JP3362664B2 - Condition monitoring device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The invention described in the claims of the present application is suitable for use in avoiding a situation in which a driver of a vehicle falls into a dozing state. The present invention relates to a state monitoring device that performs a warning and issues a necessary warning according to the detection result.

[0002]

2. Description of the Related Art While a vehicle is traveling, it is extremely important to avoid a situation in which the driver of the vehicle becomes less awake and falls asleep. From such a viewpoint, the degree of awakening of the driver while the vehicle is traveling is detected by a method of observing the operating state of the driver, the behavior of the vehicle, etc., and as a result, the degree of awakening of the driver decreases and If the driver is likely to fall asleep, the driver should be warned.
Various researches have been conducted on a state monitoring system for preventing a snooze.

In such a state monitoring system,
For example, charge coupled device (CCD)
The state of the eyes of the driver of the vehicle is observed by an image pickup means such as a small video camera (CCD camera) in which the vehicle is used. Then, based on the monitoring data about the state of the eyes of the driver formed by performing image processing on the observation results obtained by it, it is determined whether or not the driver is about to fall asleep, When a state in which the driver is about to fall asleep is detected, for example, a buzzer sounds an alarm.

Regarding the determination as to whether or not the driver is about to fall into a dozing state in such a case, for example,
As described in JP-A-6-266981, a long-time time-series average value of the driver's eye opening, blink frequency, etc. obtained by using a CCD camera was calculated and calculated. A system in which the long-term time-series average value is used as a drowsiness determination parameter for determining whether or not the driver is about to fall into a drowsiness state, or is disclosed in JP-A-7-93700. As described above, the edge image of the driver's face portion obtained by using the CCD camera is collated with the image represented by the template data stored in the memory unit in advance, and the driver is determined from the pattern collation result obtained thereby. Of eyeball movements, blinking movements, etc., and the analysis results are used to determine whether or not the driver is trying to fall asleep. There has been proposed.

[0005]

As described above, the determination as to whether or not the driver of the vehicle is about to fall asleep is performed by grasping the driver's eye opening or blinking motion. In such a system, in addition to an image pickup means such as a CCD camera and an alarm means, a complicated analysis means for analyzing an image pickup output representing the driver's eye opening degree or blinking operation obtained from the image pickup means is provided. However, there is a disadvantage that the entire structure is complicated and expensive. Moreover, it takes a relatively long time to analyze the imaging output, and it is not possible to quickly determine whether or not the driver is trying to fall asleep, and the reliability of the determination result is not so high. There is also a problem that misjudgment easily occurs.

In view of the above, the invention described in the claims of the present application is relatively simple and inexpensive to obtain when it is used to avoid a situation in which a vehicle driver falls asleep. With such a configuration, whether or not the driver of the vehicle is about to fall into a dozing state due to a decrease in awakening level is promptly and accurately determined, and the driver's necessary (EN) Provided is a state monitoring device capable of appropriately issuing a warning, detecting a state of a driver of a vehicle, and issuing a necessary warning according to the detection result.

[0007]

The state monitoring device according to the invention described in any one of claims 1 to 6 in the claims of the present application is an interrogation voice data having preset contents. A first memory portion in which is stored,
A voice output unit that emits an interrogation voice based on the interrogation voice data stored in the first memory unit, and reference voice data that represents the content corresponding to a predetermined response voice to the interrogation voice that is emitted from the voice output unit. The second where is stored
Memory unit, a voice input unit to which a response voice to the interrogation voice issued from the voice output unit is input, and a response data forming unit to obtain response voice data corresponding to a voice signal representing the response voice obtained from the voice input unit And the response voice data obtained from the response data forming unit and the reference voice data stored in the second memory unit are compared with each other, and a data comparing unit for obtaining a comparison output showing the result of the comparison, and a preset data comparing unit. A third memory unit for storing warning data and an operation control unit are provided, and the operation control unit controls reading of interrogation voice data from the first memory unit and reference voice data from the second memory unit. Read control is performed, and the comparison output obtained from the data comparison unit is the response voice data obtained from the response data forming unit as the reference voice data stored in the second memory unit. The warning data is read from the third memory unit when it indicates that there is no identity or similarity, or that the voice input unit does not receive the response voice to the interrogation voice output from the voice output unit. , And is configured to perform a control for bringing a warning based on the warning data.

Particularly, in the state monitoring device according to the invention described in claim 4 or 5 in the claims of the present application, the second memory unit in which the reference voice data is stored is
The reference voice data is stored as voice data obtained based on the voice of the person who inputs the response voice to the interrogation voice emitted from the voice output unit into the voice input unit.

Further, in particular, the state monitoring apparatus according to the invention described in claim 6 of the present application has a first memory section, a voice output section, a second memory section, a voice input section, and a response. A person who is configured such that the data forming unit, the data comparing unit, the third memory unit, and the operation control unit are all mounted on a vehicle, and inputs a response voice to an interrogation voice emitted from the voice output unit to the voice input unit. Is the vehicle driver.

In the state monitoring apparatus according to the invention as set forth in any one of claims 1 to 6 in the claims of the present application configured as described above, the control by the operation control unit is also possible. Then, when the interrogation voice data is read from the first memory unit and the interrogation voice is emitted from the voice output unit, the input state of the response voice to the interrogation voice to the voice input unit is monitored. Such monitoring is
The response data forming unit obtains the response voice data corresponding to the voice signal representing the response voice from the voice input unit to which the response voice for the interrogation voice emitted from the voice output unit is input, and the response comparator forms the response voice data. The data and the reference voice data stored in the second memory unit are compared, and a comparison output representing the result of the comparison is obtained.
The operation control unit outputs the comparison output obtained from the data comparison unit,
The response voice data obtained from the response data forming unit lacks the identity or similarity with the reference voice data stored in the second memory unit, or a call issued from the voice output unit to the voice input unit. This is performed by determining whether or not it indicates that the response voice to the call voice is not input.

Then, the comparison output obtained from the data comparison unit is such that the response voice data obtained from the response data forming unit lacks the identity or similarity with the reference voice data stored in the second memory unit. Or the response voice to the interrogation voice output from the voice output unit is not input to the voice input unit, the operation control unit confirms that the input state of the response voice to the interrogation voice to the voice input unit is normal. However, for example, a person who inputs a response voice to an interrogation voice to the voice input unit is determined to be in a state where the degree of awakening is lowered and is about to fall into a dozing state, under the control of the operation control unit, The warning data is read from the third memory unit and a warning is issued based on the warning data.

By doing so, it is possible to determine whether or not the input status of the response voice to the interrogation voice to the voice input section is not normal, for example, for the interrogation voice, with a relatively simple and inexpensive structure. Whether or not the person who inputs the response voice to the voice input unit is in a state where the degree of awakening is falling and is about to fall asleep is quickly and accurately determined, and the necessary warning is appropriate according to the determination result. Will be done.

Therefore, the state monitoring device according to the invention described in any one of claims 1 to 6 in the claims of the present application is used for avoiding a situation in which the driver of the vehicle falls asleep. In such a case, it is possible to quickly and accurately determine whether or not the driver of the vehicle is about to fall into a dozing state due to a decrease in awakening level with a configuration that is relatively simple and inexpensive to obtain. Then, a necessary warning is properly given to the driver according to the determination result.

In particular, in the state monitoring device according to the invention described in claim 4 or 5 in the claims of the present application, the reference voice data stored in the second memory section is voice output. Since the response voice to the interrogation voice issued from the voice input unit is voice data obtained based on the voice of the person who inputs to the voice input unit, the comparison output obtained from the data comparison unit by the operation control unit is the response data. A malfunction is suppressed in determining whether or not the response voice data obtained from the forming unit indicates that the response voice data lacks the sameness or similarity with the reference voice data stored in the second memory unit. It will be done exactly under the circumstances.

Further, in particular, in the state monitoring device according to the invention described in claim 6 of the present application, the first memory section, the voice output section, the second memory section,
The voice input unit, the response data forming unit, the data comparison unit, the third memory unit, and the operation control unit are all configured to be mounted on the vehicle, and the response voice to the interrogation voice issued from the voice output unit is input by voice. Since the person who inputs to the section is the driver of the vehicle, whether or not the driver of the vehicle is going to fall asleep due to a decrease in the awakening degree with a configuration that is relatively easy and inexpensive to obtain. The determination is made quickly and accurately, and a necessary warning is properly given to the driver according to the determination result.

[0016]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows an example of a condition monitoring device according to the invention described in any one of claims 1 to 6 in the claims of the present application. This example is used to avoid a situation in which the driver of the vehicle falls asleep, and the entire vehicle is mounted on the vehicle.

In the example shown in FIG. 1, an operation control unit 11 is provided, and the operation control unit 11 is associated with a program data storage memory section 12 and an input operation section 13. The program data storage memory unit 12 stores program data DPR representing a control program for controlling the operation of the operation control unit 11, reads the program data DPR according to a memory control signal CM from the operation control unit 11,
It is supplied to the operation control unit 11. In addition, the input operation unit 13 is manually operated by a user such as a driver of a vehicle, and various command data DIN is accordingly received.
Is supplied to the operation control unit 11.

Further, the operation control unit 11 sends to the operation control unit 11 interrogation voice data DCA and warning data D.
The voice signal forming unit 14 to which WR is supplied is connected, and the voice signal forming unit 14 outputs the interrogation voice data DC.
Audio signal SCA or warning data DWR corresponding to A
An audio signal SWA corresponding to is obtained. The audio signal SCA or SWA from the audio signal forming unit 14 is supplied to the amplifier 1
A speaker 1 that is amplified by 5 and forms an audio output unit
6 is supplied. As a result, the speaker 16 issues a warning based on the preset calling voice represented by the calling voice data DCA or the warning data DWR. The speaker 16 is arranged, for example, near the driver's seat in the vehicle.

Further, in the example shown in FIG. 1, a microphone 17 forming a voice input section is input with a response voice to a preset voice call voice represented by the voice call data DCA from the speaker 16. Are provided, for example, as those arranged near the driver's seat in the vehicle. From the microphone 17, a voice signal SSP corresponding to the response voice to the interrogation voice is obtained,
The audio signal SSP is amplified by the amplifier 18 and supplied to the operation control unit 11.

The operation control unit 11 includes an interrogation data storage memory unit for storing interrogation voice data DCA having preset contents, and an audio signal SSP supplied to the operation control unit 11 through the amplification unit 18. Based on the audio signal detecting unit for detecting, the response data forming unit for obtaining the response audio data corresponding to the audio signal SSP detected by the audio signal detecting unit, and the speaker 16 based on the audio signal SSP detected by the audio signal detecting unit. Reference voice forming unit for forming reference voice data representing the content corresponding to a predetermined response voice to the interrogation voice emitted from the reference voice storing unit for storing the reference voice data formed by the reference data forming unit. Section, the response voice data obtained from the response data forming section, and the reference data stored in the reference data storage memory section. It compares the audio data, and a data comparing portion for obtaining a comparing output representing the result of the comparison,
A warning data storage memory unit in which preset warning data DWR is stored is built in. The voice data DCA and the warning data DWR supplied from the operation control unit 11 to the voice signal forming unit 14 are read from the interrogation data storage memory unit and the warning data storage memory unit, respectively.

Under such a situation, the interrogation voice emitted from the speaker 16 has, for example, the content of "Are you awake?", And therefore the interrogation data incorporated in the operation control unit 11 is used. The interrogation voice data DCA stored in the storage memory section is "Are you awake?"
It is said that it represents an interrogation voice having the content. Further, it is predetermined that the response voice to the interrogation voice has the content of “Yes. I'm awake.”. Accordingly, the motion control unit 1
The reference voice data stored in the reference data storage memory unit built in No. 1 represents the content of “Yes. I'm awake.” Corresponding to the response voice. further,
When the interrogation voice is emitted from the speaker 16, the driver of the vehicle is prearranged to emit the response voice having the preset content "Yes. I'm awake."

When the reference voice data is stored in the reference data storage memory section built in the operation control unit 11, the vehicle driver who is the user of the example shown in FIG. Speak a voice with the content "Yes. I'm awake." As a result, a voice signal SSP corresponding to the voice of the driver having the content “Yes. I'm awake.” Is obtained from the microphone 17, and the voice signal SSP is amplified by the amplifier 18 and supplied to the operation control unit 11. It

In the operation control unit 11, the audio signal SSP supplied through the amplification section 18 is detected by the audio signal detection section and supplied to the reference data forming section. In the reference data forming unit, the audio signal SS
Based on P, reference voice data representing the content corresponding to the response voice having the content "Yes. I'm awake." Is formed, and the reference voice data is supplied to and stored in the reference data storage memory unit.

Under the circumstances, the motion control unit 11 is
In response to the command data DIN instructing the interrogation operation from the input operation unit 13, the interrogation operation state is set according to the control program represented by the program data DPR read from the program data storage memory unit 12.

Under such an interrogation operation state, the operation control unit 11 first reads out the interrogation voice data DCA having the preset contents from the interrogation data storage memory section, and Interrogation voice data DC
A is sent to the audio signal forming unit 14. As a result, the interrogation voice data DCA is displayed from the speaker 16.
For example, a preset interrogation voice "Are you awake?" Is issued as an alert to the driver of the vehicle. The interrogation voice from the speaker 16 is repeatedly emitted at predetermined time intervals when the vehicle is in a traveling state, for example.

When the driver 16 is in a sufficient awakening state when the interrogation voice is emitted from the speaker 16, for example, a preset response voice "Yes. I am awake." Is correctly emitted. Then, the response voice is input to the microphone 17, and the voice signal SSP corresponding to the response voice "Yes. I'm awake." Is obtained from the microphone 17. On the other hand, speaker 1
When the call voice is emitted from 6, the driver of the vehicle
When there is a possibility that the degree of awakening may decrease and the user may fall asleep, the preset response voice "Yes. I am awake." Is not correctly uttered, and the response voice that was not correctly uttered is the microphone 17 Is input to the microphone 17 to obtain a voice signal SSP corresponding to the response voice that is not correctly uttered, or no response voice is uttered at all.
No response voice is input to the microphone 17, and the voice signal SSP corresponding to the response voice cannot be obtained from the microphone 17.

At this time, when the voice signal SSP is obtained from the microphone 17, the voice signal SSP obtained from the microphone 17 is amplified by the amplifier 18 and supplied to the operation control unit 11. In the operation control unit 11, the audio signal SSP supplied through the amplification unit 18 is detected by the audio signal detection unit and is supplied to the response data formation unit. In the response data forming unit, based on the voice signal SSP, a response voice indicating the content corresponding to the response voice having the content "Yes. I'm awake." Or the response voice not correctly uttered. Data is formed.

Then, in the data comparison unit, the response voice data formed by the response data forming unit is compared with the reference voice data read from the reference data storage memory unit, and the comparison output showing the result of the comparison is output. Is obtained from the data comparison section. The comparison output obtained from the data comparison unit is such that the response voice data formed by the response data forming unit has the same or similarness as the reference voice data stored in the reference data storage memory unit. , "Yes. I'm awake." By the driver of the vehicle.
The preset response voice is correctly uttered and the response voice is input to the microphone 17, or the response voice data formed by the response data forming unit is the reference voice data stored in the reference data storage memory unit. Lacking in identity or similarity with
This means that the preset response voice "Yes. I'm awake." Was not correctly uttered by the driver of the vehicle, and the response voice that was not correctly uttered was input to the microphone 17.

On the other hand, when the voice signal SSP corresponding to the response voice is not obtained from the microphone 17 even when the interrogation voice is emitted from the speaker 16, the response voice data by the response data forming unit in the operation control unit 11 is obtained. Is not formed. Therefore, in such a case, in the data comparison unit, there is no response voice data to be compared with the reference voice data read from the reference data storage memory unit, and the comparison obtained from the data comparison unit. The output indicates that the response sound is not input to the microphone 17.

In this case, in the operation control unit 11, the comparison output obtained from the data comparison unit is the same as the reference voice data in which the response voice data obtained from the response data forming unit is stored in the reference data storage memory unit. Or lack of similarity, or microphone 17
When it is indicated that the response voice to the interrogation voice emitted from the speaker 16 is not input, it is determined that the driver of the vehicle is in a situation where the degree of awakening is lowered and the driver may fall asleep. Then, the operation control unit 11
Reads out the warning data DWR from the warning data storage memory unit and sends it to the audio signal forming unit 14 so that the speaker 16 issues a warning based on the warning data DWR. As a result, the driver is returned to a sufficient wakeful state from a situation in which the degree of wakefulness is reduced and the driver may fall asleep due to the warning issued from the speaker 16.
It prevents you from falling asleep.

The operation control unit 11 in the above-described example shown in FIG. 1 is constructed by using, for example, a microcomputer, and an example of a program executed by the microcomputer under the interrogation operation state. ,
This will be described with reference to the flowchart of FIG.

In the program shown in the flowchart of FIG. 2, after the start, in step 21, it is determined whether or not the interrogation timing at which the interrogation voice should be emitted from the speaker 16 has come. As a result, if the interrogation timing has not arrived, the determination in step 21 is repeated. When the interrogation timing has arrived, in step 22, the interrogation voice data DCA is read from the interrogation data storage memory unit to output the voice signal. It is sent to the forming unit 14.

Next, at step 23, it is judged whether or not the data comparison output judgment timing for judging the comparison output obtained from the data comparison section has come. As a result, if the data comparison output determination timing has not come, the determination in step 23 is repeated, and when the data comparison output determination timing has arrived, the process proceeds to step 24.

In step 24, it is determined whether or not the comparison output obtained from the data comparison section indicates that the response sound to the interrogation sound emitted from the speaker 16 is not input to the microphone 17.
As a result, when the comparison output obtained from the data comparison unit indicates that the response sound to the interrogation sound emitted from the speaker 16 is input to the microphone 17, the process proceeds to step 26, and the data comparison unit outputs The obtained comparison output is output to the microphone 17 through the speaker 1.
When it is indicated that the response voice to the interrogation voice issued from the sixth set is not input, the process proceeds to step 25.

In step 26, the comparison output obtained from the data comparison unit lacks the identity or similarity of the response voice data obtained from the response data forming unit with the reference voice data stored in the reference data storage memory unit. Determine whether or not it represents something. As a result of the judgment in step 26, the comparison output obtained from the data comparison unit lacks the identity or similarity of the response voice data obtained from the response data formation unit with the reference voice data stored in the reference data storage memory unit. If it indicates that the response voice data obtained from the response data forming unit is the reference voice data stored in the reference data storage memory unit, the process proceeds to step 25. When it indicates that it is not lacking in identity or similarity with, the process returns to step 21.

In step 25, the warning data DWR is read out from the warning data storage memory section and sent to the audio signal forming section 14, and thereafter the procedure returns to step 21.

In the example shown in FIG. 1 described above, the warning based on the warning data DWR sent from the operation control unit 11 is made to be issued from the speaker 16, but the warning based on the warning data DWR is Speaker 1
Instead of being emitted from the speaker 6, it may be emitted from a speaker provided separately from the speaker 16, emitted from a buzzer, emitted from a light emitting body, caused by vibration of a vibrating body, or the like.

In the example shown in FIG. 1, the operation control unit 11 includes an interrogation data storage memory unit, a voice signal detection unit, a response data formation unit, a reference data formation unit, a reference data storage memory unit, and a data comparison unit. Although a unit and a warning data storage memory unit are built in, these do not necessarily have to be built in the operation control unit 11.

Further, the example shown in FIG. 1 is used for avoiding a situation where the driver of the vehicle falls asleep, and the whole is mounted on the vehicle. The state monitoring device according to the invention described in any one of claims 1 to 6 in the range is not limited to such an example, and can take other various usage modes.

[0040]

As is apparent from the above description, the condition monitoring device according to the invention described in any one of claims 1 to 6 in the claims of the present application is relatively simple and inexpensive. With the configuration obtained in (1), it is determined whether or not the input state of the response voice to the interrogation voice is abnormal in the voice input unit. Then, it can be quickly and accurately determined whether or not the user is about to fall into a dozing state, and the necessary warning can be appropriately given according to the determination result.

Therefore, the state monitoring device according to the invention described in any one of claims 1 to 6 in the claims of the present application is used for avoiding a situation in which the driver of the vehicle falls asleep. In this case, it is possible to quickly and accurately determine whether or not the driver of the vehicle is about to fall into the dozing state due to the reduced awakening degree, with a configuration that is relatively simple and inexpensive to obtain. Therefore, it is possible to properly issue a necessary warning to the driver according to the determination result.

In particular, according to the state monitoring apparatus of the present invention as set forth in claim 4 or claim 5 of the present application, the reference voice data stored in the second memory unit is the voice output unit. Since the response voice to the interrogation voice generated from is voice data obtained based on the voice of the person who inputs to the voice input unit, the comparison output obtained from the data comparison unit by the operation control unit is the response data formation. Whether or not the response voice data obtained from the section indicates that the response voice data lacks the sameness or similarity with the reference voice data stored in the second memory unit, the malfunction is suppressed. You can do exactly with.

Further, in particular, according to the state monitoring device of the invention described in claim 6 of the present application, the first memory unit, the voice output unit, the second memory unit, the voice input unit. The response data forming unit, the data comparing unit, the third memory unit and the operation control unit are configured to be mounted on the vehicle as a whole, and a response voice to the interrogation voice output from the voice output unit is input to the voice input unit. Since the person who does is a driver of the vehicle, it is possible to determine whether or not the driver of the vehicle is in a state of falling asleep and falling into a dozing state with a configuration that is relatively simple and inexpensive to obtain, It can be performed quickly and accurately, and a necessary warning can be properly given to the driver according to the determination result.

[Brief description of drawings]

FIG. 1 is a block configuration diagram showing an example of a state monitoring device according to the invention described in any one of claims 1 to 6 in the claims of the present application.

FIG. 2 is a flowchart showing an example of a program executed by a microcomputer which constitutes the operation control unit in the example shown in FIG.

[Explanation of symbols]

11 ... Operation control unit, 12 ... Program data storage memory unit, 13 ... Input operation unit, 14 ...
Audio signal forming unit, 15, 18 ... Amplifying unit, 16 ...
Speaker, 17 ... Microphone

Claims (6)

(57) [Claims] 1. A first memory unit for storing interrogation voice data having preset contents, and a voice output for emitting interrogation voice based on the interrogation voice data stored in the first memory unit. Section, a second memory section for storing reference voice data representing content corresponding to a predetermined response voice to the interrogation voice emitted from the voice output section, and the interrogation voice emitted from the voice output section. A voice input unit for inputting a voice response to the voice response, a response data forming unit for obtaining voice response data corresponding to a voice signal representing the voice response obtained from the voice input unit, and a voice response obtained from the data forming unit. A data comparison unit that compares the data with the reference voice data stored in the second memory unit and obtains a comparison output representing the result of the comparison; and a preset warning. And a read control of the interrogation voice data from the first memory section and a read control of the reference voice data from the second memory section. The comparison output obtained from the data comparison unit is such that the response voice data obtained from the response data forming unit lacks the identity or similarity with the reference voice data stored in the second memory unit, or , When the voice input unit indicates that the response voice to the interrogation voice issued from the voice output unit is not input, the warning data is read from the third memory unit, and the warning based on the warning data is issued. A state monitoring device configured to include an operation control unit that controls the state. 2. The condition monitoring apparatus according to claim 1, wherein the voice output unit emits a warning voice based on the warning data when the warning data is read from the third memory unit. 3. A first memory section, a second memory section, a response data forming section, a data comparing section, and a third memory section are built in the operation control section. 2. The condition monitoring device described in 2. 4. The second memory unit stores the reference voice data as voice data obtained based on a voice of a person who inputs a response voice to the interrogation voice issued from the voice output unit into the voice input unit. The condition monitoring device according to claim 1 or 3, characterized in that: 5. Reference audio data is formed based on an audio signal obtained from an audio input unit, and the reference audio data is supplied to a second memory unit and stored in the second memory unit. The condition monitoring apparatus according to claim 4, further comprising a reference data forming unit. 6. A first memory section, a voice output section, a second memory section, a voice input section, a response data forming section, a data comparing section,
The whole of the third memory unit and the operation control unit is mounted on the vehicle, and the person who inputs the response voice to the interrogation voice emitted from the voice output unit to the voice input unit is the driver of the vehicle. Claims 1 to 5 characterized by
The condition monitoring device according to any one of 1.