A hybrid real-time LVC simulation test system

Abstract

The present invention discloses a hybrid real-time LVC simulation test system, comprising a plurality of hybrid real-time simulation entities, wherein the hybrid real-time simulation entities have both an object model interface used in an Ethernet environment and a shared memory interface used in a VMIC reflective memory communication environment, and a strong real-time simulation interaction buffer. The object model interface transmits data using an object model interaction method based on a TENA-like middleware and has millisecond-level real-time performance. The shared memory interface transmits data in a VMIC reflective memory environment using shared memory and has microsecond-level real-time performance. The strong real-time simulation interaction buffer is a memory area, in which a strong real-time simulation interaction request, interaction data and a wall clock value at the interaction time are written to the strong real-time simulation interaction buffer to form a strong real-time interaction data packet.

Description

Technical Field

[0001] The invention relates to a hybrid real-time LVC simulation test system, belonging to the technical field of simulation. Background Art

[0002] LVC simulation combines live, virtual, and constructed simulations. Implementing LVC simulation requires underlying communication based on a distributed simulation support architecture, and employs technologies such as gateways, adapters, and wrappers to encapsulate and integrate heterogeneous simulation resources. LVC systems require that the simulation system be clocked at a 1:1 speed. TENA is a typical LVC architecture used internationally, while TISA and Josim are popular domestically.

[0003] In LVC combat simulation applications, there are numerous simulated entities. If all simulation loops are strictly real-time, insufficient network communication capabilities to support time synchronization can cause system lockup and prevent logical time from advancing. Furthermore, combat processes such as radar countermeasures and terminally guided intercepts suffer from reduced accuracy and lack of reliability in conventional non-real-time simulation loops. Therefore, it is necessary to develop a simulation test system that balances the accuracy of strict real-time simulation with the flexibility of non-real-time simulation. Summary of the Invention

[0004] The technology of the present invention solves the problem: overcomes the shortcomings of the existing technology and provides a hybrid real-time LVC simulation test system, which can more effectively utilize the strong real-time simulation network, while reducing the strong real-time coupling of the system and supporting larger-scale system-level combat simulation.

[0005] The technical solution of the present invention is a hybrid real-time LVC simulation test system, comprising a plurality of hybrid real-time simulation entities; the hybrid real-time simulation entities simultaneously have an object model interface used in an Ethernet environment and a shared memory interface used in a VMIC reflective memory communication environment; and further comprise a strong real-time simulation interaction buffer;

[0006] The object model interface transmits data using an object model interaction method based on TENA-like middleware, with millisecond-level real-time performance;

[0007] The shared memory interface uses shared memory to transmit data in a VMIC reflective memory environment, and has microsecond-level real-time performance.

[0008] The strong real-time simulation interaction buffer is an area in the memory. It simulates the entity's own status update or the strong real-time simulation interaction process of other simulation entities, and writes strong real-time simulation interaction requests, interaction data and the wall clock value at the interaction time into it to form a strong real-time interaction data packet.

[0009] The hybrid real-time simulation entity represents a combat unit entity during the system integration combat test and has two states: normal advancement mode and strong real-time advancement mode. The characteristic of the normal advancement mode is that the interaction between combat units during the system combat simulation has low real-time requirements, including command and control and space movement; the characteristic of the strong real-time advancement mode is that the interaction between combat units during the system combat simulation has high real-time requirements, including terminal guidance, shooting, and radar illumination.

[0010] The process of a single simulation step in the normal advancement mode includes: managing a strong real-time buffer, determining whether there is strong real-time interaction, waiting, determining whether the logic time can be advanced, normal simulation calculation, and updating the simulation entity state.

[0011] The specific process of managing the strong real-time buffer is: judging whether there is a strong real-time simulation interaction requirement with other simulation entities based on the post-simulation state of the simulation entity, and if so, writing the data required for the interaction request into its own strong real-time simulation buffer; on the other hand, responding to the strong real-time interaction requirements of other simulation entities written into the strong real-time buffer.

[0012] The specific process of determining whether strong real-time interaction exists is: when it is found that the strong real-time buffer is not empty, triggering the strong real-time advancement state.

[0013] The process of the strong real-time advancement mode includes: obtaining the current wall clock, calculating the fitting function of its own simulation state, strong real-time simulation interaction, and determining whether to exit the strong real-time interaction; the strong real-time advancement mode is not controlled by logical time and only runs according to the wall clock. The timestamp of the simulation data frame is the wall clock time.

[0014] The specific process of obtaining the current wall clock is: obtaining the current wall clock from the current simulation node.

[0015] The specific process of calculating the self-simulation state fitting function is: using the self-historical simulation state data to construct a cubic spline curve function of the simulation state, which can return values ​​for both extrapolation and internalization at the current moment.

[0016] The specific process of the strong real-time simulation interaction is as follows: processing the strong real-time interaction data packets in the strong real-time interaction buffer zone; if it is a strong real-time interaction initiated by itself, the input state of the strong real-time interaction is calculated according to its own simulation state fitting function and the wall clock time of the strong real-time interaction data packet to form a strong real-time interaction data packet, and the data packet is written into the strong real-time interaction buffer zone of the simulation entity that passively undergoes the strong real-time simulation interaction; if it is a strong real-time simulation interaction that occurs passively, the simulation calculation is also completed according to the current wall clock state obtained by its own simulation state fitting function, and the interaction result is written into the strong real-time interaction buffer zone of the simulation entity that initiates the interaction.

[0017] The specific process of judging whether to exit strong real-time interaction is: judging whether the strong real-time interaction buffer of the current entity is empty, and if so, exiting strong real-time interaction and switching to normal simulation calculation.

[0018] Beneficial effects of the present invention:

[0019] (1) Dynamically organize strong real-time simulation interactions without causing the read-write lock congestion problem caused by fixed strong real-time simulation interaction loops, thereby improving the efficiency and scale of the simulation system.

[0020] (2) Using general real-time simulation loops to handle simulation interactions with low interaction density and low real-time requirements saves valuable strong real-time simulation resources.

[0021] (3) While increasing the scale of the simulation system, it also takes into account the interactive accuracy required for strong real-time simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 Schematic diagram of the hybrid real-time simulation system of the present invention;

[0023] Figure 2 Schematic diagram of the hybrid real-time simulation process of the present invention. DETAILED DESCRIPTION

[0024] The present invention provides a hybrid real-time LVC simulation test system, which includes a hybrid real-time simulation entity having both an object model interface used in an Ethernet-like environment and a shared memory interface used in a VMIC reflective memory communication environment, and further comprising a strong real-time simulation interaction buffer. The object model interface uses an object model interaction method based on a TENA-like middleware to transmit data, and has millisecond-level real-time performance. The shared memory interface uses shared memory to transmit data in a VMIC reflective memory environment, and has microsecond-level real-time performance. The strong real-time simulation interaction buffer is an area in the memory where the simulation entity's own status update or the strong real-time simulation interaction process of other simulation entities can be written with strong real-time simulation interaction requests, interaction data, and the wall clock value at the interaction time to form a strong real-time interaction data packet.

[0025] A hybrid real-time simulation entity is a simulation application that represents a combat unit during system-of-systems integrated combat testing. It operates in two modes: normal advancement mode and strong real-time advancement mode. Normal advancement mode is characterized by lower real-time requirements for interactions between combat units during system-of-systems combat simulations, including command and control, space movement, and other aspects. Strong real-time advancement mode is characterized by higher real-time requirements for interactions between combat units during system-of-systems combat simulations, including terminal guidance, firing, and radar illumination.

[0026] In the normal advancement mode, the process of a single simulation step includes managing the strong real-time buffer, determining whether there is strong real-time interaction, waiting, determining whether the logical time can be advanced, normal simulation calculation, and updating the simulation entity status.

[0027] The step of managing the strong real-time buffer, on the one hand, determines whether it involves strong real-time simulation interaction requirements with other simulation entities based on the post-simulation status of the simulation entity. If so, writes the data required for the interaction request into its own strong real-time simulation buffer; on the other hand, it can respond to the strong real-time interaction requirements of other simulation entities written into the strong real-time buffer.

[0028] Determine whether there is a strong real-time interaction step, and trigger the strong real-time advancement state if it is found that the strong real-time buffer is not empty.

[0029] The strong real-time push mode process includes obtaining the current wall clock, calculating the simulation state fitting function, performing strong real-time simulation interaction, and determining whether to exit strong real-time interaction. Strong real-time push mode is not controlled by logical time and operates solely according to the wall clock. The timestamp of the simulation data frame is the wall clock time.

[0030] The step of obtaining the current wall clock is to obtain the current wall clock from the current simulation node. The step of calculating the self-simulation state fitting function is to use the self-historical simulation state data to construct the cubic spline curve function of the simulation state, which can return values ​​for both extrapolation and internalization of the current moment. The step of strong real-time simulation interaction is to process the strong real-time interaction data packets in the strong real-time interaction buffer. For the strong real-time interaction initiated by itself, simulation operation is performed. The input state of the strong real-time interaction is calculated according to the self-simulation state fitting function and the wall clock moment of the strong real-time interaction data packet, forming a strong real-time interaction data packet and writing it into the strong real-time interaction buffer of the simulation entity that passively undergoes strong real-time simulation interaction. For the strong real-time simulation interaction that occurs passively, the simulation calculation is also completed according to the current wall clock state obtained by the self-simulation state fitting function, and the interaction result is written into the strong real-time interaction buffer of the simulation entity that initiates the interaction.

[0031] Determine whether to exit the strong real-time interaction step and whether the strong real-time interaction buffer of the current entity is empty. If it is empty, exit the strong real-time interaction and enter normal simulation calculation.

[0032] The present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments:

[0033] like Figure 1 、 2As shown, a hybrid real-time LVC simulation test system includes hybrid real-time simulation entities such as simulation entities A, B, and C. These hybrid real-time simulation entities have both an object model interface used in an Ethernet environment and a shared memory interface used in a VMIC reflective memory communication environment. In addition, they also include a strong real-time simulation interaction buffer. The object model interface uses an object model interaction method based on a TENA-like middleware to transmit data, with millisecond-level real-time performance. The shared memory interface uses shared memory to transmit data in a VMIC reflective memory environment, with microsecond-level real-time performance. The strong real-time simulation interaction buffer is an area in the memory where the simulation entity's own status update or the strong real-time simulation interaction process of other simulation entities can be written. Strong real-time simulation interaction requests, interaction data, and the wall clock value at the interaction time can be written to form a strong real-time interaction data packet.

[0034] Simulation entities A, B, and C represent a combat unit during the system integration combat test, with two states: normal propulsion mode and strong real-time propulsion mode. Simulation entity A is a ground-based radar station, and simulation entities B and C are combat aircraft.

[0035] During the simulation process, the ground-based radar station of simulation entity A undergoes normal simulation calculations. In the step of managing the strong real-time buffer, it finds that there is simulation entity B in its field of view. It needs to establish a strong real-time simulation interaction of radar detection type with simulation entity B, so it writes the data required for the interaction request into its own strong real-time simulation buffer.

[0036] Subsequently, when simulation entity A determines whether there is a strong real-time interaction step, it finds that the strong real-time buffer is not empty, triggers the strong real-time advancement state, and releases the logical time control. In the step of obtaining the current wall clock, the current wall clock is obtained from the current simulation node. In the step of calculating the self-simulation state fitting function, the cubic spline curve function of the radar beam simulation state is constructed using its own historical simulation state data. In the strong real-time simulation interaction step, the strong real-time interaction data packet written by simulation entity A is processed and simulation operations are performed. The input state of the strong real-time interaction is calculated based on the self-simulation state fitting function and the wall clock time of the strong real-time interaction data packet, forming a strong real-time interaction data packet and writing it into the strong real-time interaction buffer of simulation entity B.

[0037] Simulation entity B detects a data packet in the strong real-time interaction buffer, triggering the strong real-time advance state and releasing logical time control. In the step of obtaining the current wall clock, the current wall clock is obtained from the current simulation node. In the step of calculating the self-simulation state fitting function, a cubic spline curve function for its own motion simulation state is constructed using its own historical simulation state data. Simulation entity B completes the simulation calculation based on the current wall clock state obtained by its own simulation state fitting function and writes the interaction results to the strong real-time interaction buffer of simulation entity A.

[0038] Simulation entity A determines that the interaction has been completed, exits the strong real-time interaction, and switches to normal simulation calculation.

[0039] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Anyone familiar with this technology can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be based on the definition of the claims.

Claims

1. A hybrid real-time LVC simulation test system, characterized by: The invention comprises a plurality of hybrid real-time simulation entities; the hybrid real-time simulation entities have both an object model interface used in an Ethernet environment and a shared memory interface used in a VMIC reflective memory communication environment; and a strong real-time simulation interaction buffer; The object model interface transmits data using an object model interaction method based on TENA-like middleware, with millisecond-level real-time performance; The shared memory interface uses shared memory to transmit data in the VMIC reflective memory environment, with microsecond real-time performance; The strong real-time simulation interaction buffer is an area in the memory, in which the strong real-time simulation interaction request, interaction data and the wall clock value at the interaction time are written to simulate the entity's own state update or the strong real-time simulation interaction process of other simulation entities to form a strong real-time interaction data packet; The hybrid real-time simulation entity represents a combat unit entity during the system integration combat test and has two states: normal advancement mode and strong real-time advancement mode. The characteristic of the normal advancement mode is that the interaction between combat units during the system combat simulation has a lower real-time requirement, including command and control and space movement; the characteristic of the strong real-time advancement mode is that the interaction between combat units during the system combat simulation has a higher real-time requirement, including terminal guidance, shooting, and radar illumination. The process of a single simulation step in the normal advancement mode includes: managing a strong real-time buffer, determining whether there is a strong real-time interaction, waiting, determining whether the logic time can be advanced, normal simulation calculation, and updating the simulation entity state; The specific process of managing the strong real-time buffer is: judging whether there is a strong real-time simulation interaction requirement with other simulation entities according to the simulation entity's post-simulation state, and if so, writing the data required for the interaction request into its own strong real-time simulation buffer; on the other hand, responding to the strong real-time interaction requirements written into the strong real-time buffer by other simulation entities; The specific process of determining whether there is strong real-time interaction is as follows: when it is found that the strong real-time buffer is not empty, triggering the strong real-time advancement state; The process of the strong real-time advancement mode includes: obtaining the current wall clock, calculating the simulation state fitting function, strong real-time simulation interaction, and determining whether to exit the strong real-time interaction. The strong real-time advancement mode is not controlled by logical time and only runs according to the wall clock. The timestamp of the simulation data frame is the wall clock time. The specific process of obtaining the current wall clock is: obtaining the current wall clock from the current simulation node.

2. A hybrid real-time LVC simulation test system according to claim 1, characterized in that: The specific process of calculating the self-simulation state fitting function is: using the self-historical simulation state data to construct a cubic spline curve function of the simulation state, which can return values ​​for both extrapolation and internalization at the current moment.

3. The hybrid real-time LVC simulation test system according to claim 1, characterized in that: The specific process of the strong real-time simulation interaction is as follows: processing the strong real-time interaction data packet in the strong real-time interaction buffer, and for the strong real-time interaction initiated by itself, calculating the input state of the strong real-time interaction according to the simulation state fitting function of itself and the wall clock time of the strong real-time interaction data packet, forming a strong real-time interaction data packet, and writing it into the strong real-time interaction buffer of the simulation entity that passively undergoes the strong real-time simulation interaction; For passive strong real-time simulation interaction, the simulation calculation is also completed according to the current wall clock state obtained by its own simulation state fitting function, and the interaction result is written into the strong real-time interaction buffer of the simulation entity that initiated the interaction.

4. The hybrid real-time LVC simulation test system according to claim 1, characterized in that: The specific process of judging whether to exit strong real-time interaction is: judging whether the strong real-time interaction buffer of the current entity is empty, and if so, exiting strong real-time interaction and switching to normal simulation calculation.

Images