JP2001142385A - System and method for distributed simulation - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a system and method for distributed simulation capable of easily reproducing an optional picture seen by a player when simulation is executed at the time of restoring the internal state of simulation. SOLUTION: This system/method is provided with a training environment simulator 2 for simulation, a player terminal 3 for displaying a simulated result in real time on a player 7, a recording and reproducing device 4 for recording and reproducing simulation and an integrated control device 1 for executing integrated control of the whole system. The device 4 has a check point file 410 for retaining the whole contents of simulation at a prescribed time on the basis of a recording and retaining condition 405 previously set by the operator 6 and an event log file 411 for retaining the difference of a simulated state until execution of next retention.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a distributed simulation method and a distributed simulation system,
In particular, the present invention relates to a distributed simulation method and a distributed simulation system in which one or more players share the same virtual training environment while training screens on different display terminals to train a target task.

[0002]

2. Description of the Related Art FIG.
No. (Simulator that can be reproduced arbitrarily)
A block diagram showing a conventional simulation device that can quickly reproduce a saved state at an arbitrary time by periodically saving the entire internal state of the simulation and the operation history of the operator to a file during the execution of the simulation. It is.

In FIG. 1, reference numeral 500 denotes a data storage control device for controlling a recording operation of a state data recording device and an operation input data recording device, which will be described later; Recording device, 50
Reference numeral 2 denotes an operation device that is operated by an operator to generate operation input data, 503 denotes a simulation device that executes a simulation operation in accordance with the operation input data, 504 denotes a display device that displays various processes and simulations, and 505 denotes operation input data. Is recorded in the operation input data recording device.

The operation will be described. At the time of simulation, the simulation operation device 503 executes a simulation operation based on each data from the operation device 502 operated by the operator, and periodically outputs a synchronization signal at a minimum operation interval. The data storage control device 500 receives the synchronization signal, executes the processing program stored in the device, activates the state data recording device 501 and the operation input data recording device 505, and stores the data in the simulation operation device 503. Is recorded. After the simulation is completed, when the reproduction start time is input from the operator to the data storage control device 500, the simulation processing device 503 sets the state data corresponding to the input reproduction start time as an initial value, and sets the operation input data to Input as time-series input data and reproduce simulation.

[0005] When a distributed simulation system for training is constructed using this conventional simulation apparatus, in order to reproduce a state at an arbitrary logic time, the data recording cycle based on the synchronization signal output from the simulation operation apparatus 503 is required. What is necessary is just to make it small enough.

[0006]

In the conventional simulation apparatus as described above, when a distributed simulation system for training is constructed using the same, it is necessary to restore the internal state of the simulation at an arbitrary logic time after the training. However, it is necessary to make the data recording cycle sufficiently short, but in that case, the processing required for storage affects the execution of the simulation, so that the data recording cycle cannot be made shorter than a certain value. There was a point.

Another problem when the data recording cycle is made sufficiently short is that the amount of data to be recorded becomes large.

For these reasons, the recording cycle of data cannot be shortened. Therefore, if the recorded result is reproduced in a time series and displayed as an animation, the state of the time that has not been saved cannot be displayed or interpolation processing is performed. For,
There is a problem that it is difficult to reproduce a screen similar to the screen seen by the trainee (player) during training.

SUMMARY OF THE INVENTION The present invention has been made to solve such a problem, and determines a "recording condition" for determining whether or not to record, and a timing for writing to a file, with respect to an internal state of a simulator to be recorded. A "save condition" can be set, and only a difference between a checkpoint file that saves all system states at a certain logical time and a change in the internal state of the system after the checkpoint file is saved is saved. An event log file is provided, and by using these two types of files, during the simulation, the effect of the recording and saving process on the simulation process is small, and the amount of data to be recorded can be kept low. The reproduction process can be performed efficiently, and the reproduction of the screen similar to the screen seen by the player at the time of the simulation is performed reliably and in a satisfactory manner. And to provide a distributed simulation method and system capable to.

[0010]

SUMMARY OF THE INVENTION The present invention provides a simulating means for simulating a virtual environment, one or more display means for displaying a result simulated by the simulating means in real time, and a means for recording and storing the contents of the simulation. Record storage condition setting means for setting the conditions of the above, checkpoint file means for storing all the contents of the simulation in the checkpoint file based on the conditions, and from the time when the checkpoint file is stored to the time when the next storage is performed Using the event log file means for storing only the difference in the change of the simulation content between the event log file and the saved checkpoint file and event log file, reproducing the simulation content at an arbitrary time Playback means for displaying on the display means and integrated management means for performing integrated management of the entire system; Is an example was distributed simulation system.

Further, the integrated management means designates the reproduction start time and the stored checkpoint file and event log file for the reproduction means and the display means, and synchronizes the data output by the reproduction means with the logical time. And a control mechanism for displaying on the display means.

A recording condition setting unit for setting a recording condition for determining which of the simulated contents is to be recorded in the checkpoint file and the event log file, respectively; And a storage condition setting unit for setting storage conditions for determining the timing of writing to the event log file.

In addition, the apparatus further comprises a checkpoint buffer for temporarily storing data before storing the data in the checkpoint file, and an event log buffer for temporarily storing data before storing the data in the event log file. The simulated status change at each time is written to the above-mentioned checkpoint buffer and simultaneously written to the event log buffer, and the status change of the data of the same attribute at the next time is overwritten by the checkpoint buffer, and the event log buffer Let's write additional
By repeating the above processing, the data is written to the event log file when the recording storage condition for the event log file is satisfied, and is written to the checkpoint file when the recording storage condition for the checkpoint file is satisfied.

An identifier indicating a changed state and a value indicating a simulated new state are stored in the event log file at each time.

In the event log file, a value indicating a past state at a time immediately before the simulation is stored for each time.

A reproducing unit for receiving a reproduction start time designated on the integrated management unit; a retrieval unit for retrieving a checkpoint file stored at the latest time before the start time; Using the identifier of the event log file after the save time of the searched checkpoint file and the value indicating the new state, the state of the start time is reproduced by reproducing the simulated state in the forward direction, and the reproduction start state And a playback start state creation unit for creating a time synchronization signal output from the integrated management means,
A reproduction unit that transmits a change in the content of the simulation at each time from the event log file and performs reproduction by displaying the change on the display unit that also operates in synchronization with the time synchronization signal from the integrated management unit. I have.

A reproducing unit for receiving a reproduction start time specified on the integrated management unit; a retrieval unit for retrieving a checkpoint file stored at the latest time after the start time; Using the identifier of the event log file before the save time of the searched checkpoint file, the value indicating the new state, and the value indicating the past state, the state of the start time is reproduced by reproducing the simulated state in the reverse direction. The reproduction start state creation unit that reproduces and creates the reproduction start state, and the contents of the simulation at each time from the identifier of the event log file and the value indicating the new state in synchronization with the time synchronization signal output from the integrated management means. And a reproduction unit for reproducing by transmitting the change of the information and displaying the same on a display unit operating in synchronization with the time synchronization signal from the integrated management unit. Eteiru.

Also, a plurality of checkpoint file means and event log file means are provided in the same number, respectively, and different recording conditions are given to each of them by the recording and storage condition setting means to store the contents of the simulation.

Also, one reproducing means is provided for each pair of a plurality of checkpoint file means and event log file means, and the reproducing means are synchronized and stored under different recording conditions. By reproducing the checkpoint file and the event log file, the same virtual environment is reproduced.

Also, by setting different storage periods as storage conditions by the storage setting condition setting means for each of the plurality of checkpoint file means and event log file means provided, the recording conditions having different storage timings are different. The data is saved.

Further, the present invention provides a simulation step of simulating a virtual environment, a display step of displaying a result simulated by the simulation step on one or more display devices in real time, and a step of recording and storing the contents of the simulation. A record storage condition setting process for setting conditions, a checkpoint file process for storing all the contents of the simulation in a checkpoint file based on the conditions, and a checkpoint file process from the time when the checkpoint file is saved to the time when the next save is performed. An event log file process of storing only a difference between changes in the contents of the simulation between the event log files, and a display device that reproduces the contents of the simulation at an arbitrary time by using the stored checkpoint file and the event log file. And the integrated pipe for outputting the time synchronization signal for integrated management of the entire system A step, a distributed simulation method with a.

In the integrated management step, the reproduction start time and the stored checkpoint file and event log file are designated for the reproduction step and the display step, and the data output in the reproduction step is synchronized with the logical time. And a control mechanism for displaying on the display device.

The recording and storage condition setting step includes a recording condition setting step of setting recording conditions for determining which of the simulated contents is to be recorded in the checkpoint file and the event log file, respectively. And a storage condition setting step of setting storage conditions for determining the timing of writing to the event log file.

A checkpoint buffer step for temporarily storing data before storing the data in the checkpoint file;
An event log buffering step of temporarily storing data, wherein the simulated state change at each time is written to the checkpoint buffer and simultaneously to the event log buffer, and data of the same attribute at the next time is written. Regarding status changes, overwrite in the checkpoint buffer, write additionally in the event log buffer, repeat the above process, write to the event log file when the record storage condition for the event log file is satisfied, and write to the checkpoint file. When the above record storage condition is satisfied, the data is written to the checkpoint file.

In the event log file process, an identifier indicating a changed state and a value indicating a simulated new state are stored in the event log file at each time.

In the event log file step, the event log file further includes, at each time,
A value indicating a past state at a time immediately before the simulation is stored.

The reproducing step includes a receiving step of receiving a designated reproduction start time, a retrieval step of retrieving a checkpoint file stored at the newest time before the start time, and a retrieved checkpoint file. Using the identifier of the event log file after the save time and the value indicating the new state, the state of the start time is reproduced by reproducing the simulated state in the forward direction,
A playback start state creation step of creating a playback start state;
By synchronizing with the time synchronization signal output in the integrated management process, transmitting the change of the simulated content at each time from the event log file, and displaying the change on the display device which also operates in synchronization with the time synchronization signal And a reproduction step of performing reproduction.

The reproducing step includes a receiving step of receiving a designated reproduction start time, a retrieval step of retrieving a checkpoint file stored at the latest time after the start time, and a retrieved checkpoint file. Using the identifier of the event log file before the save time, the value indicating the new state, and the value indicating the past state, the simulated state is reproduced in the reverse direction to reproduce the state at the start time, and reproduced. A reproduction start state creation step for creating a start state, and a change in the contents of the simulation at each time are transmitted from the identifier of the event log file and the value indicating the new state in synchronization with the time synchronization signal output in the integrated management process. A reproduction step of performing reproduction by displaying on a display device also operating in synchronization with the time synchronization signal. That.

Also, a plurality of checkpoint files and event log files are provided in the same number, and different recording conditions are given to each of them in the recording storage condition setting step to store the simulated contents.

Also, one playback device is provided for each pair of the checkpoint file and the event log file provided in plurals, and the playback devices are synchronized with each other.
By reproducing the checkpoint file and the event log file stored under different recording conditions, the same virtual environment is reproduced.

Also, by designating different storage periods as storage conditions in the storage condition setting step for each of the plurality of checkpoint files and event log files, recording data having different storage timings can be obtained. Will be saved.

[0032]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiment 1 FIG. 1 is a block diagram showing a configuration of a distributed simulation system according to Embodiment 1 of the present invention. In FIG. 1, reference numeral 1 denotes an integrated management device that is operated by an operator and controls each of devices 2 to 4 described below to perform a distributed simulation by emphasizing each other, and 2 denotes a training environment simulation device that performs a simulation simulation of training. 3, a player terminal operated by the player; 4, a recording / reproducing device for recording and reproducing the status of the training environment simulator during training; and 5, a local area network (hereinafter, referred to as a network) connecting these devices 1-4. , LA
N).

In FIG. 1, reference numeral 6 denotes an operator who operates the integrated management apparatus 1, and performs operations of the entire training system such as starting and ending of training, starting and ending of recording, and starting and ending of reproduction. Reference numeral 7 denotes a player who performs training by operating the player terminal 3, sees the status of the virtual training environment on the animation display device 311 on the player terminal 3, and inputs a command using the input device 310. Participate in training.

Each of the devices 1 to 4 is connected via a LAN 5 as shown in the figure, and under the control of the distributed simulation control device 100 on the integrated management device 1, the distributed simulation control device 200 on each of the devices 2 to 4 is controlled. , 300, and 400 to form a distributed simulation system by performing cooperative processing with each other.

Here, in FIG. 1, the training environment simulation device 2, the player terminal 3, and the recording / reproducing device 4 are connected one by one, but as shown in FIG.
A system is configured by connecting one or more devices according to the purpose and operation.

The distributed simulation control function is described in, for example, IEEE 1516 (Draf
t standard for Modeling and Simulation HLA-Feder
ate Interface Specification-, http: //www.sisostds.
org / stdsdev / hla /, 1999), a distributed simulation connection architecture HLA (High Level Arc)
hitecture) specifications or equivalent.

In the HLA specification, each device is RTI
(Run Time Infrastructure). In FIG. 1, the entire distributed simulation control mechanism 100, 200, 300,
400 corresponds to the RTI, and 100 on the integrated management device 1 performs overall management and control.

The application programs on each device include the control programs 101, 201, 301, 401
Each control program also operates under the management of the external device, and functions of communication and synchronization with external devices.

Here, when the distributed simulation control mechanism is used as shown in FIG.
01, 201, 301, and 401 only need to communicate or call functions with the distributed simulation control mechanisms 100, 200, 300, and 400. In this case, it is necessary to be aware of the partner device that performs communication and synchronization. Absent.

On the other hand, when the distributed simulation control mechanism is not used as in the prior art, that is, for example, when the HLA
If the specifications are not used, for example, the control program uses various communication mechanisms prepared by the system such as a TCP / IP socket communication mechanism. In this case, each control program explicitly specifies a communication partner. Since it is necessary to perform data exchange by designating, or to explicitly designate a time synchronization partner, time synchronization is performed, and the structure of the control program becomes complicated. Therefore, it is desirable to use a distributed simulation control mechanism.

Although the HLA specification has been described as an example here, another similar distributed simulation connection specification may be used, and the same applies in that case.

Next, the structure of each of the devices 1 to 4 will be described in more detail.

First, the integrated management device 1 includes a distributed simulation control mechanism 100, a control program 101, an integrated management program 102, and a data area 103. Since the distributed simulation control mechanism 100 has been described above, the description is omitted here. The data area 103 stores various data required by the integrated management device 1.

The integrated management program 102 manages the entire distributed simulation system, and starts / stops / ends the entire system, sets recording / storage conditions for the recording / reproducing device 4, and starts / stops recording by an input operation from the operator 6. This is a program for controlling stop, start / stop of reproduction, and the like.

The control program 101 controls the operation of each application program on the integrated management device 1 as described above. That is, all controls closed in the integrated management apparatus 1 are performed under the control of only the control program 101. If communication or synchronization with another device is required, a request is issued to the distributed simulation control mechanism 100 via the control program 101, and each of the devices 2 to
4 cooperate with each other to control the entire system.

Next, the training environment simulation device 2 includes a distributed simulation control mechanism 200, a control program 201, a training environment simulation program 202, and a data area 20.
3 is comprised.

The training environment simulation program 202 simulates a virtual training environment required for training under the control of the control program 201. The simulation result is written in the data area 203 by the training environment simulation program 202. The time synchronization with other devices and the data transmission are performed by the distributed simulation control mechanism 200.

When the training environment simulating device 2 is composed of two or more devices, each distributed environment of the plurality of training environment simulating devices 2 is used similarly to the time synchronization and data exchange between the devices 1 to 4 described above. The simulation control mechanism 200 synchronizes time and exchanges data between the devices, and simulates one virtual training environment with the plurality of training environment simulation devices 2 as a whole.

The player terminal 3 includes a distributed simulation control mechanism 300, a control program 301, an input control program 302, an animation display program 303,
It comprises a data area 304, an input device 310, and an animation display device 311.

Under the control of the control program 301, the input control program 302 allows the player 7
Process input using. The distributed simulation control 200 synchronizes the time with another device and transmits data according to the input contents, and provides a means for the player 7 to participate in the training environment simulated by the training environment simulation device 2.

The animation display program 30
3 is the player 7 under the control of the control program 301.
The animation display device 311 is provided with a function of displaying the status of the virtual training environment on the animation display device 311.
Simulation information necessary for displaying the screen of the player 7 is notified one by one by the distributed simulation control management mechanism 300 and written into the data area 304.
Therefore, the animation display program 303 may display an animation using only information in the same device 3.

The recording / reproducing apparatus 4 takes a record of the internal state simulated by the training environment simulating apparatus 2 during execution of the training and saves it in a file. A process for reproducing a situation is performed. As shown in the figure, a distributed simulation control mechanism 400, a control program 401, a recording program 40
2, a playback program 403, a data area 404, a checkpoint file 410, and an event log file 411. In the data area 404, a record storage condition 405 is provided as shown in the figure.

The recording program 402 uses the results simulated by the training environment simulation device 2 received by the distributed simulation control mechanism 400 to record each piece of internal information during the execution of the training in accordance with the recording and storage conditions 405, and stores the information in a file. save. Note that the record storage conditions 405 are set by the operator 6 via a network.

That is, at the time of recording execution, the operator 6 who operates the integrated management device 6 before the execution sets the recording preservation conditions 405 stored in the data area 404 via the network. It is assumed that the recording program records and saves according to the storage condition 405. As the recording / storing condition 405, for the internal state of the simulation to be recorded, “recording conditions” for determining which is to be recorded and “storing conditions” for determining the timing of writing to a file can be set.

The result of recording and storage includes a checkpoint file 410 for storing all the internal states of the simulation at a certain logical time, and a checkpoint file 41.
0 and then the next checkpoint file 410
Is stored in the event log file 411 that stores only the difference between changes in the internal state until the file is saved.

The reproduction program 403 restores the internal state of the simulation using the past recording file, and the result is transmitted to the other apparatuses 1 to 3 via the distributed simulation control mechanism 400.

Next, the operation according to the above configuration will be described. First, the basic processing contents and data flow in the distributed simulation system for training will be described. For simplicity of explanation, we will use a simulation system for training air traffic control as an example. FIGS. 2 and 3 show a training distributed simulation system in which a part of FIG. 1 is simplified for explaining the operation.
FIG. 2 shows the time of training execution, and FIG. 3 shows the time of training regeneration.

The system shown in FIGS. 2 and 3 comprises an integrated management device 1, a training environment simulation device 2, a player terminal 3, and a recording / reproducing device 4.

The contents of the training are as follows.
Above, simulate the motion of an aircraft flying in the training airspace. The player 7 looks at the animation display of the aircraft in flight displayed on the screen of the player terminal 3,
Training of air traffic control instructions shall be conducted by sending air traffic control instructions to aircraft.

First, the operation in the case of performing training while recording will be described with reference to FIG.

Before starting the training, the operator 6 in FIG.
By operating the integrated management device 1, conditions for recording and storing training are set. The recording storage conditions 405 include four checkpoint file recording conditions 4051 (not shown here, see FIG. 7; the same applies hereinafter), checkpoint file storage conditions 4052, event log file recording conditions 4053, and event log file storage conditions 4054. There are types. Each recording / storing condition is stored in the recording / reproducing device 4 (reference numeral 405 in FIG. 1). A specific example of each condition will be described later.

At the time of training execution, the operator 6 operates the integrated management device 1 to start execution of each device 2, 3, 4. When the execution is started, the control programs (101, 201, 301, and 401 in FIG. 1) on each device and the distributed simulation control mechanism (100, 200, 300, and
400) operates as follows.

Here, in order to make the flow of data easy to understand, FIGS. 4 and 5 are time charts showing the flow of data between the devices. Hereinafter, description will be made with reference to FIG. 2, FIG. 4 and FIG.

On the training environment simulation device 2, as shown in FIG. 4, among the internal states of the simulation simulated by the training environment simulation program 202 on the device 2,
The control program 201 is notified of a list of data that can be provided to the outside of the control program 201 (1000).
01 declares to the distributed simulation control mechanism 200 (1001). Then, the distributed simulation control mechanism 200 broadcasts to the distributed simulation control mechanisms 100, 300, and 400 on all the constituent devices 1, 3, and 4 (1002, 1003, and 1004).

In the HLA specification, the communication 1001 is called “publish”, and is defined in the standard specification as a data for each device to declare to the distributed simulation control mechanism data that can be released to other devices. ing.

In this example, the distributed simulation control mechanism 200 is different from the other distributed simulation control mechanisms 10.
Broadcast to 0, 300, 400, but 1
001 communication is first sent to the distributed simulation control mechanism 100 of the integrated management apparatus 1, and the same applies even if the distributed simulation control mechanism 100 broadcasts to the distributed simulation control mechanisms 300 and 400 of the remaining other apparatuses 3 and 4. That is.

Further, the training environment simulation program 202 notifies the control program 201 of a list of data that the apparatus 2 desires to receive among the internal states of the simulation simulated on the other apparatuses 1, 3, and 4. (1005),
The control program 201 declares to the distributed simulation control mechanism 200 (1006). Then, the distributed simulation control mechanism 200 transmits all the constituent devices 1, 3, 4
Simulation control mechanisms 100, 300,
Broadcast to 400 (1007, 1008,
1009).

In the HLA specification, the communication 1006 is called a “subscribe”, and among the data obtained from another device, the data which the device is interested in is declared to the distributed simulation control mechanism as a standard specification. Stipulated.

After completion of the above communication, the training environment simulation device 2 notifies the control program 201 by the training environment simulation program 202 that processing at the next time is possible (1010). 201 notifies the distributed simulation control mechanism 200 (1011).
Then, the distributed simulation control mechanism 200 notifies the distributed simulation control mechanism 100 on the integrated management device 1 that manages the time synchronization of the entire distributed simulation (1012).

After transmitting the signal 1011, the control program 201 waits for a notice of permission to proceed to the next time. The distributed simulation control mechanism 200 continues to cooperate with the distributed simulation control devices 100, 300, and 400 on the other devices 1, 3, and 4.

In the HLA specification, the communication 1011 is called a "time advance request".

On the player terminal 3, a list of internal states of the simulation necessary for displaying on the device 3 is
The control program 301 notifies the distributed simulation control mechanism 300 (1013, referred to as "subscribe" in the HLA specification). Broadcast to the distributed simulation control mechanisms 100, 200, 400 (1014, 1015, 1016).

Further, among the input commands of the player 7,
A list of commands that can be provided to the outside of the device 3 is provided by the control program 301 to the distributed simulation control mechanism 3.
Notification to 00 (1017, "open" in HLA specifications)
) And notify that processing at the next time is possible (1
021, which is referred to as “time progress request” in the HLA specification). Note that the distributed simulation control mechanism 300
When the communication of 017 is received, all the constituent devices 1, 2, 4
Simulation control mechanisms 100, 200,
400 (1018, 1019, 1
020) Also, upon receiving the communication of 1021, it notifies the distributed simulation control mechanism 100 on the integrated management device 1 that manages the time synchronization of the entire distributed simulation (1022).

On the recording / reproducing apparatus 4, as shown in FIG. 5, according to the recording conditions given by the operator 6 described above, a list of the internal states of the simulation to be recorded is distributed by the control program 401 to the distributed simulation. Notification to the control mechanism 400 (1023, HLA
In the specification, this is called “subscription”), and a notification that processing at the next time is possible (1027, “time advance request” in the HLA specification) is made. Also in these cases, similarly to the above, upon receiving the communication of 1023, the distributed simulation control mechanism 400 broadcasts it to the distributed simulation control mechanisms 100, 200, and 300 on all the constituent devices 1, 2, and 3. (1024, 1025, 1026) When the communication of 1027 is received, it notifies the distributed simulation control mechanism 100 on the integrated management device 1 that manages the time synchronization of the entire distributed simulation (1028).

Next, the flow of data between the devices will be described. After the start of the training, the distributed simulation control mechanism 100 of the integrated management apparatus 1 notifies each of the apparatuses 2, 3, and 4 of the next logical time at which the time can advance (1029, 103).
0, 1031, referred to as “time advance permission” in the HLA specification). Each of the received devices 2, 3, and 4 performs processing until the permitted logical time (2000, 2001, 200
2) After the time advance request, wait for the next time advance permission to arrive.

The training environment simulation apparatus 2 simulates the motion of the aircraft for one time in the processing indicated by 2003, and notifies the distributed simulation control mechanism 200 of the changed state as a result of the simulation (1032, HLA). According to the specification, "status change notification" (update attribute value) is issued.

In this example, if it is the player terminal 3 and the recording / reproducing device 4 that want to know this state change, the distributed simulation control mechanism 200 sends the state to the distributed simulation control mechanisms 300 and 400 of the respective devices 3 and 4. Notify that there has been a change (1033 and 1034, "status reflection notification" (reflect attribute value) in the HLA specification)
I do.

When the player terminal 3 receives the state change,
In the process indicated by 004, the situation is displayed as an animation on the animation display device (311 in FIG. 1).

As described above, the training environment simulating apparatus 2 simulates the training environment every time as indicated by the arrow 150 shown in FIG. 2, and the result is notified to the player terminal 3 and the animation-displayed screen is displayed on the player terminal 3. 7 can be seen.

Similarly, the recording / reproducing apparatus 4 performs the state recording processing in the processing indicated by 2005, and when the storage conditions are satisfied, the checkpoint file (reference numeral 4 in FIG. 1)
10) or an event log file (reference numeral 41 in FIG. 1)
Save in 1).

Here, the checkpoint file 410
Is a file storing all the states specified in the checkpoint file recording conditions among the internal states of the training environment simulation device 2 at that time.

The event log file 411 is
This is a file in which a change record of the internal state of the simulation between the time when the checkpoint file 410 is stored and the time when the next checkpoint file 410 is stored is stored.

Next, the operation in the case of regenerating after the training is performed will be described with reference to FIG. After the training is executed and before the reproduction is started, the operator 6 in FIG. 3 operates the integrated management apparatus 1 to input a logical time at which the reproduction is started, and the reproduction start time is written in the recording / reproducing apparatus 4. When the reproduction start time is T (start), for example, T (start)>
= Search the checkpoint file 410 stored at the time T (chk_saved) satisfying the condition of T (chk_saved). Next, the event log file 411 saved from the time T (chk_saved) is searched. From these two files, the internal state of the simulation at time T (start) is restored.

Next, the recording / reproducing apparatus 4 uses the restored internal state of the simulation to perform the state update by the recording / reproducing apparatus 4 instead of the state updating by the training environment simulation apparatus 2 at the time of training execution. A list of data that can be provided to the outside of the device 4 is provided by the control program 401 so that the distributed simulation control mechanism 400
(1035, "open" in HLA specifications)
). Next, the distributed simulation control mechanism 400
Broadcasts to the distributed simulation control mechanisms 100, 200, and 300 on all the constituent devices 1, 2, and 3 (1036, 1037, 1038).

On the other hand, on the player terminal 3, a list of internal states of the simulation necessary for displaying on the device 3 is stored in the control program
To the distributed simulation control mechanism 300 (1039, called a subscription in the HLA specification). Next, the distributed simulation control mechanism 300 controls the distributed simulation control mechanism 10 on all the constituent devices 1 and 2.
Broadcast to 0,200 (1040,104
1).

When the above-mentioned preprocessing is completed on each device (1, 3, 4), the integrated management device 1 notifies the devices 4 and 3 of the next executable time (1042, 1043,
In the HLA specification, this is called "time advance permission." Then, on the recording / reproducing apparatus 4, the state of the simulation is restored up to the permitted time by using the checkpoint file and the event log file in the processing of 2006,
The internal state having the state change notifies the distributed simulation control mechanism 400 of the state change (1043,
"Status change notification" in the HLA specification).

Next, the distributed simulation control mechanism 40
0 notifies the distributed simulation control mechanism 300 of the player terminal 3 of this state change (1044, HL
In the A specification, this is referred to as a “state reflection notification”), and the player terminal 3 that has received the notification performs an animation display on the display terminal in the process of 2007.

During the execution of the reproduction, it is assumed that the reproduction can be paused or another logical time can be specified again by the operation of the operator 6 on the integrated management device 1.

Recording / reproducing that can be set for the recording / reproducing device 4
For example, the following can be considered as storage conditions.

1) Recording condition of checkpoint file: A state selected by the operator 6 and desired to be displayed during reproduction.

2) Checkpoint file storage conditions-When an event selected by the operator 6 in advance occurs. -When the system load is low. -When the amount of memory used for recording checkpoint files exceeds a certain limit.・ When the total number of recorded events exceeds a certain number.

3) Event log file recording conditions: When the amount of change from the previous time exceeds a certain ratio.

4) Event log file storage conditions: Every hour. -When the number of recorded events since the last save to a file exceeds a certain number. -When the amount of memory used for event log recording exceeds a certain limit.

Various other recording and storage conditions can be considered.
This apparatus is characterized in that various recording and storage conditions can be handled by designating conditions that can be parameterized or formulated.

As described above, according to the present invention, for the internal state of the simulator to be recorded, "recording conditions" for determining whether or not to record and "storing conditions" for determining the timing of writing to a file are defined. A checkpoint file that stores all system states at a certain logical time; and an event log file that stores changes in the internal state of the system after the checkpoint file is stored. Using the type of file, there is an effect that during the training, there is little effect on the simulation processing due to the record preservation processing, and after the training, the reproduction processing can be performed efficiently.

Embodiment 2 FIG. 7 is an enlarged view showing the configuration of the recording / reproducing device 4 of the present invention. In the data area 404 in the recording / reproducing apparatus 4, a recording / storing condition 405, a checkpoint buffer 406, and an event log buffer 407 are provided as shown in the figure. For simplification of the drawing, illustration of the control program 401 and the like shown in FIG. 1 is omitted here. The record storage conditions 405 include:
As described in the first embodiment, the checkpoint file recording condition 4051, the checkpoint file storage condition 4052, and the event log file recording condition 4053.
And event log file storage conditions 4054. The other configuration is the same as that described in the first embodiment, and a description thereof will not be repeated.

The checkpoint buffer 406 is used to temporarily store the internal state of the simulation in which the state has changed every time. When the checkpoint file storage condition 4052 is satisfied, the checkpoint buffer 410 stores the checkpoint buffer 410. Written out. Note that one entry in the buffer is provided for one internal state of the simulation, and the latest state can be stored every time by overwriting.

[0098] The event log buffer 407 stores
The point used to temporarily store the internal state of the simulation in which the state has changed is the same as that of the checkpoint buffer 406, but there is a plurality of entries at each time for one internal state of the simulation. The points are different. The event log buffer 407 is written to the event log file 411 when the event log storage condition 4054 is satisfied.

FIG. 8 shows a checkpoint buffer 406.
An example of the configuration will be described. As shown in the figure, the checkpoint buffer 406 includes a header portion 40600 indicating the logical time of the simulation to be saved, and a main body portion 40 storing the set of the identifier, attribute, and value of the object to be saved.
610. The header 40600 includes data on the logical time and the pointer as shown in the figure.

FIG. 9 shows an example of the configuration of the event log buffer 407. As shown in the figure, the event log buffer 4
07 includes a header portion 40700 for storing the time at which the event occurred, and a main body portion 40710 for storing a set of an identifier, an attribute, an event type, and a changed value of the object at which the event occurred.

As in the conventional example described above, when there is no checkpoint buffer, execution of the entire system needs to be temporarily interrupted in order to collect the internal state of the simulation when writing to the checkpoint file.
If interrupted at a significant point during training, there is a significant impact on training.

However, according to the present invention, the latest state is stored in the checkpoint buffer by overwriting every time by providing the checkpoint buffer. Since it is not necessary to collect the internal state, and it is possible to simply perform writing to a file, there is no need to interrupt the system, and the effect on training is small. Furthermore, by setting the storage conditions efficiently, it is possible to further avoid adverse effects on training.

Embodiment 3 In this embodiment, an example of an event type stored in the main body portion 40710 of the event log buffer 407 described in the second embodiment will be described. The table of 41110 in FIG. 10 shows an example of the type of event to be recorded in the event log. As can be seen from Table 41110, NEW, MO
An event log is composed of three types of events D and TRM. The meaning of each event type is as follows.

NEW: An object indicated by an identifier and an attribute has been generated. MOD: The value of the object indicated by the identifier and the attribute has been changed. Save only the changed value. TRM: The object indicated by the identifier and the attribute has disappeared.

[0105] Reference numerals 41111 and 41112 in FIG. 10 show examples of writing to the event log buffer and the event log file. One value is stored for each attribute, and the value of the new state after the change is stored. This allows
In this embodiment, the log file can be played back with the time advanced in the forward direction (time increasing direction) during playback.

Embodiment 4 In this embodiment, another example of the event type stored in the main body portion 40710 of the event log buffer 407 described in the second embodiment will be described. The table of 41120 in FIG. 11 shows an example of the type of event to be recorded in the event log. As can be seen from Table 41120, NEW, MO
An event log is composed of three types of events D and TRM. The meaning of each event type is as follows.

NEW: An object indicated by an identifier and an attribute has been generated. MOD: The value of the object indicated by the identifier and the attribute has been changed. Save the old and new values. TRM: The object indicated by the identifier and the attribute has disappeared.

[0108] Reference numerals 41121 and 41122 in Fig. 11 show examples of writing to the event log buffer and the event log file. The value to be stored for each attribute is a pair of the value before change and the value after change. This makes it possible to play the log file forward in the reverse direction (time decreasing direction) during playback. Become.

Embodiment 5 FIG. FIG. 12 shows an example of a checkpoint file and an event log file of the present invention created at each logical time during recording. 12, reference numerals 4101 to 4104 denote checkpoint files, 4111 to 4113 denote event log files, 10
Are seven simulation target objects a to g.

Here, checkpoint file 410
1, 4102, 4103, and 4104 are created at logical times T0, T20, T30, and T40, respectively. The event log file stores the logical time T
20, T30 and T40, respectively, T0 to T2
0, T20 to T30, and the state change of each object during T30 to T40, respectively.
111, 4112, and 4113. Here, it is assumed that the configuration of the event and the configuration of the event file conform to FIG.

Here, at the time of executing reproduction, the operator 6
It is assumed that an input is made to start reproduction from time 0. At this time,
The recording / reproducing apparatus 4 searches the stored checkpoint file for a checkpoint file older than the time T10, and as a result, selects the file 4101.

Next, the event log file whose recording has been started from time T10 is searched, and as a result, 4111 is selected. Subsequently, the time T is calculated using 4101 and 4111.
The internal state of the simulation at 10 is restored, and the subsequent reproduction processing is started.

As described above, according to the present invention, at a time satisfying a predetermined condition, the checkpoint file is stored, the state at that time can be restored, and the event log file can be used. Since the state at an arbitrary time after this time can also be restored, an effect is obtained that a screen similar to the screen seen by the player 7 during training can be easily reproduced.

Embodiment 6 FIG. In this embodiment, the configuration of the event and the configuration of the event file in FIG. 12 are based on FIG.

Here, it is assumed that the operator 6 inputs the reproduction from the time T15 in FIG. 12 during the reproduction.
At this time, the recording / reproducing device 4 searches the stored checkpoint file for the checkpoint file having the smallest difference in the logical time from the time T15, and as a result, the file 4102 at the time T20 is selected.

Next, the event log file including the time T15 is searched, and as a result, the event log file 4111 of T0 to T20 is selected. Subsequently, the internal state of the simulation at the time T20 in the checkpoint file 4102 is restored, and then the state is restored in the reverse direction one by one using the event log file 4111.

In this embodiment, an example has been described in which restoration is performed in the reverse direction. However, the present invention is not limited to this case.
If the checkpoint file having the smallest logical time difference is a checkpoint file older than the time T15 at which reproduction is to be started, the state is restored in the forward direction one time at a time.

As described above, according to this embodiment of the present invention, the checkpoint file is saved at a time satisfying a predetermined condition, and the state at that time can be restored. Since the log file can also be used to restore the state at any time before and after this time, a screen similar to the screen that the player 7 saw during training can be easily reproduced, and further, convenience is provided. The effect that the property is improved is obtained.

Embodiment 7 FIG. FIG. 13 shows a system configuration example when two recording / reproducing devices are connected. In the figure, reference numerals 4a and 4b denote two recording / reproducing apparatuses, which are completely identical in function. Checkpoint files 410a and 410b and event log files 411a and 411b are recorded and stored, respectively. For example, it is assumed that recording conditions are different as follows.

1) Recording condition of checkpoint file 410a and event log file 411a of recording / reproducing apparatus 4a: identifier number of object to be recorded / stored is 0
2) Checkpoint file 41 of recording / reproducing device 4b
0b and the recording condition of the event log file 411b: the identifier number of the object to be recorded / stored is 100 or later

As described above, according to the present embodiment, different information is recorded for each recording / reproducing device 4, so that the data to be stored on each recording file is different from the case where one recording / reproducing device is used. The effect is obtained that the amount can be reduced and the load of processing required for storage at the time of recording can be dispersed.

Embodiment 8 FIG. It is assumed that two recording / reproducing devices are connected as shown in FIG. 13 and recording is performed under different recording conditions such as the recording conditions 1) and 2) described above on each recording / reproducing device.

Two checkpoint files and two event log files are recorded according to the recording / storing conditions.

Before starting the execution of the reproduction, the two recording / reproducing apparatuses make preparations for the execution of the reproduction to the distributed simulation apparatus using the method described above, based on the files recorded and stored.

When the integrated management device 1 instructs the two recording / playback devices to start playback, the distributed simulation management mechanism on each device synchronizes the time between the devices and performs the playback process. .

The state of the simulation during the reproduction can be seen on the player terminal 3. The reproduction is executed on one recording / reproducing apparatus and the reproduction is executed on two apparatuses. There is no difference in the display to be reproduced, and it is recognized on the player terminal 3 as if it is being executed on one device.

As described above, the files recorded and stored by the plurality of recording / reproducing apparatuses are recorded on the same system by simultaneously reproducing the plurality of recording / reproducing apparatuses and operating in synchronization with the operation of the distributed simulation management mechanism. And the processing for reproduction can be load-balanced.

Embodiment 9 FIG. As shown in FIG. 13, two recording / reproducing devices are connected, and unlike the above embodiment, the same recording conditions are used on each recording / reproducing device.
However, the storage conditions are different, for example, as follows.

3) Recording conditions for the checkpoint file 410a and the event log file 411a of the recording / reproducing apparatus 4a: the recording start time is from logical time zero, and the recording interval is every 100 logical times.

4) Recording conditions of the checkpoint file 410b and the event log file 411b of the recording / reproducing apparatus 4b: the recording start time is 50 logical times, and the recording interval is every 100 logical times.

Then, on each recording / reproducing device, the checkpoint file is stored every 100 logical times, but when the data of both recording / reproducing devices are combined, 5 points are obtained.
A checkpoint file for every 0 logical time is available.

Accordingly, the effect of shortening the time required for cueing in the cueing process during reproduction can be obtained as compared with the case of a single recording / reproducing apparatus.

[0133]

According to the present invention, a simulating means for simulating a virtual environment, at least one display means for displaying a result simulated by the simulating means in real time, and conditions for recording and saving the contents of the simulation are set. A record storage condition setting means, a check point file means for storing all the contents of the simulation based on the conditions in a check point file, and a simulation from a time when the check point file is saved to a time when the next save is performed. Using the event log file means for storing only the difference in the change of the contents of the event log file in the event log file, and the contents of the simulation at an arbitrary time by using the saved checkpoint file and event log file, and displaying them on the display means Playback means for performing
Integrated management means for performing integrated management of the entire system;
Since it is a distributed simulation system equipped with a simulator, it is possible to set recording and preservation conditions for recording and preserving the internal state of the simulator to be recorded, and checkpoints for preserving all system states at a certain logical time A file and an event log file for saving changes in the internal state of the system after saving the checkpoint file. Since saving and playback are performed using these two types of files, the frequency of saving to a file can be reduced. During the training, there is little effect on the simulation processing by the record preservation processing, and the effect is obtained that the reproduction processing can be performed efficiently after the training.

Further, the integrated management means designates the playback start time and the stored checkpoint file and event log file for the playback means and the display means, and synchronizes the data output by the playback means with the logical time. Since the display means has a control mechanism for displaying it on the display means, the display means displays the state of the reproduction means changing every time on the screen by monitoring the time in synchronization. An effect is obtained that a display screen having the same content as that at the time of conducting a logically accurate training can be easily reproduced.

A recording condition setting unit for setting a recording condition for determining which of the simulated contents is to be recorded in the checkpoint file and the event log file, respectively; And a storage condition setting unit for setting storage conditions that determine the timing of writing to the event log file, so that the data can be efficiently saved by appropriately setting the conditions according to the purpose and the like, Good reproduction of the simulation can be performed.

Further, a checkpoint buffer for temporarily storing data before storing it in the checkpoint file, and an event log buffer for temporarily storing data before storing the data in the event log file are further provided. The simulated status change at each time is written to the above-mentioned checkpoint buffer and simultaneously written to the event log buffer, and the status change of the data of the same attribute at the next time is overwritten by the checkpoint buffer, and the event log buffer is overwritten. Let's write additional
The above process is repeated to write to the event log file when the record storage condition for the event log file is satisfied, and to write to the checkpoint file when the record storage condition for the checkpoint file is satisfied. When writing to a file and an event log file, it is not necessary to collect the internal state of the simulation that interrupts the system, and it is possible to perform the work just by writing to the file, so that the influence on the execution of the simulation can be further reduced. The effect is obtained.

Further, the identifier indicating the changed state and the value indicating the simulated new state are stored in the event log file for each time, so that the time is shifted in the forward direction (time And the log file can be played back.

Further, since the value indicating the past state of the time immediately before the simulation is stored for each time in the event log file, the time is reversed in the reproduction (the time decreases). Direction) and the log file can be reproduced.

[0139] Further, the reproducing means includes: a receiving section for receiving a reproduction start time designated on the integrated management means; a retrieval section for retrieving a checkpoint file stored at the newest time before the start time; Using the identifier of the event log file after the save time of the searched checkpoint file and the value indicating the new state, the state of the start time is reproduced by reproducing the simulated state in the forward direction, and the reproduction start state And a playback start state creation unit for creating a time synchronization signal output from the integrated management means,
A reproduction unit that transmits a change in the content of the simulation at each time from the event log file and performs reproduction by displaying the change on the display unit that also operates in synchronization with the time synchronization signal from the integrated management unit. At a time that satisfies the conditions given in advance, the checkpoint file is saved and the state can be restored at that time. Furthermore, the state at any time after this time can be restored using the event log file. Can be performed, so that an effect is obtained that a screen similar to the screen seen at the time of executing the simulation can be easily reproduced.

[0140] Further, the reproducing means includes: a receiving section for receiving a reproduction start time designated on the integrated management means; a retrieval section for retrieving a checkpoint file stored at the latest time after the start time; Using the identifier of the event log file before the save time of the searched checkpoint file, the value indicating the new state, and the value indicating the past state, the state of the start time is reproduced by reproducing the simulated state in the reverse direction. The reproduction start state creation unit that reproduces and creates the reproduction start state, and the contents of the simulation at each time from the identifier of the event log file and the value indicating the new state in synchronization with the time synchronization signal output from the integrated management means. And a reproduction unit for reproducing by transmitting the change of the information and displaying the same on a display unit operating in synchronization with the time synchronization signal from the integrated management unit. At a time that satisfies the conditions given in advance, the checkpoint file is saved, the state at that time can be restored, and any time before and after this time can be set using the event log file. Can be restored, so that a screen similar to the screen seen at the time of executing the simulation can be easily reproduced, and the effect of improving convenience can be obtained.

Further, the same number of checkpoint file means and event log file means are provided in a plural number, respectively, and different recording conditions are given to each of them by the recording / storing condition setting means to store the contents of the simulation. By recording different information for each file means, the amount of data to be stored in each file can be reduced as compared with the case of using one file means. The effect that the load of the processing required can be distributed.

Further, one reproducing means is provided for each pair of a plurality of checkpoint file means and event log file means, and the reproducing means are synchronized and stored under different recording conditions. The same virtual environment is played by playing back the checkpoint file and the event log file that have been played back. The recorded data can be reproduced on the same system, and the processing for reproduction can be load-balanced.

By specifying different storage periods as storage conditions by the recording storage condition setting unit for each of the plurality of checkpoint file units and event log file units, recording at different storage timings is performed. Since the data is saved, combining the data of multiple files shortens the saving cycle, increases the information, and shortens the time required for cueing during playback. Can be

Further, the present invention provides a simulation step for simulating a virtual environment, a display step for displaying the result simulated by the simulation step on one or more display devices in real time, and a method for recording and storing the contents of the simulation. A record storage condition setting process for setting conditions, a checkpoint file process for storing all the contents of the simulation in a checkpoint file based on the conditions, and a checkpoint file process from the time when the checkpoint file is saved to the time when the next save is performed An event log file process of storing only a difference between changes in the contents of the simulation between the event log files, and a display device that reproduces the contents of the simulation at an arbitrary time by using the stored checkpoint file and the event log file. And the integrated pipe for outputting the time synchronization signal for integrated management of the entire system And the process, the recording and preservation conditions for recording and preserving the internal state of the simulator to be recorded can be set, and all the system states at a certain logical time can be stored. A checkpoint file to be saved and an event log file to save changes in the internal state of the system after the checkpoint file is saved. Since the two types of files are used for saving and playback, the frequency of saving to the file During the training, there is little effect on the simulation processing due to the record preservation processing, and the effect is obtained that the reproduction processing can be performed efficiently after the training.

In the integrated management process, the playback start time and the stored checkpoint file and event log file are designated for the playback process and the display process, and the data output in the playback process are synchronized with the logical time. Since the display device has a control mechanism for displaying it on the display device, the display device displays the state of the reproduction process changing every time on the screen by monitoring the time in synchronization. An effect is obtained that a display screen having the same content as that at the time of conducting a logically accurate training can be easily reproduced.

The recording condition setting step includes a recording condition setting step of setting recording conditions for determining which of the simulation contents is to be recorded in the checkpoint file and the event log file, respectively. And a storage condition setting step of setting storage conditions that determine the timing of writing to the event log file, respectively, so that the data can be efficiently saved by appropriately setting the conditions according to the purpose and the like. Good reproduction of the simulation can be performed.

A checkpoint buffer step for temporarily storing data before storing the data in the checkpoint file, and a checkpoint buffering step for storing the data in the event log file.
An event log buffering step of temporarily storing data, wherein the simulated state change at each time is written to the checkpoint buffer and simultaneously to the event log buffer, and data of the same attribute at the next time is written. Regarding status changes, overwrite in the checkpoint buffer, write additionally in the event log buffer, repeat the above process, write to the event log file when the record storage condition for the event log file is satisfied, and write to the checkpoint file. Since the data is written to the checkpoint file when the above record storage conditions are satisfied, it is not necessary to collect the internal state of the simulation that interrupts the system when writing to the checkpoint file and event log file. Because performed only effort to write the file, there is an advantage that it is possible to further reduce the impact on the execution of the simulation.

In the event log file process, an identifier indicating a changed state and a value indicating a simulated new state are stored in the event log file at each time. The effect is obtained that the log file can be reproduced with the time advanced in the forward direction (time increasing direction).

Further, in the event log file step, the event log file further includes:
Since the value indicating the past state of the time immediately before the simulation is stored, it is possible to obtain the effect that the log file can be reproduced by moving the time in the reverse direction (time decreasing direction) during reproduction. .

The reproducing step includes a receiving step of receiving a designated reproduction start time, a retrieval step of retrieving a checkpoint file stored at the newest time before the start time, and a retrieval step of the retrieved checkpoint file. Using the identifier of the event log file after the save time and the value indicating the new state, the state of the start time is reproduced by reproducing the simulated state in the forward direction,
A playback start state creation step of creating a playback start state;
By synchronizing with the time synchronization signal output in the integrated management process, transmitting the change of the simulated content at each time from the event log file, and displaying the change on the display device which also operates in synchronization with the time synchronization signal And a playback step for performing playback, so that at a time that satisfies the conditions given in advance, a checkpoint file is saved, the state can be restored at that time, and furthermore, this event can be restored using the event log file. Since the state at an arbitrary time after the time can also be restored, an effect is obtained that a screen similar to the screen seen at the time of executing the simulation can be easily reproduced.

The reproducing step includes a receiving step of receiving a designated reproduction start time, a retrieval step of retrieving a checkpoint file stored at the latest time after the start time, and a retrieval step of the retrieved checkpoint file. Using the identifier of the event log file before the save time, the value indicating the new state, and the value indicating the past state, the simulated state is reproduced in the reverse direction to reproduce the state at the start time, and reproduced. A reproduction start state creation step for creating a start state, and a change in the contents of the simulation at each time are transmitted from the identifier of the event log file and the value indicating the new state in synchronization with the time synchronization signal output in the integrated management process. A reproduction step of performing reproduction by displaying on a display device also operating in synchronization with the time synchronization signal. Therefore, at a time that satisfies the condition given in advance, the checkpoint file is saved, the state at that time can be restored, and further, using the event log file, any time before and after this time can be set. Since the state can be restored, a screen similar to the screen seen at the time of executing the simulation can be easily reproduced, and the effect that convenience is improved can be obtained.

Further, the same number of checkpoint files and event log files are provided, respectively, and the simulated contents are stored by giving different recording conditions to each of them in the recording storage condition setting step. So
By recording different information for each file, the amount of data to be stored in each file can be reduced as compared with the case where one file is used, and the processing load required for storage during recording is also reduced. Can be dispersed.

Further, one playback device is provided for each pair of a plurality of checkpoint files and event log files, and the playback devices are synchronized with each other.
Since the same virtual environment is reproduced by reproducing the checkpoint file and the event log file stored under different recording conditions, a plurality of reproducing apparatuses are simultaneously reproduced and operated in synchronization of time. As a result, it is possible to reproduce data recorded and stored in a plurality of files on the same system, and it is possible to obtain an effect that the processing for reproduction can be load-balanced.

By specifying different storage periods as storage conditions in the storage condition setting step for each of the plurality of checkpoint files and event log files, recording data having different storage timings can be obtained. Since the data is stored, when the data of a plurality of files are combined, the storage cycle is shortened, the information is increased, and the time required for cueing at the time of reproduction can be shortened.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a distributed simulation system according to the present invention.

FIG. 2 is a schematic block diagram showing a case where a simulation is performed (at the time of training execution) while saving the internal state of the simulation in the system of FIG. 1;

FIG. 3 is a schematic block diagram showing a case where a simulation internal state is reproduced (during training reproduction) in the system of FIG. 1;

FIG. 4 is a time chart showing a data flow in a case where a simulation is performed (at the time of training execution) while the internal state of the simulation is stored in the system of FIG. 1;

FIG. 5 is a time chart subsequent to FIG. 4 showing a flow of data in a case where the simulation is performed while the internal state of the simulation is stored in the system of FIG. 1 (during training).

FIG. 6 is a time chart showing a data flow when reproducing the internal state of the simulation (during training reproduction) in the system of FIG. 1;

FIG. 7 is a block diagram showing a configuration of a recording / reproducing device of a distributed simulation system according to Embodiment 2 of the present invention.

FIG. 8 is a configuration diagram showing a configuration example of a checkpoint buffer provided in a recording / reproducing device according to Embodiment 2.

FIG. 9 is a configuration diagram showing a configuration example of an event log buffer provided in the recording / reproduction device according to the second embodiment.

FIG. 10 is a diagram illustrating types of events recorded in an event log file and a configuration example of the event log file according to the third embodiment of the present invention.

FIG. 11 is a diagram illustrating types of events recorded in an event log file and a configuration example of the event log file according to the fourth embodiment of the present invention.

FIG. 12 is an explanatory diagram showing a checkpoint file and an event log file created at each logical time according to the fifth to sixth embodiments of the present invention.

FIG. 13 is a block diagram showing a configuration of a distributed simulation system when a plurality of recording / reproducing apparatuses according to Embodiments 7 to 9 of the present invention are connected.

FIG. 14 is a block diagram showing a configuration of a conventional simulation device.

[Description of Signs] 1 integrated management device, 2 training environment simulation device, 3 player terminal, 4 recording / reproducing device, 5 LAN, 6 operator, 7 player (trainee), 100, 200, 30
0,400 distributed simulation control mechanism, 101,
201, 301, 401 control program, 102 integrated management program, 103, 203, 304, 404
Data area, 202 Training environment simulation program, 302
Input control program, 303 animation display program, 310 input device, 311 animation display device, 402 recording program, 403 reproduction program, 406 checkpoint buffer, 407
Event log buffer, 410 checkpoint file, 411 event log file.

Claims (22)

[Claims] 1. A simulation means for simulating a virtual environment, at least one display means for displaying a result simulated by the simulation means in real time, and a record for setting conditions for recording and saving the contents of the simulation Storage condition setting means; checkpoint file means for storing all of the contents of the simulation in a checkpoint file based on the conditions; and a time period from the time when the checkpoint file is saved to the time when the next save is performed. An event log file means for storing only the difference in the change of the simulation content in an event log file; and using the saved checkpoint file and the event log file to reproduce the simulation content at an arbitrary time. Playback means for displaying on the display means, and an integrated pipe for integrated management of the entire system A distributed simulation system, comprising: 2. The integrated management means designates a playback start time and the stored checkpoint file and event log file to the playback means and the display means, and outputs the playback means and the event log file. 2. The distributed simulation system according to claim 1, further comprising a control mechanism for displaying data on said display means in synchronization with a logical time. 3. A recording condition setting unit for setting a recording condition for deciding which of the contents of the simulation is to be recorded in the checkpoint file and the event log file, respectively. 3. The distributed simulation system according to claim 1, further comprising: a storage condition setting unit configured to set a storage condition that determines a timing of writing to the checkpoint file and the event log file. 4. 4. A checkpoint buffer for temporarily storing data before storing in the checkpoint file, and an event log buffer for temporarily storing data before storing in the event log file. Further, the state change at each time of the simulation is written to the event log buffer at the same time as being written to the checkpoint buffer, and the state change of data having the same attribute at the next time is overwritten by the checkpoint buffer. In the event log buffer, additional writing is performed, and the above processing is repeated. When the above-described recording storage condition regarding the event log file is satisfied, the above-described event storage is performed. With above check The distributed simulation system according to claim 1, wherein the system is written to a point file. 5. The event log file according to claim 1, wherein an identifier indicating a changed state and a value indicating a simulated new state are stored at each time. The distributed simulation system according to any one of the above. 6. The distributed simulation system according to claim 5, wherein a value indicating a past state at a time immediately before the simulation is stored for each time in the event log file. . 7. The reproduction means, wherein the reproduction means receives a reproduction start time designated on the integrated management means, and retrieves the checkpoint file stored at the latest time before the start time. A search unit and the start time by reproducing the simulated state in the forward direction using the identifier of the event log file and the value indicating the new state after the storage time of the searched checkpoint file. Reproduce the state of
A playback start state creation unit for creating a playback start state; and synchronizing with a time synchronization signal output from the integrated management means, transmitting a change in the simulated content every time from the event log file, and 6. The distributed simulation system according to claim 5, further comprising: a reproducing unit that performs reproduction by displaying on the display unit that operates in synchronization with the time synchronization signal from the integrated management unit. 8. The reproduction means, wherein the reproduction means receives a reproduction start time specified on the integrated management means, and retrieves the checkpoint file stored at the latest time after the start time. Using the search unit and the identifier of the event log file before the storage time of the searched checkpoint file, the value indicating the new state, and the value indicating the past state, reproduce the simulated state in the reverse direction The reproduction start state is reproduced by reproducing the state of the start time, and a reproduction start state generation unit for generating a reproduction start state is synchronized with a time synchronization signal output from the integrated management means. A change in the content of the simulation at each time is transmitted from the identifier and the value indicating the new state, and the time synchronization signal is also transmitted from the integrated management means. Distributed Simulation system according to claim 6, characterized in that it and a reproducing unit for reproducing by displaying on said display means operating synchronously. 9. The same number of the checkpoint file means and the plurality of the event log file means are provided, and a different recording condition is given to each of them by the recording / save condition setting means. 9. The distributed simulation system according to claim 1, wherein the distributed simulation is performed. 10. A reproducing unit is provided for each pair of the checkpoint file unit and the event log file unit provided in plurals, and the reproducing units are synchronized to set different recording conditions. 10. The distributed simulation system according to claim 9, wherein the same virtual environment is reproduced by reproducing the checkpoint file and the event log file saved in step (a). 11. A storage timing for each of a plurality of checkpoint file means and a plurality of event log file means provided by the recording / storage condition setting means as different storage conditions. The distributed simulation system according to claim 9, wherein different pieces of recorded data are stored. 12. A simulation step for simulating a virtual environment, and a result simulated by the simulation step is stored in real time as one
A display step of displaying on more than one display device; a record storage condition setting step of setting conditions for recording and storing the contents of the simulation; and storing all the contents of the simulation in a checkpoint file based on the conditions. A checkpoint file process, and an event log file process of storing only a difference in the change of the content of the simulation from the time at which the checkpoint file was saved to the time of the next save to an event log file; A reproducing step of reproducing the simulated content at an arbitrary time and displaying the simulated content at an arbitrary time on the display device by using the checkpoint file and the event log file, and outputting a time synchronization signal to perform integrated management of the entire system. A distributed simulation method, comprising: an integrated management step for performing. 13. In the integrated management step, a reproduction start time, the stored checkpoint file and the event log file are designated for the reproduction step and the display step, and output in the reproduction step. 13. The distributed simulation method according to claim 12, further comprising a control mechanism for displaying data on the display device in synchronization with a logical time. 14. A recording condition setting step for setting a recording condition for determining which of the contents of the simulation is to be recorded in the checkpoint file and the event log file, respectively. 14. The distributed simulation method according to claim 12, further comprising: a storage condition setting step of setting storage conditions for determining timings of writing the check point file and the event log file. 15. A checkpoint buffer step for temporarily storing data before storing the data in the checkpoint file; and an event log buffer step for storing data temporarily before storing the data in the event log file. The state change at each time of the simulation is simultaneously written to the event log buffer at the same time as being written to the checkpoint buffer. Regarding the state change of data having the same attribute at the next time, the checkpoint buffer Overwrite and additionally write in the event log buffer, repeat the above process, write to the event log file when the above record storage condition for the event log file is satisfied, and satisfy the above record storage condition for the checkpoint file At the time 15. The distributed simulation method according to claim 12, wherein the data is written to a checkpoint file. 16. In the event log file step, an identifier indicating a changed state and a value indicating a simulated new state are stored in the event log file at each time. Claims 12 to 15
3. The dispersion simulation method according to claim 1. 17. In the event log file step, a value indicating a past state of a time immediately before the simulation is stored for each time in the event log file. Item 18. The dispersion simulation method according to Item 16. 18. The reproducing step, comprising: a receiving step of receiving a designated reproduction start time; a retrieval step of retrieving the checkpoint file stored at the newest time before the start time; Using the identifier of the event log file and the value indicating the new state after the save time of the checkpoint file, the state of the start time is reproduced by reproducing the simulated state in the forward direction. ,
A reproduction start state creation step of creating a reproduction start state, and synchronizing with a time synchronization signal output in the integrated management step, transmitting a change in the content of the simulation at each time from the event log file, and 17. The dispersion simulation method according to claim 16, further comprising: a reproduction step of performing reproduction by displaying on the display device operating in synchronization with a time synchronization signal. 19. The reproducing step, comprising: a receiving step of receiving a designated reproduction start time; a retrieval step of retrieving the checkpoint file stored at the latest time after the start time; By using the identifier of the event log file before the storage time of the checkpoint file, the value indicating the new state, and the value indicating the past state, the simulation state is reproduced in the reverse direction, thereby A reproduction start state creation step of reproducing a state of a start time to create a reproduction start state; and synchronizing with the time synchronization signal output in the integrated management step, the identifier of the event log file and the new state are A change in the content of the simulation at each time is transmitted from the indicated value, and the same operation is performed in synchronization with the time synchronization signal. Distributed Simulation method according to claim 17, characterized in that and a reproduction step of reproducing by displaying on a serial display device. 20. The same number of checkpoint files and the same number of event log files are provided, respectively, and the simulated contents are stored by giving different recording conditions to each of them in the recording storage condition setting step. 20. The distributed simulation method according to claim 12, wherein the distributed simulation is performed. 21. A reproduction device is provided for each pair of the checkpoint file and the event log file provided in plurals, and the reproduction devices are synchronized and stored under different recording conditions. 21. The distributed simulation method according to claim 20, wherein the same virtual environment is reproduced by reproducing the created checkpoint file and the event log file. 22. Specifying different storage cycles as storage conditions in the recording and storage condition setting step for each of the plurality of checkpoint files and the event log files provided, so that the storage timings are different. 22. The distributed simulation method according to claim 20, wherein recorded data is stored.