JP2017060583A - Doze determination system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a doze determination system capable of determining whether or not a user will have a doze within a predetermined time from when the user starts to drive.SOLUTION: A doze determination system 100 includes: a first calculation unit 30b for calculating a sleeping latent time Ta before a user M driving based on biological information for the user M detected by a biological information detection unit 11; a third calculation unit 30d for calculating a deep sleep maximum duration time Tb based on the biological information; a fourth calculation unit 30e for calculating a first feature amount A by dividing the sleeping latent time Ta by a first coefficient α; a fifth calculation unit 30f for calculating a second feature amount B by dividing a second coefficient β by the deep sleep maximum duration time Tb; and a doze determination unit 30g for determining that the user M will have a doze within a predetermined lapse time D from when the user M starts to drive when it is determined that at least one of the first feature amount A and the second feature amount B is larger than a predetermined reference value C.SELECTED DRAWING: Figure 3

Description

  The present invention relates to a dozing determination system.

  Conventionally, as disclosed in Patent Document 1, the sleep time and sleep of the driver are detected from the operating time of the fluorescent lamp, the TV off time, information on the snoring of the driver, and external noise detected from the driver's home. A system for determining the quality of the driver and estimating the sleepiness of the driver has been proposed.

Japanese Patent No. 4211841

  In the system disclosed in Patent Document 1, when it is determined that the driver is drowsy, it is not possible to determine whether or not to fall asleep within a predetermined time from the start of driving due to this drowsiness.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a dozing determination system that can determine whether or not to fall asleep within a predetermined time from when the user starts driving. To do.

  In order to solve the above-mentioned problem, the invention of the dozing determination system according to claim 1 is a system for determining dozing when a user drives, a living body information detecting unit for detecting the living body information of the user, Based on the biometric information detected by the biometric information detection unit, in sleep before driving, which is sleep before driving of the user, sleep onset that is an elapsed time from a time when the user enters a sleep posture to a sleep onset time Based on the first calculation unit that calculates the latency, the second calculation unit that calculates the sleep depth in the sleep before driving based on the biological information detected by the biological information detection unit, and the second calculation unit Based on the sleep depth in the pre-driving sleep, the sleep depth is shallower than a predetermined reference value and the sleep is deeper than the shallow sleep. A third calculation unit that calculates the duration of the deep sleep lasted when divided into deep sleep deeper than a quasi-value, and the sleep latency calculated by the first calculation unit is divided by a first coefficient A fourth calculation unit that calculates the first feature value, a fifth calculation unit that calculates the second feature value by multiplying the reciprocal of the duration calculated by the third calculation unit by a second coefficient, When it is determined that at least one of the first feature value calculated by the fourth calculation unit and the second feature value calculated by the fifth calculation unit is larger than a predetermined reference value, the user A dozing determination unit that determines that the user is falling asleep within a predetermined elapsed time from the start of the driving.

  The inventor of the present invention includes a first feature amount obtained by dividing a user's sleep sleep latency by a first coefficient, a second feature amount obtained by multiplying the reciprocal of the user's deep sleep maximum duration by a second coefficient, and the user driving It has been found that there is a correlation between the time from the start of sleep and the elapsed time from the start of dozing of the user. Then, the inventor of the present invention, when at least one of the first feature value and the second feature value is larger than a predetermined reference value, within a predetermined elapsed time from when the user starts driving, I found out to fall asleep. Based on this knowledge, the dozing determination unit determines that at least one of the first feature value and the second feature value is larger than a predetermined reference value, and within a predetermined elapsed time from when the user starts driving. In addition, it is determined that the user is falling asleep. Accordingly, it is possible to provide a dozing determination system that can determine whether or not the user falls asleep within a predetermined time from the time when the user starts driving.

It is the graph showing the relationship between the 1st feature-value A and the 2nd feature-value B in sleep before a test subject's driving | operation, and the time from the driving | operation start after sleep of a test subject to the start of dozing. It is a graph showing the relationship between sleep depth and time in sleep before driving of a subject. It is a block diagram of a dozing determination system. FIG. 4 is a flowchart of a “slumber determination process” executed by the snooze determination apparatus 30 of FIG. 3. It is an example of a sleep determination formula.

(Outline of activity determination)
The inventor of the present invention, as shown in the graph of FIG. 1, from the first feature amount A and the second feature amount B in sleep before driving of subjects A to I, and from the start of driving of subjects A to I to the start of dozing The test was repeated about the relationship with the time. Here, the sleep before driving of the test subjects A to I refers to the previous sleep of the driving test subjects A to I.
As shown in the following formula (1), the first feature amount A is a value obtained by dividing the sleep latency Ta of the subjects A to I by a first coefficient α (for example, 20) and normalized.
A = Ta / α (1)
A: First feature amount Ta: Sleeping latency α: First coefficient Note that, as shown in FIG. 2, the sleeping latency Ta is the time at which subjects A to I enter a state of going to sleep, This is the elapsed time from the floor time to the sleep start time. In addition, bedtime means the time when the user M lay down on the bedding, for example.

The second feature amount B is a value normalized by multiplying the reciprocal of the deep sleep maximum duration Tb of the subjects A to I by a second coefficient β (for example, 25) as shown in the following equation (2).
B = 1 / Tb × β (2)
B: second characteristic amount Tb: deep sleep maximum duration β: second coefficient Deep sleep maximum duration Tb (the duration described in the claims) is the relationship between subjects A to I as shown in FIG. In 1 sleep, deep sleep is the most continuous time.
Deep sleep means sleep that is deeper than shallow sleep that is shallower than a predetermined reference value, that is, sleep that has deeper sleep than the predetermined reference value. In the present embodiment, when the sleep depths of subjects A to I are classified into REM sleep, sleep first depth, sleep second depth, sleep third depth, and sleep fourth depth as sleep becomes deep, REM sleep. The first sleep depth and the second sleep depth are shallow sleep, and the third sleep depth and the fourth sleep depth are deep sleep.

As shown in FIG. 2, during one sleep of the subjects A to I, a sleep cycle in which the sleep depth of the subjects A to I becomes deeper or shallower with a predetermined cycle is repeated a plurality of times. Then, after a plurality of sleep cycles, subjects A to I awaken.
In addition, the number attached | subjected to A-I represents the order by which the 1st feature-value A and the 2nd feature-value B about each test subject A-I were measured on the day.

  It can be said that the longer the sleep latency Ta, that is, the greater the first feature amount A, the lower the quality of sleep of the user M, and the easier it is for the user M to fall asleep after the start of driving. In addition, it can be said that the shorter the deep sleep maximum duration Tb, that is, the greater the second feature amount B, the lower the quality of the sleep of the user M, and the easier it is for the user M to fall asleep after starting driving. Then, as shown in FIG. 1, the inventor of the present invention has a correlation between the first feature amount A and the second feature amount B and the elapsed time from the start of driving of the subjects A to I to the start of dozing. I found out. Then, the inventor of the present invention starts the driving of the subjects A to I when at least one of the first feature value A and the second feature value B is larger than a predetermined reference value C (for example, 1). It was found that subjects A to I fall asleep within a predetermined elapsed time D (for example, 16 minutes) from the time.

The term “sleeping” as used herein means that the sleepiness level based on the evaluation method of NEDO (New Energy and Industrial Technology Organization) is sleepiness level 5. The NEDO evaluation method is a method for estimating sleepiness based on facial expressions of subjects A to I taken by a video camera. The sleepiness level based on the NEDO evaluation method is shown below.
Sleepiness level 1: Not sleepy at all (The movement of the line of sight is fast and frequent. The blinking is active at a stable cycle of about 2 times every 2 seconds and is accompanied by movement of the body.)
Sleepiness level 2: Slightly sleepy (lips are open. Gaze movement is slow)
Sleepiness level 3: Sleepy (blinks occur slowly and frequently. Mouth movement. Re-sitting. Hand on face.)
Sleepiness level 4: It seems to be quite sleepy (There are blinks that seem to be conscious. Yawning occurs frequently, deep breathing is also seen. Blinks and slow movement of eyes.)
Sleepiness level 5: Very sleepy (closes eyelids, head tilts forward, head falls back)

The first coefficient α, the second coefficient β, and the reference value C are appropriately set according to the value of the predetermined elapsed time D.
As described above, when at least one of the first feature amount A and the second feature amount B is larger than the reference value C, the subject within a predetermined elapsed time D from the time when the subjects A to I start driving. The dozing determination system 100 of the present embodiment invented based on the knowledge that A to I doze will be described below.

(Configuration of dozing system)
Below, the structure of the dozing determination system 100 is demonstrated using FIG. The dozing determination system 100 includes a biological information detection device 10, a mobile terminal device 20, and a dozing determination device 30.

  The biological information detection apparatus 10 includes a biological information detection unit 11 and a control unit 15. The biological information detection unit 11 is a sensor that detects at least one of body motion vibration and pulse that are biological information of the user M, and outputs the detected information to the control unit 15. The biological information detection unit 11 is, for example, a pressure sensor such as a piezoelectric element or a capacitance type, a radio wave sensor such as a radar Doppler sensor, or an acceleration sensor. In the present embodiment, the biological information detection unit 11 is provided on the upper surface portion of a bedding 19 such as a mat that the user M sleeps lying on the upper surface when sleeping. When the user M lies down on the bedding 19, the bedding 19 is repeatedly pressed by the body movement vibration and pulsation of the user M. The pressure generated by repeatedly pressing the bedding 19 propagates to the biological information detection unit 11, the biological information detection unit 11 is repeatedly pressed, and pressure is repeatedly applied to the biological information detection unit 11. Then, the biological information detection unit 11 detects the repeatedly applied pressure as the biological information (body motion information, pulsation information) of the user M, and outputs the detection result to the control unit 15.

  The control unit 15 is connected to the biological information detection unit 11. The control unit 15 includes a biological information transmission unit 15a that wirelessly communicates with a wireless communication unit 20a of the mobile terminal device 20 described later. The biological information transmission unit 15 a transmits the biological information detected by the biological information detection unit 11 to the wireless communication unit 20 a of the mobile terminal device 20.

  The mobile terminal device 20 is a device carried by the user M. The mobile terminal device 20 includes a so-called smartphone or tablet PC. The portable terminal device 20 includes a wireless communication unit 20a and a biological information storage unit 20b. The wireless communication unit 20a communicates wirelessly with the biological information transmission unit 15a and a biological information reception unit 30a described later. The communication method between the wireless communication unit 20a and the biological information transmission unit 15a and the communication method between the wireless communication unit 20a and the biological information reception unit 30a include wireless LAN, Bluetooth (registered trademark), infrared communication, and the like. The biological information storage unit 20b stores the biological information transmitted from the biological information transmission unit 15a in association with the time when the biological information is detected.

  The dozing determination device 30 is mounted on the vehicle 200 that the user M drives. The dozing determination device 30 includes a biological information receiving unit 30a, a first calculation unit 30b, a second calculation unit 30c, a third calculation unit 30d, a fourth calculation unit 30e, a fifth calculation unit 30f, a dozing determination unit 30g, and a notification unit. 30h. The biometric information receiving unit 30a communicates with the radio communication unit 20a of the mobile terminal device 20, and receives the biometric information of the user M associated with the detection time transmitted by the radio communication unit 20a.

  The processes executed by the first calculation unit 30b, the second calculation unit 30c, the third calculation unit 30d, the fourth calculation unit 30e, the fifth calculation unit 30f, the dozing determination unit 30g, and the notification unit 30h are shown in FIG. This will be described in detail with reference to a flowchart.

  The vehicle 200 is provided with a notification device 50. The notification device 50 notifies the user M of information, and includes, for example, a display device such as a car navigation device, a speaker, a display, and an indicator.

(Dozing determination process)
Hereinafter, the “sleeping determination process” executed by the sleep detection device 30 will be described with reference to the flowchart shown in FIG. 4. When the ignition of the vehicle 200 is turned on, the “sleeping determination process” starts, and the program proceeds to step S11.
If it is determined in step S11 that the dozing determination device 30 has received the biometric information of the user M associated with the detection time via the biometric information receiving unit 30a (step S11: YES), the program is stepped. Proceed to S12. On the other hand, when the dozing determination device 30 determines that the biometric information of the user M associated with the detection time has not been received (step S11: NO), the process of step S11 is repeated.

  In step S <b> 12, the first calculation unit 30 b first receives biometric information when the user M lies on the bedding 19 and starts to bed based on the biometric information of the user M associated with the detection time. The detected time is recognized as the bedtime of the user M. And the 1st calculating part 30b recognizes the sleep start time of the user M based on the biometric information matched with the detection time. Specifically, the first calculation unit 30b determines whether or not the user M has fallen asleep based on the body motion vibration of the user M. Note that any method may be used for the determination of sleep onset. For example, the calculation may be performed by a method disclosed in Japanese Patent Application Laid-Open No. 2014-70766. That is, first, the number of body movements per unit time of the user M lying on the bedding 19 is calculated based on the body movement which is the biological information detected by the biological information detection unit 11. Based on the calculated number of body movements, the sleep state and the awake state are determined. Specifically, the number of fluctuations in the load applied to the biological information detection unit 11 is converted into the number of body movements, and the number of body movements is stored. Then, the number of body movements is converted into an index Z representing the sleep state from the sleep determination formula illustrated in FIG. If the index Z exceeds 0, it is determined that the user M is in an awake state. If the index Z is 0 or less, it is determined that the user M is asleep and is in a sleep state. And the 1st calculating part 30b calculates the sleep onset latency Ta which is the elapsed time from bedtime to the sleep start time.

  In step S13, the 2nd calculating part 30c calculates the sleep depth in the last sleep of the user M based on the biometric information matched with the detection time. For example, the second calculation unit 30c calculates that the user M is at a deeper sleep depth as the body motion vibration of the user M per unit time is smaller. Note that a method for calculating the sleep depth based on the biological information of the user M is a known technique as disclosed in JP-A-2002-291710, JP-A-2009-153550, JP-A-2011-67241, and the like. Because there is, explanation here is omitted.

In step S14, the third calculation unit 30d calculates the above-described deep sleep maximum duration Tb of the user M based on the sleep depth of the user M calculated by the second calculation unit 30c.
In step S15, the fourth computation unit 30e computes the first feature amount A by dividing the sleep onset latency Ta computed by the first computation unit 30b by the first coefficient α using the above equation (1). .
In step S16, the fifth calculation unit 30f calculates the second feature amount B by multiplying the reciprocal of the maximum sleep duration Tb of the user M by the second coefficient β using the above equation (2).

  In step S17, when the dozing determination unit 30g determines that at least one of the first feature value A calculated in step S15 and the second feature value B calculated in step S16 is greater than the reference value C. (Step S17: YES), the program proceeds to Step S18. On the other hand, when the dozing determination unit 30g determines that both the first feature value A calculated in step S15 and the second feature value B calculated in step S16 are equal to or less than the reference value C, ( Step S17: NO), the program proceeds to Step S19.

In step S18, the dozing determination unit 30g determines to fall asleep within a predetermined elapsed time D from the time point when the user M starts driving.
In step S19, the dozing determination unit 30g determines that the user M does not fall asleep within the predetermined elapsed time D from the time when the user M starts driving, and ends the “sleeping determination process”.

In step S <b> 20, the notification unit 30 h causes the notification device 50 to notify the user M of information related to dozing.
The following (1) to (3) are included for information related to the dozing of the user M.
(1) Information indicating that the user M will fall asleep within the predetermined elapsed time D from the start of driving (2) Information indicating that the user M must be driven with sufficient caution (3) Information indicating that the user should not drive or the notification unit 30h May be an embodiment in which the notification device 50 that is a car navigation device is informed of the following (4) and (5) information related to the dozing of the user M.
(4) User M's rest timing calculated from the time when the user M starts to fall asleep (predetermined elapsed time D) and the expected time to reach the destination, and the rest that is present on the travel route Location information (5) Information on a resting place within a range that the user M can reach from the start of driving until the predetermined elapsed time D elapses, a route to the resting place, and a route from the resting place to the destination Step S20 When the process ends, the “sleeping determination process” ends.

(Effect of this embodiment)
As is clear from the above description, the inventor of the present invention has the first feature amount A obtained by dividing the sleep sleep latency Ta in the sleep before the driving of the user M by the first coefficient α, and the sleep before the driving of the user M. There is a correlation between the second feature amount B obtained by multiplying the reciprocal of the deep sleep maximum duration Tb by the second coefficient β and the elapsed time from the time when the user M starts driving until the user M begins to fall asleep. I found out. Then, the inventor of the present invention, when at least one of the first feature amount A and the second feature amount B is larger than the predetermined reference value C, the predetermined elapsed time D from the time when the user M starts driving. Within, we found that user M is asleep. Based on this knowledge, when the dozing determination unit 30g determines that at least one of the first feature value A and the second feature value B is larger than the reference value C (step S17 in FIG. 4: YES), the user It is determined that the user M falls asleep within a predetermined elapsed time D from the time when M starts driving (step S18 in FIG. 4). Thereby, the dozing determination system 100 which can determine whether the user M falls asleep within the predetermined elapsed time D from the driving | operation start can be provided.

  When the dozing determination unit 30g determines that the user M will fall asleep within the predetermined elapsed time D from the time when the user M starts driving (step S18 in FIG. 4), the notification unit 30h notifies the notification device 50 to the user. Information about M's doze is notified. Thereby, the user M can perceive in advance that he / she sleeps while driving, and can avoid an accident caused by sleeping while driving.

  The biological information detection unit 11 is provided in the bedding 19 where the user M lies down and sleeps during sleep. For this reason, the biometric information of the user M is detected by the biometric information detection unit 11 provided in the bedding 19 simply by the user M lying on the bedding 19 during sleep. For this reason, it is not necessary for the user M to attach the biological information detection unit before sleeping, and the user M does not feel bothered before sleeping, compared to the biological information detection unit attached to the body of the user M. Sleeping is not hindered.

(Other embodiments)
In the above embodiment, the biological information of the user M acquired by the biological information detection device 10 is input to the dozing determination device 30 via the mobile terminal device 20 carried by the user M. However, there may be an embodiment in which the biometric information of the user M acquired by the biometric information detection apparatus 10 is input to the dozing determination apparatus 30 via a storage medium such as a memory card or a USB memory. Alternatively, the biometric information of the user M acquired by the biometric information detection device 10 may be an embodiment in which the biometric information of the user M is input to the doze determination device 30 via a mobile data communication line.

  The biological information detection unit 11 may be provided on a wristwatch or wristband worn on the wrist of the user M. Alternatively, the biological information detection unit 11 may be a plurality of electrodes directly attached to the chest of the user M or a plurality of electrodes provided on clothes worn by the user M and in contact with the user M. In this embodiment, the heartbeat of the user M is detected as pulsation information based on the potential difference between the plurality of electrodes in contact with the user M. Alternatively, the biological information detection unit 11 may be a photoelectric pulse wave sensor that detects the pulse wave of the user M as the pulsation information by detecting the volume change of the blood vessel of the fingertip or earlobe of the user M.

  The first calculation unit 30b, the second calculation unit 30c, the dozing determination unit 30g, and the notification unit 30h may be an embodiment provided in the biological information detection device 10 or the portable terminal device 20.

  Embodiment in which the information that the user M falls asleep within a predetermined elapsed time D from the time when the user M starts driving is transmitted from the dozing determination device 30 to the operation management system or the operation manager of the transportation company or taxi company. However, there is no problem. In this embodiment, when information indicating that the user M is falling asleep within a predetermined elapsed time D from the time when the user M starts driving is transmitted to the operation management system or the operation manager, the operation management system or the operation manager. Shows the information on the resting place within the range that the user M can reach from the start of driving until the predetermined elapsed time D elapses, the route to the resting place, and the route from the resting place to the destination to the dozing determination device 30. The information is transmitted to the notification device 50. Alternatively, the operation manager notifies the user M of the above information.

  In the embodiment described above, the biological information detection unit 11 is provided on the upper surface portion of the bedding 19 such as a mat or the like that the user M sleeps lying on the upper surface when sleeping. However, the biological information detection unit 11 may be an embodiment provided in a vehicle seat driven by the user M. In this embodiment, when the user M stays in the vehicle before driving, the sleep sleep latency Ta and the deep sleep maximum duration Tb are detected by the biological information detection unit 11 provided on the seat, and the user M starts driving. It is determined whether or not the user M falls asleep within a predetermined elapsed time D from the time point when the user did.

  In the present embodiment, the first feature amount A and the second feature amount B are calculated based on the sleep sleep latency Ta and the deep sleep maximum duration Tb of the previous sleep of the user M who is driving, and the user M performs the driving. It is determined whether or not the user M falls asleep within a predetermined elapsed time D from the start time. However, the first feature amount A and the second feature amount B are calculated based on the sleep sleep latency Ta and the deep sleep maximum duration time Tb before and after the driving user M, and the user M performs driving. It may be an embodiment in which it is determined whether or not the user M falls asleep within a predetermined elapsed time D from the start time. Note that, as in this embodiment, based on the sleep sleep latency Ta and deep sleep maximum duration Tb of the user M driving last time, the user M within a predetermined elapsed time D from the time when the user M started driving. In the embodiment in which M determines whether or not to fall asleep, it is highly accurate to determine whether or not the user M will fall asleep within a predetermined elapsed time D from the time when the user M started driving. it can.

  DESCRIPTION OF SYMBOLS 11 ... Living body information detection part, 19 ... Bedding, 30b ... 1st calculating part, 30c ... 2nd calculating part, 30d ... 3rd calculating part, 30e ... 4th calculating part, 30f ... 5th calculating part, 30g ... Dozing determination Part, 30h ... notification part, 50 ... notification apparatus, 100 ... slumber determination system

Claims (3)

A system for determining dozing when a user drives,
A biological information detection unit for detecting the biological information of the user;
Based on the biometric information detected by the biometric information detection unit, in sleep before driving, which is sleep before driving of the user, sleep onset that is an elapsed time from a time when the user enters a sleep posture to a sleep onset time A first calculation unit for calculating the latency;
Based on the biological information detected by the biological information detection unit, a second calculation unit that calculates sleep depth in the sleep before driving;
Based on the sleep depth in the sleep before driving calculated by the second calculation unit, the sleep depth is shallower than a predetermined reference value and the sleep is deeper than the shallow sleep than the predetermined reference value. A third calculation unit that calculates the duration of the deep sleep when it is divided into deep deep sleep,
A fourth computing unit that computes a first feature amount by dividing the sleep latency calculated by the first computing unit by a first coefficient;
A fifth calculation unit that calculates a second feature amount by multiplying a reciprocal of the duration calculated by the third calculation unit by a second coefficient;
When it is determined that at least one of the first feature value calculated by the fourth calculation unit and the second feature value calculated by the fifth calculation unit is larger than a predetermined reference value, the user A dozing determination system comprising: a dozing determination unit that determines that the user is falling asleep within a predetermined elapsed time from the start of the driving. An informing device for informing information about the user's dozing;
A notifying unit for notifying the notifying device of information related to the user's dozing when the dozing determining unit determines that the user is falling asleep within the predetermined elapsed time from the time when the user starts the driving; The dozing determination system according to claim 1, comprising:   The dozing determination system according to claim 1 or 2, wherein the biological information detection unit is provided in a bedding on which the user lies down and sleeps during sleep.

Images