JP4915735B2 - Fatigue judgment system and fatigue judgment method - Google Patents

Abstract

A fatigue judgment system and a fatigue judgment method, which can decide the degree of fatigue of a driver precisely without imposing a spiritual burden on the driver. The fatigue judgment system comprises a unit (e.g., a seat pressure sensor (4)) for detecting the position fluctuations of a driver (2), and a unit (e.g., a fatigue judgment unit (13A)) for judging it from the position fluctuations (including the frequency of fluctuating the position) detected by the unit (4) whether or not the fatigue of the driver (2) exceeds a predetermined level (or a threshold value or more).

Description

  The present invention relates to a technique for determining whether or not a driver driving a vehicle such as a truck (cargo vehicle) is tired.

When driving a freight car such as a truck, it is a tough condition at present, such as being forced to drive for a long time. In addition, various regulations that will be determined in the near future (for example, speed regulations) are expected to cause drivers who drive lorries to continue to receive more severe mental stress.
Under such circumstances, it is not difficult to imagine that a driver who drives for a long time accumulates a great deal of physical and mental fatigue.

Under such circumstances, it is desirable to take appropriate measures before the driver's fatigue reaches a dangerous level in order to achieve the top priority proposition of “road safety”.
Here, since there is a large individual difference in “fatigue”, in order to suppress driver fatigue, it is impractical and effective to set uniform standards for vehicle mileage, driving time, etc. It is expected to be difficult to apply.

Also, drivers who have been engaged in the operation of freight cars for many years often drive at a pace that they consider “not tired” based on their own experience. It is expected that there are almost no cases where the standard set by various regulations is consistent with the driver's empirical pace.
In such a case, if the driver complies with uniformly defined standards, the mental burden on the driver increases, and on the contrary, the feeling of fatigue increases, which is undesirable from the viewpoint of traffic safety.

On the other hand, if you measure the driver's fatigue and take measures such as giving a warning if the measured fatigue exceeds the threshold, you can take measures according to the situation of each driver, The problem as described above does not occur.
However, with regard to fatigue, mental factors have a great influence and do not necessarily involve physical changes. Therefore, it is very difficult to determine whether or not you are fatigued.

Here, in order to detect fatigue, a sensor may be attached to the driver's body, but the driver feels mental stress when the sensor is attached. Therefore, it is desirable to measure parameters corresponding to fatigue in a manner that does not contact the driver.
As a parameter that can be measured in a non-contact manner and that reflects the fatigue felt by the driver, the eyelid opening of the driver is constantly monitored by, for example, a CCD camera, and the eyelid opening is a predetermined value. A technique has been proposed that warns a driver that the driver is determined to be drowsy due to fatigue when the following conditions continue.
Alternatively, a technique has been proposed in which a CCD camera detects that the driver's line-of-sight movement has decreased, and gives a warning to the driver when the line-of-sight movement falls below a predetermined value.

  However, the above two techniques have a problem that the position of the driver's eyes varies greatly depending on the size of the body, and it is difficult to focus the CCD camera. Further, the change in the opening degree of the eyelid has a large individual difference, and the accuracy problem of the data to be detected has not been solved. Furthermore, many drivers feel psychological stress when they are constantly observing their line of sight with a CCD camera, and such psychological stress may have an adverse effect on driving.

As another conventional technique, for example, a technique for determining a driver's fatigue state from a driver's voice has been proposed (see Patent Document 1).
In such technology, it is necessary to communicate with the driver and receive voice. However, applying the technology described above (determining the fatigue state from voice) to drivers who dislike talking while driving or drivers who are distracted by surrounding conversations while driving On the other hand, it is dangerous for traffic safety.
JP 2000-113347 A

  The present invention has been proposed in view of the problems of the prior art described above, and a fatigue determination system capable of accurately determining the degree of fatigue of a driver without causing a mental burden on the driver and The purpose is to provide a fatigue assessment method.

  In the fatigue determination system of the present invention, the driver (2) is fatigued from the unit (for example, the seat pressure sensor 4) that detects the change in the posture of the driver (2) and the frequency of the posture change detected by the unit (4). And a unit (40) for determining whether or not the driver has changed, the unit for detecting the change in the posture change of the driver is a pair of pressure sensors (4; two in total) provided on the driver seat (3). The determination unit (40) includes a detection result difference (LR), a sum (L + R), and a detection result change difference (dL-dR) of the left and right pressure sensors (4). A standard deviation is obtained, and each of the difference (LR), the sum (L + R), and the difference in change (dL-dR) is set to an abnormal value (for example, 3 times) greater than an integral multiple (for example, 3 times) of the standard deviation. Three times the standard deviation σ, greater than 3σ Defined as a numerical value), the number of times the abnormal value appears per fixed time is counted, and the sum (L + R) is calculated from the number of times the abnormal value of the difference (LR) appears (c (LR)). (C (L + R)) and the number of times the abnormal value difference (dL−dR) has appeared (c (dL−dR)) are subtracted (“c (L−R)”. ) -C (L + R) -c (dL-dR) "), the subtraction result (" c (LR) -c (L + R) -c (dL-dR) ") takes a positive value, And when there exists a tendency to increase, it has the function to determine that the driver (2) is fatigued (Claim 1).

  The fatigue determination method of the present invention includes a step (S42) of detecting a change in the posture of the driver (2) and a step (S51 to S54) of determining whether or not the driver is fatigued from the detected frequency of the posture change. In the step (S42) of detecting a change in the posture change of the driver, a detection signal is received from a pair of left and right pressure sensors (4; a total of two) on the driver seat (3), In the determination steps (S51 to S54), the respective standard values of the difference (LR), the sum (L + R), and the difference (dL-dR) between the detection results of the left and right pressure sensors (4) are detected. Deviation (σ) is calculated, and each of the difference (LR), sum (L + R), and change amount difference (dL-dR) is an integer multiple of standard deviation (σ) (eg, 3 times; 3σ). A large value is an abnormal value (for example, 3 times the standard deviation σ (3σ )), The number of times that the abnormal value has appeared per certain time is counted, and the number of times that the abnormal value of the difference (LR) has appeared (c (LR)), The number of times that the abnormal value of the sum (L + R) has appeared (c (L + R)) and the number of times that the abnormal value of the difference in amount of change (dL−dR) has appeared (c (dL−dR)) are subtracted (“c (L−R) −c (L + R) −c (dL−dR) ”) and the subtraction result (“ c (LR) −c (L + R) −c (dL−dR) ”) is“ positive ” If the value is taken and tends to increase, it is determined that the driver is tired (claim 2).

As a result of various studies, the inventor, for example, when the driver is fatigued to the extent that he / she feels “drowsiness”, the driver moves his / her neck, arms, shoulders, changes his / her position, etc. We found that the frequency of By repeating various experiments, it was confirmed that the driver's fatigue could be accurately determined by observing the driver's posture fluctuation frequency.
The present invention having the above-described configuration (inventions of claims 1 and 2) is configured based on such knowledge. In other words, it is determined whether the driver is tired from the frequency by detecting whether the driver's posture changes, such as turning the neck, arms, shoulders, or changing the sitting position. I can do it.

  Furthermore, according to the present invention, the number of times that an abnormal value of the difference (LR) between the detection results of the left and right pressure sensors has appeared from the detection signals from the pressure sensors (two in total) provided on the driver seat. (C (L−R)), the number of times that an abnormal value of sum (L + R) has appeared (c (L + R)), and the number of times that an abnormal value of the difference in detection result difference (dL−dR) has appeared (c (dL -DR)), and from the number of times the abnormal value of the difference (LR) appears (c (LR)), the number of times the abnormal value of the sum (L + R) appears (c (L + R)) and The result of subtracting the number (c (dL-dR)) of occurrence of an abnormal value of the difference in change amount (dL-dR) ("c (LR) -c (L + R) -c (dL-dR)" )), It is determined whether the driver is tired or not, so the left and right (two in total) pressure sensor signals are processed. Only that, it is possible to detect the fatigue state of the driver. Therefore, three or more pressure sensors and acceleration sensors are not required, the number of sensors can be reduced, and it can be reliably determined whether or not the driver is in a fatigue tendency.

In addition, by appropriately setting the time for obtaining the standard deviation (σ) and the number of occurrences of abnormal values (c (LR), c (L + R), c (dL-dR)), the driving situation It is possible to determine the fatigue tendency corresponding to the above.
Furthermore, if such a configuration is adopted, it is only necessary to provide two pressure sensors on the driver seat, and it is not necessary to wear a special device on the driver's body or observe the driver with a monitor or the like. There is no discomfort for the target driver.
And since measurement and control are comparatively simple, complicated and large-scale signal processing equipment is not required, and the whole system can be reduced in size.

  Embodiments of the present invention will be described below with reference to the accompanying drawings.

First, a first embodiment will be described with reference to FIGS.
In FIG. 1, the fatigue determination system indicated by the symbol J <b> 1 as a whole system has a seat seat surface when the driver 2 is seated on the back side (inside) of the seat seat surface 31 of the seat 3 for the driver 2 of the large truck 1. The seat pressure sensor 4 for detecting the seat pressure applied to the seat 31 is built in, and the acceleration sensor 5 is built in the seat back 32 or the seat cushion 31.

  Further, the fatigue determination system J1 receives the data from the seat pressure sensor 4 via the interface 6 that relays the signal from the seat pressure sensor 4 and the signal from the acceleration sensor 5, and stores the data via the interface 6. Seat pressure data storage means 7 and acceleration data storage means 8 for receiving data from the acceleration sensor 5 via the interface 6 and storing the data.

  Further, the fatigue determination system J1 distinguishes the vehicle swing caused by yawing or rolling of the vehicle from the seat pressure change based on the driver's fatigue based on the data from the seat pressure data storage means 7 and the acceleration data storage means 8. For this purpose, a “vehicle rocking influence eliminating unit” 9 that eliminates an acceleration change when the vehicle rocks is provided.

  Further, the fatigue determination system J1 includes an attitude variation unit 10, a database 11 serving as storage means, a threshold determination unit 12, a fatigue determination unit 13A, and a display / alarm means 14.

  The posture change determination unit 10 determines the amount of posture change in which the driver 2 has moved his / her waist purely based on the seat pressure data from which the influence of the vehicle swing has been removed by the vehicle swing influence eliminating unit 9. The determined posture fluctuation amount is stored in the database 11 and sent to the fatigue determination unit 13.

Here, it is conceivable that the threshold fluctuation amount, which is a threshold, that is, a determination criterion for determining “fatigue”, varies depending on wrinkles by the driver or individual differences.
It is clear from the inventor's research that there is a significant correlation between the frequency of changing postures and the amount of change, and fatigue with the driver. However, specific threshold values can be determined for each driver or case-by-case.・ It will be determined by case.

  In the case shown in FIG. 1, data related to posture change is sent from the posture change determination unit 10 to the database 11, and the threshold determination unit 12 combines the data related to posture change with the driver's impression (sense), memory, etc. A threshold (threshold in the frequency of posture change) is determined as to whether or not the feeling of fatigue has reached the danger zone.

  As described above, it is desirable that the threshold is determined for each driver. However, a fixed value stored in advance in the storage means (database) can be adopted as the threshold value.

  In the fatigue determination system J1 of the first embodiment having the above-described configuration, when the driver 2 turns his neck, sits down, or changes its posture, the pressure (seat pressure) applied to the seat surface 31 of the seat 3 varies. Therefore, the pressure (seat pressure) applied to the seat seat surface 31 is detected by the seat pressure sensor 4, and the frequency and how much the posture is changed based on the detected fluctuation of the “pressure applied to the seat seat surface 31 (seat pressure)”. The posture variation determining unit 10 calculates the posture variation amount indicating whether the posture has been changed.

  Here, as is clear from FIG. 3 and FIG. 4, the driver's fatigue is closely related to the frequency of changing the posture and the amount of change. 3 and 4, the horizontal axis indicates the passage of time, the amplitude in the vertical axis direction indicates the amount of change in posture, and the posture changes when the amplitude in the vertical axis direction is large. . If the frequency of the amplitude in the vertical axis direction increases frequently, the frequency of posture change also increases.

As compared with the non-fatigue state (normal time) shown in FIG. 3, in the fatigue state (fatigue time) shown in FIG. 4, the frequency of increasing the amplitude in the vertical axis direction clearly increases. .
This is not simply a fluctuation in pressure but represents a so-called “speed of fluctuation”. For example, if frequency analysis is performed by Fast Fourier Transform (FFT), it is predicted that a certain range of frequencies appears when the driver turns the neck or the like.

  Here, the pressure (seat pressure) applied to the seat 3 also fluctuates due to the swinging and other operations of the vehicle 1 itself. Accordingly, the swinging and other operations of the vehicle 1 itself are irrelevant to the fatigue of the driver 2 and need to be eliminated.

  Therefore, it is detected by the acceleration sensor 5 that the vehicle 1 itself swings and other movements, and from the “variation in pressure (seat pressure) applied to the seat seat surface 31” detected by the seat pressure sensor 4, The “swinging and other operations of the vehicle 1 itself” detected by the acceleration sensor 5 is excluded. By doing so, the pressure fluctuation caused only by the fluctuation of the posture of the driver 2 is extracted from “the fluctuation of the pressure applied to the seat seat surface 31 (sitting pressure)” detected by the seat pressure sensor 4.

  Next, based on FIG. 2, the fatigue determination control in the first embodiment will be described with reference to FIG.

  First, in step S1, it is determined whether or not the operation has been started. If the operation has been started (YES in step S1), the process proceeds to the next step S2, and the operation has not yet started. If (NO at step S1), the loop of step S1 is repeated.

Here, when the driver turns his head, sits down, or changes his / her posture, the pressure applied to the seat (seat pressure) varies.
Therefore, in step S2, the detection signal from the seat pressure sensor 4 is received (input) to the seat pressure data storage means 7 via the interface 6. On the other hand, the detection signal data from the acceleration sensor 5 is received by the acceleration data storage means 8 through the interface 6.

  In the next step S3, the vehicle swinging influence eliminating unit 9 receives the data from the seating pressure data storage means 7 and the data from the acceleration data storage means 8, and from the acceleration data in the data other than the driver's body, For example, the influence of the vehicle yawing or bouncing is excluded, and the data from which the influence of the vehicle swing is eliminated is input to the attitude fluctuation determination unit 10 to calculate and determine the attitude fluctuation amount. Then, the process proceeds to the next step S4.

In step S4, it is determined whether or not the posture fluctuation amount is greater than or equal to a threshold value, that is, whether or not fatigue exceeds a determined limit.
If the posture variation amount is equal to or greater than the threshold value (YES in step S4), the process proceeds to step S5. On the other hand, if the posture variation amount is less than the threshold value (NO in step S4), the process returns to step S2, and step S2 and subsequent steps are repeated again. .

In step S5, for example, the display / alarm unit 14 displays a warning message such as “You are tired. Let's take a break early”, and then proceeds to step S6.
Here, as the warning, it is more effective if it is performed by the voice of a person that the driver himself can listen to, for example, the voice of a spouse, child, lover or the like.

  In step S6, it is determined whether or not the control is to be ended. If the control is to be ended (YES in step S6), the control is ended as it is. On the other hand, if the control is continued (NO in step S6), the process returns to step S2, and step S2 and the subsequent steps are repeated.

  Next, a modification of the first embodiment will be described with reference to FIG.

  In the first embodiment shown in FIG. 1 to FIG. 4, the entire system J1 for judging driver fatigue is mounted on an automobile (cargo car) 1 and all the fatigue judgment is performed on its own. It is also possible to apply and determine whether the driver is tired or not at a processing center located at a remote location of the travel point of the automobile (cargo automobile) 1.

  In the modification of the first embodiment shown in FIG. 5 (the entire fatigue determination system is indicated by J12), the seat pressure sensor 4, the acceleration sensor 5, the interface 6, the seat pressure data storage means 7, the acceleration data means 8, the vehicle swinging. The equipment which added the communication unit 20A to the influence exclusion means 9, the posture variation determination unit 10, and the display and alarm means 14 is the vehicle side equipment U1, while the communication unit is the database 11, the threshold determination unit 12, and the fatigue determination unit 13A. The equipment to which 20B is added is divided, for example, as equipment U2 of the processing center on the operation management side.

Then, the attitude variation data determined by the vehicle-side equipment U1 is transmitted to the provider 30 by the communication unit 20A, for example, by wireless (LAN) L1.
The provider 30 transmits, for example, to the processing center equipment U2 by the existing high-speed communication means L2, and the fatigue determination unit 13A of the processing center compares the sent data with the threshold value stored in the database 11 to determine the degree of fatigue. Determine.
When it is determined that the user is clearly fatigued and “fatigue is determined”, the result is sent back to the communication unit 20A on the target vehicle side by returning the communication means L2 and L1, and the in-vehicle display / alarm means 14 It is configured to give the driver an indication of being “fatigue” and an alarm.

The functions and operations of the individual equipments 4 to 10 in the vehicle-side equipment U1 and the functions and operations of the individual equipments 11 to 13A in the equipment U2 of the processing center are substantially the same as those in the first embodiment of FIGS. The same.
The determination of the threshold is also performed in the processing center equipment U2 in the same manner as the fatigue determination.

In the modified example J12 of the first embodiment, in a transportation company having a plurality of vehicles, a part of the equipment (equipment on the processing center side) is shared by dividing the equipment into the in-vehicle equipment U1 and the processing center equipment U2. It is possible to prevent the in-vehicle unit from becoming huge.
Further, by performing information processing at the processing center, it is possible to obtain effects such as improved accuracy of determination, improved speed of processing, and thorough safety management.

  Here, in the modified example of FIG. 5, the vehicle-side equipment U1 is provided with the vehicle swing influence eliminating unit 9 and the posture fluctuation determining unit 10 to remove the influence of the swing of the vehicle 1 itself from the measurement result by the seat pressure sensor 4. Then, the signal is transmitted to the equipment U2 of the processing center on the operation management side. On the other hand, although not shown, the vehicle swing influence eliminating unit 9 and the posture variation determining unit 10 are provided on the equipment U2 side of the processing center on the operation management side, and the seat pressure sensor 4 is installed on the processing center side U2. It is possible to remove the influence of the vehicle 1 itself from the measurement result.

Next, a second embodiment will be described with reference to FIGS.
In the fatigue determination systems J1 and J12 of the first embodiment, the frequency and amount of change in the posture of the driver 2 are detected by the seat pressure sensor 4 (and the acceleration sensor 5) to determine the degree of fatigue of the driver 2. ing.
On the other hand, in the fatigue determination system of the second embodiment shown in FIGS. 6 to 8 (indicated by reference numeral J2), the driver 2 is determined from the blood flow of the driver 2. The configuration of the second embodiment will be described below based on FIG.

  In FIG. 6, the fatigue determination system J2 of the second embodiment irradiates various portions of light such as wrists and necks of the driver 2 where the arteries can be observed with the light irradiation means and reflects the portions where the arteries can be observed. A light irradiating / reflecting light receiving means 15 for detecting light using various light receiving elements is installed in the vicinity of the driver 2.

  Further, the fatigue determination system J2 of the second embodiment includes an interface 6 that relays reflected light data from the light irradiation / reflected light receiving unit 15, and a reflected light data fluctuation amount calculation unit that calculates a fluctuation amount of the reflected light data. 16, a blood flow data determination unit 17 that obtains blood flow data from the reflected light data fluctuation amount, a database 11, a threshold determination unit 12, a fatigue determination unit 13 </ b> B, and a display / alarm unit 14.

In the second embodiment having the above-described configuration, the light irradiation / reflected light receiving unit 15 irradiates various portions of light such as the wrist and neck of the driver 2 with the light irradiation / reflected light receiving unit 15 and reflects the light at the portion where the artery can be observed. Light is detected by using the various light receiving elements incorporated in the light irradiation / reflected light receiving unit 15.
Since the reflected light data measured by the light receiving element changes depending on the period of blood flow (pulse) and the blood flow volume, the reflected light data fluctuation amount calculation unit 16 obtains the fluctuation amount of the reflected light data, and from the fluctuation amount. The blood flow rate or heart rate can be obtained by the blood flow data determination unit 17.

Here, the blood flow rate has a close relationship with the heart rate, and when the heart rate decreases, the blood flow rate also decreases. It is well known that heart rate decreases when sleepy.
That is, as shown in FIG. 8, when subjects (drivers) A and B start driving at 9:00 am and continue driving for 24 hours, the body temperature starts at about 19-20 o'clock (at night and night). A decrease in blood flow, that is, a decrease in blood flow appears, and after 2 to 4 hours, the heart rate has started to decrease.
Then, from 2 o'clock the next day to 5 o'clock at dawn, the degree of arousal falls to the lowest level. The driver A suddenly has a large heart rate peak at 6 o'clock. This is because the driver A returns to an awake state for some reason, for example, when the vehicle is abnormally approaching the vehicle ahead or the guardrail. Therefore, it seems that the heart rate suddenly increased.
It should be noted that the above-mentioned rise in heart rate (large peak at 6 o'clock) is caused by the influence of human circadian rhythm (human daily rhythm) caused by the rising sun in addition to returning to the awake state as described above. Is also considered as the factor.

  As described above, if the blood flow rate of the driver decreases (decreases), the driver's fatigue level reaches a dangerous level, and the driver feels sleepy.

  Here, since the pulse is also a parameter indicating the blood flow or the heart rate, instead of acquiring the reflected light data by the combination of the light irradiation means and the light receiving element, an acceleration sensor or the like is attached to the wristband or the like. Thus, it is possible to measure the pulse of the driver 2 and determine the fatigue of the driver 2 from the measured pulse.

  Next, based on FIG. 7, the fatigue determination control in the second embodiment will be described with reference to FIG.

  First, in step S11, it is determined whether or not the operation has been started. If the operation has been started (YES in step S11), the process proceeds to the next step S12, and the operation has not yet started. If (NO at step S11), the loop of step S11 is repeated.

    In step S12, the light irradiation right reflected light receiving unit 15 irradiates the pulsation portion of the blood vessel of the driver 2 and receives the reflected light.

  In the next step S13, the received reflected light data is taken into the reflected light data fluctuation amount calculation unit 16, and the fluctuation amount of the reflected light is calculated. In the next step S14, the blood flow data determination unit 17 determines blood flow data such as a heart rate from the amount of reflected light fluctuation obtained by calculation.

In the next step S15, the fatigue determination unit 13B determines whether or not the blood flow or the body temperature that is a parameter of the blood flow is equal to or higher than a threshold value.
If the blood flow is equal to or greater than the threshold (YES in step S15), the process proceeds to the next step S16. If the blood flow is less than the threshold (NO in step S15), the process returns to step S12, and step S12 and subsequent steps are repeated.

  In step S16, the display / alarm unit 14 displays, for example, “You are tired. Take a rest as soon as possible” and a warning voice, and then proceed to step S17.

  In step S17, it is determined whether or not the control is to be ended. If the control is to be ended (YES in step S17), the control is ended as it is. On the other hand, if the control is to be continued (NO in step S17), the process returns to step S12, and step S12 and subsequent steps are repeated again.

Next, a modification of the second embodiment will be described with reference to FIG.
In the second embodiment shown in FIGS. 6 to 8, the whole system for determining driver fatigue is mounted on an automobile (cargo automobile), but the modification (in the second embodiment) of FIG. The information communication technology is applied to determine whether the driver is tired or not at a processing center located at a remote location of the vehicle (cargo vehicle) 1 travel point.

  In the modification of the second embodiment of FIG. 9 (the entire fatigue determination system is denoted by reference numeral J22), the light irradiation / reflected light receiving unit 15, the interface 6, the reflected light data fluctuation amount calculating unit 16, and the blood flow data determining unit 17 are used. The equipment that newly added the communication unit 20A to the display and alarm means 14 is the vehicle-side equipment U11, while the equipment that has the communication unit 20B added to the database 11, the threshold determination unit 12, and the fatigue determination unit 13B is operated, for example, It is divided as equipment U21 of the processing center on the management side.

Then, the blood flow data calculated by the vehicle-side equipment U11 is transmitted to the provider 30 by the communication unit 20A, for example, by wireless (LAN) L1.
The provider 30 transmits to the processing center equipment U21 by, for example, the existing high-speed communication means L2, and the processing center fatigue determination unit 13B compares the sent data with the threshold stored in the database 11 to determine the degree of fatigue. Determine.
When it is determined that the user is clearly fatigued and “fatigue is determined”, the result is sent back to the communication unit 20A on the target vehicle side by returning the communication means L2 and L1, and the in-vehicle display / alarm means 14 It is configured to give a warning to the driver.

Regarding the functions and operations of the individual equipments 15, 6, 16, 17, and 14 in the vehicle-side unit U1, and the functions and operations of the individual equipments 11 to 13B in the equipment U22 of the processing center, the second of FIGS. This is substantially the same as the embodiment.
Note that the threshold value is also determined in the processing center equipment U22 in the same manner as the fatigue determination.

In the modification J22 of the second embodiment, in a transportation company having a plurality of vehicles, a part of the equipment (equipment on the processing center side) is shared by dividing it into the in-vehicle equipment U11 and the equipment U21 of the processing center, It is possible to prevent the in-vehicle unit from becoming huge.
Further, by performing information processing at the processing center, it is possible to obtain effects such as improved accuracy of determination, improved speed of processing, and thorough safety management.

  In the modification of FIG. 9, although not shown, the reflected light data fluctuation amount calculation unit 16 and the blood flow data determination unit 17 are provided in the equipment U21 of the operation management processing center, and the equipment of the operation management processing center is provided. At U2, blood flow data can be determined to determine the heart rate.

Next, a third embodiment will be described with reference to FIG.
The third embodiment of FIG. 10 is an embodiment in which the first embodiment J1 that determines the fatigue level based on the change in the posture of the driver 2 and the second embodiment J2 that determines the fatigue level based on the blood flow rate of the driver 2 are combined. It is.
In other words, this is an embodiment in which the degree of fatigue is determined based on the change in the posture of the driver 2 and the blood flow volume of the driver 2.
The configuration of the third embodiment will be described below based on FIG.

  In FIG. 10, the fatigue determination system J3 of 3rd Embodiment is comprised from the equipment U12 by the side of a vehicle, and the personal computer 60 for fatigue determination of the processing center by the side of operation management.

The vehicle-side equipment U <b> 12 is equipped with a display / alarm means 14 connected to the in-vehicle computer 50 including the data logger 51 and the communication unit 52 and the communication unit 52.
In addition, the seat pressure sensor 4 similar to that of the first embodiment is built in the back side (inside) of the seat seat surface 31 of the seat 3 for the driver 2, and the seat back 32 or the seat cushion 31 is the same as that of the first embodiment. A similar acceleration sensor 5 is incorporated.
Further, a light irradiation / reflection light receiving means 15 similar to that of the second embodiment is installed in the vicinity of the driver 2.

  The vehicle-side equipment U12 includes a detection signal processing unit 18 and a reflected light data processing unit 19, and the detection signal processing unit 18 processes detection signals that are data from the seat pressure sensor 4 and the acceleration sensor 5. The reflected light data from the light irradiation / reflected light receiving means 15 is processed by the reflected light data processing unit 19 and transmitted to the data logger 51 of the in-vehicle microcomputer 50, and the one-end data is collected by the data logger 51. Has been.

  The data collected by the data logger 51 is transmitted to the provider 30 via the communication unit 52, for example, by wireless (wireless internet) N1, and further, the data is transmitted from the provider 30 to the processing center on the operation management side by the Internet N2. To the personal computer 60.

  The computer 60 includes a computer main body 61, a keyboard 62 as an input device, and a monitor 63, which receives data from the vehicle side and compares the received data with a threshold value stored in a database (not shown) of the computer main body 61. Thus, it is determined whether or not the driver 2 is tired.

  Then, the determination result is sent back to the communication unit 52 equipped on both sides via the Internet N2 and the wireless N1, and a warning is given to the driver 2 by the display and alarm means 14 near the driver 2.

For the control, the control shown in FIG. 2 (first embodiment) (determining the degree of fatigue based on changes in posture) and the control shown in FIG. 7 (second embodiment) (determining the degree of fatigue based on blood flow), Each of them is performed independently, and if any of them determines that the driver's fatigue has reached a predetermined level or more, a warning is given.
However, a warning should be issued to the driver 2 at the same time both when the degree of fatigue is determined by a change in posture (FIG. 2) and when the degree of fatigue is determined by blood flow (FIG. 7). It is also possible to configure so that the driver 2 is warned only when it is determined that.

  That is, some drivers may dislike increasing the frequency of warnings. If the fatigue level is determined by the posture change (FIG. 2) and the fatigue level is determined by the blood flow rate (FIG. 7), the warning frequency may increase, so the warning frequency increases. It is also assumed that there will be a negative impact on the mental stability of the type of driver who hates. In such a case, when it is determined that “Warning” should be given by both the method of determining the fatigue level based on the change in posture (FIG. 2) and the method of determining the fatigue level based on the blood flow rate (FIG. 7). Only when it is judged that the driver has become tired to a very dangerous stage, the warning is given, but otherwise the warning is not given, and the driver's discomfort is prevented and the mental stability is considered. .

  As described above, the measurement results and reflected light data of the seat pressure sensor 4 and the acceleration sensor 5 processed by the cargo vehicle 1 are, for example, wireless N1 via an information network (wireless LAN) N2 such as the Internet. And sent to the personal computer 60 of the processing center (business office). At the processing center, determination of threshold values and determination of driver fatigue are performed.

  However, although not shown, in-vehicle unit U12 also performs determination of threshold values and fatigue determination of driver 2 in addition to processing of measurement results of seat pressure sensor 4 and acceleration sensor 5 and processing of reflected light data. It is also possible to configure.

  Next, a fourth embodiment will be described with reference to FIGS.

  The fourth embodiment of FIGS. 11 to 13 includes a system (first embodiment) for determining the degree of fatigue due to a change in the posture of the driver, and a system (first embodiment) for determining the degree of fatigue due to the blood flow of the driver. In addition to the above, the embodiment further combines a system for determining a driver's fatigue degree based on an amount (angle) of a “deviation” between the direction in which the vehicle should travel and the direction in which the vehicle actually travels. .

  That is, the fatigue determination system J4 of the fourth embodiment whose overall configuration is shown in FIG. 11 is added to the configuration of the third embodiment of FIG. 10 by adding a GPS system 80, and further to a steering system, for example, a part of the steering column. A steering angle sensor 70 provided and an error processing unit 90 for handle operation are provided.

Here, the direction in which the vehicle should travel is determined from GPS data, the direction in which the vehicle is actually traveling is determined from the detection result of the rudder angle sensor 70, and both of these determined values are used for error handling of the steering wheel operation. The difference (absolute value) between the two is applied to the unit 90 and expressed as a dimension of the above-described “deviation” amount, for example, an angle (see the VH line in FIG. 12). When fatigued, the above-mentioned “deviation” amount (VH line) tends to increase.
Alternatively, the direction in which the vehicle should travel may be determined based on road data in the digital map data.
Here, since it is considered that a large steering wheel operation is performed in various situations when the vehicle is at a low speed, it is configured to determine the “direction in which the vehicle should travel” when the vehicle speed is a certain speed or more. I can do it.

  Next, referring also to FIG. 11 based on the flowchart of FIG. 13, a control method for determining the degree of fatigue based on the amount of “deviation” between the direction in which the vehicle should travel and the direction in which the vehicle actually travels Will be described.

  First, in step S21, it is determined whether or not the operation is started. If the operation is started (YES in step S21), the process proceeds to the next step S22, and the operation is not yet started. If (NO in step S21), the loop of step S21 is repeated.

  In step S <b> 22, GPS data is received by the GPS system 80 and steering angle information is also received from the steering angle sensor 70. In the next step S23, the direction in which the vehicle should originally travel is determined from the GPS data, and the actual vehicle traveling direction is determined based on information from the steering angle sensor 70.

  In the next step S24, it is determined whether the absolute value of the value obtained by subtracting the actual vehicle traveling direction from the direction in which the vehicle should travel is equal to or greater than a threshold value. If the absolute value of the value obtained by subtracting the actual vehicle traveling direction from the direction in which the vehicle should travel is equal to or greater than the threshold value (YES in step S24), the process proceeds to step S25, and the actual vehicle traveling direction is determined from the direction in which the vehicle is traveling. If the absolute value of the subtracted value is less than the threshold value (NO in step S24), the process returns to step S22, and step S22 and subsequent steps are repeated again.

In step S25, the display / alarm unit 14 displays, for example, "You are tired. Take a break early" and a warning voice, and then the process proceeds to step S26.
In step S24, the frequency at which the “value obtained by subtracting the actual vehicle traveling direction from the direction in which the vehicle should travel” (deviation) becomes greater than or equal to a certain value is analyzed, and a warning is issued when the predetermined threshold is added. You may comprise.

  In step S26, it is determined whether or not the control is to be ended. If the control is to be ended (YES in step S26), the control is ended as it is. On the other hand, if the control is to be continued (NO in step S26), the process returns to step S22, and step S2 and subsequent steps are repeated again.

  Regarding the control of the fourth embodiment as a whole, the control shown in FIG. 2 (determining the degree of fatigue by posture change), the control shown in FIG. 7 (determining the degree of fatigue by blood flow), and the control shown in FIG. The degree of fatigue is determined independently based on the amount of “deviation” between the direction in which the vehicle should travel and the direction in which the vehicle is actually traveling). It is set to give a warning if it is determined that the

However, in consideration of the case where the driver may dislike increasing the frequency of warnings, a method for determining the fatigue level based on changes in posture (FIG. 2) and a method for determining the fatigue level based on blood flow (FIG. 2) 7) and two of the above-described methods for determining the degree of picking up based on the “deviation” amount (FIG. 13), the driver is warned when it is determined that the driver should be warned. It is also possible to configure so as to.
Alternatively, when the degree of fatigue is determined by posture change (FIG. 2), when the degree of fatigue is determined by blood flow (FIG. 7), and when the degree of pick-up is determined by the above-described “deviation” amount (FIG. 13). In all of the above, it is possible to configure so that the driver is warned only when it is determined that the driver should be warned.

As described above, also in FIG. 11, the fatigue determination means is displayed as the personal computer 60 of the processing center (business establishment) located at a remote place from the travel point of the cargo vehicle 1 subject to the driver's fatigue check. .
The measurement results and reflected light data of the seating pressure sensor 4 and the acceleration sensor 5 processed by the cargo vehicle 1 are, for example, wireless N1 via an information network (wireless LAN) such as the Internet N2, etc. To the personal computer 60 of the office). At the processing center, determination of threshold values and determination of driver fatigue are performed.

  Also in the fourth embodiment of FIG. 11, although not shown, it is possible to configure the in-vehicle unit U <b> 13 to determine the threshold and determine the driver's fatigue.

  In the first to fourth embodiments and the modifications described above, it is possible to combine a system and a method for detecting the driver's body temperature and determining the degree of fatigue from the detected body temperature fluctuation.

Here, it is known that body temperature decreases during sleep. Therefore, when the body temperature falls below the threshold, at least the driver accumulates fatigue and feels “sleepy”, and there is a possibility that the patient is “sleeping” at worst.
It is possible to further improve safety by combining the detection of such a state with the above-described embodiments and modifications.

  More specifically, the control method shown in FIG. 14 can be performed. Below, the method of the display determination control of FIG. 14 is demonstrated.

First, in step S31, it is determined whether or not the operation is started. If the operation is started (YES in step S31), the process proceeds to the next step S32, and the operation is not yet started. If (NO in step S31), the loop of step S31 is repeated.

    In step S32, after measuring the body temperature of the driver 2, in step S33, it is determined whether the body temperature is equal to or higher than a threshold value. If the body temperature is equal to or higher than the threshold value (YES in step S33), the process proceeds to step S34. If the body temperature is lower than the threshold value (NO in step S33), the process returns to step S32, and step S32 and subsequent steps are repeated again.

  In the next step S34, the display / alarm unit 14 displays, for example, "You are tired. Take a break early" and a warning voice, and then the process proceeds to step S35.

  In step S35, it is determined whether or not the control is to be ended. If the control is to be ended (YES in step S35), the control is ended as it is. On the other hand, if the control is to be continued (NO in step S35), the process returns to step S32, and step S32 and subsequent steps are repeated again.

  Here, as described above, it is preferable to detect the body temperature in a non-contact manner in consideration of the presence of a driver who feels uncomfortable about attaching the sensor to the body surface.

  The inventors, while studying a series of fatigue determination systems and fatigue determination methods, embedded only two pressure sensors on the left and right sides of the driver seat, and processed information from this pair of left and right pressure sensors, By determining the processed values, the technology for determining that the driver is in a fatigue tendency has been established.

FIG. 15 is a graph summarizing data when a fatigue determination test is performed using the fatigue determination technique.
In the graph of FIG. 15, the horizontal axis indicates time, and the left vertical axis indicates that the driver abnormally moves the body in one minute, the pressure sensor receives the abnormal movement, and the received data is processed as described below. The number of times that the value (abnormal value) obtained in 1 minute was counted is scaled (corresponding to a bar graph), and the right side of the vertical axis is calibrating the heart rate of the driver (corresponding to a line graph).
According to FIG. 15, when the abnormal value is “positive” and tends to increase (portion indicated by arrow A in FIG. 15), the heart rate gradually decreases (arrow in FIG. 15). B)), that is, the driver shows that the arousal level is gradually decreasing.
In FIG. 15, the portion where the abnormal value protrudes (the portion indicated by the arrow C) is temporarily returned to the driver whose waking has been lowered temporarily (for example, by the headlight of an oncoming vehicle). It shows that.

  16 to 18 show a fifth embodiment of the technique (fatigue judgment system) obtained by the above research. The fifth embodiment will be described below with reference to the drawings.

  In FIG. 16, the fifth embodiment (the entire fatigue determination system is indicated by a symbol J5) includes a pair of seat pressure sensors (pressure sensors) 4 embedded in the left and right sides of the seat surface 31 of the driver seat 3; The control unit 40 is a control means, and the display / alarm means 14 is provided in front of the driver's seat.

  The control unit 40 includes an interface 6, a seating pressure data storage unit 41, a standard deviation calculation unit 42, a comparison unit 43, a counting unit (counter) 44, a timing unit (timer) 45, and a calculation unit (subtraction unit). 46, database 11 and fatigue determination unit 13C.

The interface 6 relays the seat pressure information from the left and right seat pressure sensors 4 (the data value of the left seat pressure is L and the right seat pressure is R) to the storage means 41.
Here, when the seating pressure information L and R is stored in the seating pressure data storage unit 41, in addition to the values of L and R, the “pressure (L−R)”, “ It is also preferably stored in the form of “sum (L + R)” and “difference in change amount of detection value (dL−dR)”.

  The standard deviation calculation unit 42 extracts “difference (L−R)”, “sum (L + R)”, and “difference in detection value change (dL−dR)” from the stored data. The standard deviation σ of each of “difference (LR)”, “sum (L + R)”, and “difference in change in detected value (dL−dR)” is obtained.

The comparison unit 43 extracts “difference (LR)”, “sum (L + R)” and “difference in detection value change (dL−dR)” between the left and right detection values from the seat pressure data storage means 41, Further, the standard deviation σ of those “difference (LR)”, “sum (L + R)”, and “difference in change in detected value (dL−dR)” obtained by calculation by the standard deviation calculation unit 42 is derived, Each standard deviation σ is tripled, and the tripled value 3σ is taken as an abnormal value.
Then, the magnitudes of “difference (LR)”, “sum (L + R)”, “difference in detection value change (dL−dR)” and three times the standard deviation σ (3σ) are compared. , “Difference (LR)”, “sum (L + R)”, and “difference in change amount of detection value (dL−dR)” that are larger than 3σ are extracted and used as abnormal values. .

  The counting unit (count unit) 44 counts the number of times the abnormal value appears per predetermined time. The timing unit 45 is means for allocating the predetermined time according to the driving state.

  The arithmetic unit (subtraction unit) 46 calculates the number of times c (L + R) that the “sum” abnormal value appears from the number c (L−R) that the abnormal value “difference” appears, and the “difference in change amount”. By subtracting the number of times C (dL−dR) at which the abnormal value has appeared, “C (LR) −C (L + R) −C (dL−dR)” is obtained. The obtained “C (LR) −C (L + R) −C (dL−dR)” is also stored in the database 11.

  The fatigue determination unit 13C has a tendency that “C (L−R) −C (L + R) −C (dL−dR)” takes a value of “positive” and increases (subtraction result of the previous cycle and If it is larger than that), it is determined that the driver is tired and a warning to that effect is given to the driver via the display / warning means 14.

Next, a fatigue determination control method according to the fifth embodiment will be described with reference to FIGS. 17 and 18.
First, in step S41, the timer 45 is activated to receive the output signals L and R from the left and right pressure sensors 4 (step S42), and "difference (LR)", "sum (L + R)", " The difference in the detected value change amount (dL−dR) ”is calculated in the seat pressure data storage means 41 (step S43).

  In the next step S44, the standard deviation calculation means 42 sets the standard deviation σ for each of the “difference (LR)”, “sum (L + R)”, and “difference in detected value variation (dL−dR)”. Calculate and compare “difference (LR)”, “sum (L + R)”, “difference in detected value variation (dL−dR)” with 3 times the standard deviation σ and 3σ (step S45). .

  Next, it is determined whether each of “difference (L−R)”, “sum (L + R)”, and “difference in change amount of detection value (dL−dR)” exceeds 3σ (step If it exceeds (YES in step S46), the process proceeds to step S47, and if it is 3σ or less (NO in step S46), the process proceeds to step S48 as it is.

  In step S47, each of “difference (LR)”, “sum (L + R)”, and “difference in change amount of detection value (dL−dR)” is counted, and the difference “3 ( L−R) ”exceeds the number of times C (LR),“ sum (L + R) ”exceeds the number of times C (L + R), and“ difference in detection value change (dL−dR) ”. Let C (dL-dR) be the exceeding number. Then, the number of times exceeding those is sequentially added.

  In step S48, it is determined whether or not a predetermined time has elapsed. If it has elapsed (YES in step S48), the process proceeds to A, that is, the flow of FIG. 18, and if it has not elapsed (NO in step S48). ), Return to step S42, and repeat step S42 and subsequent steps.

In step S51 of FIG. 18, the following subtraction, that is, “C (LR) −C (L + R) −C (dL−dR)” is obtained from the count number, and the result becomes a “positive” value. Whether or not “C (L−R) −C (L + R) −C (dL−dR)”> 0 is determined.
If the value is “positive” (YES in step S51), the process proceeds to step S52. If the value is 0 or “negative” (NO in step S51), the process proceeds to step S54.

  In step S52, it is determined whether or not “C (LR) −C (L + R) −C (dL−dR)” has an increasing tendency (YES in step S52). Proceed to S53. On the other hand, if not increasing (NO in step S52), the process proceeds to step S54.

  In step S53, it is determined that the driver 2 is in a fatigue tendency, a warning is given to the driver 2 via the display / warning means 14, and then the process proceeds to step S55. In step S54, it is determined that there is no fatigue tendency, and the process proceeds to step S55.

  In step S55, the control unit 40 determines whether or not to end the control. If the control unit 40 ends the process (YES in step S55), the control unit 40 ends the process as it is and does not end the process (NO in step S55). That is, it returns to step S41 of FIG. 17, and repeats after step S41 again.

  The fifth embodiment of FIGS. 16 to 18 is an embodiment in which all the devices related to the fatigue determination system are mounted on the vehicle. On the other hand, as in the embodiments shown in FIGS. 5 and 9 to 11, it is possible to utilize a wireless LAN or a wireless Internet.

According to the fatigue determination system of the fifth embodiment configured as described above, the following effects can be obtained.
(A) The fatigue state of the driver 2 can be detected simply by providing a pair of left and right pressure sensors 4 (two in total) and processing the signals. Therefore, three or more pressure sensors or acceleration sensors are not required.
(B) By determining whether “C (L−R) −C (L + R) −C (dL−dR)” has a positive value and tends to increase, the driver can It is possible to reliably determine whether or not there is a tendency to fatigue.
(C) It is only necessary to provide two pressure sensors on the driver 2, and it is not necessary to attach a special device to the driver's body or to observe the driver 2 with a monitor or the like. So that you do n’t feel uncomfortable.
(D) Fatigue tendency determination corresponding to the driving situation can be made by appropriately setting the time for obtaining the standard deviation σ and the number of times C (LR), C (L + R), and C (dL-dR). It becomes possible.
(E) Since measurement and control are relatively simple, a complicated and large-scale signal processing facility is not required, which contributes to downsizing of the entire system.
(F) By utilizing a wireless LAN, a wireless Internet, or the like, a device (on-vehicle device) installed in an automobile can be reduced in size.

It should be noted that the illustrated embodiment is merely an example, and is not a description to limit the technical scope of the present invention.
For example, in addition to the illustrated embodiment, the amount of “deviation” between the direction in which the vehicle should travel and the direction in which the vehicle is actually traveling in the first embodiment (determination of fatigue level due to changes in driver posture). It is also possible to combine systems for determining the degree of fatigue.
Further, the second embodiment (determining the degree of fatigue based on the blood flow rate) is combined with a system for determining the degree of fatigue based on the amount of “deviation” between the direction in which the vehicle should travel and the direction in which the vehicle is actually traveling. Is also possible.
Furthermore, it is possible to combine these with a system (FIG. 14) that detects body temperature and determines the degree of fatigue.

  In addition to this, for example, the first modification of the first embodiment and the second modification of the first embodiment can be combined. Alternatively, the first modification of the first embodiment may be combined with the second embodiment.

Alternatively, in the present invention, the surface temperature of the driver's face (for example, the cheek on the side other than the window side) is detected by an infrared sensor, the body temperature is estimated from the detection result, and whether the driver feels drowsy from the estimated body temperature It can be configured to determine whether or not.
Specifically, in FIG. 6, for example, the unit indicated by reference numeral 17 is a unit that estimates body temperature from reflected light data, and the unit indicated by reference numeral 13B is a unit that determines whether or not the person feels sleepy from the estimated body temperature. Toss. In FIG. 7, the control may be performed by setting step S <b> 14 as a “body temperature estimation step” and step S <b> 15 as a logic circuit step “body temperature ≦ threshold?”.
Here, the contents can be combined with the above-described present invention (or each embodiment of the present invention).

  The present invention not only determines the fatigue of drivers who drive automobiles such as cargo cars, but also fatigue of operators who operate other vehicles such as motorcycles, trains and airplanes, and various machines such as construction machines and machine tools. It can also be applied to determination.

The block diagram which showed the whole structure of 1st Embodiment of this invention. The flowchart explaining the fatigue determination control method of 1st Embodiment of this invention. It is a characteristic view which shows the relationship between the driver's fatigue, the posture change frequency, and the amount of fluctuation, and is a diagram showing a state in which the driver is not particularly fatigued. It is a characteristic figure which shows the relationship between the driver's fatigue, the frequency of a posture change, and the amount of change, and is a figure of the state which is in particular fatigued. The block diagram which showed the structure of the modification of 1st Embodiment. The block diagram which showed the whole structure of 2nd Embodiment of this invention. The flowchart explaining the fatigue determination control method of 2nd Embodiment of this invention. The characteristic view which shows the relationship between the driver's fatigue, the change of body temperature, and the change of heart rate. The block diagram which showed the whole structure of 3rd Embodiment of this invention. The block diagram which showed the structure of the modification of 3rd Embodiment. The block diagram which showed the structure of the modification of 4th Embodiment. FIG. 4 is a characteristic diagram showing a relationship between driver fatigue and a difference between a direction in which the vehicle should travel and a direction in which the vehicle is actually traveling. The flowchart explaining a part of fatigue determination control method in connection with 4th Embodiment of this invention. The flowchart which showed the method of detecting a dozing state other than 1st Embodiment-4th Embodiment. The graph which shows the test data performed in order to confirm in establishing 5th Embodiment of this invention. The block diagram which showed the whole structure of 5th Embodiment of this invention. The flowchart explaining a part of fatigue determination control method in connection with 5th Embodiment of this invention. The flowchart explaining the remaining part of the fatigue determination control method in connection with 5th Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Vehicle 2 ... Driver 3 ... Seat 4 ... Seat pressure sensor / pressure sensor 5 ... Acceleration sensor 6 ... Interface 7 ... Seat pressure data storage means 8 ... Acceleration Data storage means 9 ... vehicle swing influence eliminating unit 10 ... posture variation determining unit 11 ... database 12 ... threshold determining units 13A, 13B, 13C ... fatigue determining unit 14 ... display / cumulative Alarm means 15 ... Light irradiation right reflected light receiving unit 16 ... Reflected light data fluctuation calculation unit 17 ... Blood flow data determining unit 18 ... Detection signal processing unit 19 ... Reflected light data processing unit 20A, 20B ... Communication unit N1 ... Wireless / wireless Internet N2 ... Information network / Internet 30 ... Provider 40 ... Control unit 41 ... Seat pressure data storage means 42 ... Standard deviation calculation unit 43 ... Comparison unit 44 ... Count unit 46 ... Calculation unit / subtraction unit 50 ... In-vehicle computer 60 ... PC 70 for judging fatigue at the processing center ... steer angle sensor 80 ... GPS system

Claims (2)

A unit that detects a change in the posture of the driver, the unit including a unit that detects a change in the posture of the driver, and a unit that determines whether the driver is tired based on the frequency of the posture change detected by the unit. Includes a pair of left and right pressure sensors provided on the driver seat, and the determination unit obtains a standard deviation of a difference between the detection results of the left and right pressure sensors, a sum, and a difference in the amount of change in the detection results. For each of the difference of the sum, the amount of change, a value larger than an integral multiple of the standard deviation is defined as an abnormal value, the number of times the abnormal value appears per fixed time is counted, and the number of times the abnormal value of the difference appears From the above, the number of times that the abnormal value of the sum appears and the number of times that the abnormal value of the difference in the amount of change appears are subtracted, and the subtraction result takes a “positive” value and tends to increase. Is Fatigue system for judging the driver is characterized by having a a function of determining fatigued. A step of detecting a change in the posture of the driver, and a step of determining whether or not the driver is tired from the frequency of the detected posture change. A detection signal is received from a pair of left and right pressure sensors provided on the seat, and in the step of determining, the difference between the detection results of the left and right pressure sensors, the sum, and the standard deviation of the difference in the amount of change in the detection results are obtained, A value larger than an integer multiple of the standard deviation is defined as an abnormal value for each of the difference, sum, and change amount, and the number of times the abnormal value appears per certain time is counted, and the abnormal value of the difference appears. The number of times that the abnormal value of the sum appears and the number of times that the abnormal value of the difference in the amount of change appears are subtracted from the number of times, and the result of the subtraction tends to take a positive value and increase. If dry Fatigue determination method over to and determines that fatigued.

Images