CN117669710B - Multi-behavior tree decision scheme aggregation method and device for game countermeasure task - Google Patents

Multi-behavior tree decision scheme aggregation method and device for game countermeasure task Download PDF

Info

Publication number
CN117669710B
CN117669710B CN202410142819.5A CN202410142819A CN117669710B CN 117669710 B CN117669710 B CN 117669710B CN 202410142819 A CN202410142819 A CN 202410142819A CN 117669710 B CN117669710 B CN 117669710B
Authority
CN
China
Prior art keywords
execution unit
nodes
decision
action
unit structure
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.)
Active
Application number
CN202410142819.5A
Other languages
Chinese (zh)
Other versions
CN117669710A (en
Inventor
李渊
刘运韬
李晟泽
章杰元
张峰
顾孔静
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
National Defense Technology Innovation Institute PLA Academy of Military Science
Original Assignee
National Defense Technology Innovation Institute PLA Academy of Military Science
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by National Defense Technology Innovation Institute PLA Academy of Military Science filed Critical National Defense Technology Innovation Institute PLA Academy of Military Science
Priority to CN202410142819.5A priority Critical patent/CN117669710B/en
Publication of CN117669710A publication Critical patent/CN117669710A/en
Application granted granted Critical
Publication of CN117669710B publication Critical patent/CN117669710B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/243—Classification techniques relating to the number of classes
    • G06F18/24323—Tree-organised classifiers
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/04—Inference or reasoning models
    • G06N5/041—Abduction

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Mathematical Physics (AREA)
  • Computational Linguistics (AREA)
  • Computing Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Computer And Data Communications (AREA)

Abstract

The invention provides a multi-behavior tree decision scheme aggregation method and device facing game countermeasure tasks, and relates to the technical field of computers, wherein the method comprises the following steps: constructing a behavior tree set corresponding to each behavior tree; constructing a plurality of protocol candidate object sets corresponding to the behavior tree set; for a set of reduction candidates in the set of behavior trees, steps a-c are performed separately: step a, respectively carrying out decision-making according to element conventions on the condition nodes of each execution unit structure; step b, respectively carrying out decision control element protocol on the action nodes of each execution unit structure; step c, aggregating the condition nodes and the action nodes of each execution unit structure to obtain an aggregation execution unit corresponding to the protocol candidate object set; and determining an aggregate behavior tree corresponding to the behavior tree set according to the aggregate execution units corresponding to each protocol candidate object set in the behavior tree set. The method and the device can improve the generation efficiency and generalization capability of the game countermeasure decision scheme.

Description

Multi-behavior tree decision scheme aggregation method and device for game countermeasure task
Technical Field
The invention relates to the technical field of computers, in particular to a multi-behavior tree decision scheme aggregation method and device for game countermeasure tasks.
Background
The behavior tree is a hierarchical node tree structure, and large-scale decision tasks are effectively organized through layering and modularization, so that the behavior tree is a common decision method in game countermeasure scenes. The decision scheme generation method based on the behavior tree is widely applied to the relevant fields of game countermeasure decisions such as social management, intelligent transportation, economy, military and the like. The prior art generally generates a decision scheme based on a single behavior tree, but the scheme is more prone to describe a decision process of single game countermeasure, and has no adaptability to scene change, so that the generated decision scheme has poor generalization capability. For example, a strategy of an unmanned plane game countermeasure scene for avoiding enemy investigation unmanned plane is that enemy is assailed from north, and enemy is assailed to south, but in an actual scene, enemy is assailed from south, and the generated decision scheme for assailing to south is inapplicable.
To cope with this problem, aggregating a plurality of tree structures to generate a decision scheme is an effective solution. In the existing aggregation method, a generalized decision tree structure is generated by aiming at multiple decision learning in a scene, or node-level aggregation is carried out on a plurality of single decision trees to form an aggregated decision tree structure. However, the generation of the decision tree is seriously dependent on complete historical statistical data, long-time training optimization is needed, the aggregation efficiency is low, and the decision mode of the decision tree is mainly selected by a single-moment decision corpus and is not matched with the requirement that the decision scheme completely covers the full-moment decision process.
Disclosure of Invention
Aiming at the problems existing in the prior art, the invention provides a multi-behavior tree decision scheme aggregation method and device facing game countermeasure tasks.
The invention provides a multi-action tree decision scheme aggregation method facing game countermeasure tasks, which comprises the following steps:
Constructing a behavior tree set corresponding to each behavior tree according to the execution unit structures of a plurality of behavior trees facing game countermeasure tasks, wherein the behavior tree set comprises the execution unit structures of each behavior tree;
Constructing a plurality of protocol candidate object sets corresponding to the action tree sets according to at least one time interval corresponding to the action tree sets, wherein each protocol candidate object set comprises a plurality of execution unit structures for controlling the same agent in the same time interval in each action tree;
And (c) respectively executing the steps a-c for the protocol candidate object set in the behavior tree set:
step a, according to the types of decision basis elements in the condition nodes of each execution unit structure in the protocol candidate object set, respectively carrying out decision basis element protocol on the condition nodes of each execution unit structure;
step b, according to the types and interrelationships of decision control elements in the action nodes of each execution unit structure in the protocol candidate object set, respectively carrying out decision control element protocol on the action nodes of each execution unit structure;
Step c, aggregating the condition nodes and the action nodes of each execution unit structure to obtain an aggregate execution unit corresponding to the protocol candidate object set;
and determining an aggregate behavior tree corresponding to the behavior tree set according to the aggregate execution units corresponding to each protocol candidate object set in the behavior tree set.
According to the method for aggregating multi-action tree decision schemes facing game countermeasure tasks provided by the invention, decision basis element conventions are respectively carried out on the condition nodes of each execution unit structure according to the types of the decision basis elements in the condition nodes of each execution unit structure in the rule candidate object set, and the method comprises the following steps:
when the types of the decision-making basis elements in the condition nodes of each execution unit structure in the protocol candidate object set are the same, according to the quantized values of the decision-making basis elements in the condition nodes of each execution unit structure, the quantized value threshold and the conditions determined by the corresponding condition nodes, carrying out decision-making basis element protocol on the condition nodes of each execution unit structure;
Or alternatively
And when the types of the decision basis elements in the conditional nodes of at least two execution unit structures in the specification candidate object set are different, determining that the decision basis elements in the conditional nodes of the execution unit structures cannot be combined.
According to the multi-action tree decision scheme aggregation method facing game countermeasure tasks provided by the invention, the decision basis element rule is carried out on the condition nodes of each execution unit structure according to the quantized value, the quantized value threshold and the conditions determined by the corresponding condition nodes of the decision basis element in the condition nodes of each execution unit structure, and the method comprises at least one of the following steps:
under the condition that the conditions determined by the condition nodes of the execution unit structures are smaller than the relation, determining a rule result of the decision basis element in the condition nodes of the execution unit structures according to the minimum quantized value threshold in the quantized value thresholds of the decision basis element in the condition nodes of the execution unit structures;
Or alternatively
Under the condition that the conditions determined by the condition nodes of the execution unit structures are all greater than the relation, determining a rule result of the decision basis element in the condition nodes of the execution unit structures according to the largest quantized value threshold in the quantized value thresholds of the decision basis element in the condition nodes of the execution unit structures;
Or alternatively
And under the condition that the condition nodes of the execution unit structures determine that the condition nodes of the execution unit structures have a smaller relationship and a larger relationship, determining that the decision basis elements in the condition nodes of the execution unit structures cannot be combined.
According to the multi-action tree decision scheme aggregation method facing game countermeasure tasks, the decision basis element types comprise at least one of the following: regional maneuver dominant elements; target hit dominance element.
According to the method for aggregating multi-action tree decision schemes facing game countermeasure tasks provided by the invention, decision control element conventions are respectively carried out on action nodes of each execution unit structure according to types and interrelationships of decision control elements in the action nodes of each execution unit structure in the rule candidate object set, and the method comprises the following steps:
according to the types and interrelationships of decision control elements in the action nodes of each execution unit structure in the protocol candidate object set, the following operations are executed for the protocol candidate object set:
When the type of the decision control element in the action node of each execution unit structure in the protocol candidate object set comprises area elements, determining a protocol result of the area elements in the action node of each execution unit structure according to the union of the area elements in the action node of each execution unit structure when the inclusion relationship or the intersection relationship exists between the area elements in the action node of each execution unit structure; or when the inclusion relationship or the intersection relationship does not exist between the area elements in the action nodes of the execution unit structures, determining that the area elements in the action nodes of the execution unit structures cannot be combined;
Or alternatively
Determining a specification result of the hit target element in the action node of each execution unit structure according to the hit target element in the action node of each execution unit structure when the hit target element in the action node of each execution unit structure is the same under the condition that the type of the decision control element in the action node of each execution unit structure in the specification candidate object set comprises the hit target element; or when at least two hit target elements are included in the action nodes of each execution unit structure, determining that the area elements in the action nodes of each execution unit structure are not combinable.
According to the multi-action tree decision scheme aggregation method facing game countermeasure tasks, the types of the decision control elements comprise at least one of the following: a region element; striking the target element.
According to the multi-action tree decision scheme aggregation method facing game countermeasure tasks provided by the invention, the aggregation is performed on the condition nodes and the action nodes of each execution unit structure to obtain the aggregation execution units corresponding to the protocol candidate object set, and the aggregation execution units comprise at least one of the following:
Determining the condition nodes and the corresponding action nodes of the aggregation execution units corresponding to the reduction candidate object sets according to the reduction results of the decision basis elements in the condition nodes of the execution unit structures and the reduction results of the area elements in the action nodes of the execution unit structures;
Or alternatively
Determining the condition nodes and the corresponding action nodes of the aggregation execution units corresponding to the reduction candidate object sets according to the reduction results of the decision basis elements in the condition nodes of the execution unit structures and the reduction results of the hit target elements in the action nodes of the execution unit structures;
Or alternatively
Under the condition that decision basis elements in the condition nodes of each execution unit structure can be combined and action nodes of each execution unit structure cannot be combined, combining the decision basis elements in the condition nodes of each execution unit structure, and determining condition nodes of an aggregate execution unit and corresponding parallel nodes, wherein the action nodes of each execution unit structure are hung below the parallel nodes;
And under the condition that the decision-making basis elements in the conditional nodes of the execution unit structures cannot be combined, determining a plurality of sequential nodes of the aggregate execution unit, and hanging one uncombinable conditional node and a corresponding action node under each sequential node.
The invention also provides a multi-action tree decision scheme aggregation device facing game countermeasure tasks, which comprises:
The first construction module is used for constructing a behavior tree set corresponding to each behavior tree according to the execution unit structures of the behavior trees facing the game countermeasure task, wherein the behavior tree set comprises the execution unit structures of the behavior trees;
The second construction module is used for constructing a plurality of protocol candidate object sets corresponding to the action tree sets according to at least one time interval corresponding to the action tree sets, and each protocol candidate object set comprises a plurality of execution unit structures which are used for controlling the same agent in the same time interval in each action tree;
the execution module is used for executing the steps a-c for the protocol candidate object set in the behavior tree set respectively:
step a, according to the types of decision basis elements in the condition nodes of each execution unit structure in the protocol candidate object set, respectively carrying out decision basis element protocol on the condition nodes of each execution unit structure;
step b, according to the types and interrelationships of decision control elements in the action nodes of each execution unit structure in the protocol candidate object set, respectively carrying out decision control element protocol on the action nodes of each execution unit structure;
Step c, aggregating the condition nodes and the action nodes of each execution unit structure to obtain an aggregate execution unit corresponding to the protocol candidate object set;
and the determining module is used for determining an aggregate behavior tree corresponding to the behavior tree set according to the aggregate execution unit corresponding to each protocol candidate object set in the behavior tree set.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the multi-action tree decision scheme aggregation method facing game countermeasure tasks when executing the program.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a multi-behavioral tree decision-scheme aggregation method oriented to a game challenge task as any one of the above.
The invention also provides a computer program product comprising a computer program which when executed by a processor implements a multi-behavioral tree decision-scheme aggregation method for game-oriented challenge tasks as described in any of the above.
The invention provides a multi-behavior tree decision scheme aggregation method and device for game countermeasure tasks, which are characterized in that a plurality of behavior tree sets corresponding to a plurality of behavior trees are constructed to generate a plurality of protocol candidate object sets corresponding to the behavior tree sets, and each protocol candidate object set comprises a plurality of execution unit structures for controlling the same agent in the same time interval in each behavior tree; and then, the condition nodes and the action nodes of each execution unit structure are aggregated to obtain an aggregated execution unit, and an aggregated action tree is determined according to the aggregated execution unit corresponding to each protocol candidate object set in the action tree set, namely, the condition nodes and the action nodes are analyzed layer by node analysis on a plurality of action trees, the decision elements of the condition nodes and the action nodes are analyzed, and a node aggregation process with multi-action tree layer by layer granularity is carried out to obtain a game countermeasure decision scheme after multi-action tree aggregation, so that the generation efficiency and the generalization capability of the game countermeasure decision scheme can be improved.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a multi-action tree decision scheme aggregation method for game countermeasure tasks provided by the invention;
FIG. 2 is a second flow chart of a method for aggregating multiple action tree decision schemes for game challenge-oriented tasks according to the present invention;
FIG. 3 is a schematic diagram of a region element merge provided by the present invention;
FIG. 4 is a schematic diagram of three unmanned aerial vehicle game countermeasure behavior tree decision schemes provided by the invention;
FIG. 5 is a schematic diagram of element specifications in an unmanned aerial vehicle game against aggregation behavior tree decision scheme provided by the invention;
FIG. 6 is a schematic diagram of an unmanned aerial vehicle game against aggregation behavior tree decision scheme provided by the invention;
FIG. 7 is a schematic diagram of a multi-action tree decision scheme aggregation device for game challenge-oriented tasks according to the present invention;
fig. 8 is a schematic diagram of the physical structure of the electronic device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention provides a multi-behavior tree decision scheme aggregation method and device facing game countermeasure tasks, which are described below with reference to the accompanying drawings.
Fig. 1 is a schematic flow chart of a multi-action tree decision scheme aggregation method for game countermeasure tasks, as shown in fig. 1, the method includes steps 101-104, wherein:
step 101, constructing a behavior tree set corresponding to each behavior tree according to the execution unit structures of a plurality of behavior trees facing game countermeasure tasks, wherein the behavior tree set comprises the execution unit structures of each behavior tree.
It should be noted that the multi-action tree decision scheme aggregation method facing game countermeasure tasks provided by the invention can be applied to game countermeasure scenes. The execution subject of the method may be a multi-treelet decision scheme aggregation device for game countermeasure tasks, such as an electronic device, a server, or a control module in the device for executing the multi-treelet decision scheme aggregation method for game countermeasure tasks, where the electronic device may include, but is not limited to, a mobile phone, a tablet computer, or a desktop computer.
Optionally, a plurality of behavior trees facing game countermeasure tasks are obtained, and a behavior tree set corresponding to the behavior trees is built, namely, an execution unit structure of the multi-behavior tree is built.
For example, fig. 2 is a second flow chart of a multi-action tree decision scheme aggregation method for game countermeasure tasks provided by the present invention; as shown in fig. 2, in step 201, an execution unit structure of the multi-behavior tree is constructed. Building a behavior tree set for n behavior treesWherein the formalized representation or formalization of each behavior tree is defined asI takes a value of 1 to n, the behavior tree/>By/>Each dictionary is structured, and each dictionary may also be referred to as a behavior tree execution unit. Behavior tree/>Chinese dictionary structure/>Comprising conditional nodes/>And action node/>The agent controlling the execution unit is in the condition node/>By action node/>, under defined conditionsA control strategy is generated.
It should be noted that, the behavior tree includes a multi-layer structure, but the node that ultimately generates the control decision is the leaf node of the lowest layer, i.e. the execution unit, so the formalized definition of the behavior tree in the present invention omits the higher-layer structure above the leaf node and includes only the execution unit structure.
Step 102, constructing a plurality of protocol candidate object sets corresponding to the action tree sets according to at least one time interval corresponding to the action tree sets, wherein each protocol candidate object set comprises a plurality of execution unit structures for controlling the same agent in the same time interval in each action tree.
Optionally, the decision scheme generated by the game countermeasure action tree has two basic elements, namely an agent and a decision time which are controlled, and the two basic elements in each action tree have the same execution unit and belong to a protocol candidate object set.
For example, as shown in FIG. 2, a set of multi-behavior tree reduction candidates is constructed, step 202. The specific implementation method for constructing the plurality of protocol candidate object sets corresponding to the behavior tree set comprises the following steps:
Dividing the overall decision time of a behavior tree into There are N time intervals in total; all time intervals are then traversed, at each/>Traversing all agents within an intervalFor each agent/>Acquiring execution units in all the behavior trees for controlling the intelligent agent in the time interval to form a behavior tree protocol candidate object setAfter traversing all time intervals, forming at mostSet of individual protocol candidates/>Protocol candidate set/>Including elements/>。
Step 103, for the set of specification candidate objects in the behavior tree set, executing steps a-c respectively:
and a step a of respectively carrying out decision-making element conventions on the condition nodes of each execution unit structure according to the types of the decision-making element in the condition nodes of each execution unit structure in the convention candidate object set.
Optionally, the decision-making depends on the type of the element including at least one of: regional maneuver dominant elements; target hit dominance element.
Optionally, the implementation manner of performing the decision-making element convention on the condition nodes of each execution unit structure according to the type of the decision-making element in the condition nodes of each execution unit structure in the rule candidate object set may include at least one of the following:
In the mode 1, when the types of the decision-making basis elements in the condition nodes of each execution unit structure in the protocol candidate object set are the same, according to the quantized values, quantized value thresholds and conditions determined by the corresponding condition nodes of the decision-making basis elements in the condition nodes of each execution unit structure, the decision-making basis element protocol is carried out on the condition nodes of each execution unit structure.
And 2, determining that the decision basis elements in the condition nodes of the execution unit structures are not combinable when the types of the decision basis elements in the condition nodes of at least two execution unit structures in the specification candidate object set are different.
Optionally, the implementation manner of making the decision-making element specification on the condition node of each execution unit structure according to the decision-making element quantization value, the quantization value threshold and the condition determined by the corresponding condition node in the condition node of each execution unit structure may include at least one of the following:
And (c) determining a rule result of the decision basis element in the condition nodes of each execution unit structure according to the minimum quantized value threshold in the quantized value thresholds of the decision basis element in the condition nodes of each execution unit structure under the condition that the conditions determined by the condition nodes of each execution unit structure are smaller than the relation.
And b, determining a rule result of the decision basis element in the condition nodes of each execution unit structure according to the largest quantized value threshold in the quantized value thresholds of the decision basis element in the condition nodes of each execution unit structure under the condition that the conditions determined by the condition nodes of each execution unit structure are all greater than the relation.
And c, determining that the decision basis elements in the condition nodes of the execution unit structures cannot be combined under the condition that the condition nodes of the execution unit structures have a smaller relationship and a larger relationship.
For example, as shown in FIG. 2, step 203, candidate conditional node specification. For a set of reduction candidatesElement/>The reduction of the condition node is performed, and the reduction process involves the reduction of the decision basis element in the condition node. The decision basis elements in a game countermeasure scenario typically include two elements, regional maneuver advantages and target hit advantages. In the invention, the conditional node quantifies the decision basis element, and the use/>Representing regional maneuver dominance values, use/>The target hit advantage value is represented as such,Indicating that if the zone maneuver predominance is greater than/>Can execute the subsequent actions of the execution unit,/>Indicating that if the target hit dominance value is greater than/>The subsequent actions of the execution unit can be performed.
The specifications for the regional maneuver advantage value or the target hit advantage value are numerical specification processes. For the decision basis elements in the conditional nodes of each execution unit structure, if the judging relations between all the dominant values and the threshold values are smaller than the relations or larger than the relations, namelyThe maximum dominance value or the minimum dominance value can be taken as the rule result of the decision basis element, namely/>Or/>If the relation exists between the greater relation and the lesser relation, marking the decision basis element in the conditional node of each execution unit structure as the branch construction flow to be executed.
And b, respectively carrying out decision control element protocol on the action nodes of each execution unit structure according to the types and the interrelationships of the decision control elements in the action nodes of each execution unit structure in the protocol candidate object set.
Optionally, the type of the decision control element includes at least one of: a region element; striking the target element.
Optionally, according to the type and the interrelationship of the decision control element in the action node of each execution unit structure in the protocol candidate object set, the implementation manner of respectively implementing the decision control element protocol on the action node of each execution unit structure may include:
according to the types and interrelationships of decision control elements in the action nodes of each execution unit structure in the protocol candidate object set, the following operations are executed for the protocol candidate object set:
When the type of the decision control element in the action node of each execution unit structure in the protocol candidate object set comprises area elements, determining a protocol result of the area elements in the action node of each execution unit structure according to the union of the area elements in the action node of each execution unit structure when the inclusion relationship or the intersection relationship exists between the area elements in the action node of each execution unit structure; or when the inclusion relationship or the intersection relationship does not exist between the area elements in the action nodes of the execution unit structures, determining that the area elements in the action nodes of the execution unit structures cannot be combined;
Or alternatively
Determining a specification result of the hit target element in the action node of each execution unit structure according to the hit target element in the action node of each execution unit structure when the hit target element in the action node of each execution unit structure is the same under the condition that the type of the decision control element in the action node of each execution unit structure in the specification candidate object set comprises the hit target element; or when at least two hit target elements are included in the action nodes of each execution unit structure, determining that the area elements in the action nodes of each execution unit structure are not combinable.
For example, as shown in FIG. 2, step 204, candidate action node specifications. For a set of reduction candidatesElement/>The provision of the mobile node is made, and the provision process involves provision of decision control elements in the mobile node. Decision control elements in a gaming countermeasure scenario typically include a region element and a hit target element.
FIG. 3 is a schematic diagram of the merging of zone elements provided by the present invention, as shown in FIG. 3, for the specification of zone elements, if there is a rule comprisingOr intersect/>The relationship is that the result of the specification of the area element is the union/>Otherwise, the area elements which cannot be combined are reserved and marked as the branch construction flow to be executed.
For the specification of the hit target element, the hit targets in the specification candidate object set are the sameDirectly targeting the target as a reduction result; different for hit targetsIf so, then it is marked as needed to execute the branch construction flow.
And c, aggregating the condition nodes and the action nodes of each execution unit structure to obtain an aggregate execution unit corresponding to the protocol candidate object set.
Optionally, the implementation manner of aggregating the condition nodes and the action nodes of each execution unit structure to obtain the aggregate execution units corresponding to the protocol candidate object set includes at least one of the following:
Determining condition nodes and corresponding action nodes of an aggregation execution unit corresponding to the protocol candidate object set according to a protocol result of a decision basis element in the condition nodes of each execution unit structure and a protocol result of an area element in the action node of each execution unit structure;
Mode (2), determining the condition nodes and the corresponding action nodes of the aggregation execution units corresponding to the protocol candidate object set according to the protocol results of the decision basis elements in the condition nodes of the execution unit structures and the protocol results of the hit target elements in the action nodes of the execution unit structures;
In the mode (3), under the condition that the decision basis elements in the condition nodes of the execution unit structures can be combined and the action nodes of the execution unit structures cannot be combined, combining the decision basis elements in the condition nodes of the execution unit structures, and determining the condition nodes of the aggregate execution units and the corresponding parallel nodes, wherein the action nodes of the execution unit structures are hung below the parallel nodes;
And (4) determining a plurality of sequence nodes of the aggregate execution unit under the condition that the decision basis elements in the condition nodes of the execution unit structures cannot be combined, and hanging one uncombinable condition node and a corresponding action node under each sequence node.
For example, as shown in FIG. 2, an aggregate behavior tree execution unit is generated, step 205. This step may include element combining and branching processes.
The element merging flow refers to the current specification candidate objectBoth conditional nodes and action nodes in (a) can be merged, i.eMerging to form a condition node and a corresponding action node, and generating an aggregate execution unit/>。
The element branch flow refers to the condition node and the decision element of the action node are marked as uncombinable, and are divided into two cases, wherein the first condition node can be combined and the action node cannot be combined, and at the moment, the combined decision forms a single condition node according to the elementConcurrent generation of parallel nodes, underslung of marked uncombinable multiple action nodes/>Aggregation execution unit for forming aggregation behavior tree; Secondly, the condition nodes can not be combined, and any action node can be used, at the moment, the execution unit generates a plurality of sequence nodes, and one marked uncombinable condition node and the corresponding action node are hung below each sequence node to form an aggregation execution unit of an aggregation action tree.
After executing step 205, as shown in FIG. 2, step 206 is executed to determine whether to traverse to the set of specification candidatesEnd of (2); if not, repeating steps 203-205 until the set of reduction candidates/>After the traversal of (a), step 207 is performed.
Step 104, determining an aggregate behavior tree corresponding to the behavior tree set according to the aggregate execution units corresponding to the protocol candidate object sets in the behavior tree set.
Optionally, as shown in step 207 in fig. 2, an aggregated behavior tree decision scheme is formed according to the aggregate execution units corresponding to each of the protocol candidate sets in the behavior tree set.
The invention provides a multi-behavior tree decision scheme aggregation method facing game countermeasure tasks, which comprises the steps of constructing behavior tree sets corresponding to a plurality of behavior trees, generating a plurality of protocol candidate object sets corresponding to the behavior tree sets, wherein each protocol candidate object set comprises a plurality of execution unit structures for controlling the same agent in the same time interval in each behavior tree; and then, the condition nodes and the action nodes of each execution unit structure are aggregated to obtain an aggregated execution unit, and an aggregated action tree is determined according to the aggregated execution unit corresponding to each protocol candidate object set in the action tree set, namely, the condition nodes and the action nodes are analyzed layer by node analysis on a plurality of action trees, the decision elements of the condition nodes and the action nodes are analyzed, and a node aggregation process with multi-action tree layer by layer granularity is carried out to obtain a game countermeasure decision scheme after multi-action tree aggregation, so that the generation efficiency and the generalization capability of the game countermeasure decision scheme can be improved.
The invention provides a multi-behavior tree decision scheme aggregation method for game countermeasure tasks, namely a game countermeasure scheme generation method based on multi-behavior tree aggregation, and relates to the field of game countermeasure, behavior tree and decision scheme generalization. Aiming at the defects of the prior art, the method designs the decision basis element protocol mode of conditional node aggregation by analyzing a plurality of behavior trees layer by layer and node by node, designs the decision control element protocol mode of action node aggregation, and realizes the common node aggregation process of the multi-behavior tree layer granularity based on the decision control element protocol mode, thereby aggregating the multi-behavior tree to generate a game countermeasure decision scheme, solving the problems of low generation efficiency and weak generalization capability of the traditional decision scheme based on a tree structure and filling the blank of the conventional method for generating the decision scheme by aggregating the multi-behavior tree.
Here, taking an unmanned aerial vehicle air game countermeasure (3 vs 3) simulation scenario as an example, the multi-behavior tree decision scheme aggregation method facing game countermeasure tasks provided by the invention is described.
The case is based on 3 decision schemes, as shown in fig. 4, the behavior tree of each decision scheme can be distinguished by time intervals, in fig. 4, the time intervals are distinguished by parallel nodes of a second layer, each parallel node represents one time interval, each time interval comprises 3 execution units, and each execution unit controls one unmanned aerial vehicle. Thus, a set of protocol candidates may be generated from execution units of the same time interval and the same agent. In this case, there are 9 protocol candidate sets, and the execution units in the same vertical row in fig. 4 are the same protocol candidate set.
FIG. 5 is a schematic diagram of element specifications in an unmanned aerial vehicle game against aggregation behavior tree decision scheme provided by the invention; FIG. 6 is a schematic diagram of an unmanned aerial vehicle game against aggregation behavior tree decision scheme provided by the invention; see fig. 5 and 6:
for the specification number 1 candidate set, the condition nodes are the same, the action nodes respectively go to the two areas A1 and A3, and it can be observed from fig. 4 that the areas A1 and A3 overlap, so that area element combination can be performed, the areas A1 and A3 are combined into the area A5, and the combined execution unit is obtained, as shown in fig. 5 and fig. 6.
For the protocol number 2 and 3 candidate object sets, the condition node and the action node are the same, and can be directly combined, and the combination result is shown in fig. 5 and 6.
For the reduction candidate object sets of the number 4, the number 5 and the number 6, the corresponding condition nodes in the three decision schemes comprise two types of decision basis elements, so that the condition nodes are marked as the need to execute a branch construction flow, namely the need to mark the generation of branches in an aggregation execution unit. The branch construction flow comprises the following steps: generating a selection node, and hanging two sequence nodes, wherein each sequence node is hung with one type of condition node and corresponding action node, and the protocol results are shown in fig. 5 and 6.
For the reduction candidate object sets of the number 7, the number 8 and the number 9, the condition nodes are the same, each action node comprises the reduction of two reduction targets, the condition nodes and a sequence node can be generated in an execution unit of the aggregated behavior tree, the action nodes for reducing the two different reduction targets are hung down, and the reduction results are shown in fig. 5 and 6.
Thus, the construction of the aggregated behavior tree decision scheme is completed, and the aggregated result of the multi-behavior tree decision scheme oriented to game countermeasure tasks is shown in fig. 5 and 6.
According to the multi-behavior tree decision scheme aggregation method for the game countermeasure task, the decision elements of the condition nodes and the action nodes are analyzed, and an aggregation process is developed layer by layer node by node, so that a game countermeasure decision scheme after multi-behavior tree aggregation is obtained. The method can improve the generation efficiency and generalization capability of the game countermeasure decision scheme, and can be effectively applied to game countermeasure decision scenes such as unmanned systems, intelligent transportation, power grid dispatching, strategic planning and the like.
The multi-tree decision scheme aggregation device for game countermeasure tasks provided by the invention is described below, and the multi-tree decision scheme aggregation device for game countermeasure tasks described below and the multi-tree decision scheme aggregation method for game countermeasure tasks described above can be correspondingly referred to each other.
Fig. 7 is a schematic structural diagram of a multi-action tree decision scheme aggregation device for game countermeasure tasks according to the present invention, and as shown in fig. 7, the multi-action tree decision scheme aggregation device 700 for game countermeasure tasks includes: a first building block 701, a second building block 702, an execution block 703 and a determination block 704; wherein,
A first construction module 701, configured to construct a behavior tree set corresponding to each behavior tree according to execution unit structures of a plurality of behavior trees facing game countermeasure tasks, where the behavior tree set includes the execution unit structures of each behavior tree;
a second construction module 702, configured to construct a plurality of protocol candidate object sets corresponding to the behavior tree sets according to at least one time interval corresponding to the behavior tree sets, where each protocol candidate object set includes a plurality of execution unit structures in each behavior tree that control the same agent in the same time interval;
an execution module 703, configured to execute steps a-c for the set of reduction candidates in the set of action trees, respectively:
step a, according to the types of decision basis elements in the condition nodes of each execution unit structure in the protocol candidate object set, respectively carrying out decision basis element protocol on the condition nodes of each execution unit structure;
step b, according to the types and interrelationships of decision control elements in the action nodes of each execution unit structure in the protocol candidate object set, respectively carrying out decision control element protocol on the action nodes of each execution unit structure;
Step c, aggregating the condition nodes and the action nodes of each execution unit structure to obtain an aggregate execution unit corresponding to the protocol candidate object set;
and the determining module 704 is configured to determine an aggregate behavior tree corresponding to the behavior tree set according to the aggregate execution units corresponding to each of the protocol candidate object sets in the behavior tree set.
The invention provides a multi-behavior tree decision scheme aggregation device facing game countermeasure tasks, which is characterized in that a plurality of behavior tree sets corresponding to behavior trees are constructed to generate a plurality of protocol candidate object sets corresponding to the behavior tree sets, and each protocol candidate object set comprises a plurality of execution unit structures for controlling the same agent in the same time interval in each behavior tree; and then, the condition nodes and the action nodes of each execution unit structure are aggregated to obtain an aggregated execution unit, and an aggregated action tree is determined according to the aggregated execution unit corresponding to each protocol candidate object set in the action tree set, namely, the condition nodes and the action nodes are analyzed layer by node analysis on a plurality of action trees, the decision elements of the condition nodes and the action nodes are analyzed, and a node aggregation process with multi-action tree layer by layer granularity is carried out to obtain a game countermeasure decision scheme after multi-action tree aggregation, so that the generation efficiency and the generalization capability of the game countermeasure decision scheme can be improved.
Optionally, the execution module 703 is specifically configured to:
when the types of the decision-making basis elements in the condition nodes of each execution unit structure in the protocol candidate object set are the same, according to the quantized values of the decision-making basis elements in the condition nodes of each execution unit structure, the quantized value threshold and the conditions determined by the corresponding condition nodes, carrying out decision-making basis element protocol on the condition nodes of each execution unit structure;
Or alternatively
And when the types of the decision basis elements in the conditional nodes of at least two execution unit structures in the specification candidate object set are different, determining that the decision basis elements in the conditional nodes of the execution unit structures cannot be combined.
Optionally, the execution module 703 is specifically configured to at least one of the following:
under the condition that the conditions determined by the condition nodes of the execution unit structures are smaller than the relation, determining a rule result of the decision basis element in the condition nodes of the execution unit structures according to the minimum quantized value threshold in the quantized value thresholds of the decision basis element in the condition nodes of the execution unit structures;
Or alternatively
Under the condition that the conditions determined by the condition nodes of the execution unit structures are all greater than the relation, determining a rule result of the decision basis element in the condition nodes of the execution unit structures according to the largest quantized value threshold in the quantized value thresholds of the decision basis element in the condition nodes of the execution unit structures;
Or alternatively
And under the condition that the condition nodes of the execution unit structures determine that the condition nodes of the execution unit structures have a smaller relationship and a larger relationship, determining that the decision basis elements in the condition nodes of the execution unit structures cannot be combined.
Optionally, the decision-making depends on the type of the element including at least one of: regional maneuver dominant elements; target hit dominance element.
Optionally, the execution module 703 is specifically configured to:
according to the types and interrelationships of decision control elements in the action nodes of each execution unit structure in the protocol candidate object set, the following operations are executed for the protocol candidate object set:
When the type of the decision control element in the action node of each execution unit structure in the protocol candidate object set comprises area elements, determining a protocol result of the area elements in the action node of each execution unit structure according to the union of the area elements in the action node of each execution unit structure when the inclusion relationship or the intersection relationship exists between the area elements in the action node of each execution unit structure; or when the inclusion relationship or the intersection relationship does not exist between the area elements in the action nodes of the execution unit structures, determining that the area elements in the action nodes of the execution unit structures cannot be combined;
Or alternatively
Determining a specification result of the hit target element in the action node of each execution unit structure according to the hit target element in the action node of each execution unit structure when the hit target element in the action node of each execution unit structure is the same under the condition that the type of the decision control element in the action node of each execution unit structure in the specification candidate object set comprises the hit target element; or when at least two hit target elements are included in the action nodes of each execution unit structure, determining that the area elements in the action nodes of each execution unit structure are not combinable.
Optionally, the type of the decision control element includes at least one of: a region element; striking the target element.
Optionally, the execution module 703 is specifically configured to at least one of the following:
Determining the condition nodes and the corresponding action nodes of the aggregation execution units corresponding to the reduction candidate object sets according to the reduction results of the decision basis elements in the condition nodes of the execution unit structures and the reduction results of the area elements in the action nodes of the execution unit structures;
Or alternatively
Determining the condition nodes and the corresponding action nodes of the aggregation execution units corresponding to the reduction candidate object sets according to the reduction results of the decision basis elements in the condition nodes of the execution unit structures and the reduction results of the hit target elements in the action nodes of the execution unit structures;
Or alternatively
Under the condition that decision basis elements in the condition nodes of each execution unit structure can be combined and action nodes of each execution unit structure cannot be combined, combining the decision basis elements in the condition nodes of each execution unit structure, and determining condition nodes of an aggregate execution unit and corresponding parallel nodes, wherein the action nodes of each execution unit structure are hung below the parallel nodes;
And under the condition that the decision-making basis elements in the conditional nodes of the execution unit structures cannot be combined, determining a plurality of sequential nodes of the aggregate execution unit, and hanging one uncombinable conditional node and a corresponding action node under each sequential node.
Fig. 8 is a schematic diagram of an entity structure of an electronic device according to the present invention, as shown in fig. 8, the electronic device may include: processor 810, communication interface (Communications Interface) 820, memory 830, and communication bus 840, wherein processor 810, communication interface 820, memory 830 accomplish communication with each other through communication bus 840. Processor 810 can invoke logic instructions in memory 830 to perform a multi-behavioral tree decision scheme aggregation method for game-oriented countermeasure tasks, the method comprising:
Constructing a behavior tree set corresponding to each behavior tree according to the execution unit structures of a plurality of behavior trees facing game countermeasure tasks, wherein the behavior tree set comprises the execution unit structures of each behavior tree;
Constructing a plurality of protocol candidate object sets corresponding to the action tree sets according to at least one time interval corresponding to the action tree sets, wherein each protocol candidate object set comprises a plurality of execution unit structures for controlling the same agent in the same time interval in each action tree;
And (c) respectively executing the steps a-c for the protocol candidate object set in the behavior tree set:
step a, according to the types of decision basis elements in the condition nodes of each execution unit structure in the protocol candidate object set, respectively carrying out decision basis element protocol on the condition nodes of each execution unit structure;
step b, according to the types and interrelationships of decision control elements in the action nodes of each execution unit structure in the protocol candidate object set, respectively carrying out decision control element protocol on the action nodes of each execution unit structure;
Step c, aggregating the condition nodes and the action nodes of each execution unit structure to obtain an aggregate execution unit corresponding to the protocol candidate object set;
and determining an aggregate behavior tree corresponding to the behavior tree set according to the aggregate execution units corresponding to each protocol candidate object set in the behavior tree set.
Further, the logic instructions in the memory 830 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product, the computer program product including a computer program, the computer program being storable on a non-transitory computer readable storage medium, the computer program when executed by a processor being capable of executing the multi-behavior tree decision scheme aggregation method for game countermeasure tasks provided by the above methods, the method comprising:
Constructing a behavior tree set corresponding to each behavior tree according to the execution unit structures of a plurality of behavior trees facing game countermeasure tasks, wherein the behavior tree set comprises the execution unit structures of each behavior tree;
Constructing a plurality of protocol candidate object sets corresponding to the action tree sets according to at least one time interval corresponding to the action tree sets, wherein each protocol candidate object set comprises a plurality of execution unit structures for controlling the same agent in the same time interval in each action tree;
And (c) respectively executing the steps a-c for the protocol candidate object set in the behavior tree set:
step a, according to the types of decision basis elements in the condition nodes of each execution unit structure in the protocol candidate object set, respectively carrying out decision basis element protocol on the condition nodes of each execution unit structure;
step b, according to the types and interrelationships of decision control elements in the action nodes of each execution unit structure in the protocol candidate object set, respectively carrying out decision control element protocol on the action nodes of each execution unit structure;
Step c, aggregating the condition nodes and the action nodes of each execution unit structure to obtain an aggregate execution unit corresponding to the protocol candidate object set;
and determining an aggregate behavior tree corresponding to the behavior tree set according to the aggregate execution units corresponding to each protocol candidate object set in the behavior tree set.
In yet another aspect, the present invention provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method for aggregating a multi-treelike decision scheme for game-oriented countermeasure tasks provided by the methods described above, the method comprising:
Constructing a behavior tree set corresponding to each behavior tree according to the execution unit structures of a plurality of behavior trees facing game countermeasure tasks, wherein the behavior tree set comprises the execution unit structures of each behavior tree;
Constructing a plurality of protocol candidate object sets corresponding to the action tree sets according to at least one time interval corresponding to the action tree sets, wherein each protocol candidate object set comprises a plurality of execution unit structures for controlling the same agent in the same time interval in each action tree;
And (c) respectively executing the steps a-c for the protocol candidate object set in the behavior tree set:
step a, according to the types of decision basis elements in the condition nodes of each execution unit structure in the protocol candidate object set, respectively carrying out decision basis element protocol on the condition nodes of each execution unit structure;
step b, according to the types and interrelationships of decision control elements in the action nodes of each execution unit structure in the protocol candidate object set, respectively carrying out decision control element protocol on the action nodes of each execution unit structure;
Step c, aggregating the condition nodes and the action nodes of each execution unit structure to obtain an aggregate execution unit corresponding to the protocol candidate object set;
and determining an aggregate behavior tree corresponding to the behavior tree set according to the aggregate execution units corresponding to each protocol candidate object set in the behavior tree set.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (10)

1. A multi-behavior tree decision scheme aggregation method facing game countermeasure tasks is characterized by comprising the following steps:
According to the execution unit structures of a plurality of behavior trees facing game countermeasure tasks, a behavior tree set corresponding to each behavior tree is constructed, the behavior tree set comprises the execution unit structures of each behavior tree, and the game countermeasure tasks are unmanned plane game countermeasure tasks;
constructing a plurality of protocol candidate object sets corresponding to the action tree sets according to at least one time interval corresponding to the action tree sets, wherein each protocol candidate object set comprises a plurality of execution unit structures for controlling the same agent in the action tree in the same time interval, and the agent is an unmanned plane;
And (c) respectively executing the steps a-c for the protocol candidate object set in the behavior tree set:
step a, according to the types of decision basis elements in the condition nodes of each execution unit structure in the protocol candidate object set, respectively carrying out decision basis element protocol on the condition nodes of each execution unit structure;
step b, according to the types and interrelationships of decision control elements in the action nodes of each execution unit structure in the protocol candidate object set, respectively carrying out decision control element protocol on the action nodes of each execution unit structure;
Step c, aggregating the condition nodes and the action nodes of each execution unit structure to obtain an aggregate execution unit corresponding to the protocol candidate object set;
and determining an aggregate behavior tree corresponding to the behavior tree set according to the aggregate execution units corresponding to each protocol candidate object set in the behavior tree set.
2. The method for aggregating multiple action tree decision schemes for game countermeasure tasks according to claim 1, wherein the performing decision-making element rules on the conditional nodes of each execution unit structure according to the types of decision-making element in the conditional nodes of each execution unit structure in the rule candidate object set includes:
when the types of the decision-making basis elements in the condition nodes of each execution unit structure in the protocol candidate object set are the same, according to the quantized values of the decision-making basis elements in the condition nodes of each execution unit structure, the quantized value threshold and the conditions determined by the corresponding condition nodes, carrying out decision-making basis element protocol on the condition nodes of each execution unit structure;
Or alternatively
And when the types of the decision basis elements in the conditional nodes of at least two execution unit structures in the specification candidate object set are different, determining that the decision basis elements in the conditional nodes of the execution unit structures cannot be combined.
3. The method for aggregating multiple action tree decision schemes for game countermeasure tasks according to claim 2, wherein the decision-making element rule is performed on the condition nodes of each execution unit structure according to the conditions determined by the quantized value, the quantized value threshold and the corresponding condition nodes of the decision-making element in the condition nodes of each execution unit structure, and the method comprises at least one of the following:
under the condition that the conditions determined by the condition nodes of the execution unit structures are smaller than the relation, determining a rule result of the decision basis element in the condition nodes of the execution unit structures according to the minimum quantized value threshold in the quantized value thresholds of the decision basis element in the condition nodes of the execution unit structures;
Or alternatively
Under the condition that the conditions determined by the condition nodes of the execution unit structures are all greater than the relation, determining a rule result of the decision basis element in the condition nodes of the execution unit structures according to the largest quantized value threshold in the quantized value thresholds of the decision basis element in the condition nodes of the execution unit structures;
Or alternatively
And under the condition that the condition nodes of the execution unit structures determine that the condition nodes of the execution unit structures have a smaller relationship and a larger relationship, determining that the decision basis elements in the condition nodes of the execution unit structures cannot be combined.
4. A multi-action tree decision scheme aggregation method for game-oriented countermeasure tasks according to claim 2 or 3, characterized in that the decision basis element types include at least one of the following: regional maneuver dominant elements; target hit dominance element.
5. The method for aggregating multiple action tree decision schemes for game countermeasure tasks according to claim 1, wherein the performing decision control element conventions for the action nodes of each execution unit structure according to the types and interrelationships of decision control elements in the action nodes of each execution unit structure in the convention candidate object set respectively includes:
according to the types and interrelationships of decision control elements in the action nodes of each execution unit structure in the protocol candidate object set, the following operations are executed for the protocol candidate object set:
When the type of the decision control element in the action node of each execution unit structure in the protocol candidate object set comprises area elements, determining a protocol result of the area elements in the action node of each execution unit structure according to the union of the area elements in the action node of each execution unit structure when the inclusion relationship or the intersection relationship exists between the area elements in the action node of each execution unit structure; or when the inclusion relationship or the intersection relationship does not exist between the area elements in the action nodes of the execution unit structures, determining that the area elements in the action nodes of the execution unit structures cannot be combined;
Or alternatively
Determining a specification result of the hit target element in the action node of each execution unit structure according to the hit target element in the action node of each execution unit structure when the hit target element in the action node of each execution unit structure is the same under the condition that the type of the decision control element in the action node of each execution unit structure in the specification candidate object set comprises the hit target element; or when at least two hit target elements are included in the action nodes of each execution unit structure, determining that the area elements in the action nodes of each execution unit structure are not combinable.
6. The game countermeasure task oriented multi-action tree decision scheme aggregation method of claim 5, wherein the types of decision control elements include at least one of: a region element; striking the target element.
7. The method for aggregating multiple behavior tree decision schemes for game countermeasure tasks according to claim 1, wherein the aggregating the condition nodes and the action nodes of each execution unit structure to obtain an aggregate execution unit corresponding to the protocol candidate set includes at least one of:
Determining the condition nodes and the corresponding action nodes of the aggregation execution units corresponding to the reduction candidate object sets according to the reduction results of the decision basis elements in the condition nodes of the execution unit structures and the reduction results of the area elements in the action nodes of the execution unit structures;
Or alternatively
Determining the condition nodes and the corresponding action nodes of the aggregation execution units corresponding to the reduction candidate object sets according to the reduction results of the decision basis elements in the condition nodes of the execution unit structures and the reduction results of the hit target elements in the action nodes of the execution unit structures;
Or alternatively
Under the condition that decision basis elements in the condition nodes of each execution unit structure can be combined and action nodes of each execution unit structure cannot be combined, combining the decision basis elements in the condition nodes of each execution unit structure, and determining condition nodes of an aggregate execution unit and corresponding parallel nodes, wherein the action nodes of each execution unit structure are hung below the parallel nodes;
And under the condition that the decision-making basis elements in the conditional nodes of the execution unit structures cannot be combined, determining a plurality of sequential nodes of the aggregate execution unit, and hanging one uncombinable conditional node and a corresponding action node under each sequential node.
8. A multi-action tree decision scheme aggregation device for game countermeasure tasks, comprising:
the system comprises a first construction module, a second construction module and a first control module, wherein the first construction module is used for constructing a behavior tree set corresponding to each behavior tree according to the execution unit structures of a plurality of behavior trees facing game countermeasure tasks, the behavior tree set comprises the execution unit structures of the behavior trees, and the game countermeasure tasks are unmanned plane game countermeasure tasks;
The second construction module is used for constructing a plurality of protocol candidate object sets corresponding to the action tree sets according to at least one time interval corresponding to the action tree sets, wherein each protocol candidate object set comprises a plurality of execution unit structures for controlling the same agent in the same time interval in each action tree, and the agent is an unmanned plane;
the execution module is used for executing the steps a-c for the protocol candidate object set in the behavior tree set respectively:
step a, according to the types of decision basis elements in the condition nodes of each execution unit structure in the protocol candidate object set, respectively carrying out decision basis element protocol on the condition nodes of each execution unit structure;
step b, according to the types and interrelationships of decision control elements in the action nodes of each execution unit structure in the protocol candidate object set, respectively carrying out decision control element protocol on the action nodes of each execution unit structure;
Step c, aggregating the condition nodes and the action nodes of each execution unit structure to obtain an aggregate execution unit corresponding to the protocol candidate object set;
and the determining module is used for determining an aggregate behavior tree corresponding to the behavior tree set according to the aggregate execution unit corresponding to each protocol candidate object set in the behavior tree set.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the multi-behavioral tree decision-scheme aggregation method for game-oriented challenge tasks of any one of claims 1 to 7 when the program is executed by the processor.
10. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor implements a multi-treeing decision scheme aggregation method for game-oriented challenge tasks according to any of claims 1 to 7.
CN202410142819.5A 2024-02-01 2024-02-01 Multi-behavior tree decision scheme aggregation method and device for game countermeasure task Active CN117669710B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410142819.5A CN117669710B (en) 2024-02-01 2024-02-01 Multi-behavior tree decision scheme aggregation method and device for game countermeasure task

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410142819.5A CN117669710B (en) 2024-02-01 2024-02-01 Multi-behavior tree decision scheme aggregation method and device for game countermeasure task

Publications (2)

Publication Number Publication Date
CN117669710A CN117669710A (en) 2024-03-08
CN117669710B true CN117669710B (en) 2024-05-10

Family

ID=90075437

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410142819.5A Active CN117669710B (en) 2024-02-01 2024-02-01 Multi-behavior tree decision scheme aggregation method and device for game countermeasure task

Country Status (1)

Country Link
CN (1) CN117669710B (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113704252A (en) * 2021-07-23 2021-11-26 建信金融科技有限责任公司 Rule engine decision tree implementation method and device, computer equipment and computer readable storage medium
CN115617821A (en) * 2021-07-14 2023-01-17 福建省天奕网络科技有限公司 Behavior tree-based task system data updating method and system
CN115900433A (en) * 2022-12-08 2023-04-04 北京理工大学 Decision method of multi-agent unmanned countermeasure system based on SWOT analysis and behavior tree
CN116108618A (en) * 2022-11-21 2023-05-12 江南机电设计研究所 Method and device for backtracking policy data of air defense system
CN116306941A (en) * 2023-02-20 2023-06-23 南京航空航天大学 Unmanned aerial vehicle multi-round attack and defense game decision method based on decision tree
CN116522606A (en) * 2023-04-07 2023-08-01 北京仿真中心 Simulation system, method, equipment and medium based on combat behavior tree
WO2023142076A1 (en) * 2022-01-29 2023-08-03 西门子股份公司 Workflow control method, apparatus and system, and medium and program product
WO2023142587A1 (en) * 2022-01-27 2023-08-03 腾讯科技（深圳）有限公司 Virtual object control method and apparatus, device, medium, and program product
CN117369521A (en) * 2023-12-04 2024-01-09 中国科学院自动化研究所 Method, device and equipment for generating behavior tree model path for unmanned aerial vehicle decision

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115617821A (en) * 2021-07-14 2023-01-17 福建省天奕网络科技有限公司 Behavior tree-based task system data updating method and system
CN113704252A (en) * 2021-07-23 2021-11-26 建信金融科技有限责任公司 Rule engine decision tree implementation method and device, computer equipment and computer readable storage medium
WO2023142587A1 (en) * 2022-01-27 2023-08-03 腾讯科技（深圳）有限公司 Virtual object control method and apparatus, device, medium, and program product
WO2023142076A1 (en) * 2022-01-29 2023-08-03 西门子股份公司 Workflow control method, apparatus and system, and medium and program product
CN116108618A (en) * 2022-11-21 2023-05-12 江南机电设计研究所 Method and device for backtracking policy data of air defense system
CN115900433A (en) * 2022-12-08 2023-04-04 北京理工大学 Decision method of multi-agent unmanned countermeasure system based on SWOT analysis and behavior tree
CN116306941A (en) * 2023-02-20 2023-06-23 南京航空航天大学 Unmanned aerial vehicle multi-round attack and defense game decision method based on decision tree
CN116522606A (en) * 2023-04-07 2023-08-01 北京仿真中心 Simulation system, method, equipment and medium based on combat behavior tree
CN117369521A (en) * 2023-12-04 2024-01-09 中国科学院自动化研究所 Method, device and equipment for generating behavior tree model path for unmanned aerial vehicle decision

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
"行为树技术的研究进展与应用";刘瑞峰 等;《计算机与现代化》;20200229(第2期);第76-88页 *

Also Published As

Publication number Publication date
CN117669710A (en) 2024-03-08

Similar Documents

Publication Publication Date Title
CN110852448A (en) Cooperative intelligent agent learning method based on multi-intelligent agent reinforcement learning
CN114721833B (en) Intelligent cloud coordination method and device based on platform service type
CN114281104B (en) Multi-unmanned aerial vehicle cooperative regulation and control method based on improved ant colony algorithm
CN112801540B (en) Intelligent collaborative architecture design method based on unmanned clusters
CN109242099A (en) Training method, device, training equipment and the storage medium of intensified learning network
CN112468347A (en) Security management method and device for cloud platform, electronic equipment and storage medium
CN115755954B (en) Routing inspection path planning method, system, computer equipment and storage medium
CN110493221A (en) A kind of network anomaly detection method based on the profile that clusters
Xiao et al. Network security situation prediction method based on MEA-BP
CN116582349A (en) Attack path prediction model generation method and device based on network attack graph
CN111445065A (en) Energy consumption optimization method and system for refrigeration group control of data center
CN111491300A (en) Risk detection method, device, equipment and storage medium
CN117669710B (en) Multi-behavior tree decision scheme aggregation method and device for game countermeasure task
Peng et al. Modeling and solving the dynamic task allocation problem of heterogeneous UAV swarm in unknown environment
CN116365519B (en) Power load prediction method, system, storage medium and equipment
Sun et al. Security Attitude Prediction Model of Secret‐Related Computer Information System Based on Distributed Parallel Computing Programming
CN117033997A (en) Data segmentation method, device, electronic equipment and medium
CN115454473A (en) Data processing method based on deep learning vulnerability decision and information security system
Tao et al. Benign: An automatic optimization framework for the logic of swarm behaviors
Jiang et al. Ai and machine learning for industrial security with level discovery method
CN113965358B (en) Network security detection method and system for comprehensive energy system
CN117633541B (en) Parameter variable controllable sample generation method for game countermeasure scheme evaluation
Zhu et al. A Chaotic Disturbance Wolf Pack Algorithm for Solving Ultrahigh‐Dimensional Complex Functions
CN117350362A (en) Depth neural network strategy network model interpretability conversion algorithm
Hu et al. A UAV Cluster's Boundary Performance Evaluation Probablity Method Based on Discrete Artificial Potential Field

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
GR01 Patent grant
GR01 Patent grant