CN111359212A - Game object control and model training method and device - Google Patents
Game object control and model training method and device Download PDFInfo
- Publication number
- CN111359212A CN111359212A CN202010106878.9A CN202010106878A CN111359212A CN 111359212 A CN111359212 A CN 111359212A CN 202010106878 A CN202010106878 A CN 202010106878A CN 111359212 A CN111359212 A CN 111359212A
- Authority
- CN
- China
- Prior art keywords
- action
- game object
- game
- state information
- preset range
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 72
- 238000012549 training Methods 0.000 title claims abstract description 30
- 230000009471 action Effects 0.000 claims abstract description 306
- 238000012545 processing Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 4
- 230000010365 information processing Effects 0.000 claims description 3
- 238000013473 artificial intelligence Methods 0.000 description 17
- 230000008569 process Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 239000003814 drug Substances 0.000 description 5
- 230000006399 behavior Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 239000008280 blood Substances 0.000 description 2
- 210000004369 blood Anatomy 0.000 description 2
- 238000001816 cooling Methods 0.000 description 2
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- A63F13/00—Video games, i.e. games using an electronically generated display having two or more dimensions
- A63F13/55—Controlling game characters or game objects based on the game progress
- A63F13/56—Computing the motion of game characters with respect to other game characters, game objects or elements of the game scene, e.g. for simulating the behaviour of a group of virtual soldiers or for path finding
Abstract
The application provides a game object control method and a game object control model training method and device, wherein the game object control method comprises the following steps: acquiring game state information in a game scene within a preset range of a game object; determining a first action corresponding to the game object based on the game state information; and acquiring a second action currently executed by the game object, and controlling the game object to execute the first action if the second action can be interrupted by the first action. According to the game state variable control method and device, the game state information in the game scene in the preset range of the game object is used as the game state variable, so that the consumption of computer resources can be reduced, and the action learning efficiency can be improved.
Description
Technical Field
The application relates to the technical field of artificial intelligence, in particular to a game object control and model training method and device.
Background
The development of AI (Artificial Intelligence) technology provides game AI robots in games with more diverse behavioral capabilities. Currently, in the process of training a game AI robot control model or controlling a game AI robot to execute an action, a full game map image is used as a part of a game state variable to obtain an action to be executed by the game AI robot next.
In The research of The applicant, it is found that in The prior art, due to The fact that The full game map image is too large, a large amount of computer resources are required in The process of training The game AI robot control model or controlling The game AI robot to perform The action, and The state space is increased sharply, so that a dimensional disaster (The current of dimensional) is caused, The action learning efficiency is reduced, and The transmission of The full game map image increases an extra bandwidth burden, so that The action learning efficiency is further reduced.
Content of application
In view of the above, an object of the present invention is to provide a game object control method and a game object model training method and apparatus, which can improve the action learning efficiency in the process of training a game AI robot control model or controlling a game AI robot.
In a first aspect, an embodiment of the present application provides a method for controlling a game object, where the method may include:
acquiring game state information in a game scene within a preset range of a game object;
determining a first action corresponding to the game object based on the game state information;
and acquiring a second action currently executed by the game object, and controlling the game object to execute the first action if the second action can be interrupted by the first action.
With reference to the first aspect, this application provides a first possible implementation manner of the first aspect, where the preset range includes a range in which a distance to the game object is smaller than a preset distance.
With reference to the first aspect, this application provides a second possible implementation manner of the first aspect, where the preset range is smaller than a range of an entire game scene.
In combination with the first aspect, embodiments of the present application provide a second possible implementation manner of the first aspect, where the first action includes a plurality of single actions in succession; the controlling the game object to perform the first action comprises: and controlling the game object to execute each single action in turn.
With reference to the first aspect, this application provides a third possible implementation manner of the first aspect, where if the first action cannot interrupt the second action and the second action includes multiple single actions in succession, the single action that is not executed by the second action is continuously executed.
With reference to the first aspect, this application provides a fourth possible implementation manner of the first aspect, where the first action includes a single action; the controlling the game object to perform the first action comprises: controlling the game object to perform a single action of the first actions.
With reference to the first aspect, this application provides a fifth possible implementation manner of the first aspect, where if the second action is not obtained and the first action includes multiple consecutive single actions, the game object is controlled to sequentially execute each single action.
With reference to the first aspect, an embodiment of the present application provides a sixth possible implementation manner of the first aspect, where after the obtaining of the second action, the method may further include:
acquiring at least one preset interrupting action capable of interrupting the second action;
determining whether the first action is one of the at least one breaking action;
and if so, determining that the first action can interrupt the second action.
With reference to the sixth possible implementation manner of the first aspect, this application example provides a seventh possible implementation manner of the first aspect, where before acquiring the at least one breaking action, the method may further include:
setting at least one breaking action for the second action;
storing the set at least one breaking action into a breaking action set.
With reference to the sixth possible implementation manner of the first aspect, this application example provides an eighth possible implementation manner of the first aspect, where if the first action can interrupt the second action and the first action includes multiple single actions in succession, the method may further include:
storing a plurality of continuous single actions included in the first action into an action queue;
the controlling the game object to perform the first action includes:
and sequentially taking out each single action from the action queue, and controlling the game object to execute the taken-out single action.
With reference to the first aspect, this application provides a ninth possible implementation manner of the first aspect, where the game state information includes at least one of: battle force information of the game object; the fighting capacity information of other game objects in the game scene in the preset range of the game object; game object position information; position information of other game objects within the game scene within a preset range of the game object.
In a second aspect, an embodiment of the present application further provides a method for training a game object control model, which may include:
acquiring game state information in a game scene within a preset range of a game object;
inputting the game state information into a game object control model to be trained, processing the input game state information through the game object control model, and outputting a first action corresponding to the game object;
and adjusting model parameters in a game object control model to be trained on the basis of revenue information generated by the game object executing the first action.
In combination with the second aspect, the present application provides a first possible implementation manner of the second aspect, where the preset range includes a range in which a distance to the game object is smaller than a preset distance.
With reference to the second aspect, the present application provides a second possible implementation manner of the second aspect, where the preset range is smaller than a range of an entire game scene.
In combination with the second aspect, the present embodiments provide a third possible implementation manner of the second aspect, where the first action includes a plurality of single actions in succession;
the adjusting model parameters in a game object control model to be trained based on revenue information generated by the game object performing the first action comprises:
adjusting model parameters in a game object control model to be trained based on revenue information generated by the game object performing each single action.
In a third aspect, an embodiment of the present application further provides a control device for a game object, including:
the acquisition module is used for acquiring game state information in a game scene in a preset range of a game object;
the determining module is used for determining a first action corresponding to the game object based on the game state information;
and the control module is used for acquiring a second action currently executed by the game object, and controlling the game object to execute the first action if the second action can be interrupted by the first action.
In a fourth aspect, an embodiment of the present application further provides a training device for a game object control model, including:
the information acquisition module is used for acquiring game state information in a game scene in a preset range of a game object;
the information processing module is used for inputting the game state information into a game object control model to be trained, processing the input game state information through the game object control model and outputting a first action corresponding to the game object;
and the parameter adjusting module is used for adjusting model parameters in the game object control model to be trained on the basis of the income information generated by the game object executing the first action.
In a fifth aspect, an embodiment of the present application further provides an electronic device, including: a processor, a storage medium and a bus, the storage medium storing machine-readable instructions executable by the processor, the processor and the storage medium communicating via the bus when the electronic device is operating, the processor executing the machine-readable instructions to perform the steps of any one of the possible implementations of the first aspect or the steps of any one of the possible implementations of the second aspect.
In a sixth aspect, this application further provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to perform the steps in any one of the possible implementation manners of the first aspect or the steps in any one of the possible implementation manners of the second aspect.
According to the game object control method and the game object control model training method, the game state information in the game scene in the preset range of the game object is acquired, the first action corresponding to the game object is determined based on the game state information, and then the game object is controlled or the game object control model is trained. Compared with the prior art that action learning efficiency is greatly reduced and computer resources are consumed too much due to the fact that actions to be executed next by the game AI robot are obtained by taking a complete game map image as a part of the game state variables, game state information in a game scene within a preset range of a game object is taken as the game state variables, consumption of computer resources can be reduced, and action learning efficiency can be improved.
Further, the control method for the game object provided by the embodiment of the present application, wherein the first action includes a plurality of continuous single actions, that is, the first action is an action having a certain logical meaning, for example, finding a hiding point to hide, finding a safe place to take medicine, getting out an attack place from the hidden place, and the like. Compared with the prior art, the game object is controlled or the game object control model is trained based on the first action, so that the game object has more real behaviors in the game, and better game experience is provided for game players.
In order to make the aforementioned objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.
FIG. 1 is a flow chart illustrating a method for controlling a game object according to an embodiment of the present disclosure;
FIG. 2 is a flow chart illustrating a method for training a game object control model according to an embodiment of the present application;
FIG. 3 is a schematic structural diagram of a control device for a game object according to an embodiment of the present disclosure;
FIG. 4 is a schematic structural diagram of an exercise device for a game object control model according to an embodiment of the present disclosure;
fig. 5 shows a schematic structural diagram of an electronic device provided in an embodiment of the present application.
Detailed Description
In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it should be understood that the drawings in the present application are for illustrative and descriptive purposes only and are not used to limit the scope of protection of the present application. The flowcharts used in this application illustrate operations implemented according to some embodiments of the present application. It should be understood that the operations of the flow diagrams may be performed out of order, and steps without logical context may be performed in reverse order or simultaneously. One skilled in the art, under the guidance of this application, may add one or more other operations to, or remove one or more operations from, the flowchart.
In addition, the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.
In order to enable a person skilled in the art to use the present disclosure, the following embodiments are given in connection with a specific application scenario "control of a game AI robot and training of a control model". It will be apparent to those skilled in the art that the general principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the application. Although the present application is described primarily in the context of "control of gaming AI robots and training of control models," it should be understood that this is merely one exemplary embodiment.
It should be noted that in the embodiments of the present application, the term "comprising" is used to indicate the presence of the features stated hereinafter, but does not exclude the addition of further features.
To facilitate understanding of the present embodiment, a detailed description will be given of a control method of a game object disclosed in the embodiments of the present application.
Referring to fig. 1, fig. 1 is a flowchart of a method for controlling a game object according to an embodiment of the present application. As shown in fig. 1, the control method of the game object may include:
s101, obtaining game state information in a game scene in a preset range of a game object.
Wherein the game state information includes at least one of: battle force information of the game object; the fighting capacity information of other game objects in the game scene in the preset range of the game object; game object position information; position information of other game objects within the game scene within a preset range of the game object.
In the specific implementation, the game object emits rays to the periphery, and then the game state information of the area near the game object (the near area is a preset range) is determined through the feedback information of the rays. The present application is not limited to this, and the game state information of the area near the game object may be acquired by other technical means.
The fighting capacity information of the game object may include, among other things, the blood volume of the game object, skill cooling time, and the like. A game object here generally refers to a game AI robot in a game scene.
In a possible embodiment, the preset range includes a range in which a distance to the game object is smaller than a preset distance.
In another possible embodiment, the preset range is smaller than the range of the entire game scene.
In the step, the complete map information is not adopted as a part of the game state variable, but the game state information of the area near the game object is adopted as a part of the game state variable, so that the state space in the action learning process is greatly reduced, the action learning efficiency is improved, the bandwidth required for transmitting the game state variable is reduced, and the action learning efficiency is further improved.
S102, determining a first action corresponding to the game object based on the game state information.
In this step, the game state information is input into a trained game object control model, the input game state information is processed by the game object control model, and a first action corresponding to the game object is output.
S103, acquiring a second action currently executed by the game object, and controlling the game object to execute the first action if the second action can be interrupted by the first action.
In this step, when the second action is interrupted by the first action, the execution of the first action is switched to. Preferably, the first action comprises a plurality of single actions in succession, that is, the first action is a sequence of actions having a certain logical meaning, such as finding a hidden place to hide, finding a safe place to take a medicine, coming out from the hidden place to attack a place, and the like. In the present embodiment, it is defined as "macro action". The controlling the game object to perform the first action includes: and controlling the game object to execute each single action in turn. Compared with the prior art that the game object does not appear like a person under a certain condition due to the fact that only a single action is adopted, and subjective experience of a game player is reduced, the game object is controlled based on the first action, so that the game object can have more real behaviors in a game, and better game experience is provided for the game player.
With reference to step S103, in one possible implementation manner, after the second action is acquired, the method further includes: and acquiring at least one preset breaking action capable of breaking the second action, judging whether the first action is one of the at least one breaking action, and if so, determining that the first action can break the second action.
With reference to the foregoing possible embodiments, before acquiring the at least one breaking action, the method further includes: and setting at least one breaking action for the second action, and storing the set at least one breaking action into a breaking action set. Wherein, the set of the breaking actions is the union of the single action and the macro action. Upon switching to performing the first action, the interrupting action set may be optionally cleared
In connection with step S103, when switching to execute the first action, the method further includes: and setting at least one breaking action for the first action, storing the at least one breaking action, and optionally storing the at least one breaking action into a breaking action set during storage so as to prepare for judging whether the subsequent action can break the first action. When executing a first action, storing a plurality of continuous single actions included in the first action into an action queue, sequentially taking out each single action from the action queue, and controlling a game object to execute the taken-out single action. In particular implementations, other ways may also be selected to cache consecutive single actions comprised by the first action. Optionally, a second action in the action queue may be emptied to free up more storage before the first action is stored in the action queue.
In one possible embodiment, if the first action is unable to interrupt the second action and the second action comprises a plurality of single actions in succession, then the single action not performed by the second action is continued.
In this embodiment, the second action is stored in the action queue before execution, wherein the second action comprises a plurality of consecutive single actions. And when the second action is not interrupted by the first action, continuing to execute the single action which is not executed by the second action in the action queue. And if the second action is a single action, executing the second action in the action queue. In particular implementations, other ways may also be selected to cache consecutive single actions comprised by the second action.
In one possible embodiment, the first action comprises a single action. The controlling the game object to perform the first action includes: and controlling the game object to execute a single action in the first actions.
In this embodiment, when a first action is executed, a single action in the first action is stored in an action queue, and the single action is taken out from the action queue, and the game object is controlled to execute the taken-out single action. In particular implementations, other ways may also be selected to cache a single one of the first actions.
In one possible embodiment, if the second action is not obtained and the first action includes a plurality of single actions in succession, the game object is controlled to execute each of the single actions in turn.
In this embodiment, if the second action is not obtained, it is not necessary to determine whether there is another action capable of interrupting the second action, and the consecutive single actions included in the first action are directly stored in the action queue, and each single action is sequentially taken out from the action queue, and the game object is controlled to execute the taken-out single action. In particular implementations, other ways may also be selected to cache consecutive single actions comprised by the first action. If the first action is a single action, directly storing the single action in the first action into an action queue, and controlling the game object to execute the extracted single action from the action queue. In particular implementations, other ways may also be selected to cache a single one of the first actions.
According to the control method of the game object, the game state information in the game scene in the preset range of the game object is acquired, the first action corresponding to the game object is determined based on the game state information, and then the game object is controlled. Compared with the prior art that action learning efficiency is greatly reduced and computer resources are consumed too much due to the fact that actions to be executed next by the game AI robot are obtained by taking a complete game map image as a part of the game state variables, game state information in a game scene within a preset range of a game object is taken as the game state variables, consumption of computer resources can be reduced, and action learning efficiency can be improved.
Further, the control method for the game object provided by the embodiment of the present application, wherein the first action includes a plurality of continuous single actions, that is, the first action is an action having a certain logical meaning, for example, finding a hiding point to hide, finding a safe place to take medicine, getting out an attack place from the hidden place, and the like. Compared with the prior art, the game object is controlled based on the first action, so that the game object has more real behaviors in the game, and better game experience is provided for game players.
Based on the same technical concept, the embodiment of the present application further provides a training method for a game object control model, which can be specifically referred to in the following embodiments.
Referring to fig. 2, fig. 2 is a flowchart illustrating a method for training a game object control model according to an embodiment of the present disclosure. As shown in fig. 2, the training method of the game object control model may include:
s201, obtaining game state information in a game scene in a preset range of a game object.
Wherein the game state information includes at least one of: battle force information of the game object; the distance between the game object and the game object is smaller than the battle force information of other game objects in a preset distance game scene; game object position information; the distance from the game object is less than the position information of other game objects in the preset distance game scene.
In the specific implementation, the game object emits rays to the periphery, and then the game state information of the area near the game object is determined through the feedback information of the rays. The present application is not limited to this, and the game state information of the area near the game object may be acquired by other technical means.
The fighting capacity information of the game object may include, among other things, the blood volume of the game object, skill cooling time, and the like. A game object here generally refers to a game AI robot in a game scene.
In a possible embodiment, the preset range includes a range in which a distance to the game object is smaller than a preset distance.
In another possible embodiment, the preset range is smaller than the range of the entire game scene.
In the step, the complete map information is not adopted as a part of the game state variable, but the game state information of the area near the game object is adopted as a part of the game state variable, so that the state space in the action training process is greatly reduced, the action learning efficiency is improved, the bandwidth required for transmitting the game state variable is reduced, and the action learning efficiency and the training efficiency of the action training framework are further improved.
S202, inputting the game state information into a game object control model to be trained, processing the input game state information through the game object control model, and outputting a first action corresponding to the game object.
In one possible implementation manner, a second action currently being executed by the game object is acquired, and if the second action can be interrupted by the first action, the game object is controlled to execute the first action.
In one possible implementation manner, a second action currently being executed by the game object is acquired, and if the first action cannot interrupt the second action, the game object is controlled to continue executing the second action.
In one possible embodiment, if the second action is not obtained, the game object is controlled to directly execute the first action.
S203, adjusting model parameters in the game object control model to be trained based on the income information generated by the game object executing the first action.
The above steps S201 to S203 are an iterative process, and a trained game object control model is finally obtained by continuously adjusting model parameters through loop iteration until a game wins or a game termination condition is triggered.
In one possible embodiment, the first action comprises a plurality of single actions in succession, and the model parameters in the game object control model to be trained are adjusted based on revenue information generated by the game object performing each single action.
In this embodiment, the first action includes a plurality of single actions in succession, that is, the first action is an action having a certain logical meaning, such as finding a hidden place to hide, finding a safe place to take a medicine, coming out from the hidden place to attack a place, and the like. In this embodiment, the game object control model is defined as a "macro action" which is different from a game object in performance under a certain condition due to the fact that only a single action is adopted in the prior art, and subjective experience of a game player is reduced.
According to the training method of the game object control model, the game state information in the game scene in the preset range of the game object is acquired, the first action corresponding to the game object is determined based on the game state information, and then the game object control model is trained. Compared with the prior art that the action learning efficiency is greatly reduced and the training resources are consumed too much due to the fact that the game object control model is trained by taking the complete game map image as a part of the game state variable, the game state information in the game scene within the preset range of the game object is taken as the game state variable, the consumption of the training resources can be reduced, and the action learning efficiency can be improved.
Further, in the training method for a game object control model provided in the embodiment of the present application, the first action includes a plurality of continuous single actions, that is, the first action is an action having a certain logical meaning, for example, finding a hidden point to hide, finding a safe place to take medicine, getting out of the hidden place to attack a place, and the like. Compared with the prior art, the game object control model is trained based on the first action, so that the game object has more real behaviors in the game, and better game experience is provided for game players.
Referring to fig. 3, fig. 3 is a schematic structural diagram of a control device for a game object according to an embodiment of the present application. As shown in fig. 3, a control device 300 for a game object according to an embodiment of the present invention includes:
an obtaining module 310, configured to obtain game state information in a game scene within a preset range of a game object;
a determining module 320, configured to determine, based on the game state information, a first action corresponding to the game object;
the control module 330 is configured to obtain a second action currently performed by the game object, and if the second action can be interrupted by the first action, control the game object to perform the first action.
Referring to fig. 4, fig. 4 is a schematic structural diagram of a training device for a game object control model according to an embodiment of the present application. As shown in fig. 4, an exercise apparatus 400 for a game object control model provided in an embodiment of the present application includes:
the information acquisition module 410 is used for acquiring game state information in a game scene within a preset range of a game object;
the information processing module 420 is configured to input the game state information into a game object control model to be trained, process the input game state information through the game object control model, and output a first action corresponding to the game object;
and the parameter adjusting module 430 is configured to adjust model parameters in the game object control model to be trained based on revenue information generated by the game object performing the first action.
An embodiment of the present application discloses an electronic device, as shown in fig. 5, including: a processor 501, a memory 502 and a bus 503, wherein the memory 502 stores machine-readable instructions executable by the processor 501, and when the electronic device is operated, the processor 501 and the memory 502 communicate with each other through the bus 503.
The machine readable instructions, when executed by the processor 501, perform any of the methods described in the previous method embodiments, and specific implementation may refer to the method embodiments, which are not described herein again.
The computer program product of the game object control method and the game object control model training method provided in the embodiments of the present application includes a computer-readable storage medium storing a non-volatile program code executable by a processor, where instructions included in the program code may be used to execute any one of the methods described in the foregoing method embodiments, and specific implementation may refer to the method embodiments, and will not be described herein again.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present application, and are used for illustrating the technical solutions of the present application, but not limiting the same, and the scope of the present application is not limited thereto, and although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope disclosed in the present application; such modifications, changes or substitutions do not depart from the spirit and scope of the exemplary embodiments of the present application, and are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (19)
1. A method of controlling a game object, comprising:
acquiring game state information in a game scene within a preset range of a game object;
determining a first action corresponding to the game object based on the game state information;
and acquiring a second action currently executed by the game object, and controlling the game object to execute the first action if the second action can be interrupted by the first action.
2. The control method according to claim 1, wherein the preset range includes a range in which a distance to the game object is less than a preset distance.
3. The control method according to claim 1, wherein the preset range is smaller than a range of an entire game scene.
4. The control method according to claim 1, characterized in that the first action includes a plurality of single actions in succession;
the controlling the game object to perform the first action includes:
and controlling the game object to execute each single action in turn.
5. The control method according to claim 1, characterized by further comprising:
if the first action cannot interrupt the second action and the second action comprises a plurality of single actions in succession, continuing to execute the single action not executed by the second action.
6. The control method according to claim 1, wherein the first action includes one single action;
the controlling the game object to perform the first action includes:
controlling the game object to perform a single action of the first actions.
7. The control method according to claim 1, characterized by further comprising:
and if the second action is not acquired and the first action comprises a plurality of continuous single actions, controlling the game object to sequentially execute each single action.
8. The control method according to any one of claims 1 to 6, characterized in that after acquiring the second action, the method further includes:
acquiring at least one preset interrupting action capable of interrupting the second action;
determining whether the first action is one of the at least one breaking action;
and if so, determining that the first action can interrupt the second action.
9. The control method according to claim 8, wherein before acquiring the at least one breaking action, the method further comprises:
setting at least one breaking action for the second action;
storing the set at least one breaking action into a breaking action set.
10. The control method of claim 1, wherein if the first action is capable of interrupting the second action and the first action comprises a plurality of single actions in succession, the method further comprises:
storing a plurality of continuous single actions included in the first action into an action queue;
the controlling the game object to perform the first action includes:
and sequentially taking out each single action from the action queue, and controlling the game object to execute the taken-out single action.
11. The control method according to claim 1, wherein the game state information includes at least one of:
battle force information of the game object; the fighting capacity information of other game objects in the game scene in the preset range of the game object; game object position information; position information of other game objects within the game scene within a preset range of the game object.
12. A method of training a game object control model, comprising:
acquiring game state information in a game scene within a preset range of a game object;
inputting the game state information into a game object control model to be trained, processing the input game state information through the game object control model, and outputting a first action corresponding to the game object;
and adjusting model parameters in a game object control model to be trained on the basis of revenue information generated by the game object executing the first action.
13. A training method as recited in claim 12, wherein the preset range includes a range in which a distance from the game object is less than a preset distance.
14. Training method according to claim 12, wherein the preset range is smaller than the range of the entire game scene.
15. Training method according to claim 12, wherein said first action comprises a plurality of single actions in succession;
the adjusting model parameters in a game object control model to be trained based on revenue information generated by the game object performing the first action comprises:
adjusting model parameters in a game object control model to be trained based on revenue information generated by the game object performing each single action.
16. A control device for a game object, comprising:
the acquisition module is used for acquiring game state information in a game scene in a preset range of a game object;
the determining module is used for determining a first action corresponding to the game object based on the game state information;
and the control module is used for acquiring a second action currently executed by the game object, and controlling the game object to execute the first action if the second action can be interrupted by the first action.
17. An exercise device for a game object control model, comprising:
the information acquisition module is used for acquiring game state information in a game scene in a preset range of a game object;
the information processing module is used for inputting the game state information into a game object control model to be trained, processing the input game state information through the game object control model and outputting a first action corresponding to the game object;
and the parameter adjusting module is used for adjusting model parameters in the game object control model to be trained on the basis of the income information generated by the game object executing the first action.
18. An electronic device, comprising: a processor, a storage medium and a bus, the storage medium storing machine-readable instructions executable by the processor, the processor and the storage medium communicating via the bus when the electronic device is operating, the processor executing the machine-readable instructions to perform the steps of the method according to any one of claims 1 to 15.
19. A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, is adapted to carry out the steps of the method according to any one of claims 1 to 15.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010106878.9A CN111359212A (en) | 2020-02-20 | 2020-02-20 | Game object control and model training method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010106878.9A CN111359212A (en) | 2020-02-20 | 2020-02-20 | Game object control and model training method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111359212A true CN111359212A (en) | 2020-07-03 |
Family
ID=71199223
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010106878.9A Pending CN111359212A (en) | 2020-02-20 | 2020-02-20 | Game object control and model training method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111359212A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109107161A (en) * | 2018-08-17 | 2019-01-01 | 深圳市腾讯网络信息技术有限公司 | A kind of control method of game object, device, medium and equipment |
| CN109843401A (en) * | 2017-10-17 | 2019-06-04 | 腾讯科技（深圳）有限公司 | A kind of AI object behaviour model optimization method and device |
-
2020
- 2020-02-20 CN CN202010106878.9A patent/CN111359212A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109843401A (en) * | 2017-10-17 | 2019-06-04 | 腾讯科技（深圳）有限公司 | A kind of AI object behaviour model optimization method and device |
| CN109107161A (en) * | 2018-08-17 | 2019-01-01 | 深圳市腾讯网络信息技术有限公司 | A kind of control method of game object, device, medium and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102360420B1 (en) | Customized models for imitating player gameplay in a video game | |
| KR102127670B1 (en) | Multiplayer video game matchmaking system and methods | |
| CN108888958B (en) | Virtual object control method, device, equipment and storage medium in virtual scene | |
| US10695677B2 (en) | Systems and methods for hardware-based matchmaking | |
| KR102397507B1 (en) | Automated player control takeover in a video game | |
| JP2020504653A (en) | Character arrangement method and device, storage medium, and electronic device | |
| US11497993B2 (en) | Method and apparatus for providing game strategy guide | |
| CN110134375B (en) | Game character behavior control method and device and readable storage medium | |
| CN112870721A (en) | Game interaction method, device, equipment and storage medium | |
| WO2019144346A1 (en) | Object processing method in virtual scene, device and storage medium | |
| CN111111203B (en) | Robot training method and device and skill releasing method and device | |
| CN115888119A (en) | Game AI training method, device, electronic equipment and storage medium | |
| US20230311002A1 (en) | Decision model training method and apparatus | |
| US20230078340A1 (en) | Virtual object control method and apparatus, electronic device, storage medium, and computer program product | |
| CN110152292B (en) | Display control method and device for jumping characters in game, storage medium and electronic equipment | |
| CN111359212A (en) | Game object control and model training method and device | |
| CN113144617B (en) | Control method, device and equipment of virtual object and computer readable storage medium | |
| JP6576544B2 (en) | Information processing apparatus, information processing method, and computer-readable storage medium | |
| CN117046111B (en) | Game skill processing method and related device | |
| CN110841283A (en) | Game role control method and device, terminal device and storage medium | |
| KR102060528B1 (en) | Method and apparatus providing user collaborative content | |
| KR102343354B1 (en) | Energry charging apparatus and method for game | |
| CN116943204A (en) | Virtual object control method and device, storage medium and electronic equipment | |
| CN116785687A (en) | Method and device for processing view angle in game, electronic equipment and storage medium | |
| JP2023071109A (en) | Game program, game system and information processing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200703 |