A game confrontation behavior decision-making method, system, terminal device and storage medium

Abstract

This application relates to the field of behavioral decision-making technology, and in particular to a method, system, terminal device, and storage medium for game-based adversarial behavioral decision-making. If the adversarial behavioral decision-making model lacks rule antecedents corresponding to factual data, then the method acquires the factual features corresponding to the factual data; identifies the factual features and obtains corresponding response strategies, where each response strategy refers to an adversarial response plan applicable to the factual features in the current game-based adversarial scenario; determines whether the feasibility probability of the response strategy meets the expected result standard; if the feasibility probability of the response strategy meets the expected result standard, then outputs the behavioral decision result corresponding to the factual data based on the response strategy; and associates the factual data and the corresponding response strategy, adding and storing them in the adversarial behavioral decision-making model. The game-based adversarial behavioral decision-making method, system, terminal device, and storage medium provided in this application can improve the efficiency of making corresponding behavioral decisions during game-based adversarial processes.

Description

Technical Field

[0001] This application relates to the field of behavioral decision-making technology, and in particular to a game-theoretic behavioral decision-making method, system, terminal device and storage medium. Background Technology

[0002] Game theory is a strategic interaction between an opponent and an ally, using their respective strengths to achieve their goals. Behavioral decision-making in this scenario primarily involves assessing the current situation and taking targeted actions. This process is well-suited for expression using production rules, where the "assessment of the current situation" represents the rule antecedent, and the "action taken" represents the rule consequent. When a fact matches a rule antecedent, the conclusion of that rule consequent is reached. Therefore, the set of production rules acts as the "brain" controlling behavior in game theory. The more fully the possible states are considered, the smarter the "brain," resulting in a better final outcome and a larger set of production rules.

[0003] When dealing with the application of large-scale production rules in game-theoretic scenarios, existing technologies typically use persistent storage media such as hard drives to store these large-scale production rules. Then, based on the current facts, the corresponding antecedents in the production rules are matched to derive the consequents of the antecedents, and finally, the corresponding behavioral decision results are obtained. However, due to the complexity and variability of the facts in game-theoretic scenarios, the rule set composed of production rules may not match the current facts, making it impossible to output the corresponding behavioral decision results. As a result, the efficiency of making corresponding behavioral decisions during game-theoretic processes is not high. Summary of the Invention

[0004] To improve the efficiency of making corresponding behavioral decisions during game-based confrontation, this application provides a game-based confrontation behavioral decision-making method, system, terminal device, and storage medium.

[0005] Firstly, this application provides a method for decision-making on adversarial game behavior, comprising the following steps:

[0006] Acquire factual data in the game confrontation, the factual data being used to represent the confrontation state during the game confrontation process;

[0007] Determine whether the rule antecedent corresponding to the stated factual data exists in the adversarial behavior decision model;

[0008] If the rule antecedent corresponding to the factual data does not exist in the adversarial behavior decision model, then the factual feature corresponding to the factual data is obtained;

[0009] Identify the factual characteristics and obtain the corresponding response strategies, wherein the response strategies refer to the confrontation response schemes applicable to the factual characteristics under the current game confrontation scenario;

[0010] Determine whether the feasibility probability of the proposed response strategy meets the expected outcome standard;

[0011] If the feasibility probability of the coping strategy reaches the expected result standard, then the behavioral decision result corresponding to the factual data is output according to the coping strategy;

[0012] Associating the factual data with the corresponding response strategies, generating corresponding target rule antecedents, and adding and storing the target rule antecedents to the adversarial behavior decision model;

[0013] If the feasibility probability of the response strategy does not meet the expected result standard, then the corresponding conservative strategy in the adversarial behavior decision model is output as the behavior decision result.

[0014] By adopting the above technical solution, when there is no rule antecedent corresponding to the current factual data in the adversarial behavior decision model, the factual features corresponding to the actual situation of the displayed factual data are obtained, and the corresponding response strategy is obtained based on the factual features. The response strategy is combined with the backup plan strategy corresponding to the current game adversarial scenario. Then, it is judged whether the feasibility probability of the current response strategy reaches the standard of achieving the expected result corresponding to the current game adversarial scenario. If it does, it means that the response strategy can be used as the corresponding behavior decision result of the factual data, and the behavior decision result is output. The above factual data and its corresponding response strategy are further associated and added to the adversarial behavior decision model. Then, it can be updated in real time based on the response strategy corresponding to the factual data in the actual game adversarial scenario, that is, the behavior decision result, thereby improving the efficiency of making corresponding behavior decision results in the game adversarial process.

[0015] Optionally, determining whether the feasibility probability of the response strategy meets the expected outcome standard includes the following steps:

[0016] If there are multiple response strategies, then obtain the historical applicable events for each response strategy;

[0017] If there are multiple historical applicable events corresponding to each of the aforementioned response strategies, then the response strategies whose number of historical applicable events reaches a preset reference number standard are obtained.

[0018] Obtain the success rate percentage of the historical applicable events corresponding to the aforementioned response strategy;

[0019] The success rate percentage is used as the probability of the feasibility of the response strategy.

[0020] If the feasibility probability reaches the expected success rate corresponding to the expected result standard, then the feasibility probability of the response strategy is determined to have reached the expected result standard.

[0021] If the feasibility probability does not reach the expected success rate corresponding to the expected result standard, then it is determined that the feasibility probability of the response strategy does not reach the expected result standard.

[0022] By adopting the above technical solution, the success rate of historically applicable events in multiple coping strategies is used as the corresponding feasibility probability. Then, it is determined whether the feasibility probability meets the expected success rate corresponding to the expected result standard of the current game confrontation scenario, thereby improving the success probability of the game confrontation behavior.

[0023] Optionally, if the feasibility probability of the coping strategy reaches the expected outcome standard, then outputting the behavioral decision result corresponding to the factual data based on the coping strategy includes the following steps:

[0024] If the feasibility probability of the response strategy reaches the expected result standard, then it is determined whether there are multiple response strategies.

[0025] If there are multiple response strategies, then obtain the risk estimate corresponding to each response strategy;

[0026] Based on the risk estimate, a decision priority is set for the corresponding response strategy, wherein the risk estimate is inversely proportional to the decision priority;

[0027] Based on the decision priority, output each of the aforementioned response strategies.

[0028] By adopting the above technical solution, the decision priority of each response strategy is set according to the risk estimate of each response strategy, thereby improving the security analysis and applicability of the response strategies in actual game confrontation scenarios.

[0029] Optionally, outputting each of the response strategies based on the decision priority includes the following steps:

[0030] If the same decision priority corresponds to multiple response strategies, then obtain the expected damage probability corresponding to each response strategy;

[0031] If the expected damage probability is the same for multiple response strategies, then obtain the damage information corresponding to each expected damage probability.

[0032] Identify the damage information and obtain the corresponding safety level;

[0033] Based on the security level, a decision priority is set for each of the response strategies, wherein the security level is proportional to the decision priority;

[0034] Based on the decision priority, output each of the aforementioned response strategies.

[0035] By adopting the above technical solution, response strategies corresponding to higher security levels can be prioritized, thereby improving the security of the applicable response strategies and the efficiency of obtaining response strategies as the results of corresponding behavioral decisions.

[0036] Optionally, identifying the damage information and obtaining the corresponding security level includes the following steps:

[0037] Identify the damage information and obtain the corresponding associated damage range;

[0038] Based on the associated damage range, the corresponding security level is set and obtained, wherein the associated damage range is inversely proportional to the security level.

[0039] By adopting the above technical solution, the corresponding security level is set according to the associated damage range of the corresponding strategy, thereby minimizing the damage of the associated damage range by combining the security level corresponding to the response strategy.

[0040] Optionally, after determining whether the rule antecedent corresponding to the factual data exists in the adversarial behavior decision model, the following steps are further included:

[0041] If the adversarial behavior decision model contains the rule antecedent corresponding to the factual data, then the corresponding factual feature is obtained based on the factual data;

[0042] Match the feature labels corresponding to the rule antecedents based on the factual features;

[0043] Identify the feature label and obtain the rule consequent corresponding to the rule antecedent;

[0044] Based on the rule consequent, the corresponding behavior decision result is output.

[0045] By adopting the above technical solution, the corresponding rule consequent can be obtained directly from the feature label, and then the corresponding behavior decision result can be output, so as to improve the efficiency of obtaining behavior decision results.

[0046] Optionally, after outputting the corresponding behavioral decision result according to the rule consequent, the method further includes the following steps:

[0047] By associating the factual features, the rule antecedents corresponding to the factual features, and the rule consequents corresponding to the rule antecedents, a corresponding traceability record is generated.

[0048] Based on the current game-playing scenario, the traceability records are archived and stored.

[0049] By adopting the above technical solution, the process of generating corresponding behavioral decision results in the behavioral decision model can be easily traced based on the traceability records, thereby improving the analysis effect of corresponding adversarial behaviors in the corresponding game adversarial scenarios.

[0050] Secondly, this application provides a game-theoretic adversarial behavior decision-making system, including:

[0051] The first acquisition module is used to acquire factual data in the game confrontation, the factual data being used to represent the confrontation state in the game confrontation process;

[0052] The first judgment module is used to determine whether the rule antecedent corresponding to the factual data exists in the adversarial behavior decision model;

[0053] If the adversarial behavior decision model does not contain the rule antecedent corresponding to the fact data, the second acquisition module is used to acquire the fact features corresponding to the fact data.

[0054] The identification module is used to identify the factual features and obtain the corresponding response strategies. The response strategies refer to the confrontation response schemes applicable to the factual features under the current game confrontation scenario.

[0055] The second judgment module is used to determine whether the feasibility probability of the response strategy meets the expected result standard;

[0056] The first output module is configured to output the behavioral decision result corresponding to the factual data based on the coping strategy if the feasibility probability of the coping strategy reaches the expected result standard.

[0057] The association module is used to associate the factual data with the corresponding response strategy, generate the corresponding target rule preconditions, and add and store the target rule preconditions to the adversarial behavior decision model.

[0058] If the feasibility probability of the response strategy does not meet the expected result standard, the second acquisition module is used to output the corresponding conservative strategy in the adversarial behavior decision model as the behavior decision result.

[0059] By adopting the above technical solution, when there is no rule antecedent corresponding to the current factual data in the adversarial behavior decision model, the second acquisition module acquires the factual features corresponding to the actual situation of the displayed factual data, and further identifies the factual features to obtain the corresponding response strategy. The response strategy is a backup plan strategy combined with the current game adversarial scenario. Then, the second judgment module judges whether the feasibility probability of the current response strategy reaches the standard of achieving the expected result corresponding to the current game adversarial scenario. If it does, it means that the response strategy can be used as the corresponding behavior decision result of the factual data, and the first output module outputs the behavior decision result. Furthermore, the association module associates the above factual data and its corresponding response strategy and adds it to the adversarial behavior decision model. Thus, it can be updated in real time based on the response strategy corresponding to the factual data in the actual game adversarial scenario, i.e., the behavior decision result, thereby improving the efficiency of making corresponding behavior decision results in the game adversarial process.

[0060] Thirdly, this application provides a terminal device, which adopts the following technical solution:

[0061] A terminal device includes a memory and a processor. The memory stores computer instructions that can be executed on the processor. When the processor loads and executes the computer instructions, it employs the aforementioned game-theoretic adversarial behavior decision-making method.

[0062] By adopting the above technical solution, computer instructions are generated by the above-mentioned game-theoretic decision-making method and stored in memory for loading and execution by the processor. Thus, a terminal device can be made based on the memory and processor for convenient use.

[0063] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution:

[0064] A computer-readable storage medium storing computer instructions, wherein when the computer instructions are loaded and executed by a processor, the aforementioned game-theoretic adversarial behavior decision-making method is employed.

[0065] By adopting the above technical solution, a game-theoretic decision-making method is used to generate computer instructions, which are then stored in a computer-readable storage medium for loading and execution by a processor. The computer-readable storage medium facilitates the reading and storage of computer instructions.

[0066] In summary, this application includes at least one of the following beneficial technical effects: when there is no rule antecedent corresponding to the current factual data in the adversarial behavior decision model, the factual features corresponding to the actual situation of the displayed factual data are obtained, and the corresponding response strategy is obtained based on the factual features. The response strategy is a backup plan strategy combined with the current game adversarial scenario. Then, it is determined whether the feasibility probability of the current response strategy reaches the standard for achieving the expected result corresponding to the current game adversarial scenario. If it does, it means that the response strategy can be used as the corresponding behavior decision result of the factual data, and the behavior decision result is output. The aforementioned factual data and its corresponding response strategy are further associated and added to the adversarial behavior decision model, so that the response strategy corresponding to the factual data in the actual game adversarial scenario can be updated in real time, thereby improving the efficiency of making corresponding behavior decision results in the game adversarial process. Attached Figure Description

[0067] Figure 1 This is a flowchart illustrating steps S101 to S108 in a game-theoretic adversarial behavior decision-making method according to this application.

[0068] Figure 2 This is a flowchart illustrating steps S201 to S206 in a game-theoretic adversarial behavior decision-making method according to this application.

[0069] Figure 3 This is a flowchart illustrating steps S301 to S304 in a game-theoretic adversarial behavior decision-making method according to this application.

[0070] Figure 4 This is a flowchart illustrating steps S401 to S405 of a game-theoretic adversarial behavior decision-making method according to this application.

[0071] Figure 5 This is a flowchart illustrating steps S501 to S502 in a game-theoretic adversarial behavior decision-making method according to this application.

[0072] Figure 6 This is a flowchart illustrating steps S601 to S604 in a game-theoretic adversarial behavior decision-making method according to this application.

[0073] Figure 7 This is a flowchart illustrating steps S701 to S702 in a game-theoretic adversarial behavior decision-making method according to this application.

[0074] Figure 8 This is a schematic diagram of a decision-making system for game-theoretic adversarial behavior according to this application.

[0075] Explanation of reference numerals in the attached figures:

[0076] 1. First acquisition module; 2. First judgment module; 3. Second acquisition module; 4. Recognition module; 5. Second judgment module; 6. First output module; 7. Association module; 8. Second output module. Detailed Implementation

[0077] The following is in conjunction with the appendix Figure 1-8 This application will be described in further detail.

[0078] This application discloses a method for decision-making in game-theoretic adversarial behavior, such as... Figure 1 As shown, it includes the following steps:

[0079] S101. Obtain factual data in the game confrontation, which is used to represent the confrontation state in the game confrontation process;

[0080] S102. Determine whether there are rule antecedents corresponding to factual data in the adversarial behavior decision-making model;

[0081] S103. If there are no rule antecedents corresponding to the factual data in the adversarial behavior decision model, then obtain the factual features corresponding to the factual data;

[0082] S104. Identify factual characteristics and obtain corresponding response strategies. Response strategies refer to the confrontation response plans applicable based on the factual characteristics in the current game confrontation scenario.

[0083] S105. Determine whether the feasibility probability of the response strategy meets the expected outcome standard;

[0084] S106. If the probability of the coping strategy's feasibility reaches the expected outcome standard, then output the behavioral decision result corresponding to the factual data based on the coping strategy;

[0085] S107. Link factual data and corresponding response strategies, and add them to the adversarial behavior decision-making model;

[0086] S108. If the probability of the coping strategy does not meet the expected outcome standard, then output the corresponding conservative strategy in the adversarial behavior decision model as the behavior decision result.

[0087] In practical application, for ease of explanation, the game-playing scenario is a simulated beach exercise. The purpose of the simulated beach exercise is to dominate the entire beach area. The factual data in step S101 refers to the adversarial state at each stage of the current game-playing scenario. For example, the adversarial states included in the simulated beach exercise are: gaining air superiority, establishing a safe zone, increasing the depth of coverage, consolidating the current coverage area, expanding the current coverage area, and dominating the entire beach area. Each of these adversarial states is represented by corresponding factual data. For another example, the factual data could be the discovery of an abnormal intruder during reconnaissance of the beach airspace; the corresponding adversarial state is gaining air superiority.

[0088] Furthermore, in order to analyze whether there are corresponding adversarial behavior decisions in the current real-time data, it is determined whether there is a rule antecedent corresponding to the factual data in the adversarial behavior decision model. The adversarial behavior decision model is established based on the current simulated adversarial scenario. The adversarial behavior decision model stores the "judgment of the current situation" which can be represented as the rule antecedent and the "action taken" which can be represented as the rule consequent in the current simulated adversarial scenario. When there is factual data that can match a rule antecedent, the conclusion of the rule consequent is obtained.

[0089] For example, in the current simulated confrontation scenario, the entire beach area is dominated. When the factual data is that an abnormal intruder is discovered during the reconnaissance of the area above the beach, the behavioral decision model corresponding to this simulated confrontation scenario analyzes and judges the factual data. The corresponding rule antecedent is to take an image of the abnormal intruder and upload it to the control center. The image of the abnormal intruder can be obtained by taking a picture of the reconnaissance camera set on the flight device. The control center refers to the link that plays a command role in the current simulated confrontation scenario. Therefore, it can be determined that the behavioral decision model of this confrontation has a rule antecedent corresponding to the factual data.

[0090] Furthermore, in order to make rapid behavioral decisions in response to emergencies, which are situations where there are no rule antecedents corresponding to factual data in the adversarial behavior decision-making model, we further identify the factual features corresponding to real-time data and obtain corresponding response strategies. Factual features refer to the key information corresponding to factual data, and response strategies refer to the general way of dealing with these factual features. The adversarial behavior decision-making model corresponding to these factual features does not have such a response method.

[0091] For example, if the factual data is the first environmental exploration of a beach and dense forest area, and the corresponding adversarial state is exploring the beach and dense forest environment, and the adversarial behavior decision model analyzes and judges that the corresponding rule antecedent is blank, then it can be determined that there is no rule antecedent corresponding to the factual data in the adversarial behavior decision model. Further, the factual characteristics corresponding to the factual data are obtained as first time, beach and dense forest, and exploration. The corresponding response strategy is to equip the appropriate detection device according to the area of ​​the beach and dense forest.

[0092] Furthermore, in order to ensure that the response strategy can achieve the corresponding expected results, it is further determined whether the feasibility probability of the response strategy reaches the corresponding expected result standard. The feasibility probability refers to the success probability of the response strategy in achieving the corresponding expected result, and the expected result standard refers to the success probability standard that the response strategy needs to achieve in order to achieve the corresponding expected result.

[0093] For example, the response strategy is to equip 3 detection devices to detect a 10-hectare beach forest. Considering the unknown situation in the beach forest, which can be identified as an interfering party, the system analyzes that the corresponding feasibility probability is 85%. The corresponding expected result standard is that the success rate of equipping 3 detection devices to detect a 10-hectare beach forest must reach 80%. The purpose of setting the expected result standard is to maximize the success of the behavioral decision corresponding to the factual data. Therefore, it can be determined that the feasibility probability of this response strategy has reached the corresponding expected result standard. Then, the behavioral decision result corresponding to the factual data of the first beach forest area environmental detection is further output as: based on equipping 3 detection devices to detect a 10-hectare beach forest.

[0094] In order to update the adversarial behavior decision model corresponding to the simulated adversarial scenario in real time, the factual data and the corresponding response strategies are further linked and added to the adversarial behavior decision model. This allows the corresponding response strategies to be directly applied when judging the same factual data again, thereby improving the efficiency of obtaining the corresponding behavior decision results.

[0095] For example, if the system analyzes and finds that the feasibility probability of the above-mentioned coping strategy is 75%, then it can be determined that the feasibility probability of the coping strategy has reached the corresponding expected result standard, and the corresponding conservative strategy in the adversarial behavior decision model is output as the behavior decision result.

[0096] The adversarial behavior decision model stores conservative strategies corresponding to all factual data. When there are no rule antecedents corresponding to factual data in the adversarial behavior decision model, and the feasibility probability of the corresponding response strategy does not reach the corresponding expected result standard, then the conservative strategy is output as the behavior decision result.

[0097] For example, if the above-mentioned coping strategy involves equipping three detection devices to detect a 10-hectare beach forest, and the corresponding feasibility probability does not meet the corresponding expected result standard, then the corresponding conservative strategy is to not apply this coping strategy, select other coping strategies, and analyze their feasibility probabilities until the selected coping strategy meets the corresponding expected result standard.

[0098] This embodiment provides a game-theoretic adversarial behavior decision-making method. When the adversarial behavior decision-making model does not have a rule antecedent corresponding to the current factual data, it acquires factual features corresponding to the actual situation of the displayed factual data, and further acquires corresponding response strategies based on these factual features. These response strategies are combined with backup scheme strategies corresponding to the current game-theoretic scenario. Then, it is determined whether the feasibility probability of the current response strategy reaches the standard for achieving the expected result corresponding to the current game-theoretic scenario. If it does, it indicates that the response strategy can be used as the behavior decision result corresponding to the factual data, and the behavior decision result is output. Furthermore, the aforementioned factual data and its corresponding response strategy are associated and added to the adversarial behavior decision-making model. Thus, it can be updated in real time based on the response strategy corresponding to the factual data in the actual game-theoretic scenario, i.e., the behavior decision result, thereby improving the efficiency of making corresponding behavior decisions during the game-theoretic process.

[0099] In one embodiment of this example, such as Figure 2 As shown, step S105, which determines whether the feasibility probability of the response strategy meets the expected outcome standard, includes the following steps:

[0100] S201. If there are multiple response strategies, obtain the historical applicable events for each response strategy;

[0101] S202. If there are multiple historical applicable events corresponding to each response strategy, then obtain the response strategies whose number of historical applicable events reaches the preset reference number standard.

[0102] S203. Obtain the success rate percentage of historical applicable events corresponding to the response strategy;

[0103] S204. Use the success rate percentage as the probability of the feasibility of the response strategy;

[0104] S205. If the probability of feasibility reaches the expected success rate corresponding to the expected result standard, then the probability of feasibility of the response strategy is determined to have reached the expected result standard.

[0105] S206. If the probability of feasibility does not reach the expected success rate corresponding to the expected result standard, then the probability of feasibility of the response strategy is determined to have not reached the expected result standard.

[0106] In practical applications, if there are multiple coping strategies, in order to comprehensively evaluate them, we need to obtain the historical applicable events corresponding to each coping strategy. Historical applicable events refer to the game confrontation scenarios in which the coping strategy was historically applicable. Since the number of historical applicable events is more convincing for the feasibility of the corresponding coping strategy, we select the coping strategies among multiple coping strategies whose number of corresponding historical applicable events reaches the preset reference number standard. The preset reference number standard refers to the standard set in advance for selecting the corresponding coping strategy based on the number of historical applicable events.

[0107] Furthermore, the success rate of historical applicable events is used as the probability of the feasibility of the corresponding response strategy. The success rate of historical applicable events refers to the probability that the response strategy is successfully applied in the corresponding historical applicable events. Successful application in the corresponding historical applicable events means that the response strategy has achieved the expected results of the corresponding historical applicable events.

[0108] Furthermore, if the probability of feasibility reaches the expected success rate corresponding to the expected outcome standard, then the probability of feasibility of the response strategy is determined to have reached the expected outcome standard. The expected success rate corresponding to the expected outcome standard refers to the success rate that the response strategy should achieve as stipulated by the expected outcome standard.

[0109] For example, if there are 10 historical events for which strategy A is applicable, with 8 successful and 2 unsuccessful, the success rate is 80%, meaning the probability of feasibility is 80%. If the expected success rate is 90% for the expected outcome standard, then it can be determined that the probability of feasibility of this strategy does not meet the expected outcome standard.

[0110] For example, if strategy A has 10 applicable historical events, 9 of which were successful and 2 were unsuccessful, then the success rate is 90%, and the feasibility of the strategy can be determined to have reached the expected result standard.

[0111] The game-confrontation behavior decision-making method provided in this embodiment uses the success rate percentage of historically applicable events among multiple response strategies as the corresponding feasibility probability, and then judges whether the feasibility probability meets the expected success rate corresponding to the expected result standard of the current game-confrontation scenario, thereby improving the success probability of the game-confrontation behavior outcome.

[0112] In one embodiment of this example, such as Figure 3 As shown, step S106, which states that if the feasibility probability of the coping strategy reaches the expected outcome standard, then the behavioral decision result corresponding to the factual data output based on the coping strategy includes the following steps:

[0113] S301. If the probability of feasibility of the response strategy reaches the expected result standard, then determine whether there are multiple response strategies;

[0114] S302. If there are multiple response strategies, obtain the risk estimate corresponding to each response strategy;

[0115] S303. Based on the risk assessment, set the decision priority of the corresponding response strategy, with the risk assessment and decision priority being inversely proportional;

[0116] S304. Output each response strategy according to the decision priority.

[0117] In practical applications, in order to comprehensively evaluate multiple response strategies and further obtain the risk estimate corresponding to each response strategy, the risk estimate refers to the risk value corresponding to the applicable response strategy in the simulated confrontation scenario. In order to deal with the corresponding factual data in the simulated confrontation scenario more efficiently, the decision priority is set according to the risk estimate corresponding to the response strategy. The risk estimate is inversely proportional to the decision priority. The higher the risk estimate, the lower the probability that the response strategy will achieve the corresponding expected result.

[0118] For example, strategy A involves deploying three detection devices to survey a 10-hectare beach and dense forest area, with a corresponding risk estimate of 86. Strategy B involves deploying three detection devices to survey the same 10-hectare beach and dense forest area, and also deploying a data receiver / follower to acquire real-time information from the three detection devices, with a corresponding risk estimate of 80. Furthermore, strategy B is assigned a priority of Level 1, while strategy A is assigned a priority of Level 2, with Level 1 having a higher priority than Level 2. Strategy B is therefore prioritized for implementation.

[0119] The game-theoretic decision-making method provided in this embodiment sets the corresponding decision priority according to the risk estimate of each response strategy, thereby improving the security analysis and applicability of the response strategies in actual game-theoretic scenarios.

[0120] In one embodiment of this example, such as Figure 4 As shown, step S304, which outputs each response strategy according to the decision priority, includes the following steps:

[0121] S401. If the same decision priority corresponds to multiple response strategies, then obtain the expected damage probability corresponding to each response strategy;

[0122] S402. If multiple response strategies correspond to the same expected damage probability, then obtain the damage information corresponding to each expected damage probability.

[0123] S403. Identify damage information and obtain the corresponding safety level;

[0124] S404. Based on the security level, set the decision priority corresponding to each response strategy, with the security level being directly proportional to the decision priority;

[0125] S405. Output each response strategy according to the decision priority.

[0126] In practical applications, to comprehensively evaluate multiple response strategies, it is necessary to further obtain the expected damage probability corresponding to each strategy. The expected damage probability refers to the probability that our side will suffer losses or damage during the application of each response strategy. For example, the response strategy is to equip three detection devices to detect a 10-hectare beach and dense forest. During the detection process, there is a certain probability that the three detection devices will encounter unexpected situations and suffer damage.

[0127] Furthermore, when multiple response strategies have the same expected damage probability, the corresponding damage information is obtained, and the corresponding security level is set according to the damage information. The damage information refers to the specific damage suffered by our side in the response strategy.

[0128] For example, strategy A involves deploying three detection devices to survey a 10-hectare beach and dense forest area, while strategy B involves deploying three detection devices to survey the same area, plus a data receiver follower. Strategy A corresponds to damage information for two detection devices, and strategy B also corresponds to damage information for two detection devices. Since both strategies aim to acquire information about the beach and dense forest, and although both involve damage to two devices, the data receiver follower in strategy B ensures the integrity of the acquired information. Therefore, strategy B has a higher security level than strategy A. Similarly, strategy B has a higher decision priority than strategy A, and strategy B is output first.

[0129] The game-theoretic behavior decision-making method provided in this embodiment can prioritize outputting response strategies corresponding to higher security levels, thereby improving the security of the applicable response strategies and the efficiency of obtaining response strategies as corresponding behavior decision results.

[0130] In one embodiment of this example, such as Figure 5 As shown, step S403, which involves identifying damage information and obtaining the corresponding safety level, includes the following steps:

[0131] S501. Identify damage information and obtain the corresponding associated damage range;

[0132] S502. Set and obtain the corresponding safety level based on the associated damage range. The associated damage range is inversely proportional to the safety level.

[0133] In practical applications, in order to comprehensively analyze damage information and accurately obtain its corresponding security level, damage information is identified and the corresponding associated damage range is obtained. The associated damage range refers to the scope of our losses or damages in the response strategy.

[0134] For example, strategy A involves deploying three detection devices to survey a 10-hectare beach and dense forest area, while strategy B involves deploying three detection devices to survey the same area, plus a data receiver follower. In strategy A, the three detection devices include one scanning robot and two pioneering robots. Only after the two pioneering robots have explored the area can the scanning robot scan the cleared terrain to obtain the corresponding detection information. If the two pioneering robots are damaged and lose their pioneering function, the scanning robot will be unable to perform terrain scanning tasks. Therefore, strategy A has a larger associated damage range, meaning a lower retention rate of scanned information and a higher information loss rate.

[0135] Furthermore, all three detection devices in Response Strategy B are pioneering scanning robots, which can acquire corresponding scanning information while pioneering. Even if two pioneering scanning robots are damaged, the acquired scanning information can still be transmitted to the corresponding data receiver follower, and the corresponding scanning information can be obtained through the relay transmission of the data receiver follower. It can be determined that the associated damage range of Response Strategy B is smaller than that of Response Strategy A, and the loss rate of scanning information is lower. Therefore, the safety level of Response Strategy B is higher than that of Response Strategy A.

[0136] The game-theoretic decision-making method provided in this embodiment sets the corresponding security level according to the associated damage range of the corresponding strategy, thereby minimizing the associated damage range by combining the security level of the response strategy.

[0137] In one embodiment of this example, such as Figure 6 As shown, after step S102, which determines whether there are rule antecedents corresponding to factual data in the adversarial behavior decision model, the following steps are also included:

[0138] S601. If there are rule antecedents corresponding to factual data in the adversarial behavior decision model, then obtain the corresponding factual features based on the factual data;

[0139] S602. Match the feature labels corresponding to the antecedents of the factual features according to the factual features matching rules;

[0140] S603. Identify feature labels and obtain the rule consequent corresponding to the rule antecedent;

[0141] S604. Based on the rule consequent, output the corresponding behavioral decision result.

[0142] In practical applications, in order to improve the efficiency of matching factual data with rule antecedents in adversarial behavior decision-making models, rule antecedents are stored in adversarial decision-making models as a data model of label attribute graphs, i.e. feature labels. The label attribute graphs include indexes, which are used to indicate the storage location of rule consequents.

[0143] If there are rule antecedents corresponding to factual data in the adversarial behavior decision model, then the corresponding factual features are obtained based on the factual data. Furthermore, the corresponding rule antecedents can be matched by matching the label attribute graph with the same attributes in the adversarial decision model based on the attribute information of the factual features. Then, the feature labels, i.e. the label attribute graph, are identified, the index is obtained, and the corresponding rule consequents are obtained according to the corresponding storage location. After confirming the rule consequents, the corresponding behavior decision results are output.

[0144] The game-theoretic behavior decision-making method provided in this embodiment, such as Figure 7 As shown, after step S604, which outputs the corresponding behavioral decision result based on the rule consequent, the following steps are also included:

[0145] S701. Associate the factual features, the rule antecedents corresponding to the factual features, and the rule consequents corresponding to the rule antecedents to generate the corresponding traceability records;

[0146] S702. Based on the current game-playing scenario, archive and store the traceability records.

[0147] In practical applications, in order to implement the identification and matching output process of the corresponding behavioral rules in the adversarial behavior decision-making model, the factual features, the rule antecedents corresponding to the factual features, and the rule consequents corresponding to the rule antecedents are associated with the factual features after the adversarial behavior decision-making model has been analyzed and judged, and corresponding traceability records are generated. Furthermore, according to the current game adversarial scenario, the traceability records formed above are archived and stored so as to facilitate the traceability of the process of relevant factual data judgment, rule matching, and behavioral decision result output in the adversarial behavior decision-making model.

[0148] The game-theoretic behavior decision-making method provided in this embodiment improves the analysis effect of corresponding adversarial behaviors in the corresponding game-theoretic scenarios by tracing the generation process of corresponding behavioral decision-making results in the behavioral decision-making model through tracing records.

[0149] This application discloses a game-theoretic adversarial behavior decision-making system, such as... Figure 8 As shown, it includes the following steps:

[0150] The first acquisition module 1 is used to acquire factual data in the game confrontation, and the factual data is used to represent the confrontation state in the game confrontation process;

[0151] The first judgment module 2 is used to determine whether there are rule antecedents corresponding to factual data in the adversarial behavior decision model;

[0152] If there is no rule antecedent corresponding to the factual data in the adversarial behavior decision model, the second acquisition module 3 is used to acquire the factual features corresponding to the factual data.

[0153] Identification module 4 is used to identify factual features and obtain corresponding response strategies. The response strategy refers to the confrontation response plan applicable to the factual features in the current game confrontation scenario.

[0154] The second judgment module 5 is used to judge whether the feasibility probability of the response strategy meets the expected result standard;

[0155] The first output module 6, if the feasibility probability of the response strategy reaches the expected result standard, is used to output the behavioral decision result corresponding to the factual data based on the response strategy.

[0156] The association module 7 is used to associate factual data with corresponding response strategies, generate corresponding target rule antecedents, and add and store the target rule antecedents to the adversarial behavior decision model.

[0157] If the feasibility probability of the response strategy does not meet the expected result standard, the second acquisition module 3 is used to output the corresponding conservative strategy in the adversarial behavior decision model as the behavior decision result.

[0158] The game-theoretic behavior decision-making system provided in this embodiment, when there is no rule antecedent corresponding to the current factual data in the adversarial behavior decision-making model, acquires the factual features corresponding to the actual situation of the displayed factual data through the second acquisition module 3, and further identifies the factual features through the identification module 4 to obtain the corresponding response strategy. The response strategy is a backup plan strategy combined with the current game-theoretic scenario. Then, the second judgment module 5 judges whether the feasibility probability of the current response strategy reaches the standard for achieving the expected result corresponding to the current game-theoretic scenario. If it does, it means that the response strategy can be used as the behavior decision result corresponding to the factual data, and outputs the behavior decision result through the first output module 6. Further, the association module 7 associates the above factual data and its corresponding response strategy and adds it to the adversarial behavior decision-making model. Thus, it can be updated in real time based on the response strategy corresponding to the factual data in the actual game-theoretic scenario, i.e., the behavior decision result, thereby improving the efficiency of making corresponding behavior decisions in the game-theoretic process.

[0159] It should be noted that the game-playing adversarial behavior decision-making system provided in this application embodiment also includes each module and / or corresponding sub-module corresponding to the logical function or logical step of any of the above-mentioned game-playing adversarial behavior decision-making methods, achieving the same effect as each logical function or logical step, which will not be elaborated here.

[0160] This application also discloses a terminal device, including a memory, a processor, and computer instructions stored in the memory and capable of running on the processor, wherein when the processor executes the computer instructions, it employs any of the game-theoretic decision-making methods described in the above embodiments.

[0161] The terminal device can be a computer device such as a desktop computer, a laptop computer, or a cloud server. The terminal device includes, but is not limited to, a processor and a memory. For example, the terminal device may also include input / output devices, network access devices, and buses.

[0162] The processor can be a central processing unit (CPU). Of course, depending on the actual use, it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc., and this application does not limit it.

[0163] The memory can be an internal storage unit of the terminal device, such as a hard disk or RAM of the terminal device, or an external storage device of the terminal device, such as a plug-in hard disk, smart memory card (SMC), secure digital card (SD), or flash memory card (FC) equipped on the terminal device. Furthermore, the memory can be a combination of internal storage units and external storage devices of the terminal device. The memory is used to store computer instructions and other instructions and data required by the terminal device. The memory can also be used to temporarily store data that has been output or will be output. This application does not limit this.

[0164] In this terminal device, any of the game-playing adversarial behavior decision-making methods in the above embodiments can be stored in the memory of the terminal device and loaded and executed on the processor of the terminal device for convenient use.

[0165] This application also discloses a computer-readable storage medium, which stores computer instructions, wherein when the computer instructions are executed by a processor, any of the game-theoretic decision-making methods described in the above embodiments are employed.

[0166] The computer instructions can be stored in a computer-readable medium. The computer instructions include computer instruction code, which can be in the form of source code, object code, executable file, or certain middleware. The computer-readable medium includes any entity or device capable of carrying computer instruction code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the computer-readable medium includes, but is not limited to, the above-mentioned components.

[0167] In this computer-readable storage medium, any of the game-playing adversarial behavior decision-making methods in the above embodiments are stored in the computer-readable storage medium and loaded and executed on the processor to facilitate the storage and application of the above methods.

[0168] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A game-theoretic adversarial behavior decision-making method, characterized in that, Includes the following steps: Acquire factual data in the game confrontation, the factual data being used to represent the confrontation state during the game confrontation process; Determine whether the rule antecedent corresponding to the stated factual data exists in the adversarial behavior decision model; If the rule antecedent corresponding to the factual data does not exist in the adversarial behavior decision model, then the factual feature corresponding to the factual data is obtained; Identify the factual characteristics and obtain the corresponding response strategies, wherein the response strategies refer to the confrontation response schemes applicable to the factual characteristics under the current game confrontation scenario; Determine whether the feasibility probability of the proposed response strategy meets the expected outcome standard; If the feasibility probability of the coping strategy reaches the expected result standard, then the behavioral decision result corresponding to the factual data is output according to the coping strategy; Associate the factual data with the corresponding response strategies and add them to the adversarial behavior decision model; If the feasibility probability of the response strategy does not meet the expected result standard, then the corresponding conservative strategy in the adversarial behavior decision model is output as the behavior decision result.

2. The game-theoretic adversarial behavior decision-making method according to claim 1, characterized in that, The process of determining whether the feasibility probability of the response strategy meets the expected outcome standard includes the following steps: If there are multiple response strategies, then obtain the historical applicable events for each response strategy; If there are multiple historical applicable events corresponding to each of the aforementioned response strategies, then the response strategies whose number of historical applicable events reaches a preset reference number standard are obtained. Obtain the success rate percentage of the historical applicable events corresponding to the aforementioned response strategy; The success rate percentage is used as the probability of the feasibility of the response strategy. If the feasibility probability reaches the expected success rate corresponding to the expected result standard, then the feasibility probability of the response strategy is determined to have reached the expected result standard. If the feasibility probability does not reach the expected success rate corresponding to the expected result standard, then it is determined that the feasibility probability of the response strategy does not reach the expected result standard.

3. The game-theoretic adversarial behavior decision-making method according to claim 1, characterized in that, If the feasibility probability of the coping strategy reaches the expected outcome standard, then outputting the behavioral decision result corresponding to the factual data based on the coping strategy includes the following steps: If the feasibility probability of the response strategy reaches the expected result standard, then it is determined whether there are multiple response strategies. If there are multiple response strategies, then obtain the risk estimate corresponding to each response strategy; Based on the risk estimate, a decision priority is set for the corresponding response strategy, wherein the risk estimate is inversely proportional to the decision priority; Based on the decision priority, output each of the aforementioned response strategies.

4. The game-theoretic adversarial behavior decision-making method according to claim 3, characterized in that, The step of outputting each of the response strategies based on the decision priority includes the following steps: If the same decision priority corresponds to multiple response strategies, then obtain the expected damage probability corresponding to each response strategy; If the expected damage probability is the same for multiple response strategies, then obtain the damage information corresponding to each expected damage probability. Identify the damage information and obtain the corresponding safety level; Based on the security level, a decision priority is set for each of the response strategies, wherein the security level is proportional to the decision priority; Based on the decision priority, output each of the aforementioned response strategies.

5. The game-theoretic adversarial behavior decision-making method according to claim 4, characterized in that, The process of identifying the damage information and obtaining the corresponding security level includes the following steps: Identify the damage information and obtain the corresponding associated damage range; Based on the associated damage range, the corresponding security level is set and obtained, wherein the associated damage range is inversely proportional to the security level.

6. The game-theoretic adversarial behavior decision-making method according to claim 1, characterized in that, After determining whether the rule antecedent corresponding to the factual data exists in the adversarial behavior decision model, the following steps are also included: If the adversarial behavior decision model contains the rule antecedent corresponding to the factual data, then the corresponding factual feature is obtained based on the factual data; Match the feature labels corresponding to the rule antecedents based on the factual features; Identify the feature label and obtain the rule consequent corresponding to the rule antecedent; Based on the rule consequent, the corresponding behavior decision result is output.

7. The game-theoretic adversarial behavior decision-making method according to claim 6, characterized in that, After outputting the corresponding behavior decision result according to the rule consequent, the following steps are also included: By associating the factual features, the rule antecedents corresponding to the factual features, and the rule consequents corresponding to the rule antecedents, a corresponding traceability record is generated. Based on the current game-playing scenario, the traceability records are archived and stored.

8. A game-theoretic adversarial behavior decision-making system, characterized in that, include: The first acquisition module (1) is used to acquire factual data in the game confrontation, wherein the factual data is used to represent the confrontation state in the game confrontation process; The first judgment module (2) is used to determine whether there is a rule antecedent corresponding to the factual data in the adversarial behavior decision model; If the rule antecedent corresponding to the fact data does not exist in the adversarial behavior decision model, the second acquisition module (3) is used to acquire the fact features corresponding to the fact data. The identification module (4) is used to identify the factual features and obtain the corresponding response strategy. The response strategy refers to the confrontation response scheme applicable to the factual features under the current game confrontation scenario. The second judgment module (5) is used to judge whether the feasibility probability of the response strategy meets the expected result standard; First output module (6): If the feasibility probability of the coping strategy reaches the expected result standard, the output module is used to output the behavioral decision result corresponding to the factual data according to the coping strategy; The association module (7) is used to associate the factual data and the corresponding response strategy, generate the corresponding target rule antecedent, and add and store the target rule antecedent to the adversarial behavior decision model; If the feasibility probability of the coping strategy does not reach the expected result standard, the second acquisition module (3) is used to output the conservative strategy corresponding to the adversarial behavior decision model as the behavior decision result.

9. A terminal device, comprising a memory and a processor, characterized in that, The memory stores computer instructions that can run on the processor. When the processor loads and executes the computer instructions, it employs a game-theoretic decision-making method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing computer instructions, characterized in that, When the computer instructions are loaded and executed by the processor, a game-theoretic behavior decision-making method as described in any one of claims 1 to 7 is employed.

Images