JPH10329574A - Drive asleep detecting device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately decide a dozing state even when drawsiness deepens and a biological signal is returned to a normal state by a method wherein when within a given time starting from a time when a detecting signal from a biological change detecting means exceeds a first reference value, the detecting signal exceeds a second reference value, it is decided that a driver in a dozing state. SOLUTION: Dozing state decision first decides whether a blood pressure property fluctuation exceeds a reference value Hb (301). When YES, a time T when the blood pressure property fluctuation exceeds a reference value Hb is stored (303). Further, in this time memory, a time counted by a counter arranged in a microcomputer is stored at an RAM. A given time T is then counted by a counter from a time when the blood pressure property fluctuation exceeds a reference value Hb, and it is decided (305) whether a low frequency component exceeds a reference value Hs in the given time. When Yes, it is decided that a driver is in a dozing state, and awakening of a driver is urged by an alarm (307).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a drowsy driving detection device for detecting a drowsy driving of a driver.

[0002]

2. Description of the Related Art Conventionally, as a device for detecting a driver falling asleep in order to prevent a driver from falling asleep during driving, there is a device disclosed in Japanese Patent Laid-Open No. 5-96971. The dozing driving detection device is configured to determine a dozing state when the vehicle behavior signal reaches a predetermined dozing determination reference value, but together with the vehicle behavior signal, the driver's body temperature or the driver Also detects the driver's biological signals such as brain waves,
The drowsiness determination reference value is changed according to the change in the biological signal, thereby improving the accuracy of the drowsiness driving detection.

[0003]

However, changes in the driver's body temperature, brain waves, etc., change greatly once the driver's drowsiness increases, and when the driver's drowsiness further increases, the change causes the driver to feel drowsiness. Since the state returns to a state substantially equivalent to a state without the state (hereinafter, referred to as a normal state), if the dozing determination reference value is changed based on the driver's biological signal as described above, the state in the normal state despite the dozing state is changed. The dozing state is determined based on the dozing determination criteria, and there is a problem that the dozing state cannot be accurately determined.

The present invention has been made in view of the above problems, and has as its object to provide a drowsiness detection device that can accurately determine a drowsiness state even if drowsiness deepens and a biological signal substantially returns to a normal state. .

[0005]

In order to achieve the above object, the following technical means are employed. According to the first aspect of the present invention, the determining means (301 to 305) counts a predetermined time from when the detection signal from the biological change detecting means (105, 107) exceeds the first reference value (Hb). And
The vehicle behavior change detecting means (101, 1
03) is characterized by determining that the vehicle is in a dozing state when the detection signal exceeds the second reference value (Hs).

When the driver falls asleep, the vehicle behavior changes within a predetermined time after a biological change occurs. Therefore, the biological change detecting means (10
5, 107) within a predetermined time after the detection signal exceeds the first reference value (Hb).
When the detection signal from (1, 103) exceeds the second reference value (Hs), it is determined that the vehicle is in a doze state, so that the biological change exceeds the first reference value (Hb) within a predetermined time. Even if it returns within one reference value (Hb), it is possible to accurately detect the drowsy driving.

[0007] In the invention described in claim 2, the first reference value (Hb) is determined by the biological change detecting means (105, 10).
7) is set based on the average value from the start of driving in the biological change detected until a predetermined time elapses, and the second reference value (Hb) is set based on the vehicle behavior change detected by the steering angle change detecting means (101, 103). Is set on the basis of the average value from the start of operation to the elapse of a predetermined time.

[0008] Since the driver usually does not feel drowsy during the predetermined time from the start of driving, the first and second reference values (Hb, Hs) are changed to the biological change and vehicle behavior at the predetermined time from the start of driving. If the setting is made based on the average value of the change, the biological change and the vehicle behavior change at the time of normal driving in which the driver does not feel drowsy can be used as the reference values. Thereby, an appropriate reference value can be set according to the driver's state, and the drowsy driving can be detected more accurately.

According to a third aspect of the present invention, the first reference value (Hb) is set to a maximum value of the biological change detected by the biological change detecting means (105, 107) during a predetermined time from the start of operation. Set based on the average of the value and the minimum value,
A second reference value (Hb) is set based on the average of the maximum and minimum values of the change in the steering angle detected by the steering angle change detection means (101, 103) during a period from the start of driving until a predetermined time has elapsed. Even in this case, the same effect as in claim 2 can be obtained.

[0010]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing a first embodiment of the present invention. FIG. 1 shows a configuration diagram of a dozing driving detection device in the present embodiment. The dozing driving detection device,
It comprises a steering angle sensor 1, an electrocardiograph 2, an input port 3, a microcomputer 4, and an alarm 5.

The steering angle sensor 1 generates a detection signal corresponding to the steering amount related to the vehicle behavior, and detects a change in the steering angle by the steering angle sensor 1. The steering angle sensor 1 is attached to a steering shaft of a vehicle. For example, a slit plate press-fitted into the steering shaft and two sets of photo interrupters attached to a column tube can be used as the steering angle sensor 1. In this case, the rotation direction and the rotation speed can be detected by using the photointerrupter which is turned on / off by being blocked or not blocked by the slit plate with the rotation of the steering shaft. And the rotation speed are used as changes in the steering angle.

The electrocardiograph 2 generates a detection signal corresponding to a heartbeat which is a biological signal of the driver.
Is used to detect the heartbeat interval. The electrocardiograph 2 only needs to be able to detect the heartbeat interval of the driver by some means, and for example, a wristwatch-type electrocardiograph 2 wound around the driver's arm can be used. Then, detection data (output signal) according to the steering angle change and the heartbeat interval detected by the steering angle sensor 1 and the electrocardiogram is input to the microcomputer 4 through the input port 3.
The microcomputer 4 incorporates a CPU, a RAM, and a counter (time measuring means), and performs a calculation by the CPU to determine a drowsy state of the driver.

When the microcomputer 4 determines that the driver is dozing, the alarm 5 prompts the driver to wake up. In addition, the alarm device 5 is to urge the driver to wake up in some form such as sound, vibration, or smell. Next, FIGS. 2 to 4 show flowcharts in the dozing state determination, and the dozing state determination will be described based on these drawings. FIG. 2 is a flowchart of the process data calculation, FIG. 3 is a flowchart of the entire dozing state detection, and FIG. 4 is a flowchart of the dozing driving determination.

First, the processing data calculation shown in FIG. 2 will be described. In this processing data calculation, calculation of each processing data used for dozing driving detection is performed. First, step 10
In step 1, the steering amount is detected. The steering amount is detected by the steering angle sensor 1.
Done by The steering angle sensor 1 generates an output signal corresponding to the steering amount, and the output signal is
Is input to

In step 103, the microcomputer 4 performs an FFT process on the output signal from the steering angle sensor 1, calculates a low frequency component of the output signal, and extracts a low frequency component of the output signal. This low frequency component becomes data relating to a change in the steering angle. The reason why the low-frequency component is extracted by performing the FFT processing on the steering amount in this manner is that the change in the steering angle within a predetermined time becomes slower as the driver's drowsiness deepens. This is because the change in the steering angle becomes remarkable due to the extraction of. In this embodiment, the change in the steering angle is 0.4 to 0.5 Hz.
Are calculated.

Next, at step 105, a heartbeat interval is detected. The heartbeat interval is detected by the electrocardiograph 2, and the electrocardiograph 2 generates an output signal corresponding to the heartbeat, and the output signal is input to the microcomputer 4. And step 10
At 7, the microcomputer 4 calculates the blood pressure fluctuation from the output signal from the electrocardiograph 2, and returns to step 101. This blood pressure fluctuation becomes data relating to the heartbeat interval. In the present embodiment, the variance of the heartbeat interval every eight beats is calculated as the blood pressure fluctuation.

Next, a description will be given of an overall flowchart of the drowsy driving detection shown in FIG. First, in step 201, the low-frequency components obtained in step 103 are integrated. Further, in the following step 203, the blood pressure fluctuation obtained in step 107 is integrated. In step 205, it is determined whether a predetermined time has elapsed from the start of operation. This predetermined time can be set arbitrarily, and can be, for example, 10 minutes. And Y
If it is es, the process proceeds to step 207. If it is no, the process returns to step 201 and the integration of each data is continued.

In step 207, an average value of the low-frequency component obtained in step 201 and the integrated value of blood pressure fluctuation obtained in step 203 within a predetermined time is obtained. Then, the process proceeds to step 209 to calculate a low frequency component reference value Hs and a blood pressure variability reference value Hb from the average value of each data calculated in step 207, and store them in the RAM. Note that the reference values Hs and Hb are set to values obtained by multiplying each average value by a constant. Here, the reference values Hs and Hb are set to twice the average values.

As described above, the reference values Hs and Hb for judging the dozing state are set based on the average value of the change in the steering angle and the average value of the blood pressure fluctuation from the start of driving until the lapse of a predetermined time. This is because the driver normally performs normal driving without feeling drowsy during the predetermined time. As a result, a reference value corresponding to the driver's state at that time can be appropriately set. Here, the reference values Hs and Hb are calculated based on the average value from the integrated value of each data. However, a value obtained by multiplying the average value of the maximum value and the minimum value of each data by a constant may be employed. Good.

Thereafter, the routine proceeds to step 211, where a drowsy driving state is detected. Subsequently, the flowchart of the dozing state determination shown in FIG. 4 will be described. Note that FIG.
The dozing state determination shown in (1) corresponds to step 211. First, in step 301, it is determined whether or not the blood pressure fluctuation exceeds the reference value Hb of the blood pressure fluctuation. And Y
If es, proceed to step 303, if No,
The processing in step 301 is performed again. Step 30
3, the time T when the blood pressure fluctuation exceeds the reference value Hb
And the process proceeds to step 305. The time is stored by storing the time counted by a counter provided in the microcomputer 4 in the RAM.

In step 305, a predetermined time T (for example, 10 minutes) is counted by a counter from the time when the blood pressure fluctuation exceeds the reference value Hb, and the low frequency component exceeds the reference value Hs within the predetermined time. Is determined. If the determination is Yes, the process proceeds to step 307, and the alarm 5 prompts the driver to wake up assuming that the driver is dozing. If No, the driver returns to step 301 because the driver is not dozing.

Next, FIG. 5 shows an example of the result of the drowsiness determination by the drowsiness driving detection device. FIG. 5 shows a case where a dozing driving was performed in the simulation, in which processing data plotting a heartbeat interval, blood pressure fluctuation and a low frequency component until a predetermined time elapses from the start of driving, and a drowsing driving detection device. It is a time chart at the time of making a driver self-report when he felt drowsy together with the experiment. In addition, the respective reference values Hb, Hb calculated from the average value of the blood pressure fluctuation and the average value of the low frequency components for 10 minutes from the start of operation.
s is indicated by a dashed line in the figure.

As shown in the figure, when the blood pressure fluctuation exceeds the reference value Hb, the driver feels drowsy, and the low frequency component is within a predetermined time after the blood pressure fluctuation exceeds the reference value Hb. Exceeds the reference value Hs. Then, as described above, when the low-frequency component related to the change in the steering angle exceeds the predetermined reference value Hs before the predetermined time elapses after the blood pressure fluctuation related to the heartbeat interval exceeds the predetermined reference value Hb, The driver has detected that the driver is dozing. In other cases, for example, if the low-frequency component exceeds the predetermined reference value Hs before the blood pressure fluctuation exceeds the predetermined reference value Hb, dozing detection is not performed.

As described above, since the biological signal is detected based on the heartbeat interval, the biological signal can be detected with a relatively simple signal. If the vehicle behavior signal fluctuates within a predetermined time after the fluctuation of the vehicle, it is determined that the vehicle will fall into a drowsy driving state. The drowsy driving state can be determined.

In this embodiment, the drowsy driving detection is performed based on the heartbeat interval as the biological signal. However, even if the skin surface temperature, skin potential, brain wave, or the like is used as the biological signal, the present embodiment differs from the above embodiment. Similar effects can be obtained. In addition, since the biological signal is obtained with a relatively simple configuration such as the heartbeat interval and the skin surface temperature, the configuration as the dozing driving detection device can be simplified.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating an overall configuration of a dozing driving detection device according to an embodiment.

FIG. 2 is a flowchart in processing data calculation of the dozing driving detecting device shown in FIG.

FIG. 3 is an overall flowchart of dozing driving detection performed by the dozing driving detecting device shown in FIG. 1;

FIG. 4 is a flowchart of a process in a dozing state determination shown in FIG. 2;

FIG. 5 is a diagram showing an experimental result using the drowsy driving detection device in FIG. 1;

[Explanation of symbols]

1 ... steering angle sensor, 2 ... electrocardiograph, 3 ... input port, 4 ... microcomputer, 5 ... alarm.

Claims (3)

[Claims] A biometric change detecting means (105, 107) for detecting a biometric change of a driver; a vehicle behavior change detecting means (101, 103) for detecting a vehicle behavior change due to steering of the driver; A time measuring means for counting; a determination to determine a dozing state of the driver based on a detection signal from the biological change detecting means (105, 107) and a detection signal from the vehicle behavior change detecting means (101, 103). A drowsy driving detection device comprising: means (301 to 305); and means (307) for urging the driver to awake when the drowsing driving of the driver is detected by the determination means (301 to 305). The determination means (301 to 305) starts counting the time from when the detection signal from the biological change detection means (105, 107) exceeds a first reference value (Hb). A predetermined time is counted by the means, and when the detection signal from the vehicle behavior change detecting means (101, 103) exceeds the second reference value (Hs) within the predetermined time, it is determined that the vehicle is in the dozing state. A drowsy driving detection device, characterized in that: 2. The first reference value (Hb) is set based on an average value of a biological change detected by the biological change detection means (105, 107) from the start of operation to a lapse of a predetermined time, and The second reference value (Hb) is set based on an average value of a vehicle behavior change detected by the vehicle behavior change detection means (101, 103) from the start of driving to a lapse of a predetermined time. 1
A drowsy driving detection device according to claim 1. 3. The first reference value (Hb) is an average value of maximum and minimum values of the biological change detected by the biological change detecting means (105, 107) during a period from the start of driving until a predetermined time has elapsed. The second reference value (Hb) is set based on the maximum value and the minimum value of the vehicle behavior change detected by the vehicle behavior change detection means (101, 103) during a period from the start of driving until a predetermined time elapses. The drowsy driving detection device according to claim 1, wherein the setting is performed based on an average value.