JP4701694B2 - Arousal level determination device and arousal level determination method - Google Patents

Description

  The present invention relates to an arousal level determination device and an arousal level determination method.

Conventionally, a technique for determining a subject's arousal level based on a heart rate has been used (see, for example, Patent Document 1). According to the technique described in Patent Document 1, a heartbeat fluctuation amount representing a fluctuation degree of a heart rate of a driver who is a subject is calculated, and when the heartbeat fluctuation amount is larger than a predetermined determination value, the driver is in a sleep state. It is determined that
Japanese Patent Laid-Open No. 7-201000

  According to the above technique, it is possible to detect the current arousal level of the subject, that is, whether or not the subject is in a dozing state. However, it is impossible to predict whether or not the arousal level of a subject who is awakened will decrease in the future.

  The present invention has been made to solve the above problems, and when a subject's arousal level is relatively high, the arousal level determination device capable of predicting whether or not the arousal level will be reduced in the future, and An object is to provide a method for determining arousal level.

A wakefulness level determination device according to the present invention acquires a heartbeat signal acquisition means for acquiring a heartbeat signal, a power spectrum of a heartbeat period time series from the heartbeat signal, divides the power spectrum into two predetermined frequency bands, Heart rate signal processing means for calculating a band power ratio of frequency bands and obtaining a time series thereof, obtaining a standard deviation for each predetermined section of the band power ratio time series and obtaining a heartbeat fluctuation section standard deviation; based on the heartbeat fluctuation interval standard deviation obtained by means heartbeat fluctuation in a awareness decrease prediction means alertness to predict whether decreases, the awareness decrease prediction means, alertness high section The average value and standard deviation of the interval standard deviation are obtained, a level threshold is set based on the average value and the standard deviation, and the heart rate fluctuation interval standard deviation is leveled in the interval. Threshold setting means to set a time threshold according to the time below the threshold value, and when the time when the heartbeat fluctuation interval standard deviation is below the level threshold continues for more than the time threshold, awakening It is determined that the degree is lowered .

The arousal level determination method according to the present invention includes a heartbeat signal acquisition step of acquiring a heartbeat signal, a heartbeat period time series power spectrum from the heartbeat signal, and dividing the power spectrum into two predetermined frequency bands. A heart rate signal processing step of calculating a band power ratio of each frequency band, obtaining a time series thereof, obtaining a standard deviation for each predetermined section of the band power ratio time series, and obtaining a heart rate fluctuation section standard deviation; based on the heartbeat fluctuation interval standard deviation obtained by the signal processing step, and a degree of awakening decreases prediction step of alertness to predict whether decreases, the awareness decrease prediction step, degree of awakening at a high interval The average value and standard deviation of the heart rate fluctuation interval of the heart rate are obtained, a level threshold is set based on the average value and the standard deviation, and the heart rate fluctuation in the above interval There is a threshold setting step that sets the time threshold according to the time during which the standard deviation falls below the level threshold, and the time when the heartbeat fluctuation interval standard deviation falls below the level threshold is greater than the time threshold When it continues, it is determined that the arousal level decreases .

  According to the wakefulness level determination apparatus or the wakefulness level determination method according to the present invention, whether or not the wakefulness level is predicted is predicted based on the heartbeat fluctuation interval standard deviation acquired from the heartbeat signal. Here, since the heartbeat fluctuation interval standard deviation acquired by the above-described calculation has a correlation with a future change in arousal level, whether or not the arousal level will be lowered in the future when the subject's arousal level is relatively high. Can be predicted.

  Set thresholds for level and time in sections with high arousal level, and determine that the level of arousal level will decrease if the time during which the heart rate fluctuation section standard deviation is below the level threshold continues for more than the time threshold level. Thus, it is possible to predict whether or not the arousal level is reduced more accurately.

  According to the present invention, a heartbeat fluctuation interval standard deviation having a correlation with a future arousal level change is acquired from the heartbeat signal, and whether or not the arousal level is reduced based on the acquired heartbeat fluctuation interval standard deviation. Since the prediction is configured, it is possible to predict whether or not the arousal level will be lowered in the future when the awakening level of the subject is relatively high.

  DESCRIPTION OF EMBODIMENTS Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. In the figure, the same reference numerals are used for the same or corresponding parts.

  First, the overall configuration of the arousal level determination device 1 according to the present embodiment will be described with reference to FIG. FIG. 1 is a block diagram showing the overall configuration of the arousal level determination device 1. Hereinafter, the case where the arousal level determination device 1 is mounted on a vehicle and a decrease in the arousal level of the driver of the vehicle is predicted will be described as an example.

  The arousal level determination device 1 predicts whether or not the driver's arousal level will be reduced in the future based on the heartbeat signal obtained from the driver, and presents the prediction result to the driver. For this purpose, the arousal level determination apparatus 1 includes a heart rate sensor 10 and an electronic control unit (hereinafter referred to as “ECU”) 20. The ECU 20 includes a heart rate signal processing unit 22, a threshold setting unit 24, and a wakefulness level decrease prediction unit. 26 is configured. In addition, the arousal level determination device 1 includes an output device 30 as means for presenting a prediction result.

  The heart rate sensor 10 is a potential type heart rate sensor that detects a pulsed voltage (cardiac potential) generated during myocardial contraction, and acquires a driver's heart rate signal (cardiac potential R wave). The heart rate sensor 10 detects a cardiac potential from an electrode attached to a steering wheel of a vehicle, for example. The heart rate sensor 10 outputs the acquired heart rate signal to the ECU 20. The heartbeat sensor 10 functions as a heartbeat signal acquisition unit and executes a heartbeat signal acquisition step.

  The ECU 20 includes a microprocessor that performs calculations, a ROM that stores a program for causing the microprocessor to execute each process, a RAM that stores various data such as calculation results, a backup RAM in which the stored contents are held by a battery, and the like. It is configured.

  The heartbeat signal processing unit 22 reads a heartbeat signal from the heartbeat sensor 10 at regular intervals, acquires a heartbeat period time series power spectrum from the heartbeat signal, divides the power spectrum into two predetermined frequency bands, The band power ratio of the frequency band is calculated to obtain the time series, the standard deviation is obtained for each predetermined section of the band power ratio time series, and the heartbeat fluctuation section standard deviation (hereinafter referred to as “heartbeat fluctuation SD (standard deviation)”) ) To get. That is, the heartbeat signal processing unit 22 functions as a heartbeat signal processing unit and executes a heartbeat signal processing step. The heart rate fluctuation SD acquired by the heart rate signal processing unit 22 is output to the arousal level decrease prediction unit 26.

  The wakefulness level decrease prediction unit 26 includes a threshold value setting unit 24, and predicts whether or not the driver's wakefulness level decreases based on the heartbeat fluctuation SD acquired by the heartbeat signal processing unit 22. To do. That is, the arousal level decrease prediction unit 26 functions as a wakefulness level decrease prediction unit, and executes a wakefulness level decrease prediction step.

  The threshold setting unit 24 obtains an average value and a standard deviation of the heart rate fluctuation SD in a section where the arousal level is high, sets a level threshold value based on the average value and the standard deviation, and beats the heart rate fluctuation in the section. The time threshold is set according to the time when the section standard deviation falls below the level threshold. That is, the threshold setting unit 24 functions as a threshold setting unit and executes a threshold setting step. The level threshold value and the time threshold value set by the threshold value setting means are used by the arousal level lowering prediction unit 26.

  The arousal level decrease predicting unit 26 determines that the arousal level is decreased when the time when the heartbeat fluctuation SD is below the level threshold value continues for a time threshold value or more.

  The output device 30 presents the awakening level decrease information input from the ECU 20 to the driver. The output device 30 includes a display for displaying character information and video information, a speaker for reproducing audio information, an actuator that vibrates a handle, a seat, and the like.

  Next, the operation of the arousal level determination device 1 and the arousal level determination method will be described with reference to FIG. FIG. 2 is a flowchart showing a processing procedure of a wakefulness level decrease prediction process by the wakefulness level determination device 1. This process is performed by the ECU 20, and is repeatedly executed at a predetermined timing from when the power source of the ECU 20 is turned on to when it is turned off.

  In step S100, a heartbeat signal is read from the heartbeat sensor 10. In subsequent step S102, a heartbeat period time series is acquired from the heartbeat signal read in step S100. Specifically, first, when the level of the heartbeat signal exceeds a predetermined threshold value, 1 is set, and when it is less than the threshold value, 0, the heartbeat signal is binarized and binarized. Get the signal. Next, the heart rate interval dt (second) is obtained from the period of the binarized signal, and the heart rate per minute is calculated by multiplying the reciprocal of the heart rate interval dt by 60. Then, the heart rate time series obtained by repeating the above calculation is interpolated to obtain a heartbeat period time series. FIG. 3 (a) shows an example of a heartbeat signal, and FIG. 3 (b) shows a binarized signal obtained by binarizing the heartbeat signal of (a). FIG. 3C shows a heartbeat cycle time series acquired from the binarized signal shown in FIG.

  Next, in step S104, the power cycle is acquired by performing FFT (Fast Fourier Transform) for each unit interval (for example, interval width of 1 minute) for the heartbeat period time series acquired in step S102. FIG. 4 shows a power spectrum obtained by performing FFT on the heartbeat period time series of FIG.

In subsequent step S106, first, the power spectrum acquired in step S104 is divided into a low frequency band (LF) having a frequency of 0.01 Hz to 0.15 Hz and a high frequency band (HF) having a frequency of 0.15 Hz to 0.50H. (See FIG. 4), and the respective bands are integrated to obtain a power area P _LOW in the low frequency band and a power area P _HIGH in the high frequency band. Next, the power area ratio (band power ratio) of both bands is calculated by the following equation (1), and the time-series change (hereinafter referred to as “heartbeat fluctuation L / H”) is acquired. FIG. 5 shows the heartbeat fluctuation L / H acquired.

  Subsequently, in step S108, the section standard deviation is obtained for each unit section TR (see FIG. 5) of the heartbeat fluctuation L / H obtained in step S106, and the time series change (heartbeat fluctuation SD) is obtained. FIG. 6 shows the obtained heartbeat fluctuation SD.

  Next, in step S110, a determination is made as to whether or not the elapsed time t from the start of operation exceeds a predetermined time tm. If the elapsed time t after the start of operation exceeds the predetermined time tm, the process proceeds to step S116. On the other hand, when the elapsed time t does not exceed the predetermined time tm, the process proceeds to step S112.

  In step S112, a determination is made as to whether or not the elapsed time t from the start of operation is a predetermined time tm. Here, when the elapsed time t after the start of operation is the predetermined time tm, the process proceeds to step S114. On the other hand, when the elapsed time t is not the predetermined time tm, that is, when the predetermined time tm has not yet elapsed after the start of operation, the process proceeds to step S100, and the processes of steps S100 to S112 described above are repeatedly executed.

In step S114, first, an average value M _TH and a standard deviation SD _TH of the heart rate fluctuation SD in the section TM where the arousal level is high are calculated. Then, the level threshold value σTH is set based on the average value M _TH and the standard deviation SD _TH , and the time threshold value tr is set according to the time during which the heartbeat fluctuation SD falls below the level threshold value σTH in the interval TM. Is set.

  In the present embodiment, the section from the start of driving until the predetermined time tm has elapsed after the start of driving is selected as the section TM having a high arousal level. This is because the degree of arousal is considered to be high immediately after the start of driving. In addition, it is preferable that the predetermined time tm for determining the length of the section TM is long, but the section TM is set so as not to include a section in which the arousal level is reduced.

Here, a method for setting the level threshold value σTH and the time threshold value tr will be described in detail with reference to FIG. First, the average value M _TH and the standard deviation SD _TH of the heart rate fluctuation SD in the section TM where the arousal level is high are calculated. Then, the level threshold value σTH is set from the average value M _TH and the standard deviation SD _TH by the following equation (2). Note that equation (2) is an example, and other functions may be used.

Next, in the section TM, a section in which the heartbeat fluctuation SD falls below the level threshold value σTH and each section length (time) Ln (n = 1, 2,..., N) are extracted. Then, a time threshold tr is set by the following equation (3a) from each extracted section length Ln. Note that the expression (3b) or another arithmetic expression may be used instead of the expression (3a).

  Returning to FIG. 2, the description will be continued. If step S110 is affirmed, in step S116, a determination is made as to whether or not the heart rate fluctuation SD is below the level threshold value σTH set in step S114 in order to predict a decrease in arousal level. . If the heartbeat fluctuation SD is less than the level threshold value σTH, the process proceeds to step S118. On the other hand, when the heartbeat fluctuation SD is greater than or equal to the level threshold value σTH, the process proceeds to step S100, and the processes of steps S100 to S110 and S116 are repeatedly executed.

  In step S118, whether or not the time during which the heart rate fluctuation SD is below the level threshold value σTH continues for the time threshold value tr or more, that is, the section TS in which the heart rate fluctuation SD is below the level threshold value σTH (FIG. A determination is made as to whether or not the section length (see 8) is equal to or greater than the time threshold tr.

  Here, when the section length of the section TS is less than the time threshold tr, the process proceeds to step S100 without determining that the driver's arousal level is reduced. Then, the processes of steps S100 to S110 and S114 and S116 are repeatedly executed.

  On the other hand, when the section length of the section TS is equal to or greater than the time threshold tr, it is determined that the driver's arousal level will decrease. In step S120, for example, information indicating that a reduction in arousal level is expected is displayed on the display of the output device 30. Thereafter, the process is exited.

According to the present embodiment, the level threshold value σTH and the time threshold value tr are set based on the average value M _TH and the standard deviation SD _TH of the heart rate fluctuation SD in the section TM where the arousal level is high. When the length of the section TS that is lower than the level threshold value σTH is equal to or longer than the time threshold value tr, it is determined that the arousal level will be reduced in the future. Here, since the heart rate fluctuation SD has a correlation with a future change in arousal level, it is possible to predict whether or not the awakening level will decrease in the future when the driver's arousal level is relatively high. .

  With reference to FIG. 9 to FIG. 12, according to the present embodiment, when the driver's arousal level is relatively high, it is possible to predict a decrease in the arousal level according to Examples 1 and 2. Shown in In Examples 1 and 2, the heart rate fluctuation SD obtained from the driver's heart rate signal and the arousal level (drowsiness level) were measured. The evaluation of the driver's drowsiness level includes "Facial expression rating for objective evaluation of arousal level, NEDO (New Energy and Industrial Technology Development Organization), Human Sensory Measurement Manual Part 1 Chapter 2 Chapter 146 Page "was used.

(Example 1)
A heartbeat fluctuation SD was obtained by detecting a heartbeat signal of the driver by a simulated running test using a driving simulator (test conditions: a round course, no other vehicle). In a section immediately after the start of driving where the degree of arousal is evaluated to be high (see FIG. 9), the average value M _TH 1 and the standard deviation SD _TH 1 of the heart rate fluctuation SD are obtained, and the level threshold value σTH1 and the time threshold value tr1 are obtained. Set. In this embodiment, the average value M _TH 1 is 7.692, the standard deviation SD _TH 1 is 3.4627, the level threshold value σTH1 is set to 4.229, and the time threshold value tr1 is set to 1 minute 10 seconds. . The above equation (2) was used to set the level threshold value σTH1. At that time, the coefficient A was set to 1. The above equation (3a) was used for setting the time threshold value tr1. At that time, N was 4.

  When the heart rate fluctuation SD was continuously acquired, as shown in FIG. 10, a section TS1 that was lower than the level threshold value σTH1 continuously for about 5 minutes and 40 seconds (more than the time threshold value tr1) was detected. At this time, the arousal level determination device 1 determines that the awakening level will decrease in the future. The sleepiness level at the time when the section TS1 was detected was about 4, and the arousal level was relatively high. After the interval TS1 was detected, the drowsiness level began to decrease after about 30 minutes, and the drowsiness level decreased to about 1 after about 40 minutes. From the above, the effectiveness of the present embodiment was confirmed.

(Example 2)
In the same manner as in Example 1, a heartbeat fluctuation SD was obtained by detecting a heartbeat signal of the driver by a simulated running test using a driving simulator (test conditions: a round course, no other vehicle). In a section immediately after the start of driving where the degree of arousal is evaluated to be high (see FIG. 11), the average value M _TH 2 and the standard deviation SD _TH 2 of the heart rate fluctuation SD are obtained, and the level threshold value σTH2 and the time threshold value tr2 are obtained. Set. In this embodiment, the average value M _TH 2 is 2.195, the standard deviation SD _TH 2 is 1.171, the level threshold value σTH2 is set to 1.023, and the time threshold value tr2 is set to 1 minute 360 seconds. . The above equation (2) was used to set the level threshold σTH2. At that time, the coefficient A was set to 1. Further, the above equation (3a) was used to set the time threshold value tr2. At that time, N was 6.

  When the heartbeat fluctuation SD was continuously acquired, as shown in FIG. 12, a section TS2 that was lower than the level threshold value σTH2 continuously for about 6 minutes and 50 seconds (more than the time threshold value tr2) was detected. At this time, the arousal level determination device 1 determines that the awakening level will decrease in the future. The sleepiness level at the time when the section TS2 was detected was about 6, and the arousal level was high. The drowsiness level began to decrease after the interval TS2 was detected, and the drowsiness level decreased to about 2 after about 50 minutes. From the above, the effectiveness of the present embodiment was confirmed.

  Although the embodiment of the present invention has been described above, the present invention is not limited to the above embodiment, and various modifications can be made. For example, the place where the arousal level determination device is mounted is not limited to the vehicle, and the subject for whom the reduction in the arousal level is predicted is not limited to the driver of the vehicle.

  Further, the method of selecting the section TM for setting the level threshold value σTH and the time threshold value tr is not limited to the above embodiment. For example, the section TM may be selected by recognizing a state with a high arousal level from the brain wave of the driver, the steering angular velocity of the steering wheel, or the movement of the line of sight.

  Furthermore, as a sensor for acquiring a driver's heartbeat signal, in addition to a potential heartbeat sensor, an infrared heartbeat sensor that detects a reflected amount of infrared light that periodically changes according to the heartbeat, and a sensor that detects a driver's blood pressure. Etc. can be used.

It is a block diagram which shows the whole structure of the arousal level determination apparatus which concerns on embodiment. It is a flowchart which shows the process sequence of the arousal level fall prediction process by the arousal level determination apparatus which concerns on embodiment. (A) is a figure showing an example of a heartbeat signal. (B) is a figure which shows the binarized signal acquired by binarizing the heart rate signal of (a). (C) is a figure which shows the cardiac cycle time series acquired from the binarization signal of (b). It is a figure which shows an example of the power spectrum of a cardiac cycle time series. It is a figure which shows an example of heartbeat fluctuation L / H. It is a figure which shows an example of heartbeat fluctuation SD. It is a figure for demonstrating the setting method of a level threshold value and a time threshold value. It is a figure for demonstrating the fall determination method of a wakefulness degree. It is a figure which shows the sleepiness level and heart rate fluctuation | variation SD of the area where arousal level is high. It is a figure which shows the relationship between heart rate fluctuation SD and arousal level fall. It is a figure which shows the sleepiness level and heart rate fluctuation | variation SD of the area where arousal level is high. It is a figure which shows the relationship between heart rate fluctuation SD and arousal level fall.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Arousal level determination apparatus, 10 ... Heart rate sensor, 20 ... ECU, 22 ... Heart rate signal process part, 24 ... Threshold value setting part, 26 ... Arousal level fall prediction part.

Claims (2)

Heart rate signal acquisition means for acquiring a heart rate signal;
A power spectrum of a heartbeat cycle time series is obtained from the heartbeat signal, the power spectrum is divided into two predetermined frequency bands, a ratio of band powers of each frequency band is calculated to obtain the time series, and the band power A heartbeat signal processing means for obtaining a standard deviation for each predetermined section of the specific time series and obtaining a heartbeat fluctuation section standard deviation;
Wakefulness decrease prediction means for predicting whether or not the wakefulness level is lowered based on the heartbeat fluctuation interval standard deviation obtained by the heartbeat signal processing means ,
The arousal level decrease predicting means includes:
An average value and a standard deviation of the heartbeat fluctuation section standard deviation in a section with a high arousal level are obtained, a level threshold is set based on the average value and the standard deviation, and the heartbeat fluctuation section standard deviation in the section is Threshold setting means for setting a time threshold according to the time below the level threshold;
The wakefulness level determination apparatus , wherein when the time during which the heartbeat fluctuation section standard deviation is below the level threshold value continues for the time threshold value or longer, it is determined that the wakefulness level is reduced . A heartbeat signal acquisition step for acquiring a heartbeat signal;
A power spectrum of a heartbeat cycle time series is obtained from the heartbeat signal, the power spectrum is divided into two predetermined frequency bands, a ratio of band powers of each frequency band is calculated to obtain the time series, and the band power A heartbeat signal processing step for obtaining a standard deviation for each predetermined section of the specific time series and obtaining a heartbeat fluctuation section standard deviation;
Based on the heart rate fluctuation interval standard deviation obtained in the heart rate signal processing step, and predicting whether or not the arousal level is reduced ,
The arousal level decrease prediction step includes:
An average value and a standard deviation of the heartbeat fluctuation section standard deviation in a section with a high arousal level are obtained, a level threshold is set based on the average value and the standard deviation, and the heartbeat fluctuation section standard deviation in the section is A threshold setting step for setting a time threshold according to the time below the level threshold;
The wakefulness determination method , wherein when the time during which the heartbeat fluctuation section standard deviation falls below the level threshold value continues for the time threshold value or longer, it is determined that the wakefulness level decreases .

Images