Multi-level military strength joint deduction system based on adaptive aggregation and deaggregation

Abstract

The invention provides a multi-level weapon strength joint deduction system based on adaptive aggregation and deaggregation, and the system comprises a tactical-level simulation engine which is used for loading a simulation scenario to construct a simulation scene, and carrying out entity-level simulation deduction according to simulation time periods and step length simulation control parameters; the battle-level simulation engine is used for loading a simulation scenario to construct a simulation scene and receiving simulation control parameters to carry out unit-level simulation deduction; the polymerization and depolymerization module is used for carrying out mutual conversion and interaction among different resolution level models and comprises a depolymerization unit and a polymerization unit; and the heterogeneous interconnection module is used for providing a mechanism and an environment for accessing each simulation heterogeneous object to the system and carrying out data publishing and ordering, and realizing data interconnection and interworking and model interoperation among simulation objects through message middleware. The method has the advantages that the real-time performance and the verisimilitude of the multi-level heterogeneous joint simulation system are effectively balanced; the problem of multi-platform, multi-resolution and large-scene joint simulation is solved.

Description

Technical Field

[0001] The present application belongs to the field of large-scale joint force simulation, and specifically relates to a multi-level joint force simulation system based on adaptive aggregation and deaggregation. Background Art

[0002] The rapid development of computer and network technology has promoted the rapid development of simulation technology. At present, modern confrontation simulation technology based on computers has been widely used in military training, tactics research, etc., and confrontation simulation systems have also rapidly developed into an indispensable platform for cultivating and training high-quality military talents and studying modern wars and military issues. With the increasing complexity of battlefields in modern warfare, new challenges have been posed to traditional confrontation simulation technology. Confrontation simulation not only needs to simulate various joint confrontation actions including land, sea, air, and space at the campaign level, but also needs to carry out entity-level fine simulation for specific equipment at the tactical level in order to study the effectiveness of various equipment in different battlefield environments. Campaign-level confrontation simulation uses low-resolution unit-level models, generally describing aggregated, independent confrontation battalions, aviation squadrons, etc.; tactical-level confrontation simulation uses high-resolution entity-level models, generally describing independent confrontation soldiers, tanks, and aircraft, etc. Through campaign-level and tactical-level joint confrontation simulation, the needs of confrontation scenario description, battlefield situation analysis, and system evaluation of different scales and granularities can be met, thereby resolving the contradiction between simulation complexity and limited resources, and meeting the understanding and information hierarchy needs of different groups of people for simulation. There are two main problems in joint confrontation simulation at the campaign and tactical levels: large scale and high precision.

[0003] In terms of large-scale force simulation, campaign-level simulation systems and tactical-level simulation systems are heterogeneous simulation systems because they have different applicable fields, different simulation focuses, different simulation object resolutions, and different simulation architectures. Therefore, large-scale confrontation simulations inevitably require interaction (including interconnection, intercommunication, and interoperability) between heterogeneous systems with different resolution models, that is, joint operation, to achieve collaborative simulation. At present, distributed interactive simulation has become an important method and means for complex confrontation simulation. The interconnection of heterogeneous systems based on HLA (High Level Architecture) has also become one of the frontier topics in the current field of complex confrontation simulation, with broad application prospects.

[0004] In terms of high-precision force simulation, multi-resolution modeling technology is considered to be the most effective method for solving complex system simulation, and it has become one of the hot topics in the field of modeling and simulation. In distributed confrontation simulation, in order to solve high-precision calculations in large-scale simulation scenarios, models of different levels are generally established to describe simulation objects of different granularities, so as to achieve the goal of balancing the simulation scale and simulation accuracy. At present, the core problem that multi-resolution modeling needs to solve is the aggregation and disaggregation of models at different levels. Especially for the interconnection, intercommunication, and interoperability of confrontation simulation systems at different levels, in terms of simulation models, its essence is to solve the effective mapping between models of different levels of resolution of the same simulation object, that is, the aggregation and disaggregation of models.

[0005] Traditional simulation problems involving models with different levels of resolution are often solved by data transmission between simulation systems. The high-resolution model is simulated first, and the simulation results are used as the simulation input of the low-resolution model. This method is not very real-time and cannot meet the needs of joint simulation in tightly coupled multi-resolution large scenarios. Therefore, each platform needs to integrate its own simulation tasks into a unified platform through heterogeneous interconnection technology. At this time, the network engine on each platform is required to realize the networking communication and interconnection and interoperability between simulation entities, and support the interconnection and interaction with heterogeneous systems using protocols such as DIS, HLA, and DDS, and finally form a feasible, verifiable, and stable heterogeneous interconnection technology solution. At the same time, in order to accurately transfer the local confrontation area elements of the campaign-level simulation to the tactical-level simulation, it is necessary to complete the deaggregation conversion of the unit-level model to the entity-level model in real time according to the deaggregation rules under different types and states during the simulation process; conversely, in order to feed back the results of the tactical-level simulation to the campaign-level simulation, it is necessary to complete the calculation of the entity-level model results to the unit-level model results in real time according to the aggregation rules under different types and states during the simulation process, so as to ensure that the simulation process is seamless, correct, and efficient. Summary of the invention

[0006] The purpose of this application is to carry out research on model adaptive aggregation and deaggregation methods and design of heterogeneous interconnection solutions, focusing on the simulation requirements of different granularities in large-scale confrontation applications and the problem of collaborative interaction between simulations of different granularities and levels faced by campaign-level / tactical-level joint simulations, to build a large-scale, high-precision campaign-level and tactical-level force joint simulation system, and to provide technical support for confrontation solution optimization, equipment system demonstration and application, system contribution rate evaluation, and single equipment performance verification.

[0007] In order to achieve the above objectives, the present application proposes a multi-level force joint deduction system based on adaptive aggregation and deaggregation, the system comprising:

[0008] Tactical-level simulation engine, used to load simulation scenarios to build simulation scenarios, and conduct entity-level simulation deduction according to simulation period and step-size simulation control parameters;

[0009] The campaign-level simulation engine is used to load simulation assumptions to build simulation scenarios, and receive simulation control parameters sent by the tactical-level simulation engine to carry out unit-level simulation deduction;

[0010] The aggregation and deaggregation module is used for mutual conversion and interaction between models of different resolution levels, including deaggregation units and aggregation units; wherein,

[0011] The deaggregation unit is used to convert the unit-level model into a unit-level model or entity-level model with a smaller granularity;

[0012] Aggregation units, used to transform from entity-level models to unit-level models; and

[0013] The heterogeneous interconnection module is used to provide a mechanism and environment for each simulation heterogeneous object to access the system and publish and subscribe to data, and to achieve data interconnection and model interoperability between simulation objects through message middleware.

[0014] As an improvement of the above system, the aggregation methods implemented by the aggregation unit include: position aggregation, state aggregation and battle damage aggregation; the deaggregation methods implemented by the deaggregation unit include: position deaggregation, state deaggregation and battle damage deaggregation.

[0015] As an improvement of the above system, the position aggregation includes geometric mean method, weighted mean method and core positioning method;

[0016] The geometric mean method is to use the geometric center of each high-resolution entity constituting the aggregate as the position of the aggregate;

[0017] The weighted average method is to set the position of the aggregate at the weighted center of each high-resolution entity;

[0018] The core positioning method takes the position of the entity that undertakes the main confrontation task as the position of the aggregate among all high-resolution entities that constitute the aggregate; when there are multiple entities that undertake the main confrontation task, the geometric center of the multiple entities that undertake the main confrontation task is taken as the position of the aggregate.

[0019] As an improvement of the above system, the state aggregation includes same-state aggregation and different-state aggregation;

[0020] The same-state aggregation is to use the states of the entities participating in the aggregation as the state of the aggregate;

[0021] The heterogeneous state aggregation includes a majority entity state determination method and a key entity state determination method; wherein,

[0022] The majority entity state determination method is to take the state of the majority of entities participating in the aggregation as the state of the aggregate;

[0023] The key entity state determination method is to use the state of the key entity among the entities participating in the aggregation as the state of the aggregate.

[0024] As an improvement of the above system, the battle damage aggregation includes battle damage aggregation of entity models of the same type and battle damage aggregation of entity models of different types; wherein,

[0025] The battle damage aggregation of the entity models of the same type is to add up the battle damage of the aggregated entities to obtain the aggregated battle damage;

[0026] The battle damage aggregation of different types of entity models includes two methods:

[0027] One method is to divide the battle area into several battle zones, calculate the combat effectiveness index of both sides in each battle zone, and then calculate the respective losses according to the simulation step size according to the Lanchester square law, and finally count the losses of both sides after the battle.

[0028] Another method is to take a weighted average based on the importance of different equipment entities in the confrontation and the degree of attack of different types of enemy weapons on one's own side during the battle to obtain the battle damage of the aggregate-level model.

[0029] As an improvement of the above system, the position deaggregation includes a polygon deaggregation method and a formation deaggregation method; wherein,

[0030] The polygon deaggregation method comprises: taking the position of the aggregate as the position of the first deaggregated entity, and then taking this as the center, first making a regular N-gon with a side length of A to obtain N vertices, and then making a regular N-gon with a side length of 2A and each side parallel to the corresponding side of the first N-gon to obtain another set of vertices, and setting the deaggregated entities to these vertex positions in turn, and so on until the number of generated entities meets the requirements; in the deaggregation process, it is necessary to consider the constraints of the geographical environment. If the calculated vertex position is not suitable for a certain entity to be deployed at this position, then searching for a nearby suitable area according to the set distance in the upper, lower, left and right areas as the actual position after deaggregation;

[0031] The formation disaggregation method comprises: taking the position of the aggregation level model as the center, calculating the position of the disaggregated entity according to the set distribution mode of the proposed disaggregated formation.

[0032] As an improvement to the above system, the state deaggregation is to use the state of the aggregate as the state of each deaggregated entity.

[0033] As an improvement of the above system, the battle damage deaggregation includes two methods, wherein:

[0034] One method is the uniform distribution method, which is to evenly distribute the battle damage of the aggregate model to each entity model, so that the sum of the battle damage of the entity model is equal to the battle damage of the aggregate model;

[0035] Another method is to make weight corrections based on the attacks on the aggregation-level model and the defense capabilities of the entities that make up the aggregation-level model, and construct a damage ratio coefficient for each entity model. When calculating the deaggregated damage, the damage of the aggregation-level model is multiplied by the damage ratio coefficients of different entity models to obtain the deaggregated damage.

[0036] As an improvement of the above system, the process of data synchronization between the tactical level simulation engine and the campaign level simulation engine also includes a time synchronization method; the time synchronization method includes:

[0037] Adding a tactical-level simulation timestamp to the simulation data released in each cycle by the tactical-level simulation engine;

[0038] The campaign-level simulation engine starts data monitoring, and when it receives a tactical-level simulation timestamp that is consistent with the simulation time, it performs time synchronization and updates the entity status to ensure that the simulation times of the two engines are synchronized.

[0039] As an improvement of the above system, the process of data synchronization between the tactical level simulation engine and the campaign level simulation engine also includes a space synchronization method; the space synchronization method includes:

[0040] The tactical level simulation engine calls the aggregation unit to calculate the position of the center point of the aggregate and publishes the pose data in each simulation cycle;

[0041] The campaign-level simulation engine starts data monitoring, receives and synchronizes posture data, and ensures that the corresponding model space positions of the two engines are consistent.

[0042] As an improvement of the above system, the spatial synchronization method further includes: when data monitoring fails, using a DR algorithm to assist in correcting the position and posture of the unit-level model data;

[0043] The DR algorithm includes: using a first-order algorithm to calculate the spatial posture and using a second-order algorithm to calculate the spatial position; wherein,

[0044] The first-order algorithm includes:

[0045] The first-order extrapolation of the pose of the unit-level model is expressed as:

[0046]

[0047] in, represents the yaw angle calculated in the kth step; represents the pitch angle calculated in the kth step; represents the azimuth calculated in the kth step; Indicates the estimation step length;

[0048] Initial values ​​of yaw, pitch, and azimuth , and According to the rotation speed of the unit-level model along the three coordinate axes in the solid coordinate system , , The calculation results are:

[0049]

[0050] The second-order algorithm includes:

[0051]

[0052] in, , and represents the position of the kth step; , , represents the initial position, , , represents the initial velocity, , , represents the initial acceleration.

[0053] As an improvement of the above system, each module of the system adopts a unified encoding method to encode the physical machine model;

[0054] The code includes type, field, country / region, category, model, system and additional fields.

[0055] Compared with the prior art, the advantages of this application are:

[0056] 1. A model adaptive aggregation and disaggregation method for campaign-level / tactical-level joint simulation is proposed, which realizes the aggregation and disaggregation of the position, state, and damage of multi-resolution models, keeps them consistent in the entities and attributes of the simulation situation, and achieves an effective balance between the real-time and realism of the multi-level heterogeneous joint simulation system.

[0057] 2. Designed a feasible, verifiable and stable heterogeneous interconnection technology solution between campaign-level / tactical-level heterogeneous simulation systems based on the HLA protocol, solving the problem of multi-platform, multi-resolution and large-scene joint simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 The figure shows the architecture of the multi-level force joint simulation system based on adaptive aggregation and deaggregation;

[0059] Figure 2 Shown is a tree-like variable network diagram;

[0060] Figure 3 Shown is the technology roadmap for adaptive aggregation and disaggregation methods;

[0061] Figure 4 The figure shows a multi-level heterogeneous interconnection framework diagram based on the HLA protocol;

[0062] Figure 5 Shown is a diagram of a time synchronization model;

[0063] Figure 6 Shown is a diagram of a position synchronization model;

[0064] Figure 7 The figure shows the effect of maneuvering along the route when both parties are aggregates in the initial state;

[0065] Figure 8 The figure shows the confrontation effect of automatically deaggregating into entity objects in the confrontation state;

[0066] Fig. 9 The figure shows the automatic aggregation effect of the red side in the final state. DETAILED DESCRIPTION

[0067] The technical solution of the present application is described in detail below with reference to the accompanying drawings.

[0068] The present application provides a multi-level force joint simulation system based on adaptive aggregation and deaggregation, whose functions include:

[0069] 1. A model adaptive aggregation and disaggregation method for campaign-level / tactical-level joint simulation is proposed, which realizes position disaggregation strongly related to terrain and confrontation status, state disaggregation based on statistical models, and battle damage disaggregation based on exchange records, providing a means for the mutual transformation of multi-resolution models;

[0070] 2. Designed a feasible, verifiable, and stable heterogeneous interconnection technology solution between campaign-level / tactical-level heterogeneous simulation systems based on the HLA protocol, solving the problem of multi-platform, multi-resolution, and large-scale joint simulation;

[0071] 3. A multi-level and multi-granular simulation strategy and system has been constructed, so that the confrontation situation is handled by the campaign-level simulation, the decomposed key actions or key nodes are taken over by the tactical-level simulation and the simulation results are fed back to the campaign-level simulation, so as to achieve the goal of large-scale and high-precision joint simulation of campaign-level and tactical-level forces.

[0072] The present application provides a multi-level force joint simulation system based on adaptive aggregation and deaggregation. It uses the multi-level heterogeneous interconnection technology based on the HLA protocol to perform data interaction between simulation engines at different simulation levels such as the tactical level and the campaign level. To address the problem of inconsistent simulation model granularity, it uses a model adaptive aggregation and deaggregation method for campaign-level / tactical-level joint simulation to perform aggregation and deaggregation to ensure the adaptation of multi-resolution models, thereby meeting the joint deduction of simulation engines at different levels, ensuring an effective balance between the real-time and realism of multi-platform, multi-resolution, and large-scene simulation systems, and achieving the purpose of large-scale, high-precision joint simulation of campaign-level and tactical forces. The system structure is as follows: Figure 1 As shown, it includes tactical-level simulation engine, campaign-level simulation engine, aggregation and deaggregation module and heterogeneous interconnection module.

[0073] The tactical-level simulation engine is used to load simulation assumptions to construct simulation scenarios, and to conduct entity-level simulation deductions based on simulation control parameters such as simulation period and step size and by calling multi-resolution models.

[0074] The campaign-level simulation engine is used to load simulation assumptions to build simulation scenarios, receive simulation control parameters sent by the tactical-level simulation engine, and call multi-resolution models to carry out unit-level simulation deductions.

[0075] The aggregation and deaggregation module is used to convert and interact between models of different resolution levels, such as unit-level models, unit-level models with smaller granularity, or entity-level models, and includes deaggregation and aggregation functions. The deaggregation function is to convert from a unit-level model to a unit-level model with smaller granularity or an entity-level model; the aggregation function is to convert from an entity-level model to a unit-level model.

[0076] The heterogeneous interconnection module provides a mechanism and environment for each simulation heterogeneous object to access the system and publish and subscribe to data based on the HLA protocol, and realizes data interconnection and model interoperability between simulation objects through the OpenDDS message middleware.

[0077] In joint simulation, the interaction between multi-resolution models and the interconnection between heterogeneous systems are core tasks with certain difficulties. Therefore, this application mainly involves two key technologies: a model adaptive aggregation and deaggregation method for campaign-level / tactical-level joint simulation, and a multi-level heterogeneous interconnection technology based on the HLA protocol.

[0078] 1. Model adaptive aggregation and disaggregation method for campaign-level / tactical-level joint simulation

[0079] Aggregation and disaggregation is a technology for converting between a unit-level model and a smaller-granularity unit-level model or entity-level model. Its essence is the conversion between models of different resolution levels, completing the interaction between models of different resolutions. Disaggregation is the conversion from a unit-level model to a smaller-granularity unit-level model or entity-level model. Aggregation is the reverse of disaggregation, which is the conversion from an entity-level model to a unit-level model. In the field of multi-resolution modeling, there is a generally recognized basic principle for model interaction, that is, when there is an interactive relationship between models, the interacting entity models are required to have matching resolution levels. According to the basic principle of model interaction, the interaction between models at different levels must go through an aggregation or disaggregation process. In order to facilitate the analysis of the principles of aggregation and disaggregation from a mathematical perspective, a tree-like variable network diagram is used to describe the aggregation process, such as Figure 2 shown.

[0080] Figure 2 In , let Y be the final output after aggregation, and the aggregation function is F(X); , , are variables describing the low-resolution model, and the aggregation functions are (X), (X), (X); … are variables describing the high-resolution model.

[0081] According to the aggregation relationship of variables, the low-resolution expression of Y can be obtained:

[0082] Y=F( , , ) (1)

[0083] The relationship between high and low resolution variables is:

[0084] = ( )

[0085] = ( )

[0086] = ( ) (2)

[0087] Substituting (2) into (1), we can obtain the high-resolution expression of Y:

[0088] Y = F( ( ), ( ), ( )) (3)

[0089] By comparing formula (1) and (3), it can be clearly seen that the high-resolution variable expression contains more information than the low-resolution variable expression. When all high-resolution variables are assigned values, each low-resolution variable can obtain a unique value according to formula (2). But conversely, when all low-resolution variables are assigned values, many sets of high-resolution values ​​that satisfy formula (2) can be obtained. This shows that many high-resolution states only correspond to one low-resolution state.

[0090] Adaptive aggregation and disaggregation requires that in the process of confrontation simulation, the triggering conditions of aggregation and disaggregation are automatically detected in real time, and the mutual transformation between the unit-level model and the entity-level model is completed according to the aggregation and disaggregation rules. The disaggregation rules of the model in different states are usually inconsistent. For example, the formation of the infantry company in the offensive state is generally an equilateral triangle, and the formation of the infantry company in the defensive state is generally an inverted triangle. The type of entity-level model will also affect the position disaggregation of the model. For example, after the tank company is disaggregated, the spacing between tanks is significantly greater than the spacing between soldiers after the infantry company is disaggregated. In addition, the terrain environment will also affect the disaggregation rules. The position and state of the entity model after disaggregation must comply with military regulations.

[0091] The technical route of the model adaptive aggregation and disaggregation method for campaign-level / tactical-level joint simulation is as follows: Figure 3 As shown, it includes position disaggregation that is strongly related to terrain and confrontation status, state disaggregation based on statistical models, and battle damage disaggregation based on exchange of fire records.

[0092] 1) Aggregation and disaggregation of locations that are strongly related to terrain and confrontation status

[0093] Aggregation and disaggregation of entity positions is the most common situation in the aggregation and disaggregation of adversarial simulation models. In simulation, it is often necessary to decompose a unit model composed of multiple entity models into a single entity, and it is also common to batch several entities into a formation to perform a certain task. During the conversion process, how to determine the entity position is the problem that position aggregation and disaggregation needs to solve.

[0094] (1) Location aggregation

[0095] Assumptions , , ... , , is n (n 2) high-resolution solid models, whose positions in the geocentric coordinate system are ( , ), , ),…,( , ), E is , , ... , The position of the aggregate formed by aggregation in the geocentric coordinate system is (x, y, z). The position of the aggregate E can be obtained according to the following different aggregation methods.

[0096] Geometric mean method:

[0097] The geometric mean method takes the geometric center of each high-resolution entity that makes up the aggregate as the position of the aggregate, that is,

[0098] , ,

[0099] Weighted average method:

[0100] The weighted average method sets the position of the aggregate at the weighted center of each high-resolution entity. When the entity object is generated, the weight of each high-resolution entity is set according to its role level in the confrontation or the pre-configuration method. , , ... , The weights in the simulation confrontation are , , ... , , then the position of the aggregate obtained by weighted average method is:

[0101] ,

[0102] Core positioning method:

[0103] The core positioning method is to use the position of the entity that is responsible for the main confrontation task as the position of the aggregate among all the high-resolution entities that make up the aggregate. The entity that is responsible for the main confrontation task may be one or more. If there is one, its position is directly used as the position of the aggregate; if there are more than one, the position of the sub-aggregate composed of these entities is first calculated using the geometric mean method, and then this position is used as the position of the entire aggregate. Assume that the entity , , ... , middle , , ... , ( ) is the entity responsible for the main confrontation task, and its positions in the geocentric coordinate system are ( , ),( , ),…,( , ) then the position of the polymer can be obtained according to the core positioning method:

[0104] , ,

[0105] The determination of the entity responsible for the main confrontation task in the core positioning method is based on a series of rules pre-established based on confrontation knowledge to constrain, such as the degree of association between entity type and confrontation task, etc. For example, if the aggregation level model performs a ground assault task, and the entity models that make up the aggregation level model include fighter entities and bomber entities, then the bomber entity can be regarded as the entity responsible for the main task, and other entities are auxiliary entities. In practical applications, it is necessary to select the appropriate core entity according to the specific application.

[0106] (2) Position disaggregation

[0107] Disaggregation of entity positions requires comprehensive consideration of multiple constraints, such as separation distance, area, and battle formation. Among them, separation distance refers to the spacing between entities after disaggregation, which can be divided into minimum separation distance and maximum limit distance. The minimum separation distance means that the distance between two entities cannot be less than this distance, and the main purpose is to ensure maneuverability and safety as much as possible; the maximum limit distance means that the distance between two entities cannot be greater than this distance, and the main purpose is to prevent the formation from being too dispersed. Area refers to the area that cannot be exceeded when the aggregation-level entities are disaggregated in the specified area. Battle formation refers to the formation that the opposing entities should maintain.

[0108] The present application provides a hexagonal disaggregation algorithm, the basic idea of ​​which is to take the position of the aggregate as the position of the first disaggregated entity, and then make a regular hexagon with a side length of A as the center, obtain six vertices, and then make a regular hexagon with a side length of 2A and each side parallel to the corresponding side of the first hexagon, and obtain another set of vertices, and set the disaggregated entities to these vertex positions in turn, and so on until the number of entities generated meets the requirements. In the disaggregation process, it is necessary to consider the constraints of the geographical environment. If the calculated vertex position is not suitable for a certain entity to be deployed at this position, then the nearby suitable area can be searched according to the upper, lower, left and right areas according to a certain distance as the actual position after disaggregation. In other embodiments, the six deformations can also be replaced by other polygons. In addition, the calculation method based on the disaggregation of formations (such as horizontal formations, echelons, wedge formations, vertical formations, snake formations, etc.) is similar to the hexagonal disaggregation algorithm. In the fixed coordinate system of the earth, the position of the disaggregated entity is calculated according to the prefabricated distribution method of the proposed disaggregation formation with the position of the aggregate level model as the center.

[0109] 2) State aggregation and disaggregation based on statistical models

[0110] The aggregation and deaggregation of entity states mainly represent the mapping process of state definitions at different abstract levels. When deaggregating, the number of states of the model will increase to improve the model description resolution. When aggregating, the states need to be compressed and aggregated to reduce the number of states of the model.

[0111] (1) State aggregation

[0112] Entity state aggregation is the process of aggregating several high-resolution entities into a low-resolution aggregate. For example, each high-resolution aircraft entity has N possible states (such as takeoff, navigation, turning, air combat, etc.). According to the theory of permutations and combinations, the states of the four aircraft after being combined will be There are many possibilities, and the aggregated entity can only be in one state at the same time, which requires state aggregation. It can be seen that there is a considerable loss of state information after aggregation.

[0113] Since the states of different entity types vary greatly and the behaviors of entities also vary greatly, it is difficult to describe, control and operate them after different entity types are aggregated. Therefore, an agreement is made on the aggregation of entity states: state aggregation can only be performed between entities of the same type.

[0114] From the perspective of whether the states of the entities participating in the aggregation are the same, the aggregation of entity states includes: aggregation of the same state and aggregation of different states. Aggregation of the same state means that the entities participating in the aggregation are in the same state under the same task. This type of aggregation is relatively simple, directly taking the states of the entities participating in the aggregation as the state of the aggregate, that is, assuming that n entities , , ... , The status are , , ... , , E is , , ... , The state of the aggregate is S. Obviously: S= = =…= Heterogeneous aggregation refers to the different states of the entities involved in the aggregation. This form is relatively complex and requires additional rules and constraints to determine the state of the aggregate. The constraint rules are generally pre-specified by military experts based on the confrontation task. Here are two of the most common methods for determining the state of aggregate entities: the majority entity state determination method and the key entity state determination method.

[0115] a) Majority Entity State Decision

[0116] The majority entity state determination method takes the state of the majority of entities participating in the aggregation as the state of the aggregate. The idea is: suppose n entities , , ... , The status are , , ... , , where the status is (1≤i≤n) has the largest number of entities, and E is , , ... , The aggregate entity formed by aggregation has a state of S, then S= .

[0117] b) Key Entity Status Determination Method

[0118] The key entity state determination method is to take the state of the key entity among the entities participating in the aggregation as the state of the aggregate. The idea is: suppose n entities , , ... , The status are , , ... , , where the key entity The state of (1≤i ≤n) is , E is , , ... , The aggregate entity formed by aggregation has a state of S, then S= The determination of key entities is similar to the determination of core entities in the core positioning method. It is constrained by a series of rules pre-established based on adversarial knowledge, such as the degree of association between entity types and adversarial tasks.

[0119] (2) State disaggregation

[0120] Entity state deaggregation is how to determine the states of each entity after deaggregation based on the state of the aggregate. Since there is only one state in the aggregate-level model in the confrontation task, but each entity after deaggregation may have different states, it is necessary to combine the specific conditions such as the task and location of the deaggregated entity and a large number of military rules as constraints, which is very complicated. Therefore, this application uses the aggregate state replacement method to simplify the processing, that is, directly taking the state of the aggregate as the state of each deaggregated entity.

[0121] 3) Battle damage aggregation and deaggregation based on fire records

[0122] Battle damage is mainly used to describe the confrontation results between the two opposing parties. In confrontation simulation, battle damage aggregation and disaggregation include aggregation and disaggregation of aggregation-level models composed of entities of the same type and aggregation and disaggregation of aggregation-level models composed of entities of different types.

[0123] (1) Battle Damage Aggregation

[0124] For the aggregation of battle damage of entity models of the same type, the battle damage of the aggregated entities is directly added up according to the aggregation relationship. For the aggregation of battle damage of entity models of different types, this application provides two methods: one is to divide the battle area into several battle zones, calculate the combat effectiveness index of both sides in each battle zone, and then calculate the respective battle damage according to the simulation step length according to the Lanchester square law, and finally calculate the battle damage of both sides after the battle; the other is to perform weighted averaging based on the importance of different equipment entities in the confrontation and the degree of attack of different types of enemy weapons on the own side during the battle, and obtain the battle damage of the aggregation-level model.

[0125] (2) Battle Damage Decomposition

[0126] Battle damage deaggregation is how to deaggregate the battle damage of the aggregation-level model into the battle damage of each entity during the confrontation process. This application provides two methods: one is the uniform distribution method, which evenly distributes the battle damage of the aggregation-level model to each entity model, so that the sum of the battle damage of the entity model is equal to the battle damage of the aggregation-level model; the other is to perform weight correction based on the attack situation of the aggregation-level model and the defense capability of the entities that constitute the aggregation-level model, and construct a battle damage ratio coefficient for each entity model. When calculating the deaggregated battle damage, the battle damage of the aggregation-level model is multiplied by the battle damage ratio coefficient of different entity models to obtain the deaggregated battle damage.

[0127] 2. Multi-level heterogeneous interconnection technology based on HLA protocol

[0128] like Figure 4 As shown in the figure, based on the HLA specification, the publish-subscribe mechanism and the central node method are adopted to connect the heterogeneous objects participating in the simulation to the system in the form of a federation, and the OpenDDS message middleware is used to realize the efficient interconnection and interoperability of data between simulation clusters. During the simulation process, each heterogeneous object subscribes to the data items of their concern; when the entity state changes in the tactical-level granularity environment, the change result is synchronized to the campaign-level granularity environment through the OpenDDS message middleware, and the data is updated to the corresponding aggregation-level model, and vice versa. In order to ensure the consistency and correctness of the object state between heterogeneous systems, the following technologies need to be implemented:

[0129] 1) Time synchronization technology

[0130] Time synchronization is a unified time advancement mechanism to ensure that the various components in the heterogeneous simulation system can operate in a coordinated manner and present the simulation results with the correct cause-effect relationship. This application adds a timestamp field to the interaction protocol between the campaign-level environment and the tactical-level environment; based on the tactical-level high-precision model, synchronize the time to the campaign-level environment, thereby achieving synchronous processing of entity states at each moment. The process is as follows: Figure 5 shown.

[0131] The specific steps are as follows:

[0132] ① Start the joint simulation and deduction system, first running tactical-level and campaign-level deductions;

[0133] ② Add tactical-level simulation timestamps to the simulation data released in each cycle of tactical-level simulation;

[0134] ③ Data monitoring is enabled in campaign-level simulations. When a tactical-level simulation timestamp consistent with the simulation time is received, time synchronization is performed and the entity status is updated to ensure time synchronization between the two simulation systems.

[0135] 2) Space synchronization technology

[0136] Spatial synchronization is to solve the problem of inconsistent spatial processing between heterogeneous objects due to the different description methods of the spatial position and posture of the simulation entity. Spatial synchronization requires that during the simulation process, the position of the same simulation entity on different platforms is consistent at the same simulation time. Therefore, not only the coordinate position of the simulation entity is updated in real time through the heterogeneous interconnected system, but also the DR (Dead Reckoning) algorithm is used to assist in correcting the position of the simulation entity.

[0137] (1) Spatial information synchronization

[0138] This application adds a location field to the interactive protocol between the campaign-level environment and the tactical-level environment; based on the tactical-level high-precision map, the entity's latitude, longitude, elevation and other location information are synchronized to the campaign-level environment, thereby realizing the synchronization of entity locations at all times. The process is as follows: Figure 6 shown.

[0139] The specific steps are as follows:

[0140] ① Start the joint simulation and deduction system to run tactical-level and campaign-level deductions;

[0141] ② Collect the position data of tactical-level simulation entities in real time, use the aggregation algorithm to calculate the center point position of the aggregate, and publish the pose data in each simulation cycle;

[0142] ③ Enable data monitoring in campaign-level simulations to receive and synchronize posture data to ensure that the corresponding model spatial positions of the two simulation systems are consistent.

[0143] 3) DR algorithm-assisted correction

[0144] In the campaign-level simulation, the unit-level model data is aggregated based on the corresponding entity-level model data in the tactical-level simulation. When data monitoring fails, the unit-level model data is corrected using the DR algorithm. The DR algorithm recursively infers the current position and attitude based on the position and velocity at the previous moment. In the simulation, the first-order algorithm is used to calculate the spatial attitude according to the simulation step size, and the second-order algorithm is used to calculate the spatial position, and the simulation result is used to correct the spatial information synchronization result.

[0145] a) First-order extrapolation algorithm

[0146] The formula for calculating the spatial position is:

[0147]

[0148] In the formula, T is the calculation step size (usually the simulation step size).

[0149] In order to calculate the attitude angle, it is necessary to calculate the rotation speed of the unit-level model along the three coordinate axes in the physical coordinate system. , , , calculate its spatial attitude angle relative to the ground coordinate system: yaw angle Ψ, pitch angle , azimuth . To do this, perform the coordinate transformation:

[0150]

[0151] Therefore, the first-order extrapolation formula of the three attitude angles is:

[0152]

[0153] 2) The second-order extrapolation algorithm of spatial position is expressed as

[0154]

[0155] in, , , is the initial position, , , is the initial velocity, , , is the initial acceleration, is the estimated step size (usually the simulation step size), To calculate the number of steps.

[0156] 3) Entity-level model interoperability technology

[0157] The models of heterogeneous systems are managed uniformly and accessed through the data bus. In order to meet the consistency and interoperability of entity-level models under different simulation platforms, it is necessary to build a unique identifier for each entity-level model or the entity-level model after the unit-level model is deaggregated. This paper adopts the seven-segment code definition to ensure the uniqueness and interoperability of models between different simulation platforms. The seven-segment code consists of type, field, country / region, category, model, series and an additional seven fields, as shown in Table 1.

[0158] Table 1 Seven-segment code design

[0159]

[0160] Through the design of the seven-segment code, each type of model has a unique identifier, thus forming a model list consisting of seven-segment codes. Heterogeneous systems can query and call the corresponding models conveniently and quickly based on the list.

[0161] A simple red-blue confrontation scenario is constructed using a multi-level force joint simulation system based on adaptive aggregation and disaggregation to verify the aggregation and disaggregation capabilities. The green box represents the confrontation area, where the two opposing sides will conduct adaptive aggregation and disaggregation. The red icon represents the red equipment and the blue icon represents the blue equipment. Before disaggregation, it is an aggregate and after disaggregation, it is a physical equipment.

[0162] like Figure 7 As shown, the red side attacks from below and the blue side attacks from above. In the initial state, neither side has completely entered the attack area and is in an aggregate state.

[0163] like Figure 8 As shown in the figure, when the red side enters the attack area and starts to attack, it is adaptively disaggregated into 4 tank entity objects, while the blue side automatically disaggregates into 3 tank entity objects due to being attacked.

[0164] like Fig. 9 As shown, after the confrontation ends, the red side continues to maneuver upwards and automatically aggregates into an aggregate after leaving the confrontation area, while the blue side remains in the final state of the confrontation because all of them have been eliminated and will no longer automatically aggregate into an aggregate.

[0165] The present application may also provide a computer device, comprising: at least one processor, a memory, at least one network interface and a user interface. The various components in the device are coupled together through a bus system. It is understood that the bus system is used to achieve connection and communication between these components. In addition to the data bus, the bus system also includes a power bus, a control bus and a status signal bus.

[0166] The user interface may include a display, a keyboard or a pointing device, such as a mouse, a trackball, a touch pad or a touch screen.

[0167] It is understood that the memory in the disclosed embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus random access memory (DRRAM). The memories described herein are intended to include, but are not limited to, these and any other suitable types of memories.

[0168] In some embodiments, the memory stores the following elements, executable modules or data structures, or a subset thereof, or an extended set thereof: an operating system and applications.

[0169] The operating system includes various system programs, such as a framework layer, a core library layer, a driver layer, etc., which are used to implement various basic services and process hardware-based tasks. The application includes various application programs, such as a media player (Media Player), a browser (Browser), etc., which are used to implement various application services. The program for implementing the method of the embodiment of the present disclosure can be included in the application.

[0170] In the above embodiment, the processor may also call a program or instruction stored in the memory, specifically, a program or instruction stored in an application program, and is used to:

[0171] Execute the steps of the above method.

[0172] The above method can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processor can be a general processor, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field programmable gate array (Field Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The above-disclosed methods, steps and logic block diagrams can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the above-disclosed method can be directly embodied as a hardware decoding processor to execute, or the hardware and software modules in the decoding processor can be combined to execute. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0173] It is understood that the embodiments described in the present application can be implemented by hardware, software, firmware, middleware, microcode or a combination thereof. For hardware implementation, the processing unit can be implemented in one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), general-purpose processors, controllers, microcontrollers, microprocessors, other electronic units for performing the functions described in the present application or a combination thereof.

[0174] For software implementation, the technology of the present application can be implemented by executing the functional modules (such as procedures, functions, etc.) of the present application. The software code can be stored in a memory and executed by a processor. The memory can be implemented in the processor or outside the processor.

[0175] The present application may also provide a non-volatile storage medium for storing a computer program. When the computer program is executed by a processor, each step in the above method embodiment can be implemented.

[0176] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present application and are not intended to limit it. Although the present application is described in detail with reference to the embodiments, a person skilled in the art should understand that any modification or equivalent replacement of the technical solution of the present application does not depart from the spirit and scope of the technical solution of the present application and should be included in the scope of the claims of the present application.

Claims

1. A multi-level force joint simulation system based on adaptive aggregation and deaggregation, characterized in that: The system comprises: Tactical-level simulation engine, used to load simulation scenarios to build simulation scenarios, and conduct entity-level simulation deduction according to simulation period and step-size simulation control parameters; The campaign-level simulation engine is used to load simulation assumptions to build simulation scenarios, and receive simulation control parameters sent by the tactical-level simulation engine to carry out unit-level simulation deduction; The aggregation and deaggregation module is used for mutual conversion and interaction between models of different resolution levels, including deaggregation units and aggregation units; wherein, The deaggregation unit is used to convert the unit-level model into a unit-level model or entity-level model with a smaller granularity; Aggregation units, used to transform from entity-level models to unit-level models; and The heterogeneous interconnection module is used to provide a mechanism and environment for each simulation heterogeneous object to access the system and publish and subscribe to data, and to achieve data interconnection and model interoperability between simulation objects through message middleware.

2. The multi-level force joint simulation system based on adaptive aggregation and deaggregation according to claim 1 is characterized in that: The aggregation methods implemented by the aggregation unit include: position aggregation, state aggregation and battle damage aggregation; the deaggregation methods implemented by the deaggregation unit include: position deaggregation, state deaggregation and battle damage deaggregation.

3. The multi-level force joint simulation system based on adaptive aggregation and deaggregation according to claim 2 is characterized in that: The position aggregation includes geometric mean method, weighted mean method and core positioning method; The geometric mean method is to use the geometric center of each high-resolution entity constituting the aggregate as the position of the aggregate; The weighted average method is to set the position of the aggregate at the weighted center of each high-resolution entity; The core positioning method takes the position of the entity that undertakes the main confrontation task among all high-resolution entities that constitute the aggregate as the position of the aggregate; when there are multiple entities that undertake the main confrontation task, the geometric center of the multiple entities that undertake the main confrontation task is taken as the position of the aggregate.

4. The multi-level force joint simulation system based on adaptive aggregation and deaggregation according to claim 2 is characterized in that: The state aggregation includes same-state aggregation and different-state aggregation; The same-state aggregation is to use the states of the entities participating in the aggregation as the state of the aggregate; The heterogeneous state aggregation includes a majority entity state determination method and a key entity state determination method; wherein, The majority entity state determination method is to take the state of the majority of entities participating in the aggregation as the state of the aggregate; The key entity state determination method is to use the state of the key entity among the entities participating in the aggregation as the state of the aggregate.

5. The multi-level force joint simulation system based on adaptive aggregation and deaggregation according to claim 2 is characterized in that: The battle damage aggregation includes battle damage aggregation of entity models of the same type and battle damage aggregation of entity models of different types; wherein, The battle damage aggregation of the entity models of the same type is to add up the battle damage of the aggregated entities to obtain the aggregated battle damage; The battle damage aggregation of different types of entity models includes two methods: One method is to divide the battle area into several battle zones, calculate the combat effectiveness index of both sides in each battle zone, and then calculate the respective losses according to the simulation step size according to the Lanchester square law, and finally count the losses of both sides after the battle. Another method is to take a weighted average based on the importance of different equipment entities in the confrontation and the degree of attack of different types of enemy weapons on one's own side during the battle to obtain the battle damage of the aggregate-level model.

6. The multi-level force joint simulation system based on adaptive aggregation and deaggregation according to claim 2 is characterized in that: The position disaggregation includes a polygon disaggregation method and a formation disaggregation method; wherein, The polygon deaggregation method comprises: taking the position of the aggregate as the position of the first deaggregated entity, and then taking this as the center, first making a regular N-gon with a side length of A to obtain N vertices, and then making a regular N-gon with a side length of 2A and each side parallel to the corresponding side of the first N-gon to obtain another set of vertices, and setting the deaggregated entities to these vertex positions in turn, and so on until the number of generated entities meets the requirements; in the deaggregation process, it is necessary to consider the constraints of the geographical environment. If the calculated vertex position is not suitable for a certain entity to be deployed at this position, then searching for a nearby suitable area according to the set distance in the upper, lower, left and right areas as the actual position after deaggregation; The formation disaggregation method comprises: taking the position of the aggregation level model as the center, calculating the position of the disaggregated entity according to the set distribution mode of the proposed disaggregated formation.

7. The multi-level force joint simulation system based on adaptive aggregation and deaggregation according to claim 2 is characterized in that: The state deaggregation is to use the state of the aggregate as the state of each deaggregated entity.

8. The multi-level force joint simulation system based on adaptive aggregation and deaggregation according to claim 2 is characterized in that: The battle damage deaggregation includes two methods, among which: One method is the uniform distribution method, which is to evenly distribute the battle damage of the aggregate model to each entity model, so that the sum of the battle damage of the entity model is equal to the battle damage of the aggregate model; Another method is to make weight corrections based on the attacks on the aggregation-level model and the defense capabilities of the entities that make up the aggregation-level model, and construct a damage ratio coefficient for each entity model. When calculating the deaggregated damage, the damage of the aggregation-level model is multiplied by the damage ratio coefficients of different entity models to obtain the deaggregated damage.

9. The multi-level force joint simulation system based on adaptive aggregation and deaggregation according to claim 1 is characterized in that: The process of data synchronization between the tactical level simulation engine and the campaign level simulation engine also includes a time synchronization method; the time synchronization method includes: Adding a tactical-level simulation timestamp to the simulation data released in each cycle by the tactical-level simulation engine; The campaign-level simulation engine starts data monitoring, and when it receives a tactical-level simulation timestamp that is consistent with the simulation time, it performs time synchronization and updates the entity status to ensure that the simulation times of the two engines are synchronized.

10. The multi-level force joint simulation system based on adaptive aggregation and deaggregation according to claim 1 is characterized in that: The process of data synchronization between the tactical level simulation engine and the campaign level simulation engine also includes a space synchronization method; the space synchronization method includes: The tactical level simulation engine calls the aggregation unit to calculate the position of the center point of the aggregate and publishes the pose data in each simulation cycle; The campaign-level simulation engine starts data monitoring, receives and synchronizes posture data, and ensures that the corresponding model space positions of the two engines are consistent.

11. The multi-level force joint simulation system based on adaptive aggregation and deaggregation according to claim 10 is characterized in that: The spatial synchronization method further comprises: when data monitoring fails, using a DR algorithm to assist in correcting the position and posture of the unit-level model data; The DR algorithm includes: using a first-order algorithm to calculate the spatial posture and using a second-order algorithm to calculate the spatial position; wherein, The first-order algorithm includes: The first-order extrapolation of the pose of the unit-level model is expressed as: ； in, represents the yaw angle calculated in the kth step; represents the pitch angle calculated in the kth step; represents the azimuth calculated in the kth step; Indicates the estimation step length; Initial values ​​of yaw, pitch, and azimuth , and According to the rotation speed of the unit-level model along the three coordinate axes in the solid coordinate system , , The calculation results are: ； The second-order algorithm includes: ； in, , and represents the position of the kth step; , , represents the initial position, , , represents the initial velocity, , , represents the initial acceleration.

12. The multi-level force joint simulation system based on adaptive aggregation and deaggregation according to claim 1 is characterized in that: Each module of the system uses a unified coding method to encode the physical machine model; The code includes type, field, country / region, category, model, system and additional fields.

Images