JP2003316248A - Training device and training scenario generation program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a training device which can easily change a scenario and can perform the maintenance. <P>SOLUTION: The training device (1) for training a trainee (P) includes a state output means (2) for presenting to the trainee the state in which the trainee is situated, an action input means (4) for inputting the action of the trainee, a database (8) for storing a plurality of independent rules relating a new state generated by the states and the actions to the state in which the trainee is situated and the action of the trainee, and a rule retrieving means (10) for retrieving from the database the rule corresponding to the states and the actions on the basis of the state presented to the trainee by the state output means and the action inputted by the action input means and for determining a new state to be then presented to the trainee. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a training device for training trainees and a training scenario generation program, and particularly, to change the behaviors according to the behaviors of trainees such as police officers and firefighters. The present invention relates to a training device and a training scenario generation program for coping with the behavior of a trainee.

[0002]

2. Description of the Related Art Police officers, firefighters, etc. need to take appropriate actions according to the situation in which they are placed in order to carry out their duties. In order to take appropriate actions according to various situations, it is necessary to perform a lot of training, but due to constraints such as training equipment, safety of training, training cost, etc., in recent years, training equipment equipped with a simulator has been used. The training done is increasing. The most difficult problem in producing such a simulator for training is when the training target includes a human and the behavior of the training subject changes depending on the behavior of the trainee. That is, in order to train trainees such as police officers and firefighters to take appropriate actions according to various actions of trainees such as criminals and victims, A scenario that accurately simulates various actions of

Currently, a simulator equipped with a conditional branching scenario is used for such a training device. Here, as an example, a conditional branch type scenario of a conventional training device for training police officers will be described. First, the trained police officer is given the task of releasing the hostage in the situation where one criminal with a gun has taken several hostages. In this situation, if a police officer holds a gun to rescue the hostage, the criminal holding the hostage may point the hostage at the gun. And if the criminal doesn't have a hostage, he may point his gun at a police officer. If a police officer moves into view of the criminal, he may try to hold hostages nearby. In this case, if the hostage is near the criminal, the criminal will be held, and if the hostage is far away, he cannot be held, and the criminal may be hidden behind the scenes.
Police officers will be trained to take appropriate actions and release hostages in such situations.

FIG. 6 is a flowchart of the above-mentioned situation by a conventional conditional branch type scenario. The behavior pattern of the criminal that can be simulated by this flowchart is (A) a police officer holding a gun (S1).
If the criminal holds the hostage (S2), the criminal points the gun at the hostage (S5), (B) the police officer holds the gun (S1)
If the criminal does not hold the hostage (S2), the criminal points the gun at the police officer (S6), (C) the police officer does not hold the gun (S1), and the police officer moves into the sight of the criminal (S3). ) If the hostage is near the criminal (S4), the criminal holds the hostage (S7), (D) the police officer does not hold a gun (S1) and the police officer moves into the criminal's field of view (S3) If you are not near the criminal (S4), the criminal escapes in the shadow (S8), (E)
If the police officer does not hold the gun (S1) and the police officer does not move within the sight of the criminal (S3), the criminal takes no new action,
There are five types. In the training device using the conditional branching scenario, the behavior of the criminal who is the training target is simulated by combining many such flowcharts.

[0005]

However, in the training device using such a conditional branching type scenario, there is a problem that it is difficult to change the scenario or add the action pattern of the criminal. For example, FIG. 7 adds the behavior pattern of the criminal, “when the policeman moves while the criminal is pointing a gun at the hostage, the criminal shoots at the hostage” to the conditional branching type scenario shown in FIG. It is a thing. Therefore, in the conditional branching type scenario, there is a problem that the entire flow chart must be rewritten in order to add one action pattern of the criminal. Similarly, when the behavior pattern of the criminal is changed, the entire flowchart needs to be rewritten. The time required for rewriting such a flowchart becomes longer as the behavior pattern becomes more complicated.

[0006] Therefore, an object of the present invention is to provide a training apparatus which can easily change and maintain a scenario.

[0007]

In order to solve the above-mentioned problems, the present invention is a training apparatus for training a trainee, and provides a trainee with a situation in which the trainee is placed. With respect to the situation where the trainee is placed and the behavior of the trainee, the situation output means to be presented, the behavior input means for inputting the behavior of the trainee, and the new situation caused by these behaviors Based on the database that stores a plurality of independent rules related to each other, and the situation presented to the trainee by the situation output means and the behavior input by the behavior input means, rules related to this situation and behavior It is characterized by having a rule search means for searching a database and determining a new situation to be presented to the trainee.

In the present invention thus constructed,
The situation assumed as training is presented to the trainee by the situation output means. The trainee takes an action to deal with the presented situation, and the action is input to the training device by the action input means. The training device has a database that stores a plurality of independent rules that relate the situation of the trainee and the behavior of the trainee to this situation and the new situation caused by the behavior. There is. The rule search means of the training device searches the database for a rule that matches the situation presented to the trainee by the situation output means and the action input to the training device by the action input means, and finds the matched rule. The new situation is output from the situation output means.

According to this configuration, the scenario used for training is generated by a plurality of independent rules stored in the database. Therefore, when adding or changing a scenario, a new rule is simply added to the database. You don't have to rewrite the whole scenario, just add or change some rules in the database. Thereby, the scenario can be easily changed and maintenance can be performed.

In the training apparatus of the present invention, the behavior presented by the situation output means to the trainee includes the behavior of the trainee whose behavior changes according to the behavior of the trainee. good. INDUSTRIAL APPLICABILITY The training device of the present invention is particularly suitable for generating a training scenario for training a human who exhibits complicated and diverse behaviors.

Further, the training apparatus of the present invention presents to the trainee the rules used in determining the behavior of the trainee in order to examine the suitability of the trainee's behavior after the training. It is preferable to further have a verification function unit.

In the present invention thus constructed,
After the training by the trainee is completed, the rule used for determining the behavior of the human being the trainee is presented to the trainee by the verification function unit. According to this configuration, after the training, the trainee can know the reason for the behavior of the human being the trainee, so it is necessary to consider after the training how the self's behavior affects the trainee. You can

Further, the training apparatus of the present invention includes a trainee database which stores a plurality of independent rules relating to the behavior of the trainee in relation to the situation of the trainee, and the trainee database. After the end, the trainee further has an intention presenting function unit for estimating and presenting the trainee's action intention by applying the action actually taken to the rule stored in the trainee database. Is good.

In the present invention thus constructed,
During the training, the action taken by the trainee is stored in the intention presenting function unit, and after the training, a rule matching the action taken by the trainee is searched from the trainee database by the intention presenting function unit. . The retrieved rule is presented to the trainee by the intention presentation function unit. According to this configuration,
It is difficult for the trainee to recognize himself / herself, and the behavioral intention of the trainee himself / herself during the training can be recognized and examined after the training.

Further, according to the present invention, in a training scenario generation program for training a trainee, an output procedure for presenting the trainee with a situation in which the trainee is placed and an action of the trainee are input. Prepare a database that stores multiple independent rules that relate the input procedure to be performed and the situation of the trainee and the behavior of the trainee to this situation and the new situation caused by the behavior. Based on the procedure to be performed, the situation presented to the trainee in the output procedure, and the behavior entered in the input procedure, the rule applicable to this situation and behavior is searched from the database and presented to the trainee next. It is characterized by causing a computer to execute a rule search procedure for determining a new situation to be performed.

[0016]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, embodiments of the present invention will be described with reference to the accompanying drawings. First, a first embodiment of the present invention will be described with reference to FIGS. FIG. 1 is a block diagram showing the configuration of a training apparatus according to the first embodiment of the present invention. Here, the present embodiment will be described by taking a training device for police officers as an example. The training device 1 according to the present embodiment inputs to the training device 1 a display 2 which is a situation output means for presenting the trainee with a situation in which the trainee P, who is the trainee, is placed, and the behavior of the trainee. Sensor 4 as action input means
And a memory 6 for temporarily storing the behavior of the trainee input from the sensor 4 and the situation currently presented to the trainee. Further, the training apparatus 1 according to the present embodiment stores in the memory 6 a database 8 that stores a plurality of rules independent of each other that relate a new situation that occurs to the situation and behavior of the trainee. Based on the stored situation presented to the trainee and the behavior input by the sensor 4, a rule applicable to this situation and behavior is searched from the database 8 and a new rule to be presented to the trainee next is displayed. It has a rule search means 10 for deciding a situation, and an image synthesizing means 12 for synthesizing an image showing a new situation to be presented to the trainee.

As the display 2 which is the situation output means, any form such as a CRT type display, a front projection type projector and a rear projection type projector capable of presenting the trainee with a picture of the situation where the trainee is placed is provided. Display can be used. Preferably, the situation output means is provided with a speaker or the like so that the trainee can be provided with information on sound effects.

In the present embodiment, as the sensor 4, a position sensor for measuring the position of the body of the trainee, a gazing point sensor for detecting the line of sight of the trainee, a voice sensor for detecting the voice of the trainee, etc. An aiming point sensor for measuring the direction of the trainee's pistol, and a trigger sensor (not shown) for detecting whether or not the pistol is triggered are used. When the training device 1 is simply configured, a keyboard, a mouse, a joystick (above, not shown), or the like may be used as the action input means. Further, an arbitrary sensor can be used as the action input means depending on the application of the training device.

The memory 6, the database 8, the rule searching means 10, and the image synthesizing means 12 are composed of a computer, a hard disk, software for operating them, and the like. The memory 6 is configured to temporarily store the situation currently presented to the trainee and the behavior of the trainee detected by the sensor 4. The contents of the memory 6 are configured to be rewritten whenever the situation presented to the trainee changes or the behavior of the trainee changes as the scenario progresses.

The database 8 is configured to store a plurality of rules independent of each other, which are associated with a situation where the trainee is placed and a new situation caused to the behavior. FIG. 2 shows an example of a part of the rules stored in the database 8. In the figure, “numbers” R1 to R4 are rule numbers, and “priority” indicates the order in which each rule is applied. The larger the value, the higher the priority. The brackets on the left side of the arrow in each rule represent a conditional clause, the brackets on the right side represent an execution part, and the execution part is executed when the true / false value of the conditional clause becomes true. Furthermore, X, Y, and Z in the rule are variables, and “cop (X)”, “hold (Y, Z)”, and the like are predicates whose true and false values are uniquely determined. Further, “&” is a logical product operator (AND), and “!” Is a logical inversion operator (N).
OT).

Therefore, the rule R1 means the action pattern of the criminal "(the policeman holds the gun) and (the criminal holds the hostage) (the criminal points the gun at the hostage)", and the rule R2 "If (the policeman holds a gun) and (the criminal does not hold the hostage) (the criminal points the gun at the policeman)", then rule R3 is "(the policeman is in the sight of the criminal) and If (the hostage is near the criminal) (the criminal holds the hostage), "rule R4 is" (the policeman is in the sight of the criminal) and (the hostage is not near the criminal).
If so (the criminal hides behind the scenes) ", the respective action patterns are meant. Also, each rule is independent of each other,
The relationship between the rules is not specified. In the present embodiment, a scenario equivalent to the flowchart shown in FIG. 6 is generated by these four rules.

The rule search means 10 searches the database 8 for a rule applicable to the current situation stored in the memory 6 and the behavior of the police officer P who is the trainee,
Is configured to determine. In the present embodiment, the program for searching the rules is written in C language. Depending on the application, programming languages such as LISP, PROLOG can also be used.

The image synthesizing means 12 is configured to synthesize a moving image showing a situation to be presented to the trainee by computer graphic and output it to the display 2. Depending on the application, a moving image of an expected pattern may be prepared by actual shooting, and the moving image may be appropriately selected and output. When the training device 1 is simply configured, the image synthesizing unit 12 may be configured to output a still image.

Next, the operation of the training apparatus 1 according to the embodiment of the present invention will be described with reference to FIG. FIG. 3 is a flowchart showing the progress of the scenario generated by the rule shown in FIG. First, in step S11, the initial state is stored in the memory 6. That is, it is the trainee himself who appears in the scenario (A),
Criminal (B), first hostage (C), second hostage (D) 4
The fact of being a person and the distance between the criminal and the first hostage (B, C) is "small", and the distance between the criminal and the second hostage (B, D) is "large". The fact that
Memorized in. At this time, an image showing the above facts is displayed on the display 2 by computer graphic.

Next, in step S12, when the trainee's policeman P moves to a position visible to the criminal, the movement is detected by the sensor 4, and the policeman (A) enters the visual field of the criminal (B). Is added to the memory 6. That is, the predicate “within the field of view (A, B)” is added to the memory 6.

Next, in step S13, the rule search means 10 searches the database 8 for a rule that matches the fact stored in the memory 6 in descending order of priority. Here, policeman (A), criminal (B), hostage (C),
Since the distance (B, C) = small, the field of view (A, B) exists in the memory 6, the condition clause of the rule R3 is satisfied. Rule R
3 is applied, the fact that the criminal (B) holds the hostage (C), that is, "holds (B, C)" is added to the memory 6 according to the execution part of the rule R3. Memory 6
When the contents of (1) are changed, the image synthesizing means 12 synthesizes the image in which the criminal holds the hostage, and is presented to the trainee by the display 2.

Next, in step S14, when the trainee holds a gun by looking at the image presented to the trainee, the motion of the trainee is detected by the sensor 4, and "the gun is held (A)" in the memory 6. ] Is added.

When the predicate "Attach a gun (A)" is added to the memory 6, in step S15, the rule R1 is applied and the criminal (B) points the gun at the hostage (C), that is, " The predicate "point the gun (B, C)" is added to the memory 6, and an image of the scene where the criminal points the gun at the hostage is displayed on the display. Here, the fact stored in the memory 6 in step S14 satisfies not only the conditional clause of the rule R1 but also the conditional clause of the rule R3, but the priority of the rule R1 is higher than that of the rule R3. , The rule R1 is applied preferentially.
By repeating the above operation, the training device 1
Scenario progresses.

Next, the change of the scenario will be described.
As described above, the scenario realized by the rules R1 to R4 is equivalent to the scenario according to the flowchart shown in FIG. In order to change this scenario to the scenario of the flowchart shown in FIG. 7, if the policeman moves while the criminal points the gun at the hostage, the criminal fires at the hostage and the hostage dies. That is, [cop (X) & move (X) & criminal (Y) & aim gun (Y, Z) & hostage (Z)] → [fire (Y, Z), death (Z)] as rule R5 All you have to do is add it to the database 8. Further, a numerical value 6000 larger than the rule R1 is designated as the priority.

When training is performed using the database 8 modified in this way, for example, when the trainee policeman moves in the situation of step S15 of FIG. 3, the conditional clause of rule R5 is satisfied. To be done. This allows rule R
5 is applied, the predicates “fire (B, C)” and “death (C)” are added to the memory 6, and the display 2 indicates that the criminal fires at the hostage and the hostage dies. The scene is projected. This scenario is equivalent to the scenario shown in the flowchart of FIG.

In the training device 1 according to the first embodiment of the present invention, a scenario for training can be generated by combining a plurality of rules independent of each other. Also, to change the scenario created, simply add a new rule to the database, no need to change the previous rule. Therefore, the training apparatus 1 according to the first embodiment of the present invention can easily change the scenario.

Next, a training device according to a second embodiment of the present invention will be described with reference to FIG. The training apparatus of the present embodiment presents the rules applied to determine the behavior of the trainee during the training to the trainee after the training of the trainee is completed, and the reason for the behavior of the trainee. Is different from the first embodiment in that it has a verification function unit for examining. Therefore,
Here, only the configuration, operation, and effect different from those of the first embodiment will be described, and description of the same points will be omitted.

The training apparatus according to the second embodiment of the present invention is
The display 2, the sensor 4, and the memory 6 shown in FIG.
In addition to the database 8, the rule searching means 10, and the image synthesizing means 12, a verification function unit (not shown) is provided.
The verification function unit (not shown) is configured to store the rules searched and applied by the rule search means 10 during the training of the trainee and display the rules on the display 2 after the training is completed. There is. The verification function unit (not shown) can be configured by a computer, a hard disk, software that operates them, and the like. Further, the applied rule may be output to a printer (not shown) after the training is completed.

Next, the operation of the training apparatus according to the second embodiment of the present invention will be described. Here, a case where the changed scenario described in the first embodiment is executed will be described. The operation from the start of the training to the end of the training by the trainee is the same as that in the first embodiment, and therefore its explanation is omitted.
As described in the first embodiment, in this scenario, the rule R3 is first applied, and then the rules R1 and R5 are applied in this order. These applied rules are stored by the verification function unit (not shown) of the training device, and after the training, they are output to the display 2 or the printer (not shown).

FIG. 4 shows an example of the output rule.
As shown in FIG. 4, the output by the verification function unit (not shown) does not simply output the rules stored in the database 8 as they are, but the rules are related to the trainee and the training. It is divided into parts related to the criminal who is the target person, and is output in the form of plain text.

According to the training apparatus of the second embodiment of the present invention, the reason for the behavior of the trainee is presented to the trainee after the training is completed, so that the trainee can identify his / her own behavior as the trainee. It is possible to verify how the behavior was affected.

Next, a training device according to a third embodiment of the present invention will be described with reference to FIG. The training device of the present embodiment has an intention presenting function unit that estimates the behavioral intention of the trainee under training and presents it to the trainee after the training of the trainee is completed. Different from Therefore, here, only the configuration, operation, and effect different from those of the first embodiment will be described, and description of the same points will be omitted.

As shown in FIG. 5, the training device 20 according to the third embodiment of the present invention includes a display 2 and a sensor 4.
, Memory 6, database 8, and rule search means 1
0, the image synthesizing means 12, a trainee database 22, and an intention presentation function unit 24. The trainee database 22 is configured to store a plurality of independent rules relating the trainee's typical behavioral patterns to the trainee's situation. Each rule is associated with a typical behavioral intention that the trainee might have when the trainee takes a certain action in the situation.

The intention presentation function unit 24 stores the situation presented to the trainee during the training by the training apparatus 20 and the behavior of the trainee detected by the sensor 4 performed for the situation. Is configured. Furthermore, the intention presentation function unit 24 searches the trainee database 22 for a rule that matches the action taken by the trainee after the training, and presents the searched rule to the trainee. Is configured to. Trainee database 2
2 and the intention presentation function unit 24 can be configured by a computer, a hard disk, software for operating them, and the like.

Next, the operation of the training device 20 according to the third embodiment of the present invention will be described. Here, a case where the scenario described in the first embodiment is executed will be described. The operation from the start of the training to the end of the training by the trainee is the same as that in the first embodiment, and therefore its explanation is omitted. As described in the first embodiment, in this scenario, first, the trainee receives the situation of step S11 in FIG.
That is, policeman (trainee himself), criminal, first hostage, second
There are four hostages, the distance between the criminal and the first hostage is “small”, and the distance between the criminal and the second hostage is “large”.
In such a situation, he is moving to a position visible to the criminal. Furthermore, the trainee is shown in FIG.
In the situation of step S13, that is, in the situation where the criminal holds the first hostage, he takes the action of holding a gun.

The behavior of the trainee for these given situations is stored in the intention presenting function unit 24 during the training of the trainee. After the training, the intention presentation function unit 24 searches the trainee database 22 for a rule that is stored in the intention presentation function unit 24 and applies to the action taken by the trainee. . Furthermore, the typical behavioral intention associated with each retrieved rule is presented to the trainee by the intention presenting function unit 24 together with each retrieved rule.

In the training device according to the third embodiment of the present invention, the trainee's behavioral intention, which is estimated from the behavior of the trainee, can be presented to the trainee after the training. In general, the trainee's behavioral intention during training is often unrecognizable to the trainee who is absorbed in the training, so it is meaningful to present the inferred intention for each behavior of the trainee. There is.

Although the preferred embodiment of the present invention has been described above, various modifications can be made to the above-described embodiment. In particular, the training target for the trainee does not necessarily have to include a human, and for example, the training device of the present invention can be used for training fire fighting activities of a fireman at a fire site where there is no victim. Can be applied. Further, in the above-described embodiment, the situation where the trainee is placed is presented to the trainee by video and audio. However, the situation where the trainee is placed is different depending on the training target to be applied. It may be presented using any of visual sense, hearing, touch, smell, and taste.

[0044]

According to the training device of the present invention, it is possible to easily change and maintain a scenario.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a training device according to a first embodiment of the present invention.

FIG. 2 is a diagram showing an example of rules stored in a database of the training apparatus according to the first embodiment of the present invention.

FIG. 3 is a flowchart showing a scenario generated by the training device according to the first embodiment of the present invention.

FIG. 4 is a diagram showing an example of a reason for a behavior of a training subject presented to a trainee after the training by the training device according to the second embodiment of the present invention.

FIG. 5 is a block diagram showing a configuration of a training device according to a third embodiment of the present invention.

FIG. 6 is a flowchart showing an example of a conditional branch type scenario used in a conventional training device.

7 is a flowchart showing an example in which a part of the flowchart shown in FIG. 6 is modified.

[Explanation of symbols] 1 Training apparatus according to the first embodiment 2 display 4 sensors 6 memory 8 database 10 Rule search means 12 Image synthesizer 20 Training apparatus according to the second embodiment 22 Trainee database 24 Intention presentation function

Claims (5)

[Claims] 1. A training device for training a trainee, comprising: a situation output means for presenting the trainee's situation to the trainee; and a behavior of the trainee. A database for storing a plurality of independent rules relating action input means and the situation where the trainee is placed and the behavior of the trainee to these situations and new situations caused by the behavior. And, based on the situation presented to the trainee by the situation output means and the behavior input by the behavior input means, search rules related to these situations and behaviors from the database, and , A training device for determining a new situation to be presented next, and a training device. 2. The training apparatus according to claim 1, wherein the situation presented to the trainee by the situation output means includes the behavior of the trainee whose behavior changes according to the behavior of the trainee. 3. The verification of presenting the rule, which was used in determining the behavior of the trainee, to the trainee after completion of the training, in order to examine the propriety of the behavior of the trainee. The training device according to claim 2, further comprising a functional unit. 4. A trainee database storing a plurality of independent rules relating to the behavior of the trainee in relation to the situation of the trainee, and a trainee database after the training, Applying an action actually taken by the trainee to a rule stored in the trainee database to infer a behavior intention of the trainee and present it. The training device according to any one of claims 1 to 3, which is characterized. 5. A training scenario generation program for training a trainee, wherein an output procedure for presenting the situation where the trainee is placed to the trainee and an action of the trainee are input. For the input procedure and the situation of the trainee and the behavior of the trainee,
A procedure of preparing a database that stores a plurality of independent rules relating to this situation and a new situation caused by the action, the situation presented to the trainee in the output procedure, and the input procedure Based on the action entered
A training scenario generation program for causing a computer to execute a rule search procedure for searching a rule applicable to this situation and action from the database and determining a new situation to be presented to the trainee.