An Agent Situation Information Enhancement Method and System Based on Knowledge Reasoning

Abstract

The present application discloses an intelligent agent situation information enhancement method and system based on knowledge reasoning. The method includes: determining the initial situation information obtained by a target intelligent agent perceiving the current confrontation environment; constructing rule knowledge for each intelligent agent in the confrontation environment based on first-order predicate logic; converting the initial situation information into a formal expression of first-order predicate logic to obtain formally compatible situation information; performing knowledge reasoning on the formally compatible situation information based on the constructed rule knowledge to obtain the enhanced situation information of the target intelligent agent; wherein, the enhanced situation information includes hidden situation information obtained through knowledge reasoning. In this way, based on the objectively observed situation information, with the knowledge of domain experts, potential but unobservable information is inferred, so as to support the analysis and decision-making of intelligent agents with more comprehensive information.

Description

Technical Field

[0001] This application relates to the field of computer technology, specifically to technical fields such as game confrontation, and particularly to an intelligent agent situation information enhancement method and system based on knowledge reasoning. Background Art

[0002] The application of intelligent agent game confrontation is very extensive, covering multiple fields, including game AI, autonomous driving, financial transactions, network security, etc., providing powerful tools and models for solving real-world problems, and having important roles and far-reaching significance especially in formulating simulation decisions, creating economic value, enhancing security, and promoting social influence and research.

[0003] In game confrontation, rule-based confrontation and machine learning model-based confrontation are two common strategy design methods. The rule-based confrontation method mainly relies on manually designed rules or strategies to guide the behavior of intelligent agents. Among them, the rules are usually formulated based on game theory, professional knowledge or experience. The advantage of the rule-based confrontation method is high transparency, and the behavior of the intelligent agent is easy to understand and explain; however, the design of the rules may be restricted and it is difficult to cope with complex, unknown or changing environments. The machine learning model-based confrontation method uses machine learning algorithms to train intelligent agents to automatically learn and optimize strategies in the game. This method can handle more complex and dynamic environments and can learn from a large amount of data. But the disadvantage is that the internal decision-making process of the model may not be transparent and professional knowledge is required to explain the behavior of the model.

[0004] Regardless of which game confrontation strategy is adopted, it is very important to obtain complete and accurate situation information. The process of intelligent agent game confrontation basically includes four parts: observation, analysis, decision-making, and action, as Figure 1 shown. The above two strategies focus on the analysis and decision-making parts, and observation, as the first link, has an important impact on analysis and decision-making. Observation mainly refers to the changes in the opponent, ourselves, and the environment objectively perceived by the intelligent agent, that is, situation information. Through observation, the intelligent agent collects information and builds a model of the current state in order to make more reasonable decisions. Therefore, whether the observed information is accurate and complete is very important and directly affects the rationality and effectiveness of the decision-making.

[0005] Due to the adversarial nature of game confrontation, the situation information obtained through intuitive observation may not be complete and accurate. Game confrontation is an adversarial strategic interaction in which the opponent and our side use their respective force units to achieve their respective goals. Both sides often, under the guidance of their respective macro strategies, comprehensively use all entities for division of labor and cooperation, hide their true intentions, and gradually advance towards their own goals. For example, in real-time strategy games, the opponent may feign an attack in one direction while actually attacking in another, and deliberately hide the cooperation relationship between important characters. If one can distinguish the true from the false in the fog of confrontation and accurately infer the hidden information, it will help to adopt effective confrontation strategies. Summary of the Invention

[0006] This application provides an intelligent agent situation information enhancement method and system based on knowledge reasoning to solve the problem that the situation information of intelligent agents directly obtained in the existing game confrontation process may be incomplete and inaccurate, thus affecting confrontation decisions.

[0007] The technical solutions are as follows:

[0008] In a first aspect, an intelligent agent situation information enhancement method based on knowledge reasoning is provided, including:

[0009] Determine the initial situation information obtained by the target intelligent agent when perceiving the current confrontation environment;

[0010] Based on first-order predicate logic, construct rule knowledge for each intelligent agent in the confrontation environment; among them, the rule knowledge constructed for each intelligent agent can formally express and reason about the intelligent agent, its capabilities, and relationships.

[0011] Convert the initial situation information into a formal expression of first-order predicate logic to obtain situation information with compatible forms.

[0012] Based on the constructed rule knowledge, perform knowledge reasoning on the situation information with compatible forms to obtain the enhanced situation information of the target intelligent agent; among them, the enhanced situation information includes hidden situation information obtained through knowledge reasoning.

[0013] In a possible implementation, the method further includes: sending the enhanced situation information to the target intelligent agent so that the target intelligent agent can use the enhanced situation information for confrontation decisions.

[0014] In a possible implementation, first-order predicate logic includes multiple basic elements: individual constants, predicates, quantifiers, and logical connectives; among them, an individual constant refers to an intelligent agent participating in game confrontation in an adversarial environment; a predicate is used to describe the properties and / or relationships of intelligent agents; quantifiers include universal quantifiers and existential quantifiers, where the universal quantifier indicates that the predicate applies to all intelligent agents, and the existential quantifier indicates that the predicate applies to at least one intelligent agent; logical connectives are used to combine atomic propositions to form more complex propositions.

[0015] In a possible implementation, different intelligent agents in the adversarial environment are divided into different role types according to their capabilities: air-attack role, ground-attack role, and escort role; among them, the main ability of the air-attack role is to attack air targets; the main ability of the ground-attack role is to attack ground targets; the main ability of the escort role is to protect other roles from being attacked.

[0016] In a possible implementation, in the game confrontation game scenario, based on first-order predicate logic, rule knowledge is constructed for each intelligent agent in the adversarial environment, including:

[0017] If the intelligent agent is a ground-attack role, then the intelligent agent has no air-attack ability, and the corresponding first rule knowledge is constructed as: ;

[0018] If the intelligent agent is an air-attack role, then the intelligent agent has air-attack ability, and the corresponding second rule knowledge is constructed as: ;

[0019] If the intelligent agent has no air-attack ability but enters the warning range of the opponent's air-attack role, then the intelligent agent has an escort role, and the corresponding third rule knowledge is constructed as: ;

[0020] If the intelligent agent has an escort role and is not observed, it is speculated that the escort role is a stealth intelligent agent, and the corresponding fourth rule knowledge is constructed as: .

[0021] In a possible implementation, the initial situation information is transformed into a formal expression of first-order predicate logic to obtain situation information with formal compatibility, specifically including: determining the first intelligent agent and the second intelligent agent of the adversarial game included in the initial situation information; based on first-order predicate logic, determining the situation information with formal compatibility including the first intelligent agent and the second intelligent agent; among them, the situation information with formal compatibility includes: first situation information: the first intelligent agent is a ground-attack role: ; second situation information: the second intelligent agent F is an air-attack role: ; Third situation information: first agent and the second agent It is an adversarial relationship: ; Fourth situation information: the first agent The distance from the second agent F: ; Fifth situation information: second agent Warning range: .

[0022] In a possible implementation, based on the constructed rule knowledge, knowledge reasoning is performed on the form-compatible situation information to obtain enhanced situation information of the target intelligent agent, specifically including:

[0023] Based on the first rule knowledge and the fifth situation information, the first reasoning information is obtained by reasoning: the first agent No anti-air attack capability: ;

[0024] Based on the second rule knowledge and the second situation information, the second reasoning information is obtained by reasoning: the second agent Has anti-air attack capability: ;

[0025] Based on the fourth situation information and the fifth situation information, the third reasoning information is obtained by reasoning: the first agent Entered the second agent Warning range: ;

[0026] Based on the first reasoning information, the second reasoning information, the third situation information, and the third reasoning information, the fourth reasoning information is obtained by reasoning: the first agent There is escort role y: ;

[0027] If the escort role y is not observed, then there is the fifth inference information: ;

[0028] Based on the first situation information, the fourth reasoning information, and the fifth reasoning information, the sixth reasoning information is inferred: the escort character y is a stealth agent: ;

[0029] The first reasoning information, the second reasoning information, the third reasoning information, the fourth reasoning information, the fifth reasoning information and the sixth reasoning information obtained by reasoning are determined as the enhanced situation information of the target intelligent agent.

[0030] In a second aspect, a computing device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the above-mentioned method when executing the computer program.

[0031] In a third aspect, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the above-described method is implemented.

[0032] In a fourth aspect, an intelligent agent situation information enhancement system based on knowledge reasoning is provided, including:

[0033] A determination module, configured to determine the initial situation information obtained by a target intelligent agent perceiving the current confrontation environment;

[0034] A construction module, configured to construct rule knowledge for each intelligent agent in the confrontation environment based on first-order predicate logic; wherein, the rule knowledge constructed for each intelligent agent can formally express and reason about the intelligent agent, its capabilities, and relationships; a conversion module, configured to convert the initial situation information into a formal expression of first-order predicate logic to obtain formally compatible situation information;

[0035] An inference module, configured to perform knowledge inference on the formally compatible situation information based on the constructed rule knowledge to obtain the enhanced situation information of the target intelligent agent; wherein, the enhanced situation information includes hidden situation information obtained through knowledge inference.

[0036] The beneficial effects of the technical solution provided by this application at least include:

[0037] As can be seen from the above technical solution, the initial situation information obtained by a target intelligent agent perceiving the current confrontation environment is determined; rule knowledge is constructed for each intelligent agent in the confrontation environment based on first-order predicate logic; the initial situation information is converted into a formal expression of first-order predicate logic to obtain formally compatible situation information; knowledge inference is performed on the formally compatible situation information based on the constructed rule knowledge to obtain the enhanced situation information of the target intelligent agent; wherein, the enhanced situation information includes hidden situation information obtained through knowledge inference. In this way, based on the objectively observed situation information, with the knowledge of domain experts, potential but unobservable information is inferred, so as to support the analysis and decision-making of the intelligent agent with more comprehensive information.

[0038] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of this application, nor is it used to limit the scope of this application. Other features of this application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0040] Figure 1 It is a schematic diagram of the intelligent agent game confrontation process in the prior art.

[0041] Figure 2 It is a schematic diagram of the steps of an intelligent agent situation information enhancement method based on knowledge reasoning proposed in the embodiments of the present application.

[0042] Figure 3 It is a schematic diagram of the technical solution of the intelligent agent game confrontation process proposed in the present application.

[0043] Figure 4 It is a schematic diagram of the game confrontation game scenario proposed in the embodiments of the present application.

[0044] Figure 5 It is a schematic diagram of the initial situation information in the embodiments of the present application.

[0045] Figure 6 It is a schematic diagram of the enhanced situation information in the embodiments of the present application.

[0046] Figure 7 It is a schematic diagram of the structure of an intelligent agent situation information enhancement system based on knowledge reasoning proposed in the embodiments of the present application.

[0047] Figure 8 It is a block diagram of the computing device in the embodiments of the present application. Detailed implementation manners

[0048] The following will describe the exemplary embodiments of the present application with reference to the accompanying drawings. Various details of the embodiments of the present application are included to assist understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described here without departing from the scope and spirit of the present application. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted below.

[0049] Obviously, the described embodiments are part of the embodiments of the present application, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts fall within the scope of protection of the present application.

[0050] It should be noted that the terminal devices involved in the embodiments of the present application may include, but are not limited to, intelligent devices such as mobile phones, personal digital assistants (PDAs), wireless handheld devices, and tablet computers; display devices may include, but are not limited to, devices with display functions such as personal computers and televisions.

[0051] In addition, the term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after.

[0052] To solve the problem that the intuitive agent situation information obtained in the existing game confrontation process may be incomplete and inaccurate, which affects the confrontation decision-making, the present application proposes an agent situation information enhancement scheme based on knowledge reasoning. The inventive concept lies in: determining the initial situation information obtained by the target agent perceiving the current confrontation environment; constructing rule knowledge for each agent in the confrontation environment based on first-order predicate logic; transforming the initial situation information into a formal expression of first-order predicate logic to obtain formally compatible situation information; performing knowledge reasoning on the formally compatible situation information based on the constructed rule knowledge to obtain the enhanced situation information of the target agent; wherein the enhanced situation information includes hidden situation information obtained through knowledge reasoning. In this way, based on the objectively observed situation information, with the help of the knowledge of domain experts, potential but directly unobservable information is inferred, so as to support the analysis and decision-making of the agent with more comprehensive information.

[0053] The solution of the present application will be described in detail through specific embodiments below.

[0054] Refer to Figure 2 As shown, it is a schematic diagram of the steps of an agent situation information enhancement method based on knowledge reasoning proposed in the embodiments of the present application. The execution subject of this method can be a software module with computing and data processing capabilities, or a computing device integrated with a similar software module.

[0055] The agent situation information enhancement method based on knowledge reasoning may include the following steps:

[0056] Step 202: Determine the initial situation information obtained by the target agent perceiving the current confrontation environment.

[0057] In the solution of this application, the target agent can be an agent that plays a decision-making role in an adversarial environment. The target agent can obtain the initial situation information of all agents in the current adversarial environment, and the initial situation information can reflect the changes of the opponent, our side, and the environment in the adversarial environment.

[0058] Step 204: Based on first-order predicate logic, construct rule knowledge for each agent in the adversarial environment; among them, the rule knowledge constructed for each agent can formally express and reason about the agent, its capabilities, and relationships.

[0059] It should be understood that first-order predicate logic includes multiple basic elements: individual constants, predicates, quantifiers, and logical connectives; among them, individual constants refer to the agents participating in the game confrontation in the adversarial environment; predicates are used to describe the properties and / or relationships of agents; quantifiers include universal quantifiers and existential quantifiers. The universal quantifier indicates that the predicate applies to all agents, and the existential quantifier indicates that the predicate applies to at least one agent; logical connectives are used to combine atomic propositions to form more complex propositions.

[0060] Construct domain expert knowledge based on first-order predicate logic. First-order predicate logic is a formal system in mathematical logic used to formally express and reason about individuals and their properties and relationships. It includes basic elements such as individual constants, predicates, quantifiers, and logical connectives. Among them, individual constants refer to specific individuals: such as "Socrates", "the Earth", etc.; predicates are used to describe the properties or relationships of individuals. Usually, individual constants in parentheses follow the predicates. For example, "Socrates is a human" can be expressed as Human("Socrates"); quantifiers are divided into universal quantifiers and existential quantifiers. The universal quantifier indicates that the predicate applies to all individuals, expressed as "for all" or "for each", and the symbol is " "; the existential quantifier indicates that the predicate applies to at least one individual, expressed as "there exists some" or "there is some", and the symbol is " "; logical connectives are used to combine atomic propositions to form more complex propositions, including conjunction ( , and), disjunction ( , or), negation ( , not), implication ( , if... then...), equivalence ( , if and only if...), etc.

[0061] Based on first-order predicate logic, the logical reasoning processes of different domain experts can be represented, so that they can be understood and executed by machines.

[0062] Optionally, in the solution of this application, different agents in the adversarial environment are divided into different role types according to their capabilities: air attack role, ground attack role, and escort role. Among them, the main ability of the air attack role is to attack air targets; the main ability of the ground attack role is to attack ground targets; the main ability of the escort role is to protect other roles from being attacked.

[0063] In the game adversarial scenario, if the agent is a ground attack role, then the agent has no air attack ability, and the corresponding first rule knowledge is constructed as: ;

[0064] If the agent is an air attack role, then the agent has air attack ability, and the corresponding second rule knowledge is constructed as: ;

[0065] If the agent has no air attack ability but enters the warning range of the opponent's air attack role, then the agent has an escort role, and the corresponding third rule knowledge is constructed as: ;

[0066] If the agent has an escort role and it is not observed, it is speculated that the escort role is a stealth agent, and the corresponding fourth rule knowledge is constructed as: .

[0067] Step 206: Convert the initial situation information into a form expressed in first-order predicate logic to obtain situation information with compatible form.

[0068] Considering that the data format of the situation information directly observed by the target agent is inconsistent and may be unstructured or semi-structured data, it is necessary to convert it into a representation form compatible with first-order predicate logic. It is mainly converted into the form of individual constants or atomic propositions in first-order predicate logic.

[0069] Optionally, in the solution of this application, convert the initial situation information into a form expressed in first-order predicate logic to obtain situation information with compatible form. Specifically, the first agent and the second agent in the adversarial game included in the initial situation information can be determined; based on first-order predicate logic, the situation information with compatible form including the first agent and the second agent is determined; among them, the situation information with compatible form can include:

[0070] First situation information: The first agent is a ground attack role: ;

[0071] Second situation information: The second agent is an air attack role: ;

[0072] Third situation information: First agent and the second agent It is an adversarial relationship: ;

[0073] Fourth situation information: the first agent With the second agent Distance between: ;

[0074] Fifth situation information: Second agent Warning range: .

[0075] Step 208: Based on the constructed rule knowledge, knowledge reasoning is performed on the form-compatible situation information to obtain enhanced situation information of the target intelligent agent; wherein the enhanced situation information includes hidden situation information obtained through knowledge reasoning.

[0076] In specific implementation, the situation information represented by step 106 can be used as input, and based on the rule knowledge constructed in step 104, new facts can be inferred to serve as enhanced situation information.

[0077] Optionally, based on the constructed rule knowledge, knowledge reasoning is performed on the form-compatible situation information to obtain enhanced situation information of the target intelligent agent, specifically including:

[0078] Based on the first rule knowledge and the fifth situation information, the first reasoning information is obtained by reasoning: the first agent No anti-air attack capability: ;

[0079] Based on the second rule knowledge and the second situation information, the second reasoning information is obtained by reasoning: the second agent Has anti-air attack capability: ;

[0080] Based on the fourth situation information and the fifth situation information, the third reasoning information is obtained by reasoning: the first agent Entered the second agent Warning range: ;

[0081] Based on the first reasoning information, the second reasoning information, the third situation information, and the third reasoning information, the fourth reasoning information is obtained by reasoning: the first agent Escort role exists : ;

[0082] Escort role If it is not observed, then there is the fifth inference information: ;

[0083] Based on the first situation information, the fourth inference information, and the fifth inference information, infer the sixth inference information: escort role It is a stealth agent: ;

[0084] Determine the first inference information, the second inference information, the third inference information, the fourth inference information, the fifth inference information, and the sixth inference information obtained by inference as the enhanced situation information of the target agent.

[0085] Optionally, after obtaining the enhanced situation information of the target agent, the enhanced situation information can also be sent to the target agent so that the target agent can use the enhanced situation information for confrontation decision-making.

[0086] Compared with the traditional agent decision-making process, the technical solution proposed in this application is as Figure 3 shown. It adds a knowledge inference engine module. It takes the situation information as input, obtains new facts through inference, and then adds them to the original situation information. The enhanced situation information is used as input and provided to the agent for decision-making. Among them, the knowledge inference engine mainly includes three parts, namely rule knowledge construction, formal representation of situation information, and situation information inference.

[0087] Take the two-player game confrontation game as an example. Assume that there are different game agents in this game. Different game agents are divided into different roles according to their abilities, specifically including: ① Air attack type role, whose main ability is to attack air targets; ② Ground attack type role, whose main ability is to attack ground targets; ③ Escort role, whose main ability is to protect other roles from being attacked. Assume that the purpose of the opponent agent is to destroy our command server, and the purpose of our agent is to protect the command server from being destroyed by the opponent. Specifically, what needs to be done is to observe the real-time situation and control our agent to make necessary action decisions, such as Figure 4 shown. It should be understood that in this embodiment, the ground attack type role is mainly illustrated by taking an unmanned aerial vehicle as an example.

[0088] Assume that the situation information obtained by the current game agent is as Figure 5 shown. Among them, it is found that an opponent ground attack type role B (such as an unmanned aerial vehicle) intends to strike our command server, and it is 180 km away from our air attack type role F and has entered the 200 km warning range of F. If only based on the situation information currently observed by the game agent, since the ground attack type role has no air attack ability, the intuitive decision should be to send the air attack type role F forward to strike.

[0089] However, in an adversarial environment, the other party often conceals the truth and shows false information. The observed situation information may not be comprehensive and accurate. Therefore, some knowledge rules are constructed using the solution of this application to enhance the situation information and assist our agent in making reasonable decisions. According to the foregoing method, the specific steps are as follows.

[0090] 1. Construction of rule knowledge

[0091] First, construct rule knowledge as shown in formulas (1)-(4).

[0092] If the game agent type is a ground attack type character, then it has no air-to-air attack ability:

[0093]

[0094] If the game agent type is an air-to-air attack type character, then it has air-to-air attack ability:

[0095]

[0096] If the game agent has no air-to-air attack ability but enters the warning range of the other party's air-to-air attack type character, then there should be an escort character:

[0097]

[0098] If there is an escort character and it is not observed, it is speculated to be a stealth agent:

[0099]

[0100] 2. Formal representation of situation information

[0101] Then, the situation information shown Figure 5 needs to be formally represented, as shown in formulas (5)-(9).

[0102] Game agent B is a ground attack type character:

[0103] Game agent F is an air-to-air attack type character:

[0104] Game agents B and F are in an enemy-ally relationship:

[0105] The distance between game agents B and F is 180 km:

[0106] The warning range of game agent F is 200 km:

[0107] 3. Situation Information Inference

[0108] Combining rules (1)-(4) and the observed situation information (5)-(9), the reasoning process is as follows:

[0109] Combining (1) and (5), it is inferred that game agent B has no anti-air attack ability:

[0110] Combining (2) and (6), it is inferred that game agent F has anti-air attack ability:

[0111] Combining (8) and (9), it is inferred that game agent B has entered F's warning range:

[0112]

[0113] Combining (10), (11), (7), and (12), it is inferred that game agent B has an escort role:

[0114]

[0115] However, since y has not been observed, there is:

[0116] Combining (5), (13), and (14), it is inferred that there is a stealth agent:

[0117]

[0118] In this way, the actual situation information, that is, the enhanced situation information, may be as Figure 6 shown. Thus, the implicit information is inferred based on knowledge, enhancing the agent's mastery of the situation. The agent can make full use of the enhanced situation information to make decisions, such as strengthening reconnaissance and anti-air capabilities, while maintaining vigilance and avoiding taking rash actions relying solely on the observed information, which may lead to losses.

[0119] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0120] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0121] It should be understood that in this application, the ground attack character B and the game agent B both represent the specific roles of the first agent B in different scenarios. Therefore, the same symbol B is used to represent them. Similarly, the air attack character F and the game agent F both represent the specific role indications of the second agent F in different scenarios. Therefore, the same symbol F is used to represent them.

[0122] Figure 7 The block diagram of the intelligent agent situation information enhancement system based on knowledge reasoning provided by an embodiment of this application is shown as Figure 7 shown. The intelligent agent situation information enhancement system 700 based on knowledge reasoning in this embodiment may include a determination module 701, a construction module 702, a conversion module 703, and an inference module 704. Among them, the determination module 701 is used to determine the initial situation information obtained by the target agent perceiving the current confrontation environment; the construction module 702 is used to construct rule knowledge for each agent in the confrontation environment based on first-order predicate logic; among them, the rule knowledge constructed for each agent can formally express and reason about the agent and its capabilities and relationships; the conversion module 703 is used to convert the initial situation information into a formal expression of first-order predicate logic to obtain formally compatible situation information; the inference module 704 is used to perform knowledge reasoning on the formally compatible situation information based on the constructed rule knowledge to obtain the enhanced situation information of the target agent; among them, the enhanced situation information includes hidden situation information obtained through knowledge reasoning.

[0123] It should be noted that part or all of the intelligent agent situation information enhancement system based on knowledge reasoning in this embodiment can be an application located on the local terminal, or can also be a functional unit such as a plug-in or software development kit (SDK) set in the application located on the local terminal, or can also be a processing engine located in the network-side server, or can also be a distributed system located on the network side.

[0124] It can be understood that the application can be a native program (nativeApp) installed on the local terminal, or can also be a web program (webApp) of the browser on the local terminal. This embodiment does not limit this.

[0125] Optionally, in a possible implementation manner of this embodiment, it further includes: a sending module, configured to send the enhanced situation information to the target agent, so that the target agent can use the enhanced situation information for confrontation decision-making.

[0126] Optionally, in a possible implementation of this embodiment, the first-order predicate logic includes multiple basic elements: individual constants, predicates, quantifiers, and logical connectives; where an individual constant refers to an intelligent agent participating in a game confrontation in an adversarial environment; a predicate is used to describe the properties and / or relationships of the intelligent agent; the quantifiers include a universal quantifier and an existential quantifier, the universal quantifier indicates that the predicate applies to all intelligent agents, and the existential quantifier indicates that the predicate applies to at least one intelligent agent; logical connectives are used to combine atomic propositions to form more complex propositions.

[0127] Optionally, in a possible implementation of this embodiment, different intelligent agents in the adversarial environment are divided into different role types according to their capabilities: air attack role, ground attack role, escort role; where the main ability of the air attack role is to attack air targets; the main ability of the ground attack role is to attack ground targets; the main ability of the escort role is to protect other roles from being attacked.

[0128] Optionally, in a possible implementation of this embodiment, in the game confrontation game scenario, when the construction module 702 constructs rule knowledge for each intelligent agent in the adversarial environment based on the first-order predicate logic, it is specifically used for:

[0129] If the intelligent agent is a ground attack role, then the intelligent agent has no air attack ability, and the corresponding first rule knowledge is constructed as: ;

[0130] If the intelligent agent is an air attack role, then the intelligent agent has an air attack ability, and the corresponding second rule knowledge is constructed as: ;

[0131] If the intelligent agent has no air attack ability, but enters the warning range of the air attack role of the other party, then the intelligent agent has an escort role, and the corresponding third rule knowledge is constructed as:

[0132] ;

[0133] If the intelligent agent has an escort role and is not observed, it is speculated that the escort role is a stealth intelligent agent, and the corresponding fourth rule knowledge is constructed as: .

[0134] Optionally, in a possible implementation of this embodiment, when the conversion module 703 converts the initial situation information into a formal expression of the first-order predicate logic to obtain situation information with formal compatibility, it is specifically used for:

[0135] Determine the first intelligent agent and the second intelligent agent participating in the adversarial game included in the initial situation information;

[0136] Based on the first-order predicate logic, situation information including the first agent and the second agent that is compatible in form is determined; wherein the situation information that is compatible in form includes:

[0137] First situation information: first agent For ground attack characters: ;

[0138] Second situation information: Second agent For anti-air attack characters: ;

[0139] Third situation information: First agent and the second agent It is an adversarial relationship: ;

[0140] Fourth situation information: the first agent With the second agent Distance between: ;

[0141] Fifth situation information: Second agent Warning range: .

[0142] Optionally, in a possible implementation of this embodiment, when the reasoning module 704 performs knowledge reasoning on the form-compatible situation information based on the constructed rule knowledge to obtain the enhanced situation information of the target agent, it is specifically used to:

[0143] Based on the first rule knowledge and the fifth situation information, the first reasoning information is obtained by reasoning: the first agent No anti-air attack capability: ;

[0144] Based on the second rule knowledge and the second situation information, the second reasoning information is obtained by reasoning: the second agent Has anti-air attack capability: ;

[0145] Based on the fourth situation information and the fifth situation information, the third reasoning information is obtained by reasoning: the first agent Entered the second agent Warning range: ;

[0146] Based on the first reasoning information, the second reasoning information, the third situation information, and the third reasoning information, the fourth reasoning information is obtained by reasoning: the first agent Escort role exists : ;

[0147] Escort role If not observed, there is a fifth piece of inference information:

[0148] Based on the first situation information, the fourth piece of inference information, and the fifth piece of inference information, the sixth piece of inference information is inferred: Escort role Is a stealth agent: ;

[0149] The first piece of inference information, the second piece of inference information, the third piece of inference information, the fourth piece of inference information, the fifth piece of inference information, and the sixth piece of inference information obtained by inference are determined as the enhanced situation information of the target agent.

[0150] In this embodiment, the initial situation information obtained by the target agent sensing the current confrontation environment is determined; based on first-order predicate logic, rule knowledge is constructed for each agent in the confrontation environment; the initial situation information is transformed into a formal expression of first-order predicate logic to obtain formally compatible situation information; based on the constructed rule knowledge, knowledge inference is performed on the formally compatible situation information to obtain the enhanced situation information of the target agent; wherein, the enhanced situation information includes hidden situation information obtained by knowledge inference. In this way, based on the objectively observed situation information, with the knowledge of domain experts, potential but unobservable information is inferred, so as to support the analysis and decision-making of the agent with more comprehensive information.

[0151] An embodiment of the present application provides a computer-readable storage medium, in which at least one instruction is stored, and the at least one instruction is loaded and executed by a processor to implement the method for enhancing the situation information of an agent based on knowledge inference as described above.

[0152] An embodiment of the present application provides an electronic device, the electronic device includes a processor and a memory, and at least one instruction is stored in the memory, and the instruction is loaded and executed by the processor to implement the method for enhancing the situation information of an agent based on knowledge inference as described above.

[0153] In the technical solution of the present application, the collection, storage, use, processing, transmission, provision, and disclosure of the user's personal information and other processing all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0154] Figure 8FIG. shows a schematic block diagram of an exemplary computing device 800 that may be used to implement embodiments of the present application. The computing device is intended to represent various forms of digital computers, such as, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The computing device may also represent various forms of mobile devices, such as, personal digital processors, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely exemplary and are not intended to limit the implementation of the present application described and / or claimed herein.

[0155] As Figure 8 shown, the computing device 800 includes a computing unit 801 that can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the computing device 800 can also be stored. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0156] A plurality of components in the computing device 800 are connected to the I / O interface 805, including: an input unit 806, such as a keyboard, a mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a magnetic disk, an optical disk, etc.; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 809 allows the computing device 800 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0157] The computing unit 801 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 executes the various methods and processes described above, such as the intelligent agent situation information enhancement method based on knowledge reasoning. For example, in some embodiments, the intelligent agent situation information enhancement method based on knowledge reasoning can be implemented as a computer software program that is tangibly contained in a machine-readable medium, such as the storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed onto the computing device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded into the RAM 803 and executed by the computing unit 801, one or more steps of the intelligent agent situation information enhancement method described above can be executed. Alternatively, in other embodiments, the computing unit 801 can be configured to execute the intelligent agent situation information enhancement method by any other suitable means (e.g., by means of firmware).

[0158] The various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0159] The program code for implementing the methods of this application can be written in any combination of one or more programming languages. These program codes can be provided to the processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing devices, such that when the program codes are executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program codes can be executed entirely on the machine, partially on the machine, as an independent software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0160] In the context of this application, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0161] To provide for interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can also be used to provide for interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic, speech, or tactile input).

[0162] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0163] A computer system may include a client and a server. The client and the server are generally far from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.

[0164] It should be understood that various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this application can be achieved, and no limitations are imposed herein.

[0165] The above specific embodiments do not constitute a limitation on the protection scope of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application shall be included within the protection scope of this application.

Claims

1. A method for enhancing agent situation information based on knowledge reasoning, characterized in that: include: Determine the initial situation information obtained by the target agent when perceiving the current confrontation environment; Based on the first-order predicate logic, rule knowledge is constructed for each intelligent agent in the adversarial environment; wherein the rule knowledge constructed for each intelligent agent can formally express and reason about the intelligent agent and its capabilities and relationships; The initial situation information is converted into a formal expression of first-order predicate logic to obtain a situation information that is compatible with the form; wherein the first-order predicate logic includes multiple basic elements: individual constants, predicates, quantifiers, and logical connectives; individual constants refer to the agents participating in the game confrontation in the confrontation environment; predicates are used to describe the properties and / or relations of agents; quantifiers include universal quantifiers and existential quantifiers, wherein the universal quantifiers indicate that the predicates are applicable to all agents, and the existential quantifiers indicate that the predicates are applicable to at least one agent; logical connectives are used to combine atomic propositions to form more complex propositions; Based on the constructed rule knowledge, knowledge reasoning is performed on the form-compatible situation information to obtain enhanced situation information of the target intelligent agent; wherein the enhanced situation information includes hidden situation information obtained through knowledge reasoning; The enhanced situation information is sent to the target intelligent agent so that the target intelligent agent uses the enhanced situation information to make confrontation decisions.

2. The method according to claim 1, characterized in that Different agents in the confrontation environment are divided into different role types according to their capabilities: air attack role, ground attack role, escort role; Among them, the main ability of the air attack type character is to attack aerial targets; the main ability of the ground attack type character is to attack ground targets; the main ability of the escort character is to protect other characters from being attacked.

3. The method according to claim 2, characterized in that In the game confrontation scenario, based on the first-order predicate logic, rule knowledge is constructed for each agent in the confrontation environment, including: If the agent is a ground attack character, then the agent has no air attack capability, and the corresponding first rule knowledge is constructed as follows: ; If the agent is an air attack character, then the agent has air attack capability, and the corresponding second rule knowledge is constructed as follows: ; If the agent has no air attack capability but enters the alert range of the opponent's air attack role, then the agent has an escort role, and the corresponding third rule knowledge is constructed as follows: ； If the agent has an escort role and it is not observed, it is inferred that the escort role is a stealth agent, and the corresponding fourth rule knowledge is constructed as follows: .

4. The method according to claim 3, characterized in that The initial situation information is converted into a formal expression of first-order predicate logic to obtain situation information that is compatible with the form, specifically including: Determine a first agent and a second agent of the confrontation game included in the initial situation information; Based on the first-order predicate logic, situation information including the first agent and the second agent that is compatible in form is determined; wherein the situation information that is compatible in form includes: First situation information: first agent For ground attack characters: ; Second situation information: Second agent For anti-air attack characters: ; Third situation information: First agent and the second agent It is an adversarial relationship: ; Fourth situation information: the first agent With the second agent Distance between: ; Fifth situation information: Second agent Warning range: .

5. The method according to claim 4, characterized in that Based on the constructed rule knowledge, knowledge reasoning is performed on the situation information compatible with the form to obtain enhanced situation information of the target intelligent agent, specifically including: Based on the first rule knowledge and the fifth situation information, the first reasoning information is obtained by reasoning: the first agent No anti-air attack capability: ; Based on the second rule knowledge and the second situation information, the second reasoning information is obtained by reasoning: the second agent Has anti-air attack capability: ; Based on the fourth situation information and the fifth situation information, the third reasoning information is obtained by reasoning: the first agent Entered the second agent Warning range: ; Based on the first reasoning information, the second reasoning information, the third situation information, and the third reasoning information, the fourth reasoning information is obtained by reasoning: the first agent Escort role exists : ; Escort role If it is not observed, then there is the fifth inference information: ; Based on the first situation information, the fourth reasoning information, and the fifth reasoning information, the sixth reasoning information is obtained by reasoning: escort role For an invisible agent: ; The first reasoning information, the second reasoning information, the third reasoning information, the fourth reasoning information, the fifth reasoning information and the sixth reasoning information obtained by reasoning are determined as the enhanced situation information of the target intelligent agent.

6. A computing device, characterized in that: The invention comprises a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the method described in any one of claims 1 to 5 when executing the computer program.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method described in any one of claims 1 to 5 is implemented.

8. An intelligent agent situation information enhancement system based on knowledge reasoning, characterized in that: include: A determination module is used to determine the initial situation information obtained by the target intelligent agent when perceiving the current confrontation environment; A construction module is used to construct rule knowledge for each intelligent agent in the adversarial environment based on first-order predicate logic; wherein the rule knowledge constructed for each intelligent agent can formally express and reason about the intelligent agent and its capabilities and relationships; A conversion module is used to convert the initial situation information into a formal expression of first-order predicate logic to obtain situation information with compatible form; wherein the first-order predicate logic includes multiple basic elements: individual constants, predicates, quantifiers, and logical connectives; individual constants refer to the agents participating in the game confrontation in the confrontation environment; predicates are used to describe the properties and / or relations of agents; quantifiers include universal quantifiers and existential quantifiers, wherein the universal quantifiers indicate that the predicates are applicable to all agents, and the existential quantifiers indicate that the predicates are applicable to at least one agent; logical connectives are used to combine atomic propositions to form more complex propositions; A reasoning module, configured to perform knowledge reasoning on the form-compatible situation information based on the constructed rule knowledge, to obtain enhanced situation information of the target intelligent agent; wherein the enhanced situation information includes hidden situation information obtained through knowledge reasoning; The sending module is used to send the enhanced situation information to the target intelligent agent so that the target intelligent agent can use the enhanced situation information to make confrontation decisions.

Images