CN113656962A - Strategic layer game deduction method based on information flow - Google Patents

Abstract

The invention discloses a strategic layer game deduction method based on information flow, and belongs to the technical field of simulation deduction. The method abstracts a strategic layer game into a complex network according to the fighting thought of an OODA (object oriented data acquisition) ring and based on technologies such as graph theory, complex network analysis and visualization, layout generation, message queues and the like, forms four types of nodes of reconnaissance, command, influence and target, constructs edges of the complex network according to reconnaissance relations, command relations, attack relations and communication relations, and models a strategic countermeasure process into information circulation and countermeasure relations in the two complex networks, so that the potential countermeasure strategies and possible victory or defeat judgment of both enemies and the my parties from a strategic level is realized, and support is provided for the establishment of military strategies. The method has the advantages of good openness and capability of supporting the development of multi-granularity simulation, can be used in combination with tactical-level deduction simulation, and greatly improves the reliability of strategic-level deduction simulation results.

Description

Strategic layer game deduction method based on information flow

Technical Field

The invention relates to the technical field of simulation deduction, in particular to a strategic layer game deduction method based on information flow.

Background

With the rapid development of the information age, the battle forms tend to be intelligent, systematized and diversified, and even the evaluation of the adversarial and my adversarial results from the macroscopic perspective becomes unpredictable. The system modeling and game deduction become very effective means for promoting the concept demonstration of the battle and the military strategy formulation, play a key role in the aspects of the demonstration verification of the military information system and the application research of the battle, and urgently need to carry out the game deduction from the perspective of information organization and application at the strategic layer aiming at the characteristics of the current information system battle.

The graph theory is a subject of study on graphs, and belongs to a branch of mathematics. The graph in graph theory consists of a number of given points, where a point represents an object, and a line connecting two points, which represents the relationship between two objects. The relations between the fighting node networks can be clearly represented by using graph theory, reconnaissance, attack and other relations are represented as arrows, and different colors and shapes are used for indicating formation and node types, so that the fighting entities on a real battlefield can be converted into an abstract graph structure, the node networks are formed, and the foundation is laid for later network analysis and fighting simulation.

Network analysis and visualization belong to the major category of network science, and the network analysis and visualization involve a plurality of technologies such as data mining, network situation awareness and interaction visualization. By using these techniques to present, analyze and count networks, it is possible to more intuitively understand network topology information and obtain the overall properties of the network. The method is particularly applied to system confrontation, so that the commander can be clear to the fighting situation of the commander, and the method plays an auxiliary reference role in action decision.

Message queuing is a technique for exchanging information between distributed applications. The technology mainly solves the problems of application coupling, asynchronous messages, traffic cutting and the like, realizes high-performance, high-availability, scalable and final consistency architecture, and is one of indispensable technologies of large-scale distributed systems. In order to realize the decoupling and flexibility of functions, the fight simulation platform comprises a plurality of components capable of independently operating. The message queue is responsible for providing data transmission pipelines for the components, so that the components can cooperate to transmit information such as battlefield conditions, action instructions and the like to each other.

Based on the technology, a strategic layer game deduction method based on information flow can be realized, but the prior art is still lack of such a system.

Disclosure of Invention

In view of the above, the invention provides an information flow-based strategic layer game deduction method, which can better reflect the relationship between battlefield situations and operational entities when two systems resist, and present, analyze and simulate the obtained operational information network, so that a decision maker can know the operational conditions, and assist a strategic commander to generate more comprehensive understanding and more accurate evaluation on the potential situations of the two parties, thereby being beneficial to making an optimal strategic decision.

In order to achieve the purpose, the invention adopts the technical scheme that:

a strategic layer game deduction method based on information flow comprises the following steps:

step S1, model preparation phase: abstracting each combat unit of the red and blue parties into a target node, and abstracting the attribute of the combat unit into a corresponding attribute node, wherein the reconnaissance attribute is abstracted into a reconnaissance node, the command attribute is abstracted into a command node, and the attack attribute is abstracted into an attack node; the attribute nodes of the same operation unit are all attributed to the target nodes of the operation unit; each node of the same combat unit has the same state attribute and signal reflection attribute, the state attribute is used for representing the current blood volume of the node, the signal reflection attribute is used for representing what type of detector the target node can be detected by, and the type of the detector comprises radar, infrared and optics; the target nodes communicate with each other through links;

step S2, operation stage: the two parties of the red and the blue carry out systematic countervailing deduction through a local area network, the deduction process is a fast rhythm round system, the two parties respectively operate own nodes and configure reconnaissance and attack tasks to form OODA packets which react in a de-circulation manner to attack enemy nodes; after each round is finished, the loss conditions of both parties are settled;

step S3, decision stage: if one party cannot establish an OODA packet with a cycle, the other party is declared a winner.

Further, the specific process of step S2 is:

step S201: three computer terminals are arranged in the local area network and are respectively a red, a blue and a white party; the red party and the blue party are two parties of a battle, the conditions of the two parties cannot be known mutually, different types of nodes can be added or deleted respectively, a reconnaissance task can be configured for a reconnaissance node, and a batting task can be configured for a batting node; the Baifang is a judge party and can specify a specific battlefield environment to be simulated and confronted, observe the real-time node network conditions of the Baifang and the Baifang, check the situation evaluation statistical indexes of the Baifang and the Baifang, the change trend of the number of various nodes and the attribute of a specific certain node, and judge the fighting condition result;

step S202: when both the red and blue parties are ready, the round starts, and in one round, the two parties can carry out the related operation and task configuration of the nodes and the links, and change the node attributes and the topological structure of the graph; after the red and blue sides finish the configuration of nodes, links and tasks, clicking to finish the configuration, namely entering a waiting stage, wherein the operation cannot be carried out in the waiting stage;

step S203: after both sides finish the turn, calculating the damage to the enemy node according to the node attribute in the OODA packet in the de-turn, if the damage is larger than the current blood volume, the enemy node is eliminated, and simultaneously, all other nodes corresponding to the node in the same operation unit also disappear.

Compared with the prior art, the invention has the following advantages:

1. the invention combines the concept of OODA (namely, Observation of Observation, judgment of organization, Decision of Decision and Action execution of the first letter of four words, also called Baudede cycle) with the fighting simulation process, and realizes a strategic layer game deduction method based on information flow based on the relevant theoretical knowledge of a complex network. The method can effectively reflect the fighting behaviors and the battlefield situations of both parties and provide the change trend of each index by abstracting the battlefield information into a graph structure and adopting a series of technologies to present, analyze and count the graph structure, thereby deducing more credible system game results.

2. The invention is based on two visual angles of OODA and target node, abstracts and analyzes the fighting behaviors of both sides, performs multi-dimensional modeling on complex events which occur in parallel on a battlefield, gives analysis statistics and variation trend of various indexes, can effectively reflect the fighting behaviors and battlefield situation of both sides, and makes the decision process become more data and efficient, thereby better assisting in command and making strategic decision makers more concentrate on the process of strategy design and dynamic game.

Drawings

Fig. 1 is a schematic diagram of the OODA relationship between different types of nodes and different types of nodes in the embodiment of the present invention.

Fig. 2 is a diagram of a predetermined battlefield environment of a red party in an embodiment of the present invention. The topology of the blue-party pre-defined battlefield environment is the same.

Detailed Description

In order to make the technical solution of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be noted that the specific embodiments described herein are only used for explaining the technical solutions of the present invention, and are not used for limiting the protection scope of the present invention.

A strategic layer game deduction method based on information flow comprises the following steps:

s1, model preparation phase: various combat units are abstracted to contain different functional nodes: the system comprises a detection node, a command node, a strike node and a target node.

The attribute of the reconnaissance node designed by the method is set as the maximum detection number and carried in a detection mode. The former refers to the maximum number of targets that can be detected simultaneously, and the latter includes three detection modes, which are LD (radar), HW (infrared), and GX (optical), respectively, corresponding to the signal reflection properties of the node. It should be noted that a probing node must have more than one probing method.

Because the scout nodes have a plurality of detection modes and the target nodes have a plurality of types of reflected signals, the detection nodes can acquire target information if and only if the types of the scout nodes are the same, and each detection mode can generate an information integrity value to represent the number of detected information. In addition, due to the fact that multiple detection modes can be carried, information detected by the scout node is diversified, and the scout node can integrate the information detected by the multiple different detection modes to form the unique information integrity of the target node.

When the reconnaissance mission is configured, the detectable objects are calculated according to factors such as distance, but because the system confronts the simulation platform and does not accurately represent the battlefield, the battlefield is represented from a battle logic perspective, and therefore the position distribution of the nodes in the simulation platform is not represented by the real distance in the battlefield environment. Therefore, the current preliminary strategy is that the detectable object is the target node owned by the enemy. However, since the maximum probing number of a scout node limits the number of target nodes that it probes simultaneously, the probing task that is actually being performed should be a non-empty subset thereof. Selecting a currently executed probe task from among the detectable targets adds the selected node to the scout probe task configuration list. Different detection accuracies of the detection modes can be set at this time: LD _ JD, HW _ JD and GX _ JD, and the value range [0, 100], wherein the detection precision influences the integrity of the obtained information, and the higher the precision is, the higher the integrity of the obtained information is. Assuming that 1 represents the probing method, 0 represents none, n probing methods are available for configuration, the probing method configuration D of the existing probing node is { D1, D2.., dn }, the reflection method of the target node is F { F1, F2.., fn }, and the corresponding probing accuracy is JD { JD1, JD 2.., jdn }, the obtained information integrity calculation formula is as follows:

the integrity of the detected information and the combat effectiveness of the attack nodes in the OODA ring are key factors for calculating damage to the target nodes of the enemy. In addition, the required number of rounds can be specified when the reconnaissance task is configured, and because the reconnaissance task cannot be executed instantly and needs several rounds of time, the configuration is more suitable for practical situations.

The command nodes designed by the method exist as the integrators and the transferors of the information. The command node can be connected with a plurality of scout nodes to obtain the scout information integrity of the command node to different detection targets. If the detection task performed by the detection node d, d is DM { t1, t 2., tn } (a plurality of detection targets, ti represents an enemy target), the command node o will obtain an information integrity list MSG { m1, m 2., mn }. If the command node is connected with a scout node set, an information integrity list set is obtained, wherein MSGset is { msg1, msg2.

Since the scout objects of different scout nodes may be the same or different enemy target nodes, information integration needs to be performed on the information integrity list set by taking a target as a unit. Setting a command node o, obtaining information integrity lists detected by two detection nodes, namely, an MSG1 ═ { m1(t1), an m2(t3), an m3(t4) }, an MSG2 ═ m1(t1), an m2(t2) and an m3(t3) }, and accumulating the information integrity of the same target node to obtain the integrated target information integrity. For the data in the hypothetical example, it will eventually be obtained that t1 (information integrity) ═ msg1.m1+ msg2.m1, t2 (information integrity) ═ msg2.m2, and so on. For the obtained information integrity list set, the above forms are added to obtain an information integrity list, MSG ═ { m1(t1), m2(t2),.. gtang, mn (tn) }. According to the information integrity calculation method of the investigation node, the obtained information integrity is a decimal of a percentage, so that for the command node, the obtained information integrity list is accumulated, wherein the information integrity of a single target may be greater than 1, so that all the information integrity greater than or equal to 1 is set to be 1 (equivalent to obtaining all the information of the target). Finally, objects with information integrity less than 50 percent (information integrity is insufficient and cannot be hit) are excluded, and all the other objects are used as selectable hitting tasks.

The hitting node designed by the method executes hitting tasks on known enemy target nodes with enough information integrity. Its special attributes are two kinds, the striking ability and the maximum striking number. The former is the hitting attribute of the node itself, and affects the final damage effect together with the information integrity; the latter refers to the maximum number of targets that can be hit simultaneously. After the command nodes integrate and screen the information integrity of the results of the scout nodes, the attacking nodes can acquire the attacking objects, namely the enemy target nodes meeting the conditions, from the superior command nodes. But since the maximum number of hits for a hit node limits the number of target nodes it hits at the same time, the hit task that is actually being performed should be a non-empty subset of them. Selecting a currently executed percussion task among the impartable targets adds the selected node to the percussion task configuration list. Like reconnaissance nodes, the striking task cannot be completed instantly, so that the striking node has an index of the number of rounds, and the striking node can damage target nodes of enemies after the number of rounds. The damage calculation mode is that if a striking point a is set, the striking capability is Aa, the information integrity of a striking task target is msg, and the blood volume of a target node is b, the damage caused by the method is calculated as follows:

BloodLoss＝b*(A_a*msg)

BloodRest＝b-BloodLoss

wherein A is_aMsg is a decimal number indicating the percentage of blood volume that can be removed. Since the target is hit many times during the fight, the initial base blood volume of the target node is used each time the damage is calculated.

The target node designed by the method represents a specific entity target in the system confrontation simulation platform, and the investigation node, the command node and the attack node of the confrontation parties can be regarded as part of the target node. A target node corresponds to a unit of action on the battlefield and includes one or more reconnaissance, command or attack nodes. For example, a ship is a target node, and a detection node, a command node and a attack node are arranged on the ship. A tank is a target node that contains an attack node. Thus, the target node is higher in the dimension in which it includes other types of nodes.

In the system countermeasure process, the investigation, command and attack node is the minimum unit for the user to execute operations, that is, the two countermeasures can operate some kind of node, for example, the investigation node of the red control target a executes the investigation task or the attack node of the control target B attacks.

The target node can be understood as the minimum unit of combat, that is, a target node as a whole, which contains the common attributes of the investigation, command or attack nodes consistent with the target node. For example, the target node a in the above figure does not reflect the signal wave, and the blood volume is 100, so that the spy node and the hit node included in the target node a do not reflect the signal wave, and their blood volumes are 100. It should be noted that the execution objects of the scout node and the attack node can only be the target node (not the command, scout or attack node, because the target node represents a specific unit of combat). If the enemy attack node performs the attack task on the target node A, when the blood volume is damaged to be less than 0, the target node A and all the nodes contained by the target node A are destroyed, because the command, reconnaissance or attack node represents an abstract function and must be attached to the target node. The target node has two basic attributes, which are a signal reflection attribute (whether a signal wave is reflected or not) and a node state (blood volume). For the former, an adversary can only detect the object by a detector of this signal type if the corresponding signal is reflected. The reflection properties correspond to the detector type and are classified into three categories, LD (radar), HW (infrared), and GX (optical). In addition, a target node comprises a signal reflection attribute of a spy, command or attack node consistent with the target node. In the latter case, after a node is hit, the state of the node decreases, and the specific properties and behaviors of the node change with the state. Similarly, a target node contains a volume of blood for a spy, command, or hit node that is consistent with the target node.

Finally, since a target node contains one or more scout, command or percussive nodes, the unique attributes of the target node inherit the respective unique attributes and behaviors of these nodes. Taking a certain target node (a ship) as an example, since it includes a reconnaissance node (radar detection equipment), a command node (command room) and a attack node (ship-borne cannonball) at the same time, the target node has the unique attributes of the three nodes it includes, namely, the maximum simultaneous detection number, the carried detection mode and the attack capability.

The communication link designed by the method connects two nodes of a specific type and serves as a link for information transmission flow. Because the flow of information is directional, the communication link is also directional. The different nodes and communication links eventually form a directed graph as shown in fig. 1. For the start node and the end node of the communication link, there are the following rules:

a) the single investigation node can only be connected to the command node, and one investigation node can only be connected to one command node, which means that the command node can obtain the investigation node information

b) The command nodes can be connected in one way to show information downward transmission (superior and inferior relation)

c) The single command node is connected with only one command node of the own party, and the command node points to the command node to indicate that only the command of one command node is received

d) The scout node can detect a plurality of enemy target nodes

e) The striking node can strike a plurality of enemy target nodes

It should be noted that the first three items can be realized by adding a communication link, but the second two items need to configure the spy task and the batting task to generate the corresponding edges.

S2, operation stage: the red and blue parties carry out system confrontation simulation in the local area network range, the red and blue parties of the simulation software can operate different types of nodes of the own party and configure reconnaissance and attack tasks, an OODA battle ring is formed to react, attack enemy nodes, and finally win victory. The software simulates the processes of action strategy selection and game in the confrontation of two parties through the cyclic behavior flow of observation, adjustment, decision and action in the actual combat process of the OODA ring body based on the view of command decision. In the specific embodiment, the simulation platform operates independently, and the red, blue and white are located in the same local area network for counter simulation. First, a typical OODA loop is shown in FIG. 1, which includes participation of scout nodes, director nodes, aggressor nodes, and target nodes. Since the network topology is more complex when more nodes and reconnaissance and attack relations are involved, in order to reduce the time spent by both parties in the initial configuration, the software is preset with four battlefield environments which can be loaded by the white parties. The preset environment selected in this example is shown in fig. 2, and the total command node of each party governs two sub command nodes, and each sub command node is responsible for three attack nodes and one scout node. In order to implement attack on enemy nodes, the own scout node needs to be configured with a scout task, then the command node transmits related information to the attack nodes, and the attack nodes are configured with attack tasks, so that the attack can be formed by settlement after the turn is finished.

The detailed process of step S2 is:

s201: the strategy layer game deduction process is a fast rhythm turn system and is operated at three computer terminals of the local area network. The operation mode is as follows: the formation is divided into red, blue and white three parties, the red party and the blue party are two parties in battle, the conditions of the two parties cannot be known mutually, different types of nodes can be added/deleted respectively, reconnaissance tasks can be configured for reconnaissance nodes, and attack tasks can be configured for attack nodes; the Baifang is a judge party and can specify a specific battlefield environment to be simulated and confronted, observe the real-time node network conditions of the two parties, check the situation evaluation statistical indexes of the two parties, the variation trend of the number of various nodes and the attribute of a specific certain node, and judge the battle condition result. The red and blue parties can only see respective node networks and cannot know the node information of the enemy. Wherein, the red party is firstly provided with a reconnaissance device 1 to reconnaissance the general command node of the enemy, and is provided with a percussion task to the charge device 1, thereby forming an OODA battle ring of 'blue party total command part-red party reconnaissance device 1-red party squad command room-red party charge device 1-blue party total command part', which can cause harm in the settlement stage. The blue side operates similarly to the red side, forming an OODA ring.

S202: when the red and blue parties are ready, the turn starts, and in one turn, the two parties can perform related operations and task configuration of nodes and links, and change the node properties and the topological structure of the graph. After the red and blue sides complete the configuration of the nodes/links/tasks, the configuration is finished by clicking, namely, the waiting stage is entered, and the operation cannot be carried out in the waiting stage. In the aspect of visualization, the moving, zooming and dragging operations of the drawing and the nodes in the drawing are supported, and links and OODA rings related to a certain node can be viewed. In addition, switching of two views, namely an OODA ring view and a target node view, is provided. The former embodies the process of battle, can express abstract battlefield elements and the flowing direction of information between the elements, and is the internal logic embodiment of the battlefield; the latter focuses on battlefield entities, with target nodes as the smallest units for describing the external appearance of the battlefield. The Bai can see the global situation and obtain the statistical information and various indexes of the battlefield through the situation perception instrument panel. In addition, the Bai Fang can also check the battle log and the attributes of any node. The red and blue party can carry out the configuration of the tasks of adding and deleting, reconnaissance and attacking nodes and links and check the related information of the nodes and the links.

S3, decision stage: after both sides finish the round, the settlement work of the countermeasures is automatically carried out. Firstly, calculating damage to an enemy target node according to the node attribute in the OODA ring, and if the damage is larger than the existing blood volume, the enemy target node is destroyed and other types of nodes contained in the target node also disappear. In the specific embodiment, the total directors of both red and blue will be damaged to some extent, but because the total directors have higher blood volume, more rounds are needed to win or lose. Assuming that the attack mode is still maintained by the red party in the next round, no more configuration is carried out, the blue party establishes another OODA ring, and one of the three target nodes of scout connection 1, squad 1 or assault connection 1 of the red party is attacked to destroy the OODA ring of the red party, so that the next round of attack of the red party fails, and the OODA ring of the blue party is not destroyed and the attack can be continued. Finally, if one party can not establish an OODA ring, the other party is declared the winner.

In summary, the whole flow should be as follows: after the turn is started, the two countersides mutually create and set various nodes and configure detection tasks for the scout nodes. After a plurality of rounds, the task is executed, and the command node integrates the integrity of the target information collected by the investigation node and configures the attack task for the attack node. After a plurality of rounds, the hitting nodes combine the information integrity and the hitting capability of the hitting nodes to damage the target nodes of the enemy. And then, the two parties adjust the strategy according to the situation of war to carry out a new round of game confrontation, and the game confrontation is repeated in a circulating way until one party can not carry out complete OODA circulating scheduling any more, namely the winning is finished.

The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above-mentioned embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications to the invention that do not depart from the spirit of the invention should be construed as within the scope of the invention as set forth in the claims that follow.

Claims (2)

1. A strategic layer game deduction method based on information flow is characterized by comprising the following steps:

step S1, model preparation phase: abstracting each combat unit of the red and blue parties into a target node, and abstracting the attribute of the combat unit into a corresponding attribute node, wherein the reconnaissance attribute is abstracted into a reconnaissance node, the command attribute is abstracted into a command node, and the attack attribute is abstracted into an attack node; the attribute nodes of the same operation unit are all attributed to the target nodes of the operation unit; each node of the same combat unit has the same state attribute and signal reflection attribute, the state attribute is used for representing the current blood volume of the node, the signal reflection attribute is used for representing what type of detector the target node can be detected by, and the type of the detector comprises radar, infrared and optics; the target nodes communicate with each other through links;

step S2, operation stage: the two parties of the red and the blue carry out systematic countervailing deduction through a local area network, the deduction process is a fast rhythm round system, the two parties respectively operate own nodes and configure reconnaissance and attack tasks to form OODA packets which react in a de-circulation manner to attack enemy nodes; after each round is finished, the loss conditions of both parties are settled;

step S3, decision stage: if one party cannot establish an OODA packet with a cycle, the other party is declared a winner.

2. The strategic layer game deduction method based on information flow as claimed in claim 1, wherein the specific process of step S2 is:

step S201: three computer terminals are arranged in the local area network and are respectively a red, a blue and a white party; the red party and the blue party are two parties of a battle, the conditions of the two parties cannot be known mutually, different types of nodes can be added or deleted respectively, a reconnaissance task can be configured for a reconnaissance node, and a batting task can be configured for a batting node; the Baifang is a judge party and can specify a specific battlefield environment to be simulated and confronted, observe the real-time node network conditions of the Baifang and the Baifang, check the situation evaluation statistical indexes of the Baifang and the Baifang, the change trend of the number of various nodes and the attribute of a specific certain node, and judge the fighting condition result;

step S202: when both the red and blue parties are ready, the round starts, and in one round, the two parties can carry out the related operation and task configuration of the nodes and the links, and change the node attributes and the topological structure of the graph; after the red and blue sides finish the configuration of nodes, links and tasks, clicking to finish the configuration, namely entering a waiting stage, wherein the operation cannot be carried out in the waiting stage;

step S203: after both sides finish the turn, calculating the damage to the enemy node according to the node attribute in the OODA packet in the de-turn, if the damage is larger than the current blood volume, the enemy node is eliminated, and simultaneously, all other nodes corresponding to the node in the same operation unit also disappear.

Images