Compared with using the Shapley solution method, it can more reasonably distribute the camp income on the basis of satisfying the individual rationality of the participants.

[0172] In one implementation manner of this embodiment, referring to Figure 9 , step S104, that is, constructing a perfect Bayesian equilibrium according to the optimal allocation result, in combination with the strategy set and the initial prior probability, specifically includes the following steps:

[0173] S701. Screen out the initial strategy set in the initial game stage from the strategy set according to the optimal allocation result.

[0174] S702. Construct a perfect Bayesian equilibrium in combination with the initial strategy set and the initial prior probability.

[0175] Among them, infer the optimal strategy selection in the initial game stage in combination with the initial strategy set and the initial prior probability, calculate the posterior probability in the initial game stage according to the Bayesian rule, and then construct the perfect Bayesian equilibrium in the initial game stage in combination with the optimal strategy selection and the posterior probability.

[0176] In one implementation manner of this embodiment, referring to Figure 10 , step S105, that is, obtaining the optimal strategy selection and the posterior probability in the initial game stage by solving the perfect Bayesian equilibrium, specifically includes the following steps:

[0177] S801. Calculate and infer the posterior probability based on the initial strategy set.

[0178] Among them, in this embodiment, based on the description in Table 2, assume that the two multi-party camps are the first camp and the second camp respectively. Calculate the posterior probability inference of the type of the participants in the second camp by the first camp based on the initial strategy set of the participants in the second camp as P(Θ B |B).

[0179] S802. Calculate and analyze the optimal strategy selection of the two multi-party camps in the initial game stage according to the inferred posterior probability.

[0180] Among them, referring to the example description in step S801, the first camp selects the optimal strategy based on the posterior probability inference P(Θ B |B) of the type of the participants in the second camp, so that the expected value of its game payoff takes the maximum value, that is, by calculating max∑P(Θ B |B)*U A (A,B,Θ B ), obtain the set of optimal strategy selections A * (B) inferred by the first camp, and A * (B) satisfies the condition The participants in the second camp foresee that the participants in the first camp will select the optimal strategy set A * (B). To make the game payoff of itself take the maximum value, by calculating maxUB (A * (B), B, Θ B ) Obtain the set B of the optimal strategy choices inferred by the second camp * (Θ B ), B * (Θ B ) satisfies the condition

[0181] S803. Combine the optimal strategy choices and the initial prior probabilities, and calculate the posterior probabilities at the initial game stage through Bayes' rule.

[0182] Among them, referring to the example descriptions in steps S801 and S802, the first camp calculates the posterior probability P(Θ B |B) at the initial game stage based on the prior probabilities and the set of strategies B of the participants in the second camp through Bayes' rule.

[0183] S804. If the inferred posterior probability does not conflict with the posterior probability, the solution of the perfect Bayesian equilibrium is completed.

[0184] Among them, in this embodiment, it is assumed that the two multi-party camps are the first camp and the second camp respectively, and the formula for the perfect Bayesian equilibrium at the initial game stage is:

[0185]

[0186] Among them, EQ(k) represents the perfect Bayesian equilibrium at the k-th game stage, k = 1 indicates that the current game stage is the initial game stage, A * (B) is the initial strategy set of the first camp; B * (Θ B ) is the initial strategy set of the second camp; is the posterior probability of the type of the participants in the second camp calculated by the first camp according to B * (Θ B ) and A * (B) and using Bayes' rule.

[0187] When the initial game stage ends, the first camp corrects its inference of the participants in the second camp through the posterior probability, that is When entering the next stage, the first camp will select the posterior probability of the perfect Bayesian equilibrium solution in the previous stage as the prior judgment of the type of the participants in camp B, that is

[0188] The implementation principle of one implementation manner of this embodiment is:

[0189] In the initial game stage, the perfect Bayesian game equilibrium is solved. The obtained perfect Bayesian equilibrium includes the initial strategy sets of the first camp and the second camp and the updated posterior probabilities of the types of the participants. The equilibrium strategy corresponds to the optimal strategy of the participants. At the same time, through the solution of the perfect Bayesian game equilibrium, the prior probability of one camp for the other camp is also updated. The posterior probability in the initial game stage will be used as the prior probability in the next stage, thus further affecting the game equilibrium solution and game result in the next stage.

[0190] The embodiment of the present invention also discloses a weapon equipment development planning strategy selection system for multi-party camp games, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the method as described above is implemented.

[0191] The implementation principle of this embodiment is as follows:

[0192] Through the retrieval of the program, based on the confrontation relationship between two multi-party camps, the camp benefits of the two multi-party camps in the initial game stage can be calculated according to the camp parameters. Then, based on the alliance cooperation relationship within the multi-party camps, the optimal distribution results of the camp benefits for all participants are calculated using the clan solution method. Thus, a perfect Bayesian equilibrium is constructed and solved to obtain the optimal strategy selection and posterior probability in the initial game stage. The posterior probability of each current game stage is used as the prior probability of the next game stage, thereby further affecting the strategy selection in the next game stage and forming a dynamic iterative process of incomplete information update - strategy selection. Finally, according to the solution of the perfect Bayesian equilibrium for each game stage, the optimal strategy selection for each game stage is obtained. Compared with the theoretical analysis of single non-cooperative or cooperative games, it can analyze the multi-party game scenario with coexistence of non-cooperation and cooperation more accurately, and thus more accurately select the weapon equipment development planning strategy in the game scenario.

[0193] Those of ordinary skill in the art should understand that the discussion of any above embodiment is only exemplary, and is not intended to imply that the protection scope of the present invention is limited to these examples; under the idea of the present invention, the technical features in the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations in different aspects of one or more embodiments of the present invention as above. For the sake of brevity, they are not provided in detail.

[0194] One or more embodiments of the present invention are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the present invention. Therefore, any omission, modification, equivalent substitution, improvement, etc. made within the spirit and principle of one or more embodiments of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for selecting a development planning strategy of weaponry and equipment in a multi-party camp game, characterized in that The method includes the following steps: Obtain the camp parameters of two multi-party camps participating in the camp game, as well as the initial prior probabilities of the multi-party camps in the camp game and the set of strategies of the weapon equipment development strategies. The initial prior probability is the probability obtained by inferring the types of participants in the opposing camp when one of the multi-party camps is in the initial game stage of the camp game; Calculate the camp benefits in the initial game stage based on the camp parameters; Use the sectarian solution method to calculate the optimal allocation results of all participants in the multi-party camps for the camp benefits; Construct a perfect Bayesian equilibrium according to the optimal allocation results, combined with the set of strategies and the initial prior probabilities; By solving the perfect Bayesian equilibrium, obtain the optimal strategy selection and posterior probability in the initial game stage. The posterior probability is the probability obtained by inferring the types of participants in the opposing camp based on the strategy selection by one of the multi-party camps in the camp game; Take the posterior probability of each current game stage as the prior probability of the next game stage, and based on the solution of the perfect Bayesian equilibrium for each game stage, obtain the optimal strategy selection for each game stage.

2. The method for selecting a development planning strategy for weapons and equipment in a multi-party camp game according to claim 1, wherein The calculation of the camp benefits in the initial game stage based on the camp parameters includes the following steps: Generate the weapon equipment system network of the multi-party camps based on the camp parameters; Combine the weapon equipment system networks of the two multi-party camps to generate a weapon equipment system confrontation network between the two multi-party camps; Evaluate and calculate the camp benefits of the multi-party camps in the initial game stage based on the weapon equipment system confrontation network.

3. The method for selecting a development planning strategy of weaponry and equipment for multi-party camp game according to claim 2, characterized in that, The camp parameters include a camp strategy set, weapon equipment parameters, and participant parameters in the multi-party camps. The generation of the weapon equipment system network of the multi-party camps based on the camp parameters includes the following steps: Generate a plurality of equipment function nodes based on the weapon equipment parameters; Generate equipment function edges between the plurality of equipment function nodes according to the participant parameters; Combine the equipment function nodes and the equipment function edges to generate the weapon equipment system network of the multi-party camps.

4. The method for selecting a development planning strategy of weaponry and equipment for multi-party camp game according to claim 2, characterized in that The evaluation and calculation of the camp benefits of the multi-party camps in the initial game stage based on the weapon equipment system confrontation network includes the following steps: Construct an association matrix between a plurality of equipment function nodes in the multi-party camps based on the weapon equipment system network; Combine the association matrix and the weapon equipment system confrontation network to calculate a plurality of combat ability loops of the multi-party camps; Aggregate the plurality of combat ability loops to calculate the camp benefits of the multi-party camps.

5. The method for selecting a development planning strategy of weapons and equipment for multi-party camp games according to claim 4, characterized in that, The combat ability loops include a general combat ability loop and a standard combat ability loop. The aggregation of the plurality of combat ability loops to calculate the camp benefits of the multi-party camps includes the following steps: Calculate the number of ability loops of the general combat ability loop and the standard combat ability loop respectively; Statistically calculate the first duration of the threat caused by the general combat ability loop to the enemy; Statistically calculate the second duration of the threat caused by the standard combat ability loop to the enemy; Calculate the camp income of the multi-party camp in combination with the number of ability rings, the first duration, and the second duration.

6. The method for selecting a development planning strategy of weaponry and equipment for multi-party camp game according to claim 1, characterized in that, The method of calculating the optimal allocation result of all participants in the multi-party camp for the camp income by using the clan solution method includes the following steps: Calculate the average alliance income of all non-empty alliances in the multi-party camp under the camp income; Solve for the clan alliances of all the participants based on the average alliance income; Determine the quasi-clan alliance of each participant; Solve for the clan core based on Pareto improvement in combination with the clan alliance and the quasi-clan alliance. The clan core is the optimal allocation result of all the participants for the camp income.

7. The method for selecting a development planning strategy of weapons and equipment for multi-party camp game according to claim 1, characterized in that The method of constructing a perfect Bayesian equilibrium according to the optimal allocation result in combination with the strategy set and the initial prior probability includes the following steps: Screen out the initial strategy set in the initial game stage from the strategy set according to the optimal allocation result; Construct a perfect Bayesian equilibrium in combination with the initial strategy set and the initial prior probability.

8. The method for selecting a development planning strategy of weaponry and equipment for multi-party camp game according to claim 7, characterized in that The method of obtaining the optimal strategy selection and posterior probability in the initial game stage by solving the perfect Bayesian equilibrium includes the following steps: Calculate the inferred posterior probability based on the initial strategy set; Calculate and analyze the optimal strategy selection of the two multi-party camps in the initial game stage according to the inferred posterior probability; Combine the optimal strategy selection and the initial prior probability, and calculate the posterior probability in the initial game stage through Bayes' rule; If the inferred posterior probability does not conflict with the posterior probability, the solution of the perfect Bayesian equilibrium is completed.

9. The method for selecting a development planning strategy of weaponry and equipment for multi-party camp game according to claim 7, characterized in that, The two multi-party camps are the first camp and the second camp respectively. The formula of the perfect Bayesian equilibrium is: Among them, A * (B) is the set of the initial strategies of the first camp; B * (Θ B ) is the set of the initial strategies of the second camp; is the posterior probability of the type of the participating party in the second camp calculated by the first camp according to B * (Θ B ) and A * using the Bayesian rule.

10. A weapon and equipment development planning strategy selection system for multi-party camp games, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 1 to 9.

Images