JP2671394B2 - Driver abnormal condition detection device - Google Patents

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an abnormal driver state detection device for detecting and notifying an abnormal mental or physical state of a driver based on the pulse rate detected by a pulse sensor.

2. Description of the Related Art Conventionally, as devices for detecting an abnormal state of a driver, those disclosed in Japanese Utility Model Laid-Open No. 62-9404 and Japanese Patent Laid-Open No. 62-15129 have been proposed. That is, these devices detect the pulse rate of the driver by the pulse sensor, and when the average value, variance, variance / average value, etc. of the pulse rate per unit time are equal to or greater than a reference value, or when a plurality of pulse sensors are used. When a plurality of detected signals detected are both above a predetermined level, the driver discriminates the excitement state or the abnormal tension state and notifies the driver.

DISCLOSURE OF THE INVENTION Problems to be Solved by the Invention However, in such a conventional device, the setting of the unit time is performed at an arbitrary timing. Therefore, due to the difference in the setting timing of the unit time, the pulse within the unit time is The numbers are very different. That is, as a method of counting the pulse rate, a method of time-sequentially counting the R wave of the electrocardiographic waveform shown in FIG. 17 (the waveform portion where the strongest pulsation is felt) as R _n , R _{n + 1} , ... , R wave interval time Tn is measured, converted into pulse rate per 60 seconds each time, and the instantaneous heart rate H shown in the following formula is measured.

H = 60 / Tn In any of the methods, the number of counts is greatly different due to the difference in the unit time setting timing. This means that, for example, in the area A shown in FIG. It can be easily understood because the value of Tn and the value of H are different from the case of counting by II.

In addition, the person to be measured is generally placed in a resting state when detecting the pulse rate. This is because in order to eliminate the fluctuation of the pulse rate due to the muscle movement, since the mental factors (excitement, tension), muscle movement, and diseases are complicatedly mixed in the factors of the fluctuation of the pulse rate. Thus, by keeping the subject in a resting state in this manner, it becomes possible to measure the pulse rate effective for judging the activity level of the cerebrum. Therefore, even when detecting the pulse rate of the driver, it is preferable for the driver to be in a resting state to judge the activity level of the cerebrum, but for the driver, the movement of the body caused by the driving action etc. However, the increase in the pulse rate due to the movement of the body is unavoidable, as is always the case.

That is, Fig. 19 is a record of pulse fluctuations during normal operation ("Railway Labor Science", '73, No. 27). From Fig. 19, pulse rate and fluctuations during operation accompanying clutch operation and brake operation are recorded. , A decrease in pulse rate due to signal stop, and a decrease in fluctuations are observed.

Therefore, when the pulse rate is increased and the driver's abnormality is notified by detecting only the pulse rate and notifying the increase in the pulse rate due to the exercise operation as in the conventional device described above, the increase is detected. Whether the cause is due to muscle exercise,
It is difficult to determine whether it is due to an actual driver abnormality,
It was not always possible to accurately announce the abnormal condition of the driver.

The present invention has been made in view of such conventional problems, and an object thereof is to provide a driver abnormal state detection device capable of detecting an abnormal state of the driver more accurately. is there.

Means for Solving the Problems In the driver abnormal state detection device of the present invention in order to solve the problems, a pulse sensor for detecting the pulse rate of the driver is provided, and the pulse rate detected by the pulse sensor. In the device for determining an abnormal state of the driver based on the above, in accordance with a combination of the calculating means for calculating the average value and the variance of the pulse rate, the magnitude of the average value and the variance value of the pulse rate, a plurality of in advance. While the weighting coefficient is set, from among the plurality of weighting coefficients, one of the weighting coefficients corresponding to the combination of the respective calculated values of the average value and the variance of the pulse rate is specified, and this is specified. Based on the weighting coefficient and the cumulative fatigue level in the past, a new cumulative fatigue level of the driver is calculated, and a fatigue index calculation means for determining an abnormal condition of the driver according to the calculation result. It is provided. In addition, a body movement sensor for detecting the movement of the body of the driver is provided, and the reference pulse rate or the reference dispersion value is changed according to the movement of the body detected by the body movement sensor, or the abnormal level is calculated and estimated, It is configured to determine the abnormal state of the driver based on the abnormal level.

Action In the above configuration, the pulse sensor detects the pulse of the driver as an electric signal, converts it into a pulse wave and outputs it to the calculating means, and the calculating means calculates the average value and variance of the pulse rate, and the fatigue calculating means Output. Then, the fatigue degree calculating means
Depending on the average value and the magnitude of the variance, one of the corresponding weighting factors is selected from among a plurality of preset weighting factors, and the weighting factor and the cumulative fatigue level up to this point are used to create a new value. Calculate the cumulative fatigue level. Then, it is determined whether or not the cumulative fatigue level exceeds a reference value, and if it exceeds, it is determined that the driver is in an abnormal state, and an alarm or the like is activated to notify this.

When the driver performs an operation such as a driving operation, the body motion sensor senses a movement of the body of the driver who performed the driving operation, and based on the movement of the driver sensed by the body movement sensor,
The reference pulse rate and the reference variance value for determining the abnormal state of the driver change. Therefore, if there is an increase in the pulse rate due to the muscle movement, that is, the driver's driving operation, the increase in the pulse rate is determined to be an abnormal state by changing the reference pulse rate or the reference variance value. There is no such thing.

Furthermore, from the movement of the driver detected by the body motion sensor,
By determining the standard driving state of the driver, the abnormal level of the pulse rate can be calculated and estimated with reference to the pulse rate in the standard driving state. Therefore, by activating the alarm when the pulse rate of the driver reaches the abnormal level, it is possible to notify the abnormality only when the driver actually enters the abnormal state.

An embodiment of the present invention will be described below with reference to the drawings. That is, as shown in FIG. 1, the driver abnormal condition detecting device 1 is provided with a pulse sensor 2. The pulse sensor 2 is a biological electrode used in medical institutions,
As shown in FIG. 2, it has a support plate 3 that is in close contact with the driver's body, and an electrode surface 4 is provided on the close contact surface of the support plate 3. One end of a lead wire 5 is connected to the electrode surface 4, and the other end of the lead wire 5 is provided with a terminal 7 connected to an input port of an arithmetic unit 6 which is an arithmetic means shown in FIG. Has been. Further, as shown in FIG. 3, the steering wheel 8 is provided with a conductive plate 9 over the entire circumference thereof, and the conductive plate 9 is a spoke 10 of the steering wheel 8.
It is grounded through a metal member embedded in the interior or the like.

As shown in FIG. 4, the calculation unit 6 has a function of measuring the time interval Ti of the pulse pulse wave output from the pulse sensor 2, a function of calculating 60 / Ti, and n 60 / Ti. It has a function of calculating an average value HR (hereinafter simply referred to as an average value HR) and a variance V _HR . On the output side of the calculation unit 6, there is provided a fatigue calculation unit 11, which is a fatigue calculation unit, and the fatigue calculation unit 11 is provided with a combination of the average value HR and the variance V _HR. Accordingly, as shown in Table 1 below, the relationship between the “state” and the “weighting coefficient” is preset, and after selecting one of the weighting coefficients, the cumulative fatigue degree Hr shown in the following formula And has a function of operating the alarm unit 12 according to the calculation result.

In addition, in the above table, each state and the weighting coefficient corresponding thereto are defined as follows.

(1) Mental excitement: For example, when an unexpected event occurs, it means a relatively long time until the pulse rises and falls to the original pulse rate, and the average value HR is large (120
Times / min) and variance V _HR is defined as small.

(2) Exercise: A state of muscular movement associated with body movement, in which the instantaneous pulse rate rises immediately after body movement and immediately returns to the original state, so when both the average value HR and the variance V _HR are large. It is defined as

(3) Standard operating state: Body movements associated with normal driving operation, average HR is medium (80 to 120 times / minute as a guide), and variance V
Defined when _HR is medium.

(4) Rest: It is defined as a state in which the vehicle is stopped and resting, the average value HR is small (80 times / minute as a guide), and the variance V _HR is also small.

In the above table, the relationship among the weighting factors α, β, γ and δ is α ≧ β>γ> δ. The reason for α ≧ β here is that it is considered that the person is already tired or is not suitable for driving in a state where psychological excitement occurs, for example, in a state where a person feels nervous, so a coefficient from “exercise” Is set to a larger value.

Next, the operation of the present embodiment having the above configuration will be described.
The description will be made according to the flowchart shown in FIG. That is, first, in step 101, the cumulative fatigue level Hr is reset,
Next, in steps 102 to 204, the instantaneous pulse rate 60 / Ti is set to n.
Increment the counter to measure individual pieces. That is, it is determined whether or not the pulse pulse shown in FIG. 4 is output (step 102), and if the pulse pulse is output, it is incremented by 1 until the timer count I reaches n (steps 103 and 104). ). And the timer count I is n
When 60 / T ₁ , 60 / T ₂ , 60 / T ₃ ... 60 / Ti is measured n times, the process proceeds to step 105, the timer count I is reset, and the average value as shown in the same step. HR and variance
Calculate V _HR and (step 105).

Next, it is judged whether or not the average value HR is larger than 120 times / minute (step 106), and if it is larger, the variance V _HR is the reference value.
It is determined whether or not it is larger than V ₀ (step 107). If both the determinations at steps 106 and 107 are large, it can be estimated from the above table that the driver is in the "exercise" state, and the weighting coefficient β is selected as the value of C in the above equation (step 108). If the determination in step 106 is large and the determination in step 107 is small, it is estimated that the driver is in the "mental agitation" state, and the weighting coefficient α is selected (step 109).

On the other hand, if the determination in step 106 is small, step 200
Then, it is determined whether or not the average value HR is 80 times / minute or more. If this determination is large, it can be estimated from the above table that the driver is in the “standard driving state”, and the weighting coefficient is determined. γ is selected (step 201), and if the determination is small,
The driver can infer from the table that he is in the "rest" state and the weighting factor δ is selected (step 20).
3).

Next, when any one of C: α, β, γ, δ in the above equation is selected in this way in steps 108, 109, 201, 202, the cumulative fatigue degree Hr is calculated from the equation using this coefficient C. Is calculated (step 203), and if the cumulative fatigue level Hr exceeds the reference fatigue level H ₀ (step 204), the alarm unit 12
Is activated to notify the driver of this.

Therefore, even if the factors of varying the pulse rate are complicatedly mixed with mental conditions and muscle movements, it is possible to accurately detect the abnormal state of the driver based on the cumulative fatigue level Hr. For the driver, the alarm sound emitted from the alarm unit 12
It is possible to realize that accumulated fatigue is occurring in one's own body, and it is possible to encourage safe driving by means such as taking a rest.

As described above, in this embodiment, the bioelectrode used in the medical institution is used as the pulse sensor 2.
In this case, the problem is the processing of grounding the living body. However, in this embodiment, since the steering wheel 8 is provided with the conductive plate 9, the driver grasps the steering wheel 8 so that the pulse sensor 2 and the living body are reliably grounded in common, and the biopotential is weak. Even if there is, there is no fear of baseline fluctuation or noise mixing in the signal processing process of detection and amplification.

FIG. 6 shows a second embodiment of the present invention, in which the reference pulse rate and the reference dispersion value, which are the reference for determining the abnormal state, are changed according to the driver's body conduction level. . That is, on the output side of the pulse sensor 2, the instantaneous heart rate H
Is provided as H = 60 / Tn (see FIGS. 18 and 19), and an instantaneous heartbeat calculating unit 13 is provided. An abnormal value determining unit 14 is provided on the output side of the instantaneous heartbeat calculating unit 13. There is.

Also, as shown in FIGS. 7 and 8, the seat belt retractor is
A seat belt 17 is wound around a take-up shaft 16 of 15, a take-up spring part 18 and a lock mechanism part 19 are provided at one end of the take-up shaft 16, and a withdrawal amount control is provided at the other end. Part 20
Is provided. The pull-out amount control unit 20 includes a gear 21 fixed to the winding shaft 16, a motor 22 linked to the gear 21, and a pull-out position memory 23, and the housing 24 of the seat belt retractor 15 has a housing 24. A body motion sensor 25 is attached. The body motion sensor 25 is a photocoupler that detects the number of moving teeth of the gear 21, and is configured to output an ON signal based on the detection of the number of moving teeth. Is provided with a body movement level determination unit 26 that discriminates and outputs the magnitude of body movement, that is, the body movement level based on the ON signal.

The output signal of the body movement level determination unit 26 is the abnormal value determination unit.
It is supposed to be input to 14. Based on the body motion level, the abnormal value determination unit 14 determines the reference heart rate H ₀ , H ₁ and H ₂ and the reference variance value V _HR0 , as shown in Table 2 and FIG. 9 below.
It is configured to change the setting of V _HR1 and input the comparison result of these values with the average value HR and the variance V _HR to the output circuit 28 of the safety drive control unit 27.

Further, in this table, an example of the actual values of the reference heart rate and the reference variance of each body movement level is shown below.

Body movement level 0: When the driver is in a resting state, the standard heart rate is about 70 beats / minute, and the standard variance value is about 10 beats / minute.

Body movement level I: When the driver performs normal steering wheel, brake, and accelerator operation, and the standard heart rate is 120 times /
Min, standard dispersion value about 20 times / min.

Body movement level II: When the driver operates the switch to move his / her back away from the seat back or twists his / her body during backward driving, the reference variance value is not set, and the reference heart rate is about 150.
Times / minute.

In the body movement level II, it is not rare that the reference variance value is not set, and it is not rare for the variance value to instantaneously change to 60 times / minute when a large motion is performed. This is because the variance value cannot be specified.

The safety drive control unit 27 is composed of an interface 29, a microcomputer 30, and the output circuit 28. The output circuit 28 is connected with a display unit 31 for notifying an abnormal state of the driver, and the interface 29 has the following: The signal of is input.

The steering signal of the turning angular velocity sensor 32 that detects the steering pattern.

Wiper signal of wiper switch 33 for detecting bad weather (rainy weather).

Light signal of the light switch 34 that detects the driving time (night).

The vehicle speed signal of the speedometer 35 that detects the traveling speed.

Time signal of clock 36 to detect driving time.

Brake 37 to detect malfunction prevention, turn signal 3
8, clutch, inhibitor switch 39 signals.

The display unit 31 includes a start indicator 40, an operating time indicator 41, a pictorial display 42, and a reset switch 43.
A buzzer (not shown) is also provided.

In the present embodiment having the above-mentioned configuration, the safety drive control unit 27 turns on the driving time indicator 41 about every 30 minutes when the vehicle is traveling based on each signal of, and displays a pictorial display 42 for 10 seconds every about 2 hours of traveling. And activate the buzzer to encourage rest. When driving at night or in the rain, 1.2 times the actual driving time is set as the driving time and displayed on the driving time indicator 41.Furthermore, when driving at a vehicle speed of 50 km / h or more, it is considered that it often appears during fatigue driving. It is determined whether or not it matches the steering operation of the pattern, and the rest is urged by the pictorial display 42 and the buzzer regardless of the driving time.

On the other hand, the instantaneous heartbeat calculating unit 13 sets the unit time T in the built-in timer as shown in the flowchart of FIG. 10 (step 201), and detects the pulse per unit time T output from the pulse sensor 2. Yes (step 202). Instantaneous heart rate calculation unit 13 further calculates the instantaneous heart rate H (step 203), computes the the average heart HR per unit time T and variance V _HR using instantaneous time heart rate H (step 204).

On the other hand, when the driver finishes wearing the seat belt 17, the motor 22 operates to wind up the amount of slack of the seat belt 17, whereby the seat belt 17 is brought into close contact with the driver, and the close contact belt position is pulled out. It is stored in the memory 23. Therefore, when the seat belt 17 is pulled out again and the gear 21 is rotated by the body movement of the driver, the body movement sensor 25 composed of a photocoupler counts the number of moving teeth of the gear 21, and the body number of the driver is determined by the number of moving teeth. It becomes possible to detect motion.

Then, the body movement level determination unit 26 determines from the detection signal of the body movement sensor 25 whether or not there is body movement of the driver (step 20
5) If there is no body movement, it is considered "rest" and the process proceeds to step 206. Accordingly, the abnormal value determination unit 14 determines that the current average heart rate HR and the variance V _HR are the reference heart rate H in the resting state.
₀ and the reference variance value V _{HR 0} or less, and if it is above, the driver is in an abnormal state.
Output an operation signal to 31. For this reason, the buzzer in the display unit 31 is activated (step 210), and an abnormal state of the driver in a resting state is notified.

Further, in the step 205, if there is a body movement of the driver, it is judged whether or not the body movement is large (step 207), and if the body movement is small, it is regarded as "standard movement" and the step 208 is performed. Proceed to. In step 207, the current average heart rate HR and variance V _HR are the reference heart rate H in the standard operating state.
₁ and whether it is within the reference variance value V _HR1 or not, and if it is above, the driver is in an abnormal state.
The operation signal is output to 31 and the buzzer is operated as described above.

Further, if it is determined in step 207 that the body movement of the driver is large, the step 20 is determined to be “large motion”.
Go to 9. In step 209, it is determined whether or not the current average heart rate HR and the variance V _HR are within the reference heart rate H ₂ and the reference variance value V _HR2 in the operating state, and if it is below, the process returns to the first step 201. If the above is true, the driver is in an abnormal state. Therefore, an operation signal is output to the display unit 31 and the buzzer is operated in the same manner.

Therefore, the driver's mode is "rest" in this way.
After being _divided into “standard motion” and “large motion”, the abnormal state of the driver will be determined based on the reference heart rate H ₀ and the reference variance value V _HR0 according to each mode.

Therefore, while taking into account the increase in pulse rate due to driving motion, when the pulse rate increases, the cause of the increase is not due to muscular exercise, but rather due to an abnormality of the actual driver. It is possible to make a distinction and accurately notify only the abnormal condition of the driver.

FIG. 11 is a reference heart rate-reference dispersion value setting characteristic diagram of the third embodiment of the present invention, in which an abnormal state is determined even when the average heart rate is low. That is, the reference heart rate set on the vertical axis is set to the reference heart rate H ₁ ′ in the “low” state, which is lower than the reference heart rate H ₀ in the resting state.

In such an embodiment, for example, the points shown,
That is, if the current average heart rate H _C of the driver exists at a point lower than the reference heart rate H ₀ at the time of “rest”, it can be judged as an abnormal state, and the heart rate of the driver decreases. It is possible to detect and notify even the abnormal state of the driver.

FIG. 12 shows a fourth embodiment of the present invention. On the output side of the pulse sensor 2, a comparator 44 for extracting an R wave (see FIG. 18) from the output signal of the pulse sensor 2 is provided. The output signal of the comparator 44 is
It is supposed to be input to 14. The abnormal value determination unit
The configurations of 14, the body movement sensor 25, the body movement level determination unit 26, the alarm unit 12, and the like are the same as those in the above embodiment.

Next, regarding the operation of the present embodiment having the above configuration,
Description will be given according to the flowchart shown in FIG. That is, in step 301, the count I and the heart rate J of the timer built in the abnormal value determination unit 14 are reset, and then in steps 302 to 304, the counter for incrementing the pulse rate J R times is incremented. I do. That is, it is determined whether or not the pulse pulse shown in FIG. 18 is output (step 302), and if the pulse pulse is output, it is incremented by 1 until the timer count I reaches R (steps 303 and 304). ). And when the timer count I is R,
In step 305, it is determined whether or not there is an interrupt.
This interrupt is a process of stopping the execution of this flowchart for t seconds when there is a large operation (see Table-2) as shown in FIG. 14, and when there is the interrupt, it continues for t seconds. After the lapse of time, the timer count I is reset (step 305). As a result, when the driver performs a large motion and the pulse rate increases, the pulse rate is measured only when the pulse rate returns to the normal state after t seconds, and as a result, the driver's standard pulse rate is measured. The operating state will be determined.

If there is no interruption, that is, if the driver is in the standard driving state, the process proceeds to step 307, the timer count I is set to 0, and the average heart rate X (J) is set.
Is calculated (step 307) to determine whether the average heart rate X (J) is larger than the reference value J ₀ (step 30).
8). If this determination is large, the average value of the average heart rate and the variance S ² are obtained from the data of the average heart rate X (J) measured above a predetermined number from the arithmetic expression shown in step 309.

Next, the value of the measured average heart rate X (J) is (
(3S) to (+ 3S) It is judged whether or not it is in the abnormal level outside the range (step 310), and if it is out of this range, it is judged that the driver is in an abnormal state and the alarm unit 12 is activated. Let At this time, as described above, this flow chart is configured to be executed only when the driver is in the normal state by the interruption, and therefore, in this normal state, the driver performs the heartbeat shown in FIG. An alarm is issued only when a few people are abnormal. Therefore, when the pulse rate increases, it is possible to determine that the cause of the increase is not due to a large motion, but due to the actual abnormality of the driver, and accurately notifies only the abnormal state of the driver. FIG. 16 shows a flowchart of the fifth embodiment of the present invention, in which the abnormal level is changed according to the running state of the vehicle. That is, step 401
Then, the timer counts I ₁ , I ₂ , I ₃ and the heart rate J ₁ , J ₂ , J ₃ for each vehicle speed built in the abnormal value determination unit 14 are reset, and then in step 402 the current Determine whether the vehicle speed is 70 km / h or more. If the vehicle speed is 70 km / h or more, it is determined that the vehicle is traveling at high speed, and the processing determination in steps 403 to 405 is performed, that is, the counter for measuring the pulse rate J ₁ during the high-speed traveling R ₁ times is incremented. Specifically, it is determined whether or not a pulse pulse is output (step 40
3) If the pulse pulse is output, timer count I
₁ is increased by one at a time until the R ₁ 404,405). When the timer count I ₁ becomes R ₁ , the process proceeds to step 406 and it is determined whether or not there is an interrupt. Similar to the fourth embodiment, the processing content of this interrupt is processing for stopping the execution of this flowchart for t seconds when there is a large operation (see Table 2). Resets the timer count I ₁ after t seconds have elapsed (step 407). As a result, when the driver performs a large motion to increase the pulse rate, the pulse rate is measured only when the pulse rate returns to the normal state after t seconds, and as a result, the pulse rate is measured during high-speed traveling. The standard driving state of the driver will be determined.

If there is no interruption, that is, if the driver is in the standard driving state, the process proceeds to step 408, the timer count I _{1 is set} to 0, and the average heart rate X ₁ (J ₁ ) during high-speed traveling is calculated. Then, it is determined whether or not the average heart rate X ₁ (J ₁ ) is larger than the reference value J ₀₁ during high-speed traveling (step 409). If this judgment is large, the average heart rate X during high-speed running over the predetermined number measured next time
₁ (J ₁₎ and the average value of ₁ from the data average heart rate of the dispersion S ² ₁
Is calculated from the arithmetic expression shown in step 410.

Next, the value of the measured average heart rate X ₁ (J ₁ ) is (
It is determined whether or not the abnormal level is outside the range of ( ₁ −3S ₁ ) to ( ₁ + 3S ₁ ) (step 411), and if it is out of this range, it is determined that the driver is in an abnormal state and an alarm is issued. Department
Activate 12

If the vehicle speed is 70 km / h or less, the process proceeds from step 402 to step 413 to determine whether the vehicle speed is 30 km / h or more (step 413). If the vehicle speed is 70 km / h or less and 30 km / h or more, it is determined that the vehicle is running normally, and the processing determination in steps 414 to 422 determines the average heart rate X ₂ (J ₂ ) Is calculated, and the average value of the average heart rate is calculated from the average heart rate X ₂ (J ₂ ) data during normal driving.
₂ and variance S ² ₂ are calculated from an arithmetic expression, and whether the measured average heart rate X ₂ (J ₂ ) is in an abnormal level outside the range of ( ₂ −3S ₂ ) to ( ₂ + 3S ₂ ). Whether or not it is outside this range is determined, and the alarm unit 12 is activated because the driver is in an abnormal state.

Further, when the vehicle speed is 30 km / h or less, it is determined that the vehicle is running in a traffic jam, and the processing determinations in steps 424 to 431 indicate that the average heart rate X ₃ (during running in a traffic jam is the same as during high speed running described above. J ₃₎ calculates the portions to determine the average heart rate X ₃ (J ₃ and the average value ₃ from the data average heart rate of) the dispersion S ² ₃ during a traffic jam running from the arithmetic expression, the average heart rate X ₃ that the measured The value of (J ₃ ) changes from ( ₃ −3S ₃ ) to ( ₃ +3
S ₃₎ out of range decides whether or not there is an abnormal level of, when deviating from this range, the driver is in an abnormal state, to operate the alarm unit 12. Therefore, it is determined whether or not the driver is in an abnormal state based on the abnormal level during high-speed traveling, normal traveling, or traffic congestion as described above. Even if the number increases, it will not be mistakenly determined to be abnormal,
Only the abnormal state of the driver in each traveling state can be accurately announced. As described above, in this embodiment, the reference value is switched depending on the running state of the vehicle, so that the driver's situation can be more accurately determined.

EFFECTS OF THE INVENTION As described above, the present invention is provided with a calculating means for calculating the average value and variance of the pulse rate of the driver, and the average value and variance of the pulse rate among a plurality of preset weighting factors. While specifying any one of the weighting factors corresponding to the combination of the respective calculated values of, the new cumulative fatigue level of the driver is calculated based on the specified weighting factor and the past cumulative fatigue level, A fatigue degree calculating means for determining an abnormal state of the driver according to the calculation result is provided. Therefore, by determining the abnormal state based on the cumulative fatigue level, it is possible to accurately detect the abnormal state of the driver even if mental conditions, muscle movements, etc. are complicatedly mixed as factors of fluctuation of the pulse rate. In addition, the driver can notice the abnormal state and realize that cumulative fatigue is occurring in his / her own body, and can drive safely.

Further, in the present invention, a body movement sensor for detecting movement of the body of the driver is provided, and a reference pulse rate for discriminating an abnormal state of the driver according to the movement of the body detected by the body movement sensor. The standard dispersion value was changed. Therefore, while considering the increase in pulse rate due to driving motion, when the pulse rate increases, it is determined that the increase is not due to muscular movement, but is due to an abnormality of the actual driver. Therefore, it is possible to accurately notify only the abnormal state of the driver.

Further, the standard driving state is discriminated based on the movement of the body detected by the body movement sensor, and the abnormal state of the driver is discriminated based on the abnormal level estimated and calculated from the pulse rate in the standard driving state. Therefore, when the pulse rate is increased, it is possible to determine whether or not the increase in the abnormal pulse rate is an abnormal level, assuming the standard operating state, and to determine the abnormal state of the driver based on the standard operating state. This enables accurate detection.

In addition, in the above-described fifth embodiment, the reference value is set and changed according to the running state of the vehicle. Therefore, it is determined whether or not the driver is in an abnormal state based on the reference value during high-speed traveling, normal traveling, traffic congestion traveling, etc., so that the heart rate of the driver increases due to the traveling state. However, it is possible to accurately notify only the abnormal state of the driver in each traveling state without being erroneously determined to be abnormal.

[Brief description of the drawings]

1 is a block diagram showing a first embodiment of the present invention, FIG. 2 is a perspective view showing a pulse sensor of the same embodiment, FIG. 3 is a perspective view showing a grounding conductive plate of the same embodiment, and FIG. FIG. 5 is an explanatory view showing the operation of the pulse sensor and the arithmetic unit of the same embodiment, FIG. 5 is a flow chart showing the operation of the same embodiment, and FIG. 6 is the second of the present invention.
FIG. 7 is a block diagram showing an embodiment, FIG. 7 is a front view showing a seat belt retractor constituting a body movement sensor of the embodiment, and FIG.
7 is a sectional view taken along line VIII-VIII of FIG. 7, FIG. 9 is a reference dispersion value-reference heart rate characteristic diagram of the same embodiment, FIG. 10 is a flow chart showing the operation of the same embodiment, and FIG. FIG. 12 is a block diagram showing a fourth embodiment of the present invention, FIG. 13 is a flow chart showing the operation of the third embodiment of the present invention, FIG. FIG. 15 is a flow chart showing an interrupt routine of the same embodiment, FIG. 15 is an explanatory diagram showing heart rate of a heartbeat abnormal person and a normal person, FIG. 16 is a flow chart showing a fifth embodiment of the present invention, and FIG. FIG. 18 is an explanatory diagram showing an example of a shape, FIG. 18 is an explanatory diagram showing an example of a heart rate change model, and FIG. 19 is an explanatory diagram showing an example of an instantaneous heart rate change model. 1 ... Driver abnormal condition detection device, 2 ... Pulse sensor, 6
...... Calculation section (calculation means), 11 …… Fatigue calculation section (fatigue calculation means), 12 …… Alarm section, 25 …… Body movement sensor.

Claims (3)

(57) [Claims] 1. A device having a pulse sensor for detecting a pulse rate of a driver, and determining an abnormal state of the driver based on the pulse rate detected by the pulse sensor, wherein an average value and variance of the pulse rate are calculated. A plurality of weighting factors are set in advance according to the combination of the calculating means for calculating and the magnitude of the average value and the variance value of the pulse rate, while the average value of the pulse rate is selected from among the plurality of weighting factors. And one of the weighting factors corresponding to the respective combinations of the calculated values of the variance and the variance are specified, and a new cumulative fatigue level of the driver is calculated based on the specified weighting factor and the past cumulative fatigue level. A driver abnormal condition detection device comprising: a fatigue degree calculator that determines an abnormal condition of the driver according to the calculation result; 2. An apparatus having a pulse sensor for detecting the pulse rate of a driver, and comparing the pulse rate detected by the pulse sensor with a reference pulse rate and a reference variance value to determine an abnormal state of the driver. In the above, a body movement sensor for detecting movement of the body of the driver is provided, and the reference pulse rate and the reference dispersion value are changed according to the movement of the body detected by the body movement sensor. Driver abnormal condition detection device. 3. A standard driving state is determined based on the movement of the body sensed by the body motion sensor, an abnormal level is calculated and estimated from the pulse rate in the standard operating state, and the driver's movement based on the abnormal level. The driver's abnormal condition detection device according to claim 2, wherein the abnormal condition is determined.