Battle processing method, device and equipment of operation unit

Abstract

The present disclosure relates to a battle processing method, device and equipment of operating units, belonging to the technical field of computer programs, and the method comprises: in response to a user's battle request for a first operating unit in a game scene, calling an agent, determining the interaction identification combination of each operating unit in the game scene; through the agent, taking the unit identification of the first operating unit, the unit identification of the second operating unit and the interaction identification combination as the first input condition, determining the interaction relationship of the camp where the first operating unit and the second operating unit are located, generating the inter-camp interaction identification; through the agent, taking the interaction identification combination and the inter-camp interaction identification as the second input condition, and taking the operation configuration chain set in the game scene as the configuration constraint condition, generating the behavior configuration identification adapted to the behavior of the first operating unit, and obtaining the battle deployment operation of the first operating unit at the target battle node in the game scene according to the behavior configuration identification.

Description

Technical Field

[0001] This disclosure relates to the technical field of computer programs, and more specifically, to a method, apparatus, and device for processing combat operations using an operating unit. Background Technology

[0002] With the rapid development of gaming applications, strategy role-playing games (SRPGs) have merged role-playing and development with turn-based strategy. Users can arrange their controlled units on a grid-based battlefield, with an AI agent controlling these units to advance the battle. However, existing AI agent-controlled units exhibit limited behavior and struggle to adapt to scenarios involving combinations of units or complex character skills, thus impacting the user's gaming experience. Summary of the Invention

[0003] One objective of this disclosure is to provide a new technical solution for the combat processing of an operating unit.

[0004] According to an aspect of this disclosure, a method for processing combat operations using an operating unit is provided, the method comprising: In response to a user's request for the first operation unit to engage in combat within the game scene, an intelligent agent is invoked to determine the interaction identifier combination of each operation unit in the game scene; wherein, the interaction identifier combination includes the movement identifier corresponding to the first operation unit and the second operation unit in the opposing camp, and the skill identifier configured under their respective movement operations. Using the intelligent agent, the interaction relationship between the camps of the first operation unit and the second operation unit is determined by the combination of the unit identifier of the first operation unit, the unit identifier of the second operation unit, and the interaction identifier, and an inter-camp interaction identifier is generated. Using the intelligent agent, the interaction identifier combination and the inter-faction interaction identifier are used as the second input conditions, and the operation configuration chain set by the game scene is used as the configuration constraint condition to generate the behavior configuration identifier adapted to the behavior of the first operation unit. Based on the behavior configuration identifier, the battle deployment operation of the first operation unit on the target battle node in the game scene is obtained. When the target battle node is reached, the first operation unit is shown to the user to perform the battle deployment operation.

[0005] Optionally, the intelligent agent includes multiple unit processing layers; the invoking of the intelligent agent to determine the interaction identifier combination of each operation unit in the game scene includes: For each target operation unit in the game scene, the character identifier and skill identifier in the unit identifier of the target operation unit are input into the corresponding unit processing layer to determine the skill identifiers configured for all target operation units under each movement identifier, so as to form an interaction identifier combination.

[0006] Optionally, the intelligent agent includes a unit attention layer and a lineup attention layer; the step of determining the interaction relationship between the first operating unit and the second operating unit's faction through the intelligent agent, using the unit identifier of the first operating unit, the unit identifier of the second operating unit, and the interaction identifier combination as the first input condition, and generating inter-faction interaction identifiers, includes: The role identifier and skill identifier of the first operation unit and the second operation unit are input into the unit attention layer to determine the faction to which each role identifier belongs, the correspondence and temporal relationship between each role identifier and each skill identifier, and generate unit self-attention identifier; The unit self-attention identifier and the interaction identifier are combined and input into the lineup attention layer to determine the interaction relationship between the first operation unit and the second operation unit in the same camp, and generate the inter-camp interaction identifier.

[0007] Optionally, the operation configuration chain set in the game scene includes a first operation configuration for releasing an object and a second operation configuration for the object affected by the released object. The first operation configuration includes the object's basic parameters, movement position, and skill release. The second operation configuration includes the object's basic parameters and affiliated faction. The intelligent agent includes an operation attention layer. The generation of behavior configuration identifiers for the first operation unit's behavior adaptation, using the interaction identifier combination and the inter-faction interaction identifier as second input conditions, and the operation configuration chain set in the game scene as configuration constraints, through the intelligent agent, includes: The interaction identifier combination and the inter-faction interaction identifier are input into the operation attention layer. The first operation configuration and the second operation configuration are used as configuration constraints to extract the association between the first operation unit and the second operation unit and the game operation in the game scene, and generate the behavior configuration identifier for the behavior adaptation of the first operation unit.

[0008] Optionally, the intelligent agent further includes a scene attention layer; the method further includes: The first attribute configuration set by each scene unit in the game scene and the second attribute configuration set by each operation unit in the game scene are input into the scene attention layer, so as to map the first attribute configuration and the second attribute configuration to the scene thumbnail associated with the game scene to obtain the scene association identifier; Based on the scene association identifier, extract the target behavior configuration identifier that is adapted to the game scene from the behavior configuration identifier; The step of obtaining the battle deployment operation of the first operation unit on the target battle node in the game scene based on the behavior configuration identifier includes: Based on the target behavior configuration identifier, the first operation unit obtains the battle deployment operation of the target battle node in the game scene.

[0009] Optionally, the first attribute configuration includes the effect set by each scene unit and the corresponding first plot identifier in the scene thumbnail; the second attribute configuration includes the movement constraints and skill constraints set by each operation unit and the corresponding second plot identifier in the scene thumbnail; the step of inputting the first attribute configuration set by each scene unit in the game scene and the second attribute configuration set by each operation unit in the game scene into the scene attention layer, so as to map the first attribute configuration and the second attribute configuration to the scene thumbnail associated with the game scene to obtain the scene association identifier, includes: The first attribute configuration set by each scene unit in the game scene and the second attribute configuration set by each operation unit in the game scene are input into the scene attention layer, so as to map the effect to the first tile corresponding to the first plot identifier of the scene thumbnail, and map the movement constraint and the skill constraint to the second tile corresponding to the second plot identifier of the scene thumbnail, so as to obtain the scene association identifier.

[0010] Optionally, obtaining the battle deployment operation of the first operation unit on the target battle node in the game scene based on the behavior configuration identifier includes: Based on the behavior configuration identifier, a battle deployment operation is obtained in which the first operation unit of the target battle node in the game scene moves to the target movement position and releases the target skill to the third operation unit of the friendly camp and / or the second operation unit of the enemy camp.

[0011] Optionally, the method further includes: Obtain a training sample set; wherein the training sample set includes training samples and sample labels; The agent is trained using the training samples to obtain a trained agent; wherein the entropy coefficient of the loss function of the agent is determined by the entropy ratio constructed from the current entropy and the maximum entropy of the loss function.

[0012] According to a second aspect of this disclosure, a battle processing apparatus for an operating unit is also provided, the apparatus comprising: The response module is used to invoke the intelligent agent to determine the interaction identifier combination of each operation unit in the game scene in response to the user's battle request for the first operation unit; wherein, the interaction identifier combination includes the movement identifier corresponding to the first operation unit and the second operation unit in the opposing camp and the skill identifier configured under their respective movement operations. The generation module is used to determine the interaction relationship between the camps of the first operation unit and the second operation unit through the intelligent agent, using the unit identifier of the first operation unit, the unit identifier of the second operation unit, and the interaction identifier combination as the first input condition, and generate an inter-camp interaction identifier. The module is used to generate a behavior configuration identifier for the behavior adaptation of the first operation unit by using the intelligent agent, the interaction identifier combination and the inter-faction interaction identifier as the second input condition, and the operation configuration chain set by the game scene as the configuration constraint condition, and to obtain the battle deployment operation of the first operation unit at the target battle node in the game scene according to the behavior configuration identifier. The display module is used to display to the user, when the target battle node is reached, the first operation unit performs the battle deployment operation.

[0013] According to a third aspect of this disclosure, a computer system is also provided, the computer system including a processor, and a method for handling operations of operational units in the implementation aspect of the computer system when the processor executes program instructions or code.

[0014] For example, the computer system also includes a memory for storing program instructions or code.

[0015] According to a fourth aspect of this disclosure, a computer-readable storage medium is also provided, wherein a computer program is stored therein, wherein the computer program is configured to execute the battle processing method of the above-described operation unit when it is run.

[0016] According to a fifth aspect of this disclosure, a computer program product is also provided, comprising a computer program that, when executed, causes a computer to perform the battle processing method steps of the above-described operation unit.

[0017] According to a sixth aspect of this disclosure, an electronic device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the battle processing method of the operation unit through the computer program.

[0018] One beneficial effect of this disclosure is that the combat processing method for operation units provided by the present invention can, upon a user submitting a combat request to an operation unit, invoke an intelligent agent to determine the specific movement position and skill release interaction identifier combination of the operation unit in the game scene. Then, through the intelligent agent, using unit identifiers such as the operation unit's character health, attack range, and attack power, along with the interaction identifier combination, the interaction relationship between the first and second operation units' factions is determined, generating inter-faction interaction identifiers. Furthermore, through the intelligent agent, using the interaction identifier combination and inter-faction interaction identifiers as second input conditions, and the operation configuration chain set in the game scene as configuration constraints, a behavior configuration identifier adapted to the first operation unit's behavior is generated. Based on the behavior configuration identifier, the combat deployment operation of the first operation unit at the target combat node in the game scene is obtained, and this combat deployment operation is displayed to the user. This adapts to scenarios with complex combinations of operation units or complex character skills, thereby improving the user's gaming experience.

[0019] Other features and advantages of the embodiments of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0020] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments of the present disclosure and, together with their description, serve to explain the principles of the embodiments of the present disclosure.

[0021] Figure 1 A schematic diagram of the hardware structure of an electronic device is shown, which can be used to implement a battle processing method of an operating unit according to an embodiment of the present disclosure. Figure 2 A flowchart illustrating a battle processing method of an operating unit according to some embodiments is shown; Figure 3 A schematic diagram of an intelligent agent according to some embodiments is shown; Figure 4 A schematic diagram of the structure of a battle processing apparatus for an operating unit according to some embodiments is shown; Figure 5 A schematic diagram of the hardware structure of an electronic device according to some embodiments is shown. Detailed Implementation

[0022] Various exemplary embodiments of this specification will now be described in detail with reference to the accompanying drawings.

[0023] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the embodiments of this specification or their application or use.

[0024] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0025] It should be noted that all actions involving the acquisition of signals, information, or data in this embodiment are carried out in compliance with the relevant data protection laws and regulations of the country where the location is situated, and with authorization from the owner of the relevant equipment.

[0026] This disclosure provides a novel combat processing scheme for operating units. It allows operating units to submit combat requests, determines the specific movement position and skill release combinations of the operating unit in the game scene, and then, through an intelligent agent, uses unit identifiers such as the operating unit's character health, attack range, and attack power, along with the combination of interaction identifiers, to determine the interaction relationship between the factions of the first and second operating units, generating inter-faction interaction identifiers. Furthermore, through the intelligent agent, using the interaction identifier combination and inter-faction interaction identifiers as second input conditions, and the operation configuration chain set in the game scene as configuration constraints, it generates behavior configuration identifiers adapted to the behavior of the first operating unit. Based on the behavior configuration identifiers, it obtains the combat deployment operation of the first operating unit at the target combat node in the game scene and displays this combat deployment operation to the user. This adapts to scenarios with complex combinations of operating units or complex character skills, thereby improving the user's gaming experience.

[0027] Figure 1 A schematic diagram of the hardware structure of an electronic device is shown, which can be used to implement a battle processing method of an operating unit according to an embodiment of the present disclosure.

[0028] The electronic device 1000 is a device capable of running computer programs. These programs can be local applications installed on the electronic device, or web applications, lightweight applications, or mini-programs, etc., without limitation. The electronic device 1000 can be a mobile phone, tablet computer, PC, etc., without limitation.

[0029] like Figure 1 As shown, the electronic device 1000 may include a processor 1101, a memory 1102, an interface device 1103, a communication device 1104, an output device 1105, an input device 1106, etc. Figure 1 The hardware configuration shown is illustrative only and is not intended to limit this disclosure, its application, or its use.

[0030] The processor 1101 executes computer programs, which can be written using instruction sets of architectures such as x86, Arm, RISC, MIPS, and SSE. The memory 1102 includes, for example, ROM (Read-Only Memory), RAM (Random Access Memory), and non-volatile memory such as a hard disk. The interface device 1103 includes, for example, a USB interface, a network cable interface, and a headphone jack. The communication device 1104 is capable of wired or wireless communication. The communication device 1104 may include at least one short-range communication module, such as any module for short-range wireless communication based on short-range wireless communication protocols such as Hilink, WiFi (IEEE 802.11), Mesh, Bluetooth, ZigBee, Thread, Z-Wave, NFC, UWB, and LiFi. The communication device 1104 may also include a long-range communication module, such as any module for WLAN, GPRS, or 2G / 3G / 4G / 5G long-range communication. The output device 1105 may include, for example, an LCD screen or touch screen, and a speaker. Input device 1106 may include, for example, a touch screen, a keyboard, a microphone, various sensors, etc.

[0031] In this embodiment, the memory 1102 of the electronic device 1000 is used to store a computer program, which is used to control the processor 1101 to operate and execute the battle processing method of the operating unit according to any embodiment of the present disclosure.

[0032] Next, as Figure 1 Taking the electronic device 1000 as an example, various embodiments of the battle processing method of the operation unit are explained.

[0033] <First Embodiment> Figure 2 A battle processing method for an operating unit according to some embodiments is shown, which may include the following steps S210 to S240: Step S210: In response to the user's request for the first operation unit to engage in combat in the game scene, the intelligent agent is invoked to determine the interaction identifier combination of each operation unit in the game scene; wherein, the interaction identifier combination includes the movement identifier corresponding to the first operation unit and the second operation unit in the opposing camp, and the skill identifier configured under their respective movement operations.

[0034] In this embodiment, the game scenario can be of various game formats, including turn-based and time-based competitive gameplay. In a turn-based competitive game scenario, the first operation unit and the second operation unit in the opposing camp can perform corresponding movement and skill release operations in each turn. For example, the first operation unit is a virtual character R1, and the second operation unit is a virtual character R2. In turn one, virtual character R1 moves to [4, 3] and releases skill S1 to virtual character R2. Virtual character R2 remains at its original position [5, 5] in turn one and releases skill S2 to itself. Here, the movement identifier of the first operation unit is [4, 3], the skill identifier of the first operation unit is S1, the movement identifier of the second operation unit is [5, 5], and the skill identifier of the second operation unit is S2. In a time-based competitive game scenario, the first operation unit and the second operation unit in the opposing camp can freely perform movement and skill release operations according to a set timeline. For example, the first operation unit is virtual character R1, the second operation unit is virtual character R2, virtual character R1 moves to [6, 6] at the 5th second, releases skill S3 on virtual character R2 at the 8th second, virtual character R2 moves to [6, 8] at the 6th second, and releases skill S4 on virtual character R1 at the 9th second.

[0035] In this embodiment, the intelligent agent can be an artificial intelligence network model, which has functions such as information extraction and information processing. The intelligent agent can be a general artificial intelligence model or an optimized or trained artificial intelligence model, and there is no limitation here.

[0036] Step S220: Using an intelligent agent, the interaction relationship between the camps of the first and second operation units is determined by using the combination of the unit identifier of the first operation unit, the unit identifier of the second operation unit, and the interaction identifier as the first input condition, and an interaction identifier between the camps is generated.

[0037] In this embodiment, the first operating unit belongs to the user's own faction, and the second operating unit belongs to the opposing faction. The operating unit of the user's own faction can be controlled by the user and an allied user, or it can be controlled solely by the user. The operating unit of the opposing faction can be controlled by an enemy user who is hostile to the user.

[0038] In this embodiment, the inter-faction interaction identifier can be an identifier that focuses on the first operation unit and the second operation unit performing movement operations and skill operations in different factions.

[0039] Step S230: Through the intelligent agent, using the combination of interaction identifiers and the interaction identifiers between factions as the second input conditions, and the operation configuration chain set in the game scene as the configuration constraint conditions, a behavior configuration identifier for the behavior adaptation of the first operation unit is generated, and the battle deployment operation of the first operation unit in the target battle node of the game scene is obtained according to the behavior configuration identifier.

[0040] In this embodiment, the behavior configuration identifier can be one or more sets of operations that the first operation unit can specifically implement in the game scene. Here, the operation set consists of movement operations and / or skill operations.

[0041] Step S240: When the target battle node is reached, the first operation unit is shown to the user to perform the battle deployment operation.

[0042] In this embodiment, the game scene may include multiple battle nodes, which may correspond to a certain round or a certain time period. Taking a battle node corresponding to a round as an example, the first operation unit is the virtual character R3. The battle deployment operation may be that the virtual character R3 moves to [x1, y1], releases skill S5 on [x2, y2] to affect the virtual character R4 of the friendly faction, and also affects the virtual character R5 of the enemy faction.

[0043] The combat processing method for operation units according to embodiments of the present invention solves the problem that existing intelligent agents controlling operation units have simple behaviors and are difficult to adapt to scenarios with combinations of operation units or complex character skills of operation units. Based on this method, when a user submits a combat request to an operation unit, an intelligent agent is invoked to determine the specific movement position and skill release interaction identifier combination of the operation unit in the game scene. Then, the intelligent agent, using unit identifiers such as the operation unit's character health, attack range, and attack power, and interaction identifier combinations, determines the interaction relationship between the first and second operation units' factions to generate inter-faction interaction identifiers. Next, the intelligent agent, using the interaction identifier combination and inter-faction interaction identifiers as second input conditions, and the operation configuration chain set in the game scene as configuration constraints, generates behavior configuration identifiers adapted to the behavior of the first operation unit. Based on the behavior configuration identifiers, the combat deployment operation of the first operation unit at the target combat node in the game scene is obtained and displayed to the user, thus adapting to scenarios with combinations of operation units or complex character skills of operation units, thereby improving the user's gaming experience.

[0044] <Second Embodiment> In this embodiment, in order to extract features related to the specific movement and skills of each operation unit from the unit identifiers of each operation unit in the game scene, and to determine the combination of interaction identifiers, the intelligent agent includes multiple unit processing layers. Through multiple unit processing layers, the movement identifiers and skill identifiers of the corresponding operation units can be processed respectively to determine the skill identifiers configured for all target operation units under each movement identifier.

[0045] In these embodiments, relative to the first embodiment described above, step S210 may include the following step S310: Step S310: For the target operation unit in each operation unit in the game scene, input the character identifier and skill identifier in the unit identifier of the target operation unit into the corresponding unit processing layer, and determine the skill identifier configured for all target operation units under each movement identifier, so as to serve as the interaction identifier combination.

[0046] In some examples, such as Figure 3 The agent shown can be divided into several operation units in the game scene, from unit1 to unitN, where N can be a positive integer greater than 1. Different operation units can correspond to different unit processing layers. Taking the unit processing layer corresponding to unit1 as an example, the basic parameters hero feas and the configured skill feas of operation unit1 are extracted, and the basic parameters hero feas and the configured skill feas are fused into a feature vector representing the skill identifier configured by operation unit1 under each movement identifier through a fully connected layer in the unit processing layer. Different unit processing layers process their respective operation units to determine the skill identifiers configured by all operation units under each movement identifier in the game scene. For example, operation unit R5 moves forward at most 5 squares with full health in round 2, that is, operation unit R5 - round 2 - full health - moves forward at most 5 squares, that is, the furthest distance is [7, 8]. After moving to [7, 8], operation unit R5 can release skills S6 and S7 on [7, 9], thereby obtaining the skill identifiers configured for all operation units in the game scene under each movement identifier.

[0047] <Third Embodiment> In this embodiment, in order to generate more accurate inter-faction interaction identifiers, the intelligent agent includes a unit attention layer and a lineup attention layer. The unit attention layer focuses on the movement and skill operations of the first and second operation units in their respective factions. The lineup attention layer focuses on the interaction relationship between the factions of the first and second operation units, thereby generating inter-faction interaction identifiers that represent the interaction relationship between the friendly and enemy factions.

[0048] In these embodiments, relative to the first embodiment described above, step S220 may include the following steps S410 and S420: Step S410: Input the role identifiers and skill identifiers of the first operation unit and the second operation unit into the unit attention layer, determine the faction to which each role identifier belongs, the correspondence and temporal relationship between each role identifier and each skill identifier, and generate unit self-attention identifiers.

[0049] In this embodiment, as Figure 3 As shown, the role identifiers and skill identifiers of the first and second operation units are input into the unit self-attention 3-layers. The unit self-attention 3-layers are constructed by stacking three identical unit self-attention layers. Based on the correspondence and temporal relationship between the faction, role identifier, and skill identifier to which each operation unit belongs, these 3-layers can obtain unit self-attention identifiers representing the movement and skill operations that each operation unit can perform in the game scene. For example, the correspondence here is that the skill identifiers corresponding to the virtual character R6 belonging to its own faction are S8 and S9. The temporal relationship is, for example, after the virtual character R7 belonging to its own faction releases skill S10 or S11, the virtual character R8 belonging to the enemy faction releases skill S12, S13, or S14.

[0050] Step S420: Input the combination of unit self-attention identifier and interaction identifier into the lineup attention layer to determine the interaction relationship between the first operation unit and the second operation unit in the same camp, and generate the inter-camp interaction identifier.

[0051] In this embodiment, as Figure 3 As shown, the combination of unit self-attention identifiers and interaction identifiers is input to the multi-cross attention 3 layers, which are constructed by stacking three identical cross attention layers. These multi-cross attention 3 layers can determine the interaction relationship between the first and second operation units based on the combination of unit self-attention identifiers and interaction identifiers, generating inter-faction interaction identifiers. For example, the interaction relationship could be that after a virtual character R9 belonging to one's own faction releases skill S15 or S16 against a virtual character R10 belonging to the enemy faction, then a virtual character R10 belonging to the enemy faction releases skill S17 against a virtual character R9 belonging to one's own faction.

[0052] In this embodiment, based on the agent's unit attention layer and lineup attention layer, legitimate relationship types between operation units are set (such as "owning", "acting on", "located in", etc.), and an independent attention head is established for each relationship type. For example, the "operation unit-skill" attention layer specifically learns the adaptation relationship between the operation unit and the carried skill, while the "skill identifier-release object" attention layer focuses on the target selection logic of skill release. This structured attention mechanism significantly reduces the computation of invalid associations and enhances the interpretability of the decision-making process.

[0053] <Fourth Embodiment> In this embodiment, in order to obtain more focused behavioral configuration identifiers related to operation commands, the operation configuration chain set in the game scene includes a first operation configuration of the release object and a second operation configuration of the object to which the release object acts. The first operation configuration includes the basic parameters of the release object, its movement position, and the release skill. The second operation configuration includes the basic parameters of the object to which the release object acts and its affiliated faction. The intelligent agent includes an operation attention layer, which focuses on operation commands to obtain matching behavioral configuration identifiers.

[0054] In these embodiments, relative to the third embodiment described above, step S230 may include the following step S510: Step S510: Input the interaction identifier combination and the inter-faction interaction identifier into the operation attention layer, use the first operation configuration and the second operation configuration as configuration constraints, extract the association between the first operation unit and the second operation unit and the game operation in the game scene respectively, and generate the behavior configuration identifier for the behavior adaptation of the first operation unit.

[0055] In this embodiment, as Figure 3 As shown, the interaction identifier combination and inter-faction interaction identifier are input into the action attention 2layers. These layers focus on the basic parameters of the releasing object, its movement position, the released skill, the basic parameters of the affected object, and its affiliated faction to generate a behavior configuration identifier for the first action unit's behavior adaptation. This behavior configuration identifier can be represented as the releasing object Ra moving to [x1, y1] and releasing skill Sa on [x2, y2], affecting both friendly faction's affected object Rb and enemy faction's affected object Rc. Here, the releasing object Ra is the first action unit.

[0056] In some examples, such as Figure 3As shown, the intelligent agent also includes a decoding layer. Behavior configuration identifiers are input to the target attention layer (TargetAttention) and the decoder (Decoder) through the decoding layer, ultimately yielding the battle deployment operations representing the specific behaviors of each operational unit in the game scene at a target time node or target round. Here, the target time node or target round corresponds to the target battle node.

[0057] <Fifth Embodiment> In this embodiment, in order to extract target behavior configuration identifiers that are more concerned with game rules from the behavior configuration identifiers, the intelligent agent also includes a scene attention layer. Through this scene attention layer, attention is focused on the attribute configurations that conform to the game rules set by the game scene, and the corresponding target behavior configuration identifiers are extracted from the behavior configuration identifiers to ensure the accuracy of the battle deployment operations displayed to the user.

[0058] In these embodiments, relative to the first embodiment described above, the method includes the following steps S610 and S620 before step S230: Step S610: Input the first attribute configuration set by each scene unit in the game scene and the second attribute configuration set by each operation unit in the game scene into the scene attention layer, so as to map the first attribute configuration and the second attribute configuration to the scene thumbnail associated with the game scene and obtain the scene association identifier.

[0059] In this embodiment, each scene unit in the game scene may include obstacles deployed by the first operation unit and / or the second operation unit, and may also include obstacles deployed by the game scene at specific battle nodes. These specific battle nodes can be set based on the complexity of the game scene. For example, if the game scene is highly complex, the specific battle node may be every battle node in the game scene; if the game scene is low-complexity, the specific battle node may be the first battle node in the game scene. Figure 3 As shown, the first attribute configuration includes obstacle traps and basic scene attribute traps. The second attribute configuration includes all hero traps. Mapping the first and second attribute configurations to a minimap associated with the game scene yields a minimap containing information related to both the first and second attribute configurations. This minimap is then processed by the VisionTransformer encoding layer in the scene attention layer to obtain the scene association identifier. The basic scene attribute traps here are, for example, pre-defined attribute configurations for the game scene, such as a 10% increase in health for warrior-type units and a 20% increase in attack power for mage-type units.

[0060] Step S620: Based on the scene association identifier, extract the target behavior configuration identifier that is adapted to the game scene from the behavior configuration identifier.

[0061] In this embodiment, as Figure 3 As shown, inputting scene association identifiers and behavior configuration identifiers into the target attention layer TargetAttention can improve the target behavior configuration identifiers that are adapted to the game scene.

[0062] Based on this, step S230 may include the following step S630: Step S630: Based on the target behavior configuration identifier, obtain the battle deployment operation of the first operation unit in the target battle node of the game scene.

[0063] In this embodiment, a unique semantic space is preserved by setting up a dedicated feature extractor and embedding layer for each type of game entity (character, skill, terrain, effect, etc.). For example, the encoder of an operation unit focuses on attribute values ​​and state changes, the skill encoder of the operation unit's skill identifier emphasizes the scope of action and conditional constraints, and the terrain encoder captures spatial layout and strategic value. This typified encoding method enables the agent to more accurately understand the essential characteristics of various operation units.

[0064] <Sixth Embodiment> In this embodiment, in order to enable the intelligent agent to integrate the constraints related to scene association identifiers, movement operations, and skill operations, the first attribute configuration includes the effect set by each scene unit and the corresponding first plot identifier in the scene thumbnail; the second attribute configuration includes the movement constraints and skill constraints set by each operation unit and the corresponding second plot identifier in the scene thumbnail, and maps the movement constraints and skill constraints to the scene thumbnail.

[0065] In these embodiments, relative to the fifth embodiment described above, step S610 may include the following step S710: Step S710: Input the first attribute configuration set by each scene unit in the game scene and the second attribute configuration set by each operation unit in the game scene into the scene attention layer, so as to map the effect to the first tile corresponding to the first plot identifier of the scene thumbnail, and map the movement constraint and skill constraint to the second tile corresponding to the second plot identifier of the scene thumbnail, so as to obtain the scene association identifier.

[0066] In this embodiment, movement constraints and skill constraints can be expressed using a unified mathematical expression, based on a spatial coordinate system and transformation mechanism, and a fast conversion algorithm between the map coordinate system and the unit coordinate system centered on the operation unit. This allows the casting range corresponding to the skill identifier and the movement distance corresponding to the movement identifier to be represented by an M×M two-dimensional grid. By encoding the skill constraints of the operation unit's casting range and the movement constraints of its movement distance into convolution kernels of specific shapes and weights, all possible target positions are quickly calculated through convolution operations. For example, a circular casting range corresponds to a circular convolution kernel, and a linear attack range corresponds to a bar convolution kernel. Based on the cascaded convolution calculation of rule constraints, the basic casting target matrix T of the operation unit is constructed. Then, the range is filtered by the convolution kernel R to generate an effective casting matrix TM. Simultaneously, based on the preset movement constraints of the operation unit, the movement matrix RM of the operation unit is constructed. Finally, multiple constraints are fused through matrix intersection operations to generate the final legal action mask, which is mapped onto the second tile corresponding to the second plot identifier in the scene thumbnail, thereby realizing the digital representation of the operation unit's behavior in the game scene.

[0067] <Seventh Embodiment> In this embodiment, in order to obtain more detailed combat deployment operations, based on the behavior configuration identifier, it can be determined that the first operation unit releases the target skill to the third operation unit of its own camp and / or the second operation unit of the enemy camp when the target combat node moves to the target movement position.

[0068] In these embodiments, relative to the first embodiment described above, step S230 may include the following step S810: Step S810: Based on the behavior configuration identifier, obtain the combat deployment operation in which the first operation unit of the target combat node in the game scene moves to the target movement position and releases the target skill to the third operation unit of the friendly camp and / or the second operation unit of the enemy camp.

[0069] In some examples, the first operation unit is the virtual character R11. The battle deployment operation can be that the virtual character R11 of the friendly faction moves to [2, 2] as the release target and releases the skill S18 on [2, 4] to affect the virtual character R12 of the friendly faction and the virtual character R13 of the enemy faction, so that the virtual character R12's health increases by 10% and the virtual character R13's attack power decreases by 20%.

[0070] <Eighth Embodiment> In this embodiment, in order to improve the accuracy of the combat deployment operations output by the agent, the agent can be trained using a training sample set.

[0071] In these embodiments, relative to the first embodiment described above, the method further includes the following steps S910 and S920 before step S210: Step S910: Obtain the training sample set; wherein the training sample set includes training samples and sample labels.

[0072] In this embodiment, the training sample set can be game battles involving real players, and the sample labels are real battle labels.

[0073] Step S920: Train the agent using training samples to obtain a trained agent; wherein, the entropy coefficient of the agent's loss function is constructed by the ratio of the current entropy and the maximum entropy of the loss function.

[0074] In this embodiment, the expression for entropy loss in the loss function of the agent is as follows: (1) (2) In equations (1) and (2), Represents the entropy value. This represents the probability of the i-th action occurring. Represents the entropy value. It is the maximum possible entropy calculated using a mask. It is the entropy output by the current agent. This represents the entropy ratio.

[0075] In this embodiment, by monitoring indicators such as policy entropy, win rate fluctuation, and tactical novelty during the training process in real time, the opponent sampling weight, learning rate scaling factor, and exploration coefficient are automatically adjusted. When policy innovation stagnation is detected, highly random operation units are proactively introduced or the policy is intentionally "downgraded" to an earlier policy version to revitalize exploration. When the policy tends to stabilize and converge, the strength of the operation units is increased and the training parameters are tightened to encourage the agent to perform refined policy optimization.

[0076] <Device Embodiment> Figure 4 A schematic diagram of the composition structure of a battle processing apparatus for an operating unit according to an embodiment of the present disclosure is shown. Figure 4 As shown, the battle processing device 400 of the operation unit includes a response module 410, a generation module 420, an acquisition module 430, and a display module 440.

[0077] The response module 410 is used to invoke the intelligent agent to determine the interaction identifier combination of each operation unit in the game scene in response to the user's battle request for the first operation unit; wherein, the interaction identifier combination includes the movement identifier corresponding to the first operation unit and the second operation unit in the opposing camp, and the skill identifier configured under their respective movement operations. The generation module 420 is used to determine the interaction relationship between the camps of the first operation unit and the second operation unit by using the intelligent agent, the unit identifier of the first operation unit, the unit identifier of the second operation unit and the interaction identifier combination as the first input conditions, and generate the interaction identifier between the camps. The module 430 is used to generate a behavior configuration identifier for the behavior adaptation of the first operation unit by using the intelligent agent, with the interaction identifier combination and the interaction identifier between the factions as the second input conditions, and the operation configuration chain set by the game scene as the configuration constraint conditions, and to obtain the battle deployment operation of the first operation unit in the target battle node of the game scene based on the behavior configuration identifier. The display module 440 is used to display the first operation unit performing the battle deployment operation to the user when the target battle node is in operation.

[0078] In some embodiments, the response module 410 is further configured to input the character identifier and skill identifier in the unit identifier of the target operation unit in each operation unit in the game scene to the corresponding unit processing layer, and determine the skill identifier configured by all target operation units under each movement identifier, so as to serve as the interaction identifier combination.

[0079] In some embodiments, the generation module 420 is further configured to input the role identifier and skill identifier of the first operation unit and the second operation unit respectively into the unit attention layer, determine the faction to which each role identifier belongs, the correspondence and timing relationship between each role identifier and each skill identifier, and generate a unit self-attention identifier; input the unit self-attention identifier and interaction identifier together into the lineup attention layer, determine the interaction relationship between the factions to which the first operation unit and the second operation unit belong, and generate an inter-faction interaction identifier.

[0080] In some embodiments, the obtaining module 430 is further configured to input the interaction identifier combination and the inter-faction interaction identifier to the operation attention layer, using the first operation configuration and the second operation configuration as configuration constraints, extract the association relationship between the first operation unit and the second operation unit and the game operation in the game scene, and generate a behavior configuration identifier for the behavior adaptation of the first operation unit.

[0081] In some embodiments, the battle processing device 400 of the operation unit includes a feature extraction module, which is used to input the first attribute configuration set by each scene unit in the game scene and the second attribute configuration set by each operation unit in the game scene into the scene attention layer, so as to map the first attribute configuration and the second attribute configuration to the scene thumbnail associated with the game scene to obtain the scene association identifier; based on the scene association identifier, the target behavior configuration identifier that is adapted to the game scene is extracted from the behavior configuration identifier; The module 430 is also used to obtain the battle deployment operation of the first operation unit on the target battle node in the game scene based on the target behavior configuration identifier.

[0082] In some embodiments, the feature extraction module is further configured to input the first attribute configuration set by each scene unit in the game scene and the second attribute configuration set by each operation unit in the game scene into the scene attention layer, so as to map the effect to the first tile corresponding to the first plot identifier of the scene thumbnail, and map the movement constraint and skill constraint to the second tile corresponding to the second plot identifier of the scene thumbnail, so as to obtain the scene association identifier.

[0083] In some embodiments, the obtaining module 430 is further configured to obtain, based on the behavior configuration identifier, a battle deployment operation in which the first operation unit of the target battle node in the game scene moves to the target movement position and releases the target skill to the third operation unit of its own camp and / or the second operation unit of the enemy camp.

[0084] In some embodiments, the battle processing device 400 of the operation unit includes a training module for acquiring a training sample set; wherein the training sample set includes training samples and sample labels; the agent is trained using the training samples to obtain a trained agent; wherein the entropy coefficient of the loss function of the agent is constructed by the ratio of the current entropy and the maximum entropy of the loss function.

[0085] <Equipment Example> Figure 5 A schematic diagram of the hardware structure of an electronic device according to some other embodiments is shown. For example... Figure 5 As shown, the electronic device 500 includes a processor 510 and a memory 520. The memory 520 stores a computer program that controls the processor 510 to operate in order to control the electronic device 500 to execute the battle processing method of the operating unit according to any embodiment of the present disclosure.

[0086] This disclosure also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements a battle processing method of an operating unit according to any embodiment of this disclosure.

[0087] This disclosure also provides a computer program product, which includes a computer program or instructions that, when executed by a processor, implement a battle processing method of an operating unit according to any of the disclosed embodiments.

[0088] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. This disclosure can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium loaded with computer-readable program instructions for causing a processor to implement any of the methods in the foregoing embodiments of this disclosure. A computer-readable storage medium may be a tangible device capable of holding and storing instructions used by an instruction execution device. For example, a computer-readable storage medium may include an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), compact disc-read-only memory (CD-ROM), digital versatile disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any combination thereof. Computer-readable storage media as used herein is not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires. The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include one or more of copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to computer-readable storage media in the respective computing / processing device.The computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source or object programs written in any combination of one or more programming languages, including object-oriented programming languages ​​(such as Smalltalk, C++, etc.) and conventional procedural programming languages ​​(such as the "C" language or similar programming languages). The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network (e.g., a local area network or a wide area network), or it may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays, or programmable logic arrays, may be customized by utilizing state information from the computer-readable program instructions to execute the computer-readable program instructions, thereby implementing various aspects of the embodiments of this disclosure. Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus, and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions. These computer-readable program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions that execute on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions. It should be noted that implementation in hardware, implementation in software, and implementation using a combination of software and hardware are all equivalent. Various embodiments of the present disclosure have been described above; the above description is exemplary and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein. The scope of this disclosure is defined by the appended claims.

Claims

1. A method for processing combat scenarios using an operational unit, characterized in that, The method includes: In response to a user's request for the first operation unit to engage in combat within the game scene, an intelligent agent is invoked to determine the interaction identifier combination of each operation unit in the game scene; wherein, the interaction identifier combination includes the movement identifier corresponding to the first operation unit and the second operation unit in the opposing camp, and the skill identifier configured under their respective movement operations. Using the intelligent agent, the interaction relationship between the camps of the first operation unit and the second operation unit is determined by the combination of the unit identifier of the first operation unit, the unit identifier of the second operation unit, and the interaction identifier, and an inter-camp interaction identifier is generated. Using the intelligent agent, the interaction identifier combination and the inter-faction interaction identifier are used as the second input conditions, and the operation configuration chain set by the game scene is used as the configuration constraint condition to generate the behavior configuration identifier adapted to the behavior of the first operation unit. Based on the behavior configuration identifier, the battle deployment operation of the first operation unit on the target battle node in the game scene is obtained. When the target battle node is reached, the first operation unit is shown to the user to perform the battle deployment operation.

2. The method according to claim 1, characterized in that, The intelligent agent includes multiple unit processing layers; the invoking intelligent agent determines the interaction identifier combination of each operation unit in the game scene, including: For each target operation unit in the game scene, the character identifier and skill identifier in the unit identifier of the target operation unit are input into the corresponding unit processing layer to determine the skill identifiers configured for all target operation units under each movement identifier, so as to form an interaction identifier combination.

3. The method according to claim 2, characterized in that, The intelligent agent includes a unit attention layer and a lineup attention layer; the process involves using the intelligent agent, with the unit identifier of the first operation unit, the unit identifier of the second operation unit, and the interaction identifier combination as the first input condition, to determine the interaction relationship between the lineups of the first operation unit and the second operation unit, and generating inter-lineage interaction identifiers, including: The role identifier and skill identifier of the first operation unit and the second operation unit are input into the unit attention layer to determine the faction to which each role identifier belongs, the correspondence and temporal relationship between each role identifier and each skill identifier, and generate unit self-attention identifier; The unit self-attention identifier and the interaction identifier are combined and input into the lineup attention layer to determine the interaction relationship between the first operation unit and the second operation unit in the same camp, and generate the inter-camp interaction identifier.

4. The method according to claim 1, characterized in that, The operation configuration chain set in the game scene includes a first operation configuration for the released object and a second operation configuration for the object affected by the released object. The first operation configuration includes the basic parameters of the released object, its movement position, and the skill it releases. The second operation configuration includes the basic parameters of the affected object and its affiliated faction. The intelligent agent includes an operation attention layer. The process involves generating a behavior configuration identifier for the first operation unit's behavior adaptation using the intelligent agent, with the interaction identifier combination and the inter-faction interaction identifier as second input conditions, and the operation configuration chain set in the game scene as configuration constraints. This includes: The interaction identifier combination and the inter-faction interaction identifier are input into the operation attention layer. The first operation configuration and the second operation configuration are used as configuration constraints to extract the association between the first operation unit and the second operation unit and the game operation in the game scene, and generate the behavior configuration identifier for the behavior adaptation of the first operation unit.

5. The method according to claim 1, characterized in that, The intelligent agent further includes a scene attention layer; the method further includes: The first attribute configuration set by each scene unit in the game scene and the second attribute configuration set by each operation unit in the game scene are input into the scene attention layer, so as to map the first attribute configuration and the second attribute configuration to the scene thumbnail associated with the game scene to obtain the scene association identifier; Based on the scene association identifier, extract the target behavior configuration identifier that is adapted to the game scene from the behavior configuration identifier; The step of obtaining the battle deployment operation of the first operation unit on the target battle node in the game scene based on the behavior configuration identifier includes: Based on the target behavior configuration identifier, the first operation unit obtains the battle deployment operation of the target battle node in the game scene.

6. The method according to claim 5, characterized in that, The first attribute configuration includes the effect set by each scene unit and the corresponding first plot identifier in the scene thumbnail; the second attribute configuration includes the movement constraints and skill constraints set by each operation unit and the corresponding second plot identifier in the scene thumbnail; the step of inputting the first attribute configuration set by each scene unit in the game scene and the second attribute configuration set by each operation unit in the game scene into the scene attention layer, so as to map the first attribute configuration and the second attribute configuration to the scene thumbnail associated with the game scene to obtain the scene association identifier, includes: The first attribute configuration set by each scene unit in the game scene and the second attribute configuration set by each operation unit in the game scene are input into the scene attention layer, so as to map the effect to the first tile corresponding to the first plot identifier of the scene thumbnail, and map the movement constraint and the skill constraint to the second tile corresponding to the second plot identifier of the scene thumbnail, so as to obtain the scene association identifier.

7. The method according to claim 1, characterized in that, The step of obtaining the battle deployment operation of the first operation unit on the target battle node in the game scene based on the behavior configuration identifier includes: Based on the behavior configuration identifier, a battle deployment operation is obtained in which the first operation unit of the target battle node in the game scene moves to the target movement position and releases the target skill to the third operation unit of the friendly camp and / or the second operation unit of the enemy camp.

8. The method according to claim 1, characterized in that, The method further includes: Obtain a training sample set; wherein the training sample set includes training samples and sample labels; The agent is trained using the training samples to obtain a trained agent; wherein the entropy coefficient of the loss function of the agent is determined by the entropy ratio constructed from the current entropy and the maximum entropy of the loss function.

9. A combat processing device for an operating unit, characterized in that, The device includes: The response module is used to invoke the intelligent agent to determine the interaction identifier combination of each operation unit in the game scene in response to the user's battle request for the first operation unit; wherein, the interaction identifier combination includes the movement identifier corresponding to the first operation unit and the second operation unit in the opposing camp and the skill identifier configured under their respective movement operations. The generation module is used to determine the interaction relationship between the camps of the first operation unit and the second operation unit through the intelligent agent, using the unit identifier of the first operation unit, the unit identifier of the second operation unit, and the interaction identifier combination as the first input condition, and generate an inter-camp interaction identifier. The module is used to generate a behavior configuration identifier for the behavior adaptation of the first operation unit by using the intelligent agent, the interaction identifier combination and the inter-faction interaction identifier as the second input condition, and the operation configuration chain set by the game scene as the configuration constraint condition, and to obtain the battle deployment operation of the first operation unit at the target battle node in the game scene according to the behavior configuration identifier. The display module is used to display to the user, when the target battle node is reached, the first operation unit performs the battle deployment operation.

10. An electronic device, characterized in that, The system includes a memory and a processor, the memory being used to store a computer program; the processor being used to execute the computer program to implement the method according to any one of claims 1 to 8.

Images