A construction method and device of a distributed LVC simulation system and a storage medium

Abstract

The application relates to the field of simulation technology, in particular to a construction method and device of a distributed LVC simulation system and a storage medium. The distributed LVC simulation system comprises a digital simulation system, a semi-physical simulation system and a flight simulator simulation system. The method comprises the following steps: dividing and merging simulation resources in each subsystem based on a preset splitting granularity, so as to obtain a plurality of component models with independent simulation functions; selecting target component models from the plurality of component models based on simulation task requirements, and combining the target component models, so as to construct a plurality of simulation entities; for an external target component model not belonging to the digital simulation system, creating a corresponding agent model in the digital simulation system; the agent model is used for bidirectional state synchronization with the corresponding external target component model in the simulation running process, and represents the external target component model to interact with the digital simulation system. The flexibility and expandability of the simulation are improved.

Description

Technical Field

[0001] This application relates to the field of simulation technology, and in particular to a method, device and storage medium for constructing a distributed LVC simulation system. Background Technology

[0002] LVC simulation is a distributed simulation technology that integrates physical (L, Live), virtual (V, Virtual), and constructive (C, Constructive) elements into a single system. It serves as a crucial supporting method and tool for military simulation technology. "Physical" refers to real personnel operating real equipment in a combat or near-combat environment, which can be achieved through a hardware-in-the-loop simulation system. "Virtual" refers to real personnel operating virtual equipment, with typical applications including flight simulators, ship operation simulators, and other simulator systems. "Constructive" refers to computer-generated virtual characters and rules that enable virtual personnel to operate virtual equipment within a simulation environment, such as various digital simulation systems.

[0003] LVC simulation not only enhances the realism, complexity, and coverage of military exercises and training, but also effectively reduces the high costs, security risks, and resource consumption associated with live-fire training. It is particularly suitable for simulation, verification, and capability assessment in emerging combat domains such as joint operations, network attack and defense, and intelligent confrontation.

[0004] For example, the invention patent application with publication number CN114861379A discloses a military training system based on LVC simulation. By constructing a heterogeneous system interconnection environment, it can support interconnection, interoperability, and interoperability with various systems such as live-fire simulation, virtual simulation, and structural simulation. It also establishes a heterogeneous system interface protocol system to support multi-interface access and access to multiple communication protocols (such as HLA, DDS, etc.), enabling network communication between independent subsystems and realizing a simulation system with multi-interface integration. Furthermore, it satisfies the multi-source data fusion of heterogeneous systems, realizes the integration of multiple data sources, and achieves data and resource sharing and reuse.

[0005] However, the heterogeneity of subsystems in LVC simulation makes cross-domain and cross-level interactions difficult, hindering the implementation of complex joint simulation tasks and severely limiting its practicality. Summary of the Invention

[0006] The main objective of this application is to provide a method, device, and storage medium for constructing a distributed LVC simulation system. To address the aforementioned technical problems, this application specifically adopts the following technical solution:

[0007] The first aspect of this application is to provide a method for constructing a distributed LVC simulation system, wherein the distributed LVC simulation system includes three subsystems: a digital simulation system, a hardware-in-the-loop simulation system, and a flight simulator simulation system; the method includes:

[0008] S101, obtain simulation resources from each subsystem;

[0009] S102, based on the preset splitting granularity, the simulation resources in each subsystem are divided and merged to obtain multiple component models with independent simulation functions;

[0010] S103, based on the simulation task requirements, select target component models from multiple component models and combine them to construct multiple simulation entities; wherein, each simulation entity is composed of target component models from a single subsystem, or is composed of a mixture of target component models from any two or three subsystems;

[0011] S104. For external target component models that do not belong to the digital simulation system, create corresponding proxy models in the digital simulation system.

[0012] The digital simulation system is used to provide simulation resources and perform global simulation deduction; the proxy model is used to perform bidirectional state synchronization with the corresponding external target component model during the simulation operation, and to interact with the digital simulation system on behalf of the external target component model.

[0013] A second aspect of this application is to provide a computer device, the device comprising:

[0014] Memory, used to store computer programs;

[0015] A processor is configured to execute the computer program and, in executing the computer program, implement the steps of the method for constructing a distributed LVC simulation system as provided in any embodiment of this application.

[0016] A third aspect of this application is that a computer-readable storage medium is also provided, the computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the construction method of a distributed LVC simulation system provided in any embodiment of this application.

[0017] Beneficial effects:

[0018] This application provides a method, device, and storage medium for constructing a distributed LVC simulation system. Through unified-granularity resource componentization, it enables flexible combination of simulation entities across subsystems as needed. Simultaneously, it introduces a proxy model as a lightweight intermediary between the digital simulation system and external heterogeneous resources, transforming the complex and diverse cross-system interactions arising from highly flexible combination into unified and structured standard communication within the digital simulation system. Therefore, while improving the flexibility and scalability of LVC simulation, it effectively reduces communication overhead and enhances simulation stability and operational efficiency.

[0019] By dividing the simulation resources of the three subsystems—digital, hardware-in-the-loop, and flight simulator—into functionally independent component models with a unified granularity, a high degree of modularization and standardization of heterogeneous simulation resources is achieved. Based on this, when constructing specific simulation tasks, components can be freely selected and mixed across subsystems according to accuracy, cost, or functional requirements. For example, a simulated aircraft can partially utilize high-fidelity hardware-in-the-loop devices, partially utilize convenient digital models, and can also incorporate human-in-the-loop simulator operation.

[0020] The highly flexible combination of simulation entities enables a high degree of simulation customization capability, improves the breadth and depth of simulation, enhances the system's adaptability to diverse application scenarios and resource utilization efficiency, and further complicates the interaction between heterogeneous systems during the simulation process, which to some extent increases the communication burden of data synchronization.

[0021] By introducing a proxy model mechanism, the state of external target component models is mirrored and synchronized through the parameter set of the proxy model. This transforms the original multi-protocol, multi-interface, point-to-point communication between the digital simulation system and various heterogeneous external systems into a unified, structured, centralized interaction between the digital simulation system and only the internal proxy model. In this way, the protocol differences, data formats, and communication logic of external systems are encapsulated within the proxy model, simplifying communication logic and reducing the complexity and communication overhead of system interaction. Furthermore, through differentiated data synchronization strategies, the real-time performance of key simulation data is ensured while effectively reducing the network transmission of unnecessary data, thus lowering the communication load.

[0022] Meanwhile, the digital simulation system can better maintain and manage the simulation process based on the proxy model. For example, when an external target component model fails due to communication interruption or functional abnormality, the latest complete parameters stored in the proxy model can be used to quickly locate and initialize a digital model with similar functions to take over the simulation task, thereby ensuring the continuous and stable operation of the global simulation and effectively reducing the risk of task interruption caused by the access of multiple external devices. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. The elements or parts in the drawings are not necessarily drawn to scale. Obviously, the drawings described below are some embodiments of this application; for those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0024] Figure 1 This is a schematic diagram of the architecture of a distributed LVC simulation system provided in an embodiment of this application;

[0025] Figure 2 This is a schematic flowchart illustrating a method for constructing a distributed LVC simulation system provided in an embodiment of this application;

[0026] Figure 3 This is a schematic diagram illustrating the distribution of component models under different functions provided in an embodiment of this application;

[0027] Figure 4 This is a schematic diagram illustrating the functionality of a digital agent model provided in an embodiment of this application;

[0028] Figure 5 This is a schematic diagram of a co-simulation scenario of an LVC system provided in an embodiment of this application;

[0029] Figure 6 This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0031] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0032] In this document, suffixes such as “module,” “part,” or “unit” used to denote elements are used only for illustrative purposes and have no specific meaning in themselves. Therefore, “module,” “part,” or “unit” may be used interchangeably.

[0033] In this document, the terms "upper," "lower," "inner," "outer," "front," "rear," "one end," and "the other end," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0034] In this document, unless otherwise explicitly specified and limited, the terms "installed," "equipped with," and "connected," etc., should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, a direct connection, or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0035] In this document, the term “and / or” includes any and all combinations of one or more of the listed related items.

[0036] In this article, the term "multiple" means two or more, that is, it includes two, three, four, five, etc.

[0037] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0038] Current common LVC simulation architectures, such as High-Level Architecture (HLA), employ an object-oriented approach to construct object models of simulation resources. This enables interoperability and resource reuse within the simulation federation, as well as separation of the application layer and the basic resource layer. Its core lies in the use and interaction of simulation models, but it has poor applicability in real-time simulation systems. Another example is the Experiment and Training Enabled Architecture (TENA), which is an architecture rather than a specific implementation tool. Its core is to integrate real-world range resources into the simulation training system, improving the reusability and composability of real-world resources in experimentation and training. However, it fails to systematically cover the needs of the three domains of LVC (i.e., physical, virtual, and construction). Therefore, existing technologies have shortcomings in unified modeling of cross-domain resources.

[0039] Based on this, embodiments of this application provide a method, device, and storage medium for constructing a distributed LVC simulation system. Through unified-granularity resource componentization, simulation entities can be flexibly combined across subsystems as needed. Simultaneously, a proxy model is introduced as a lightweight intermediary between the digital simulation system and external heterogeneous resources, transforming the complex and diverse cross-system interactions arising from highly flexible combinations into unified and structured standard communication within the digital simulation system. Thus, while improving the flexibility and scalability of LVC simulation, communication overhead is effectively reduced, and the stability and operational efficiency of the simulation are enhanced.

[0040] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0041] In this embodiment, the distributed LVC simulation system includes three subsystems: a digital simulation system, a hardware-in-the-loop simulation system, and a flight simulator simulation system. Please refer to... Figure 1 , Figure 1 This is a schematic diagram of the architecture of a distributed LVC simulation system provided in an embodiment of this application.

[0042] like Figure 1 As shown, the flight simulator simulation system is a flight simulation device that represents a real or highly immersive simulation environment with the participation of real personnel; the hardware-in-the-loop simulation system integrates real hardware with simulation models to achieve high-fidelity dynamic response; and the digital simulation system, as the global simulation hub, is responsible for task management, simulation control, and resource coordination.

[0043] A digital simulation system, also known as a numerical simulation platform or digital simulation node, comprises a model layer, a functional layer, and an application layer. The model layer includes various digital simulation models within the system, as well as proxy models for other subsystems, such as flight simulation proxy models and semi-physical proxy models. The functional layer is used for scenario editing, behavioral logic generation, and data storage within the LVC system, providing functions such as scenario editing, behavioral logic generation, and data storage. The application layer provides functions such as collaborative control, scene rendering, and performance evaluation.

[0044] The digital simulation system also serves as a deduction engine, driving the entire simulation process. Based on physical rules, behavioral models, and task scenarios, it performs real-time calculations and updates on the states of each simulated entity, coordinating the timing synchronization and interaction logic of heterogeneous components such as digital, hardware-in-the-loop, and flight simulators, enabling the global simulation to proceed efficiently and consistently according to a unified time base. For example, the digital simulation system is a standalone computer device.

[0045] A hardware-in-the-loop (HIL) simulation system, also known as a hardware-in-the-loop simulation platform or hardware-in-the-loop simulation node, includes a real-time simulator, a flight control computer, an external mission computer, and a simulation node management terminal. It simulates the aircraft's attitude and motion in space and target signals through on-the-loop simulation of physical aircraft hardware. The flight control computer may be a real avionics system, connected to sensor and actuator models to form a hardware-in-the-loop closed loop; the real-time simulator runs high-precision flight dynamics and environmental models; and the management terminal is responsible for monitoring and scheduling the status of local nodes. For example, a hardware-in-the-loop simulation system may have a large number of hardware-in-the-loop simulation nodes, with each node constructed from multiple hardware-in-the-loop simulation devices.

[0046] Flight simulator simulation systems, also known as flight simulator simulation platforms, flight simulator simulation nodes, or flight simulation equipment, consist of cockpit operation units such as flight cockpits, operation units (such as joysticks and throttles), simulation panels, and pilot-view rendering modules, forming a human-in-the-loop interactive environment. Human-in-the-loop flight simulators can realize electronic warfare in complex environments, highly maneuverable warfare, and human-in-the-loop decision-making warfare, meeting the needs of more complex and realistic scenarios required by manned / unmanned systems.

[0047] The distributed communication bus is a network connecting the three subsystems and supports three synchronization and communication mechanisms: clock synchronization, which ensures that all nodes have the same time base; DDS (Data Distribution Service) event-driven, which enables efficient and low-latency asynchronous message passing based on the data distribution service; and reflective memory stepping, which is used for shared memory communication between nodes with high real-time requirements to ensure deterministic transmission of critical data.

[0048] Please see Figure 2 , Figure 2This is a schematic flowchart illustrating a construction method for a distributed LVC simulation system provided in an embodiment of this application, such as... Figure 2 As shown in the figure, this application provides a method for constructing a distributed LVC simulation system.

[0049] S101, obtain simulation resources from each subsystem.

[0050] Specifically, in digital simulation systems, simulation resources include mathematical models used to simulate the motion characteristics, perception capabilities, adversarial behavior, and communication mechanisms of various entities (such as aircraft and radar); in hardware-in-the-loop simulation systems, simulation resources include real or simulated hardware devices (such as flight control computers, navigation modules, and communication models); and in flight simulator simulation systems, simulation resources include cockpit operating units. These resources are all parsed and encapsulated into component models with independent simulation functions based on a preset granularity.

[0051] In some embodiments, resource scanning and registration are performed on the three types of subsystems in advance.

[0052] S102, based on the preset splitting granularity, the simulation resources in each subsystem are divided and merged to obtain multiple component models with independent simulation functions.

[0053] Specifically, the preset splitting granularity refers to the pre-determined functional division scale of simulation resources. Simulation resources in each subsystem are decoupled and encapsulated based on this unified scale to obtain several component models.

[0054] For example, component models are not limited to mathematical models; they can be mathematical models in digital simulation systems, hardware devices in hardware-in-the-loop simulation systems, or human-computer interaction entities (such as cockpit control units) in flight simulator simulation systems. These component models have diverse physical forms, independent input / output interfaces, internal state representation, and independent simulation capabilities, and can be combined and used across subsystems in subsequent steps.

[0055] In some embodiments, during combat simulation, the preset granularity of the decomposition includes four functional dimensions: maneuvering, sensing, weaponry, and communication. Simulation resources for subsystems are built based on these functional dimensions to simulate the comprehensive behavior of aircraft in complex battlefield environments (such as maneuvering evasion, target detection, electronic jamming, and formation coordination), resulting in component models that can independently simulate each function. Correspondingly, the component models can be categorized by function into maneuvering models, sensing models, weaponry models, and communication models.

[0056] Among them, the maneuvering model is used to simulate the motion behavior of simulated entities, reflecting their dynamic characteristics such as position, velocity, acceleration and controlled response in space, such as the trajectory, speed and attitude changes of aircraft; the sensing model is used to simulate the functions of various detection or sensing devices, and to generate observation data on the environment or other entities, such as radar, infrared or photoelectric sensors with different detection ranges, accuracies and data output logic; the adversarial model is used to simulate offensive and defensive interactive behaviors such as weapon use and defensive measures, and to realize combat simulation and simulation; the communication model is used to simulate the information transmission mechanism between entities, such as communication protocols, link status, data format and transmission performance.

[0057] In some embodiments, the granularity of the decomposition can be flexibly set according to the simulation's requirements for the finer details of various functions and the independence of various functions in the actual application scenario. It should be understood that the granularity of the decomposition can be understood as the rules for functional decoupling and module encapsulation of simulation resources. It is used to determine the level of detail to which simulation resources are divided based on which functions, so that each division result can independently represent a certain type of runnable, synchronizeable, and composable simulation capability, while also closely matching the specific needs of the simulation scenario. For example, in high-fidelity tasks, fine-grained decomposition can be performed based on a single function; in scenarios emphasizing efficiency, closely related functions can be merged into a single unit, forming a coarse-grained encapsulation.

[0058] For example, if a simulation task has higher fidelity requirements for a certain dimension and needs to finely characterize its internal mechanisms, the granularity of the breakdown can be refined. Taking the communication dimension as an example, it can be further broken down into several functionally defined sub-units, such as message encoding / decoding, network latency and packet loss simulators, etc., thus allowing for more refined breakdown of simulation resources. In this case, the preset breakdown granularity includes multiple functional dimensions such as maneuver, sensing, countermeasures, message encoding / decoding, and network latency and packet loss simulators. For example, if certain dimensions are highly coupled and usually used synchronously in a specific real-world scenario, they can be merged into one dimension, and the breakdown granularity can be adjusted accordingly to a coarser-grained division. For instance, maneuver and communication can be merged into maneuver command. In this case, the preset breakdown granularity includes three functional dimensions: sensing, countermeasures, and maneuver command.

[0059] S103, based on the simulation task requirements, select target component models from multiple component models and combine them to construct multiple simulation entities; wherein, each simulation entity is composed of target component models from a single subsystem, or is composed of a mixture of target component models from any two or three subsystems.

[0060] Specifically, the simulation task requirements include the types of entities to be simulated that are clearly defined to achieve specific simulation goals, as well as their behavioral complexity, simulation accuracy, etc. Based on this, it can be determined whether each simulation entity needs to introduce human-in-the-loop operation, hardware-in-the-loop operation, or pure digital inference. Then, target component models can be selected from the component models according to the functional matching principle for combination.

[0061] For example, to build a high-fidelity aircraft simulation entity, hardware devices (such as flight control computers) from a hardware-in-the-loop simulation system, communication model components from a digital simulation system, and cockpit operation units from a flight simulator simulation system can be selected and combined to form a composite simulation entity spanning subsystems. If only rapid simulation of entity behavior is required, lightweight component models from a digital simulation system can be used to build a pure digital simulation entity.

[0062] In some embodiments, the component models supported under different functions exhibit differentiated characteristics based on the specific configuration capabilities and resource coverage of various subsystems. For example, the actual coverage of component models in a subsystem is shown in Table 1. Some hardware-in-the-loop simulation systems may only cover specific sensor models and can only build component models that strictly match specific models; digital simulation systems can preset multiple types of digital simulation models to cover most common models; while flight simulators may be built entirely around the cockpit layout and operating logic of a single model and can only build component models corresponding to a single model.

[0063] Table 1: Actual Coverage of Component Models in Subsystems

[0064]

[0065] Simple equipment models can be formed by scheduling maneuvering models to simulate and calculate the aircraft's operational position and situational information; it also supports scheduling four types of component models for joint assembly. For example, in the same scenario, multiple signal-level radar models available in the hardware-in-the-loop can be scheduled for detection. For signal-level radar hardware-in-the-loop models that are difficult to obtain, coarser-grained functional-level digital radar models in the digital simulation platform can also be scheduled for detection, realizing cross-resolution model joint simulation across signal level, mission level, and functional level.

[0066] For example, please see Figure 3 , Figure 3 This is a schematic diagram illustrating the distribution of component models under different functions, as provided in an embodiment of this application. Figure 3 As shown, under the mover dimension, all three types of subsystems have component models with corresponding functions. Under the sensor, weapon, and comm dimensions, only the digital simulation system and the hardware-in-the-loop simulation system have component models with corresponding functions, and the LVC simulation system also includes a corresponding proxy model.

[0067] Therefore, the simulation resources of the three subsystems of digital, semi-physical, and flight simulator are divided into functionally independent component models at a unified granularity, realizing a high degree of modularization and standardization of heterogeneous simulation resources. On this basis, when constructing specific simulation tasks, components can be freely selected and mixed and matched across subsystems according to accuracy, cost, or functional requirements.

[0068] The highly flexible combination of simulation entities enables a high degree of simulation customization capability, improves the breadth and depth of simulation, enhances the system's adaptability to diverse application scenarios and resource utilization efficiency, and further complicates the interaction between heterogeneous systems during the simulation process, which to some extent increases the communication burden of data synchronization. However, the embodiments of this application effectively reduce the network transmission of unnecessary data by introducing a proxy model mechanism, thereby reducing the communication load to a certain extent.

[0069] S104. For external target component models that do not belong to the digital simulation system, create corresponding proxy models in the digital simulation system.

[0070] Specifically, when the simulation entity includes external target component models from a hardware-in-the-loop simulation system or a flight simulator simulation system, a corresponding proxy model, also known as a digital proxy model, is automatically created within the digital simulation system. During the initialization phase, the proxy model aligns with the external target component model in terms of underlying communication protocols, interfaces, and parameter sets, and establishes a bidirectional data channel.

[0071] In some embodiments, the digital simulation system is used to provide simulation resources and perform global simulation deduction; the proxy model is used to perform bidirectional state synchronization with the corresponding external target component model during simulation operation, and to interact with the digital simulation system on behalf of the external target component model.

[0072] Specifically, during the simulation, the agent model continuously receives status data (such as position, velocity, and operation commands) sent by the external target component model, converts it into a unified data format within the digital simulation system, and synchronizes it to the global inference engine; at the same time, it also forwards the environmental feedback or control commands generated by the digital simulation system to the corresponding external target component model.

[0073] It should be understood that by introducing a proxy model mechanism, the state of the external target component model is mirrored and synchronized through the parameter set of the proxy model. This transforms the original multi-protocol, multi-interface, point-to-point communication between the digital simulation system and various heterogeneous external systems into a unified, structured, centralized interaction between the digital simulation system and only the internal proxy model. In this way, the protocol differences, data formats, and communication logic of the external systems are encapsulated within the proxy model, simplifying the communication logic and thus reducing the complexity of system interaction and communication overhead.

[0074] In some embodiments, the proxy model is configured with a set of parameters.

[0075] In some embodiments, the parameter set corresponds one-to-one with the actual operating parameters of the external target component model. The proxy model continuously receives and updates the parameters during the simulation process, ensuring that the digital simulation system always maintains the latest state of the external target component model.

[0076] In some embodiments, the method includes: selecting a target parameter type from multiple actual operating parameters of the target component model based on the independent simulation function implemented by the external target component model; wherein the target parameter type is a parameter participating in simulation inference interaction; configuring a parameter set of a corresponding proxy model according to the target parameter type; wherein the parameter values ​​in the parameter set are kept synchronized with the parameter values ​​in the corresponding actual operating parameters.

[0077] Specifically, when configuring the proxy model parameters, based on the independent simulation functions (such as sensing, maneuvering, or communication) undertaken by the external target component model, the attributes that directly participate in the simulation and deduction interaction are identified from all its actual operating parameters and defined as target parameter types; corresponding parameter fields are established in the proxy model only for these target parameter types to form a concise and effective parameter set.

[0078] It should be understood that when dividing and merging simulation resources in step S102, some non-interactive or redundant parameters may be retained to ensure the universality or integrity of the component models. The parameter configuration process of the proxy model is precisely a process of tailoring and focusing on these parameters as needed, eliminating internal variables that do not participate in cross-system interaction, and retaining only the key states that affect the global inference.

[0079] For example, the maneuvering model is used to simulate the motion simulation of an aircraft. The input is generally the real-time position of the aircraft or target. After being solved by a preset control algorithm, the real-time position, attitude, and velocity of the aircraft are output after each simulation step. The parameters that need to be mapped are shown in Table 2.

[0080] Table 2. Parameters of the Motion Model

[0081]

[0082] For example, the adversarial model is used to simulate the adversarial equipment carried by some aircraft. The timing of the adversarial action is selected through combat strategies or human intervention. The input is the strike enable, and the output is the current timing state of the adversarial model. The set of parameters that need to be mapped is shown in Table 3.

[0083] Table 3. Parameter Table for Adversarial Model

[0084]

[0085] For example, the sensor model is used to simulate the radar detection function of an aircraft or a ground command and control center. It takes target information from the simulation scenario as input, periodically calculates enemy target information detected by our target detection methods, and outputs it to our operational units. This introduces realistic detection errors and demonstrations into the simulation scenario, increasing the realism of LVC simulation. The set of parameters that need to be mapped is shown in Table 4.

[0086] Table 4 Sensing Model Parameter Table

[0087]

[0088] For example, the communication model is used to simulate the external communication functions of an aircraft and to build a communication link for cluster control. The set of parameters that need to be mapped is shown in Table 5.

[0089] Table 5 Communication Model Parameter Table

[0090]

[0091] In some embodiments, the parameter set includes: static configuration parameters and dynamic state parameters; wherein, the static configuration parameters are set during the simulation initialization phase and remain unchanged during the simulation operation; the dynamic state parameters change dynamically during the simulation operation and directly participate in the inference logic interaction.

[0092] Specifically, based on the changing characteristics of the parameters of the external target component model during the simulation process, the parameters are divided into static configuration parameters and dynamic state parameters. The changing characteristics of parameters refer to whether they remain unchanged over a long period of time during the simulation. For example, the protocol, gateway, and address parameters in the communication model are generally fixed after initialization and remain unchanged during the simulation process, constituting static configuration parameters; the position, attitude, and velocity parameters in the maneuvering model, as well as the slant range and azimuth angle parameters in the sensing model, are continuously updated as the simulation progresses and directly participate in the inference logic, constituting dynamic state parameters.

[0093] In some embodiments, the method includes: at the start of each simulation step, receiving simulation step data issued by the digital simulation system based on the proxy model, and forwarding the simulation step data to the external target component model; performing real-time step calculation based on the simulation step data according to the external target component model, and updating the actual operating parameters of the external target component model; and transmitting the updated actual operating parameters to the proxy model to update the parameter set in the proxy model.

[0094] Among them, simulation step data refers to the control and environmental information generated and distributed by the digital simulation system at the beginning of each simulation step, which is used to drive the external target component model to perform the simulation calculation of the current cycle.

[0095] Specifically, during simulation operation, a step-driven mechanism is used to synchronize the digital simulation system with the external target component model. At the start of each simulation step, the digital simulation system generates simulation step data for the current step (such as timestamps, environmental disturbances, and received messages from other entities) and sends it to the corresponding proxy model via an internal bus. Upon receiving the data, the proxy model immediately converts the simulation step data into a format recognizable by the external target component model and forwards it to the external target component model via a preset communication channel (such as DDS or reflective memory). The external target component model performs local real-time calculations (such as flight control calculations, sensor responses, or operational feedback) based on the received simulation step data, updates its internal actual operating parameters, and sends the updated parameter values ​​back to the proxy model. The proxy model then synchronously refreshes the corresponding fields in its parameter set, enabling the digital simulation system to perform global calculations based on the latest state in the next simulation cycle, thus forming a closed-loop, time-consistent collaborative simulation process.

[0096] In some embodiments, by employing differentiated data synchronization strategies, the real-time nature of key simulation data is ensured while effectively reducing the network transmission of unnecessary data and lowering the communication load.

[0097] For example, the method includes: during simulation operation, the proxy model only performs periodic or event-triggered synchronization of dynamic state parameters to ensure that the digital simulation system can efficiently obtain the real-time state of the external target component model, while static configuration parameters are used to maintain interface consistency and functional context to avoid redundant transmission.

[0098] For example, the method includes: dividing the attribute parameters in the parameter set into core parameters and monitoring parameters based on a preset synchronization priority; transmitting the parameter values ​​of the actual operating parameters corresponding to the core parameters to the proxy model in each simulation step; continuously monitoring the changes in the values ​​of the monitoring parameters; and transmitting the parameter values ​​of the actual operating parameters corresponding to the monitoring parameters to the proxy model when the value of any monitoring parameter exceeds a preset range threshold or the rate of change exceeds a preset fluctuation threshold.

[0099] Specifically, during the simulation initialization phase, synchronization priorities are set according to the degree of influence of parameters on the deduction logic, with higher priority parameters designated as core parameters and relatively lower priority parameters designated as monitoring parameters. For example, parameters that directly affect control decisions, collision detection, or task flow (such as position and velocity) are classified as core parameters; while parameters used for recording or auxiliary analysis (such as sensor internal temperature and secondary status flags) are classified as monitoring parameters.

[0100] In each simulation step, the external target component model actively transmits the latest values ​​of core parameters to the proxy model to ensure high-frequency synchronization. At the same time, it continuously monitors the changes in parameter values, and only triggers the synchronization of monitoring parameters when they exceed the preset range threshold (such as height anomalies) or the rate of change exceeds the preset fluctuation threshold (such as attitude mutations).

[0101] The preset range threshold refers to the upper and lower limits set for the monitored parameter, used to determine whether the parameter is within the normal physical or logical range, or whether there is a large numerical change. The preset fluctuation threshold refers to the maximum allowable rate of change set for the monitored parameter, such as the amount of numerical change per unit time, used to detect whether it has abrupt changes or unexpected jumps. The specific values ​​of the preset range threshold and preset fluctuation threshold can be flexibly set and adjusted according to different parameters, and are not limited here.

[0102] It should be understood that both core parameters and monitoring parameters are dynamic state parameters. Core parameters refer to attribute parameters that have a direct and immediate impact on the simulation and deduction logic in dynamic state parameters, and they adopt periodic synchronization. Monitoring parameters, on the other hand, refer to attribute parameters that do not participate in high-frequency control logic, and they adopt event-driven synchronization, transmitting only when the values ​​are abnormal or change drastically.

[0103] Therefore, dynamic state parameters in the proxy model parameter set are classified and managed based on preset synchronization priorities. This significantly reduces unnecessary data traffic while ensuring simulation stability, optimizes communication efficiency, and guarantees the real-time synchronization of critical states.

[0104] In some embodiments, when an external target component model is determined to be failed due to communication interruption, hardware failure, or response timeout, a fault-tolerant takeover mechanism is activated. The method includes: when the external target component model fails, selecting a similar alternative model with the same independent simulation function from the digital simulation system based on static configuration parameters in the proxy model; initializing the initial operating state of the similar alternative model according to dynamic state parameters in the proxy model; and taking over the simulation task in the digital simulation system based on the initialized similar alternative model.

[0105] Specifically, based on the static configuration parameters stored in the proxy model, a similar alternative model with the same independent simulation function is matched in the component library of the digital simulation system. The selected similar alternative model is initialized with the latest dynamic state parameters in the proxy model. After initialization, the similar alternative model seamlessly takes over the original external target component model within the digital simulation system and continues to participate in the global simulation.

[0106] In some embodiments, a proxy model continuously monitors the communication status and data validity between the external target component model and the target component model to achieve real-time data monitoring and health status monitoring. For example, it detects whether no status update from the external target component model has been received within a preset number of simulation steps, or determines whether the received data exceeds the physically reasonable range or contains error flags; if so, the external target component model is determined to have failed; if not, the external target component model is determined to have recovered.

[0107] In some embodiments, when the external target component model is detected to have returned to normal, the proxy model updates its parameter set based on the actual operating parameters in the similar alternative model, and synchronizes the latest parameter set saved in the proxy model to the original external target component model, smoothly switching control from the similar alternative model back to the original external target component model, and re-establishing the bidirectional synchronization channel between it and the proxy model, so that subsequent simulations can continue to utilize real hardware or human-in-the-loop input to maintain high fidelity.

[0108] In distributed LVC simulation systems, external devices (such as hardware-in-the-loops, flight simulators, etc.) are typically heterogeneous, potentially originating from different manufacturers, employing different communication protocols, and operating in varying environments. They may also experience temporary failures or communication interruptions due to hardware malfunctions, network jitter, driver anomalies, or human error. If the system directly relies on these external devices for core simulations, the loss of connection to any one device could trigger the suspension or even termination of the entire simulation task.

[0109] By introducing a proxy model mechanism, the digital simulation system used for global simulation deduction no longer directly couples with external devices, but interacts with its internal proxy model. When external devices fail, the proxy model can immediately activate a functionally equivalent replacement model in the digital simulation system, seamlessly taking over the simulation behavior based on its saved parameter set, allowing the global deduction to continue running. Once the external devices recover, it smoothly switches back to the original devices. Specifically, static configuration parameters fully preserve the functional semantics and interface characteristics of the external target component model, enabling precise matching of digital models with the same simulation capabilities. Dynamic state parameters mirror the latest operating state of the external target component model in real time, providing high-fidelity initial conditions for similar replacement models, ensuring smooth simulation continuity and physical consistency, and avoiding state jumps or abrupt behavioral changes.

[0110] The entire takeover process requires no external intervention. The parameters obtained by the proxy model as a communication intermediary are transformed into the buffer and fault tolerance capabilities of the simulation system. The original direct-connect architecture, which was highly dependent on external devices, is transformed into a loosely coupled architecture with flexible takeover capabilities. This effectively isolates the impact of the uncertainty of external devices on the global simulation task, ensuring that the system can still perform simulation simulations stably and continuously when connected to diverse, complex, and unreliable external resources. This improves the robustness and stability of the LVC simulation system in complex heterogeneous environments, while effectively reducing the risk of task interruption caused by the access of multiple external devices.

[0111] In some embodiments, the method includes: unregistering the corresponding proxy model when the simulation task of the external target component model ends or is destroyed during the adversarial process.

[0112] Specifically, when an external target component model fails due to completing a specified simulation task (such as reaching the target area or completing the task) or is deemed ineffective during the simulation (e.g., being hit, destroyed, or voluntarily exiting in an adversarial scenario), the digital simulation system sends a termination signal to the corresponding proxy model. The proxy model then stops sending and receiving data with the external target component model and releases its occupied memory resources, communication channels, and state cache. After deregistration, the proxy model no longer participates in the calculation of subsequent simulation steps, ensuring that system resources are not unnecessarily occupied.

[0113] In some embodiments, the LVC experimental system has a large number of complex distributed simulation nodes. A key challenge is encapsulating the communication interfaces of heterogeneous resources such as simulators and distributed hardware-in-the-loop simulation models, addressing the interconnection, interoperability, and interoperability issues of these resources. The system provides real-time services, is compatible with different underlying communication mechanisms, and shields the differences in LVC simulation resource communication protocols through unified service calls, providing a basic operating environment that allows the simulation engine to seamlessly and with low latency utilize various LVC simulation resources. DDS is a data-centric distributed real-time communication protocol used for sending and receiving simulation events. The publish / subscribe message queue feature ensures that DDS can flexibly and accurately transmit simulation events point-to-point, isolating and optimizing network communication. A reflective memory communication protocol is used for transmitting step information in the simulation system. The reflective memory network is a high-speed, replicated shared memory hardware network. Application software at each layer on each simulator can directly read and write data on the reflective memory through the API interface, ensuring the transmission rate and real-time performance of the simulation system. Furthermore, the reflective memory network supports different operating systems, enabling data transmission between different operating systems and greatly simplifying the data transfer process.

[0114] In some embodiments, a Base Object Model (BOM) technique is employed to construct a unified, cohesive heterogeneous component model digital proxy model. This unifies simulation objects across the three systems and provides a proxy model that can be scheduled and used by the joint simulation system, forming a proxy model construction method and communication interface covering real-world, virtual, and constructed scenarios, as well as manned / unmanned joint scenarios. Please refer to [link to relevant documentation]. Figure 4 , Figure 4 This is a schematic diagram illustrating the functionality of a digital agent model provided in an embodiment of this application, such as... Figure 4 As shown, the proxy model can be used for entity initialization and destruction, simulation event adjustment, step information interaction, node information collection, real-time data monitoring, health status monitoring, etc.

[0115] By combining the mapping relationship between the BOM logical structure and the data object model, the life cycle of the proxy model and the communication mechanism can be divided into four stages: the proxy model object creation stage, the simulation initialization stage, the simulation running stage, and the proxy model destruction stage.

[0116] In the proxy model creation phase, based on the identifiers and characteristics of the hardware-in-the-loop simulation model and the flight simulation node, a hardware-in-the-loop simulation digital proxy model and a flight simulation digital proxy model are constructed, and the communication interface is created.

[0117] During the simulation initialization phase, the digital agent model sends simulation initialization event commands to the hardware-in-the-loop simulation model and the flight simulation node through the scheduling communication interface. The hardware-in-the-loop simulation model and the flight simulation node prepare for initialization and report an initialization success event after initialization is completed. At this time, the digital agent model also enters the waiting run phase.

[0118] During the simulation run, the proxy model acquires simulation step information at each step and forwards it to the hardware-in-the-loop simulation model and flight simulation node through the distributed communication bus. It also acquires the current simulation state of the hardware-in-the-loop simulation model and flight simulation node at each step and updates its own object attribute parameters.

[0119] During the proxy model destruction phase, the proxy model deletes the registered object instances, sends the simulation end command to the hardware-in-the-loop simulation model and the flight simulation node, then closes the communication interface, releases the proxy node communication resources in the co-simulation system, and exits the LVC simulation.

[0120] For example, a manned / unmanned collaborative reconnaissance and strike scenario was designed, and the specific configuration is shown in Table 6.

[0121] Table 6: Detailed Configuration of the Test Scenario

[0122]

[0123] In the first phase, 30 Class A UAVs were deployed for collaborative area reconnaissance, forming 10 UAV swarms, each consisting of 3 UAVs. Class A UAV models could be simultaneously loaded into both hardware-in-the-loop (HIL) and digital simulation nodes (e.g., a digital functional level model for the mover and a hardware-in-the-loop signal level radar for the sensor), enabling a hybrid virtual-real simulation experiment. The leader aircraft in the subsequent 5 formations could use HIL models. Once the target was detected, the UAV swarms transmitted the target trajectory to the manned aircraft on patrol, concluding the collaborative reconnaissance mission.

[0124] In the second phase, manned / unmanned aerial vehicles (UAVs) cooperate to engage ground targets. Two manned / unmanned combat formations fly into the engagement range and interact with enemy targets. Each formation consists of one manned aircraft and two Class B UAVs. The manned aircraft is simulated using a flight simulator, while the two Class B UAVs can be composed of one digital simulation model and one semi-physical simulation model, respectively.

[0125] Please see Figure 5 , Figure 5 This is a schematic diagram of a co-simulation scenario for an LVC system provided in an embodiment of this application, in which co-simulation is achieved by flexibly scheduling component models to form a multi-granularity simulation model. Figure 5 As shown, the manned aircraft uses the move part of the flight simulator, combined with a certain type of radar in the hardware-in-the-loop simulation system as the sensor part, and the digital simulation model in the digital simulation system as the weapon part and comm part. The above four parts together form a highly realistic and easy-to-schedule manned aircraft simulation.

[0126] For example, for Class B drones, high-fidelity hardware-in-the-loop and digital simulation models can be used to form the drone's move part, a digitally simulated radar model can be used as the sensor part, a functional digital model in the digital simulation platform can be used as the weapon part, and a digital communication model can be used as the comm part to form a drone simulation model.

[0127] The method and distributed LVC simulation system of this application can be used in a wide range of general-purpose or special-purpose computing system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer terminal devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices.

[0128] Please see Figure 6 , Figure 6 This is a schematic block diagram illustrating the structure of a computer device according to an embodiment of this application. The computer device may be a terminal device or a server.

[0129] For example, the above method can be implemented as a computer program, which can be used in, for example... Figure 6 It runs on the computer device shown.

[0130] like Figure 6 As shown, the computer device includes a processor, memory, and network interface connected via a system bus, wherein the memory may include non-volatile storage media and internal memory.

[0131] Non-volatile storage media can store operating systems and computer programs. These computer programs include program instructions that, when executed, cause the processor to perform any construction method based on a distributed LVC simulation system.

[0132] The processor provides computing and control capabilities, supporting the operation of the entire computer device.

[0133] Internal memory provides an environment for the execution of computer programs in non-volatile storage media. When the computer program is executed by the processor, it enables the processor to execute any construction method based on a distributed LVC simulation system.

[0134] This network interface is used for network communication, such as sending assigned tasks.

[0135] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0136] In one embodiment, the processor is configured to run a computer program stored in memory to perform the following steps:

[0137] S101, obtain simulation resources from each subsystem;

[0138] S102, based on the preset splitting granularity, the simulation resources in each subsystem are divided and merged to obtain multiple component models with independent simulation functions;

[0139] S103, based on the simulation task requirements, select target component models from multiple component models and combine them to construct multiple simulation entities; wherein, each simulation entity is composed of target component models from a single subsystem, or is composed of a mixture of target component models from any two or three subsystems;

[0140] S104. For external target component models that do not belong to the digital simulation system, create corresponding proxy models in the digital simulation system.

[0141] The digital simulation system is used to provide simulation resources and perform global simulation deduction; the proxy model is used to perform bidirectional state synchronization with the corresponding external target component model during the simulation operation, and to interact with the digital simulation system on behalf of the external target component model.

[0142] For example, the processor is used to run a computer program stored in a memory, and is also used to implement the steps of the method provided in any embodiment of this application, which will not be repeated here.

[0143] The embodiments of this application also provide a computer-readable storage medium storing a computer program, the computer program including program instructions, and the processor executing the program instructions to implement the steps of the method provided in the embodiments of this application.

[0144] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiments, such as the hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, SmartMedia Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the computer device.

[0145] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for constructing a distributed LVC simulation system, characterized in that, The distributed LVC simulation system comprises three subsystems: a digital simulation system, a hardware-in-the-loop simulation system, and a flight simulator simulation system; the method includes: S101, obtain simulation resources from each subsystem; S102, based on the preset splitting granularity, the simulation resources in each subsystem are divided and merged to obtain multiple component models with independent simulation functions; S103, based on the simulation task requirements, select target component models from multiple component models and combine them to construct multiple simulation entities; wherein, each simulation entity is composed of target component models from a single subsystem, or is composed of a mixture of target component models from any two or three subsystems; S104. For external target component models that do not belong to the digital simulation system, create corresponding proxy models in the digital simulation system. The digital simulation system is used to provide simulation resources and perform global simulation deduction; the proxy model is used to perform bidirectional state synchronization with the corresponding external target component model during the simulation operation, and to interact with the digital simulation system on behalf of the external target component model.

2. The method according to claim 1, characterized in that, The proxy model is configured with a parameter set; the method includes: Based on the independent simulation function implemented by the external target component model, a target parameter type is selected from multiple actual operating parameters of the target component model; wherein, the target parameter type is the parameter that participates in the simulation and deduction interaction; Configure the parameter set of the corresponding proxy model according to the target parameter type; wherein the parameter values ​​in the parameter set are synchronized with the parameter values ​​in the corresponding actual running parameters.

3. The method according to claim 2, characterized in that, The parameter set includes: static configuration parameters and dynamic state parameters; wherein, the static configuration parameters are set during the simulation initialization phase and remain unchanged during the simulation operation; the dynamic state parameters change dynamically during the simulation operation and directly participate in the inference logic interaction.

4. The method according to claim 2, characterized in that, The method includes: At the start of each simulation step, the agent model receives the simulation step data issued by the digital simulation system and forwards the simulation step data to the external target component model. Based on the external target component model, real-time step calculations are performed according to the simulation step data, and the actual operating parameters of the external target component model are updated. The updated actual running parameters are transmitted to the proxy model to update the parameter set in the proxy model.

5. The method according to claim 2, characterized in that, The method includes: Based on the preset synchronization priority, the attribute parameters in the parameter set are divided into core parameters and monitoring parameters; In each simulation step, the parameter values ​​of the actual running parameters corresponding to the core parameters are transmitted to the proxy model; The system continuously monitors the changes in the values ​​of the monitored parameters; when the value of any monitored parameter exceeds a preset range threshold or the rate of change exceeds a preset fluctuation threshold, the system transmits the value of the actual operating parameter corresponding to the monitored parameter to the proxy model.

6. The method according to claim 3, characterized in that, The method includes: When the external target component model fails, a similar alternative model with the same independent simulation function is selected from the digital simulation system based on the static configuration parameters in the proxy model. The initial running state of the similar substitution model is initialized based on the dynamic state parameters in the proxy model; The simulation task continues to be performed in the digital simulation system based on the initialized similar alternative model to replace the failed external target component model.

7. The method according to claim 1, characterized in that, The method includes: When the simulation task of the external target component model ends or it is destroyed during the adversarial process, the corresponding proxy model is deregistered.

8. The method according to any one of claims 1-7, characterized in that, The preset splitting granularity includes four functional dimensions: mobility, sensing, countermeasures, and communication.

9. A computer device, characterized in that, The device includes: Memory, used to store computer programs; A processor is configured to execute the computer program and, in executing the computer program, implement the construction method of the distributed LVC simulation system as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, causes the processor to implement the construction method of the distributed LVC simulation system as described in any one of claims 1 to 8.

Images