Variable step size LVC simulation method for respectively controlling simulation step size and interaction step size

Abstract

A variable step size LVC simulation method for respectively controlling a simulation step size and an interaction step size comprises the steps that communication interconnection of a plurality of LVC simulation objects is achieved through a simulation communication network supporting a TENA-like middleware protocol, simulation propulsion of a whole system is controlled according to global logic time, and simulation propulsion comprises a state evolution process and an object interaction process; wherein the object interaction process is triggered and executed based on a period or an event according to the requirement of current LVC simulation object interaction, and is divided into three modes of complete synchronization, equal-step-size asynchronization and variable-step-size asynchronization according to the logic time advancing characteristic of the LVC simulation object participating in the interaction process; the continuous variable type state parameters and the mark type state parameters are processed respectively, and outfield parameter values needed by the current logic time are obtained; the state evolution process is triggered and executed according to evolution periodicity of the LVC simulation object, and the state parameter of the LVC simulation object at the next simulation moment is calculated according to the current state parameter and the external field parameter of the LVC simulation object.

Description

Technical Field

[0001] This invention relates to a variable step size LVC simulation method that controls the simulation step size and the interactive step size separately, belonging to the field of LVC simulation technology. Background Technology

[0002] LVC simulation refers to simulation that combines live, virtual, and construct elements. Implementing LVC simulation requires a distributed simulation support architecture for underlying communication and the use of technologies such as gateways, adapters, and wrappers to encapsulate and integrate heterogeneous simulation resources. For distributed systems containing a large number of heterogeneous simulation objects, the biggest challenge is matching the simulation step size with the information exchange.

[0003] For purely mathematical simulation systems containing a small number of objects, a fully synchronous simulation method with equal-step synchronization is typically used. However, in LVC systems, both physical objects and simulator objects have specific sampling period limitations, making it impossible to require all hardware devices to use the same simulation and interaction cycle. Existing LVC simulation systems cannot meet the interactive communication requirements as much as possible while ensuring the normal operation of the LVC simulation objects themselves. Furthermore, they struggle to resolve issues such as jagged interactive data caused by mismatches in lower logic times, and therefore lack the integrated simulation accuracy of LVC. Summary of the Invention

[0004] The technical problem solved by this invention is to overcome the shortcomings of the prior art and provide a variable step size LVC simulation method that controls the simulation step size and the interaction step size separately, so as to handle heterogeneous asynchronous interaction problems more systematically and effectively.

[0005] The technical solution of this invention is: a variable step size LVC simulation method with separate control of simulation step size and interactive step size, comprising:

[0006] Multiple LVC simulation objects are interconnected through a simulation communication network supporting TENA-like middleware protocols. The entire system simulation is driven by global logical time control, and this simulation progression includes object interaction and state evolution processes.

[0007] The object interaction process is executed based on the needs of the current LVC simulation object interaction, triggered by a period or event. According to the logical time progression characteristics of the LVC simulation object participating in the interaction process, it is divided into three modes: fully synchronous, equal step asynchronous, and variable step asynchronous. Various state parameters are processed separately to obtain the external field parameter values ​​required for the current logical time.

[0008] The state evolution process is triggered periodically according to the evolution of the LVC simulation object. The state parameters of the LVC simulation object at the next simulation moment are calculated based on the current state parameters and external field parameters of the LVC simulation object.

[0009] Preferably, the state parameters are used to describe the state of the LVC simulation object itself;

[0010] External parameters refer to parameters that come from external LVC simulation objects and affect the state parameters of the LVC simulation objects through object interaction processes.

[0011] Preferably, the state parameters include: continuous variable state parameters and flag state parameters.

[0012] Continuous variable state parameters are physical quantities that can produce related changes, including position, velocity, and attitude;

[0013] Flag-type status parameters are parameters related to device status information, including: whether the device is powered on and whether the device is working.

[0014] Preferably, complete synchronization means that the sampling of each LVC simulation object is advanced by the same step size and the step size is equal;

[0015] Equal step asynchronous means that all LVC simulation objects advance at equal steps, but the simulation steps of each object are inconsistent.

[0016] Variable step size asynchronous means that at least one of the simulation objects adopts a variable step size advancement strategy.

[0017] Preferably, for the object interaction process of continuous variable class state parameters in both fully synchronous and equal-step asynchronous cases, the specific details are as follows:

[0018] Triggered based on the judgment conditions of logical time or state parameters during the main simulation loop of the LVC simulation object;

[0019] As the logical time of the main simulation loop progresses, the corresponding state parameters of the LVC simulation object that undergoes the interaction process of the object are read.

[0020] Establish a state parameter queue with a length of n parameters, and the sampling logic interval is the longer main simulation loop logic time step in the LVC simulation object of this interaction process;

[0021] Perform a cubic spline curve fitting each time the state parameter queue changes;

[0022] When updating the external field parameters during each LVC simulation object interaction process, the coefficients of the cubic spline curve are called for interpolation / extrapolation calculation to obtain the external field parameter values ​​required for the current logic time.

[0023] When the LVC simulation object interaction process meets the completion conditions, the current object interaction process exits.

[0024] The value of n is set according to the characteristics of the interaction between the parameter and the object; when the queue length is 1, the cubic spline curve fitting condition is not met, and the first value is directly taken as the value of the external field parameter required for the current logical time.

[0025] Preferably, for the object interaction process of continuous variable class state parameters in the case of asynchronous variable step size, the specific details are as follows:

[0026] Triggered based on the judgment conditions of logical time or state parameters during the main simulation loop of the LVC simulation object;

[0027] As the logical time of the main simulation loop progresses, the corresponding state parameters of the LVC simulation object that undergoes the interaction process of the object are read.

[0028] Establish a state parameter queue with a length of m parameters. When reading the value of a state parameter, if the corresponding state parameter of the LVC simulation object that is subject to the interaction process of the object has not changed, then update one state parameter in the parameter queue. The logical timestamp of the state parameter is the logical time when the value of the state parameter is read. Otherwise, the number of parameters updated is the number of times the state parameter has changed, and the timestamp of the corresponding state parameter is the logical time when the parameter changes.

[0029] Perform a cubic spline curve fitting each time the state parameter queue changes;

[0030] Each time the external field parameters are updated during the interaction process of this object, the coefficients of the cubic spline curve are called for interpolation or extrapolation calculation to obtain the external field parameter values ​​required for the current logical time.

[0031] When the interaction process of the object meets the completion conditions, the current interaction process of the object will exit.

[0032] The value of m is set according to the characteristics of the interaction between the parameter and the object; when the queue length is 1, the cubic spline curve fitting condition is not met, and the first value is directly taken as the value of the external field parameter required for the current logical time.

[0033] Preferably, in the object interaction process of continuous variable state parameters under the asynchronous case of variable step size, a cache queue needs to be established on the side of the LVC simulation object that receives the interaction for the state parameters involved in the object interaction process, and the changed state parameter values ​​and corresponding timestamps need to be recorded.

[0034] The preferred interaction process for objects with flag-type state parameters is as follows:

[0035] During the main simulation loop of the LVC simulation object, the system is triggered based on the judgment conditions of the logic time or state parameters; the latest value that can be read from the previous logic time is taken as the value of the external field parameter.

[0036] Compared with the prior art, the present invention has the following advantages:

[0037] (1) The time-progression decoupling of the state evolution of the LVC simulation object and the object interaction process of the present invention can effectively reduce the interaction coupling of the LVC system.

[0038] (2) The present invention involves a special logic timing matching logic for each interaction mode between LVC simulation objects, which can improve the balance between interaction efficiency and interaction accuracy in the LVC simulation system as much as possible.

[0039] (3) The present invention uses the fitting curve interpolation method to achieve time matching of asynchronous heterogeneous interoperability, making the maximum use of effective data and improving the smoothness of interaction. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the LVC simulation system of the present invention;

[0041] Figure 2 This is a schematic diagram of the simulation system operation logic of the present invention. Detailed Implementation

[0042] This invention relates to a variable step size LVC simulation method that controls the simulation step size and the interaction step size separately. Multiple LVC simulation objects are interconnected through a simulation communication network that supports a TENA-like middleware protocol, and the simulation of the entire system is advanced according to global logical time control.

[0043] The simulation calculation process of an LVC simulation object is divided into two processes: state evolution and object interaction. The state evolution process is triggered periodically according to the evolution of the LVC simulation object; the object interaction process is triggered periodically or by events according to the interaction requirements of the object.

[0044] The state evolution process, i.e., the main simulation loop process of the LVC simulation object, can be calculated using a fixed-step or variable-step logical time advancement strategy, depending on the needs of the object. The calculation process involves calculating the state parameters of the object at the next simulation moment based on the current state parameters and external field parameters of the LVC simulation object. State parameters describe the state of the LVC simulation object itself, including position, velocity, attitude, operating condition, mass, and its own target characteristics. External field parameters are those from external LVC simulation objects that influence the state parameters of the LVC simulation object through object interaction, including external forces, external commands, and target characteristics of targets within the sensor's field of view.

[0045] The object interaction process is a specially designed process for each interaction between LVC simulation objects, realizing the state update of external field parameters. This process is maintained by the LVC simulation object that initiates the interaction. Object interaction processes can be handled separately for two types of state parameters: continuous variables and flag-based state parameters. Continuous variables refer to physically changing quantities such as position, velocity, and attitude; flag-based state parameters refer to state parameters such as whether the engine is working or the radar is activated.

[0046] For object interaction processes involving continuous variable state parameters, based on the logical time progression characteristics of the two LVC simulation objects participating in the interaction process, they are categorized into three types: fully synchronous, equal-step asynchronous, and variable-step asynchronous. Fully synchronous means that the two LVC simulation objects advance with equal step sizes; equal-step asynchronous means that both LVC simulation objects advance with equal step sizes, but their simulation step sizes are inconsistent; variable-step asynchronous means that at least one of the two simulation objects adopts a variable-step advancement strategy.

[0047] For object interaction processes involving continuous variable state parameters in both fully synchronous and equal-step asynchronous scenarios, the specific steps are as follows: First, triggering the interaction based on logical time or state parameter judgment conditions during the main simulation loop of the LVC simulation object. Second, as the logical time of the main simulation loop progresses, reading the corresponding state parameters of the LVC simulation object undergoing the interaction process. Third, establishing a state parameter queue with a length of n parameters, with a sampling logical interval equal to the longer logical time step of the main simulation loop between the two LVC simulation objects in the interaction process. Fourth, performing a cubic spline curve fitting each time the state parameter queue changes. Fifth, when updating the external field parameters during the object interaction process, using the coefficients of the cubic spline curve for interpolation / extrapolation calculations to obtain the external field parameter values ​​required for the current logical time. Sixth, exiting the process when the object interaction process meets the completion conditions. The value of n is set according to the characteristics of the parameter and object interaction, with a default value of 4. When the queue length is 1, the cubic spline curve fitting condition is not met, and the first value is directly taken as the external field parameter value required for the current logical time.

[0048] For the interaction process of continuous variable state parameters in the asynchronous case with variable step size, the specific steps are as follows: First, trigger the interaction based on the judgment condition of the logic time or state parameter during the main simulation loop of the LVC simulation object; Second, as the logic time of the main simulation loop advances, read the corresponding state parameters of the LVC simulation object that is subject to the interaction process; Third, establish a state parameter queue with a length of m parameters. When reading the state parameter value, if the corresponding state parameter of the LVC simulation object subject to the interaction process has not changed, update the parameters in the queue. The process involves six steps: 1. A state parameter with a logical timestamp based on the logical time of reading its value; 2. A state parameter with a logical timestamp based on the number of times it changes; 3. A cubic spline curve fitting is performed each time the state parameter queue changes; 4. Each time the object interaction process updates the external field parameters, the coefficients of the cubic spline curve are used for interpolation / extrapolation to obtain the external field parameter values ​​required for the current logical time; 5. When the object interaction process meets the completion conditions, the process exits. The value of m is set according to the characteristics of the parameter and the object interaction, with a default value of 3. When the queue length is 1, the cubic spline curve fitting condition is not met, and the first value is directly taken as the external field parameter value required for the current logical time.

[0049] In the object interaction process of continuous variable state parameters under the asynchronous case of variable step size, a cache queue needs to be established on the side of the LVC simulation object that receives the interaction for the state parameters involved in the object interaction process, to record the changed state parameter values ​​and the corresponding timestamps, and the queue length is m.

[0050] The interaction process of the flag-type state parameter object is as follows: First, it is triggered according to the judgment condition of the logic time or state parameter during the main simulation loop of the LVC simulation object; Second, the latest value that can be read from the previous logic time is taken as the value of the external field parameter.

[0051] Example:

[0052] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments:

[0053] The contents not described in detail in this specification are common knowledge to those skilled in the art.

[0054] like Figure 1The LVC simulation system shown includes a hardware-in-the-loop (HIL) radar R, a jammer simulator J, and an aircraft mathematical model M. The aircraft mathematical model M has a state evolution step size of 10ms and includes two state parameters: the position vector P (a continuous state parameter) and the target characteristic file index value RCS (a flag-type state parameter). The jammer simulator J does not have a fixed simulation time; its simulation logic time is determined based on external instructions. Each instruction updates the logic time and adjusts the jamming gain value G (a continuous state parameter). The HIL radar R has a simulation cycle of 17ms and requires initiating three object interaction processes: acquiring P, acquiring RCS, and acquiring G.

[0055] like Figure 2 The simulation process shown illustrates the object interaction process where object R obtains P from object M. This is an asynchronous, continuous variable-type state parameter object interaction process with equal step size. The specific process is as follows: When the global logic time T = 0, object R reads P(0ms) from object M. Since the queue length is 1, P(0ms) is directly used for simulation calculation. When the global logic time progresses to T = 17ms, object R reads P(10ms) from object M, establishes a queue containing P(0ms) and P(10ms), uses cubic spline fitting to obtain P(t), and calculates P(17ms) for simulation calculation. When T = 34ms, object R reads P(30ms) from object M, establishes a queue containing P(0ms)... A queue of P(10ms), P(30ms) is created, and a new P(t) is obtained by cubic spline fitting. The calculated P(34ms) is then used for simulation calculation. When T = 51ms, object R reads P(50ms) from object M and creates a queue containing P(0ms), P(10ms), P(30ms), and P(50ms). A new P(t) is obtained by cubic spline fitting, and the calculated P(51ms) is then used for simulation calculation. This process is repeated, eliminating earlier P values ​​from the queue and updating to obtain a new P(t) for iterative calculation.

[0056] In the simulation process, the object interaction process in which object R obtains the RCS from object M is the object interaction process of the flag class state parameter. The specific process is as follows: When the global logic time T = 0, object R reads the RCS (0ms) of object M and directly uses RCS (0ms) for simulation calculation; when the global logic time advances to T = 17ms, object R reads the RCS (10ms) of object M and directly uses RCS (10ms) for simulation calculation.

[0057] like Figure 2The simulation process shown is an object interaction process in which object R obtains G from object J. This process is an asynchronous, variable-step, continuous variable-class state parameter object interaction process. The specific process is as follows: When the global logic time T = 34ms, object R reads G(34ms) from object J. Since object J has not received the instruction, G(34ms) = 0, and object R does not consider G in its calculation process. When the global logic time T = 50ms, object J receives the instruction to set G to 10dB. When the global logic time T = 51ms, object R reads G(50ms) from object J and uses this value for calculation. When T = 57ms, object J receives the instruction to set G to 11dB. Subsequently, when T = 65ms, object J receives the instruction to set G to 14dB. When T = 68ms, object R reads G(57ms) and G(65ms) from object J, establishes queues G(50ms), G(57ms), and G(65ms), performs cubic spline fitting to obtain G(t), and interpolates G(68ms) for simulation calculation of object R. This process continues, eliminating earlier G values ​​in the queue and updating to obtain new G(t) for iterative calculation.

[0058] The contents not described in detail in this specification are existing technologies known to those skilled in the art.

Claims

1. A variable step size LVC simulation method with separate control of simulation step size and interactive step size, characterized in that... include: Multiple LVC simulation objects are interconnected through a simulation communication network supporting TENA-like middleware protocols. The entire system simulation is driven by global logical time control, and this simulation progression includes object interaction and state evolution processes. The object interaction process is executed based on periodic or event triggers according to the current LVC simulation object interaction requirements. According to the logical time progression characteristics of the LVC simulation objects participating in the object interaction process, it is divided into three modes: fully synchronous, equal step asynchronous, and variable step asynchronous. Various state parameters are processed separately to obtain the external field parameter values ​​required for the current logical time. The state evolution process is triggered periodically according to the evolution of the LVC simulation object. The state parameters of the LVC simulation object at the next simulation moment are calculated based on the current state parameters and external field parameters of the LVC simulation object.

2. The variable step size LVC simulation method according to claim 1, wherein the simulation step size and interactive step size are controlled separately, is characterized in that: State parameters are parameters used to describe the state of the LVC simulation object itself; External parameters refer to parameters that come from external LVC simulation objects and affect the state parameters of the LVC simulation objects through object interaction processes.

3. The variable step size LVC simulation method according to claim 1, wherein the simulation step size and interactive step size are controlled separately, is characterized in that: State parameters include: continuous variable state parameters and flag state parameters. Continuous variable state parameters are physical quantities that can produce related changes, including position, velocity, and attitude; Flag-type status parameters are parameters related to device status information, including: whether the device is powered on and whether the device is working.

4. The variable step size LVC simulation method according to claim 1, wherein the simulation step size and interactive step size are controlled separately, is characterized in that: Full synchronization means that all LVC simulation objects are sampled at equal steps and the steps are equal; Equal step asynchronous means that all LVC simulation objects advance at equal steps, but the simulation steps of each object are inconsistent. Variable step size asynchronous means that at least one of the simulation objects adopts a variable step size advancement strategy.

5. The variable step size LVC simulation method according to claim 3, wherein the simulation step size and interactive step size are controlled separately, is characterized in that: For the interaction process of continuous variable class state parameters in both fully synchronous and equal-step asynchronous cases, the specific details are as follows: Triggered based on the judgment conditions of logical time or state parameters during the main simulation loop of the LVC simulation object; As the logical time of the main simulation loop progresses, the corresponding state parameters of the LVC simulation object that undergoes the interaction process of the object are read. Establish a state parameter queue with a length of n parameters, and the sampling logic interval is the longer main simulation loop logic time step in the LVC simulation object of this interaction process; Perform a cubic spline curve fitting each time the state parameter queue changes; When updating the external field parameters during each LVC simulation object interaction process, the coefficients of the cubic spline curve are called for interpolation / extrapolation calculation to obtain the external field parameter values ​​required for the current logic time. When the LVC simulation object interaction process meets the completion conditions, the current object interaction process exits. The value of n is set according to the characteristics of the interaction between the parameter and the object; when the queue length is 1, the cubic spline curve fitting condition is not met, and the first value is directly taken as the value of the external field parameter required for the current logical time.

6. The variable step size LVC simulation method according to claim 3, wherein the simulation step size and interactive step size are controlled separately, is characterized in that: For the interaction process of continuous variable class state parameters in the case of asynchronous operation with variable step size, the specific steps are as follows: Triggered based on the judgment conditions of logical time or state parameters during the main simulation loop of the LVC simulation object; As the logical time of the main simulation loop progresses, the corresponding state parameters of the LVC simulation object that undergoes the interaction process of the object are read. Establish a state parameter queue with a length of m parameters. When reading the value of a state parameter, if the corresponding state parameter of the LVC simulation object that is subject to the interaction process of the object has not changed, then update one state parameter in the parameter queue. The logical timestamp of the state parameter is the logical time when the value of the state parameter is read. Otherwise, the number of parameters updated is the number of times the state parameter has changed, and the timestamp of the corresponding state parameter is the logical time when the parameter changes. Perform a cubic spline curve fitting each time the state parameter queue changes; Each time the external field parameters are updated during the interaction process of this object, the coefficients of the cubic spline curve are called for interpolation or extrapolation calculation to obtain the external field parameter values ​​required for the current logical time. When the interaction process of the object meets the completion conditions, the current interaction process of the object will exit. The value of m is set according to the characteristics of the interaction between the parameter and the object; when the queue length is 1, the cubic spline curve fitting condition is not met, and the first value is directly taken as the value of the external field parameter required for the current logical time.

7. The variable step size LVC simulation method according to claim 6, wherein the simulation step size and interactive step size are controlled separately, is characterized in that: In the object interaction process of continuous variable state parameters under asynchronous conditions with variable step size, a cache queue needs to be established on the side of the LVC simulation object that receives the interaction for the state parameters involved in the object interaction process, to record the changed state parameter values ​​and the corresponding timestamps.

8. The variable step size LVC simulation method according to claim 3, wherein the simulation step size and interactive step size are controlled separately, is characterized in that: The interaction process of objects with flag-type state parameters is as follows: During the main simulation loop of the LVC simulation object, the system is triggered based on the judgment conditions of the logic time or state parameters; the latest value that can be read from the previous logic time is taken as the value of the external field parameter.

Images