Driver Drowsiness Estimation Method

Abstract

To provide a driver's drowsiness estimation method which can accurately detect a sleepy or shallow sleepy state before a driver realizes.SOLUTION: A driver's drowsiness estimation method for estimating drowsiness of a driver of a vehicle comprises: a preparation step (S1) of preparing an in-vehicle camera 2 which detects a head behavior of the driver and a steering angle sensor 3 which detects a steering angle by a steering operation of the driver; a head roll angle calculation step (S3) of calculating a head roll angle based on the head behavior detected by the in-vehicle camera 2; a coherence calculation step (S4) of calculating a coherence value of the steering angle and the head roll angle detected by the steering angle sensor 3; and a drowsiness level determination step (S5, S6, S8, S9) of determining a preset drowsiness level 2 when the coherence value is 0.7 or more and less than 0.9 and determining the drowsiness level 3 at which the drowsiness is stronger than the drowsiness level 2 when the coherence value is less than 0.7.SELECTED DRAWING: Figure 7

Description

Technical Field

[0001] The present invention relates to a drowsiness estimation method for estimating the drowsiness of a vehicle driver.

Background Art

[0002] Conventionally, techniques have been proposed for estimating the biological state of a vehicle driver, particularly drowsiness (wakefulness), based on a plurality of feature amounts detected from the driver's face image, biological information, and the like. In such vehicle driving support technologies, the biological state of the driver during vehicle driving is monitored in real time and measured in a non-restrained and non-invasive manner without interfering with the view ahead of the vehicle or the operation behavior of the driving equipment.

[0003] The drowsy driving warning device of Patent Document 1 includes a behavior sensor such as a steering angle sensor, an image drowsiness level determination unit that determines an image drowsiness level and reliability based on the driver's image, a vehicle behavior drowsiness level determination unit that determines a vehicle behavior drowsiness level and reliability based on the vehicle behavior, a comprehensive drowsiness level determination unit that determines a comprehensive drowsiness level and comprehensive reliability based on the image drowsiness level and reliability and the vehicle behavior drowsiness level and reliability, and a warning level setting unit that sets a warning level based on the comprehensive drowsiness level and comprehensive reliability.

[0004] In the image drowsiness level determination unit, the drowsiness level is determined using known evaluation criteria. As shown in FIG. 8, the driver's drowsiness level is divided into five levels from 1 to 5. When the driver's actions acquired from the captured image are such that the eye movement is fast and frequent and the movement is active with body movement, it is determined as drowsiness level 1. When the lips are open and the eye movement is slow, it is determined as drowsiness level 2. When the blinking is slow and frequent, it is determined as drowsiness level 3. Also, when there are intentional blinks, the head is shaken, and yawning is frequent, it is determined as drowsiness level 4. When actions such as closing the eyelids and tilting the neck back and forth are detected, it is determined as the highest drowsiness level 5.

Prior Art Documents

Patent Document

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] The driving of a vehicle is an operation (output) after information input from the driver's vision, vestibular sense, somatic sensation, etc. is processed, and is composed of a higher-order voluntary (cognitive) movement represented by the operation behavior of driving equipment and a lower-order involuntary movement represented by maintaining the head posture, etc. The cerebral cortex that controls the higher-order voluntary function and the brainstem that controls the lower-order involuntary function are both aggregates of nerve cells, and these nerve cells output according to a predetermined firing frequency (firing characteristics). Under normal circumstances, the cortical activity coordinated by the cerebral cortex and the brainstem activity coordinated by the brainstem have a linkage that follows each other and a periodicity that operates rhythmically. Therefore, when drowsiness is regarded as a kind of slight brain movement abnormality as a function of the whole brain, it is expected that the periodic linkage between the cortical activity and the brainstem activity will be impaired by drowsiness.

[0007] The drowsy driving warning device of Patent Document 1 comprehensively determines the drowsiness level by using the movement (expression) of the photographed face and the occurrence frequency of corrective steering. However, in Patent Document 1, since the cortical activity and the brainstem activity are not evaluated from the perspective of the correlation that correlates them in the brain, there is a possibility that sufficient determination accuracy cannot be ensured. In the evaluation criteria of FIG. 8, for example, the determination of drowsiness level 4 with a high drowsiness level determines the large movements of the driver that are physically and superficially manifested. Therefore, although the determination is easy and the accuracy is high, the timing for completing the determination is late, which hinders post-processing such as warnings. On the other hand, the determination of drowsiness level 2 with a low drowsiness level can determine small movements of the driver. Although the determination timing can be made earlier, the actual determination is difficult and the accuracy tends to be low.

[0008] Moreover, both the facial expression and the corrected steering frequency are evaluated after being expressed in a manner recognizable by visual observation from the outside. In other words, they are evaluated after the driver's drowsiness level has increased to a certain extent. Therefore, it does not extract the periodic linkage between the cortical activity and the brainstem activity and evaluate the drowsiness level in time series for a minute frequency band that cannot be easily recognized from the outside. That is, it is not easy to accurately detect before the driver becomes aware of drowsiness or in a light drowsiness state.

[0009] An object of the present invention is to provide a method for estimating a driver's drowsiness and the like that can accurately detect drowsiness or a light drowsiness state before the driver becomes aware of it.

Means for Solving the Problem

[0010] The method for estimating a driver's drowsiness according to claim 1 is a method for estimating a driver's drowsiness in a vehicle, comprising a preparation step of preparing a head movement detection means for detecting a driver's head movement and a steering angle detection means for detecting a steering angle by the driver's steering angle operation, a head roll angle calculation step of calculating a head roll angle based on the head movement detected by the head movement detection means, and a coherence calculation step of calculating the coherence between the steering angle detected by the steering angle detection means and the head roll angle a coherence calculation step of calculating coherence using the steering angle and the head roll angle having a frequency band of 0.2 Hz or less and a drowsiness level determination step of determining a first drowsiness level set in advance when the coherence is equal to or greater than a first threshold value and less than a second threshold value, and determining a second drowsiness level with stronger drowsiness than the first drowsiness level when it is less than the first threshold value.

[0011] In this driver drowsiness estimation method, since there is a preparation step of preparing a head movement detection means for detecting the head movement of the driver and a steering angle detection means for detecting the steering angle by the steering operation of the driver, it is possible to reliably extract the head movement of the driver and the steering angle reflecting the steering wheel angle operation by the driver which is a higher-order voluntary movement. Since there is a head roll angle calculation step of calculating a head roll angle based on the head movement detected by the head movement detection means, it is possible to calculate the head roll angle of the driver reflecting the lower-order involuntary movement that is involuntarily executed as a brainstem activity. Since there is a coherence calculation step of calculating the coherence between the steering angle detected by the steering angle detection means and the head roll angle, it is possible to calculate the periodic linkage between the steering angle which is a cortical activity result and the head roll angle which is a brainstem activity result in a micro frequency band. When the coherence is equal to or greater than a first threshold value and less than a second threshold value, it is determined as a preset first drowsiness level, and when it is less than the first threshold value, there is a drowsiness level determination step of determining a second drowsiness level with stronger drowsiness than the first drowsiness level. Therefore, it is possible to determine the previous drowsiness or a light drowsiness state that the driver perceives based on the periodic linkage between the cortical activity and the brainstem activity.

[0012] and In the coherence calculation step, coherence is calculated using the steering angle and the head roll angle whose frequency band is 0.2 Hz or less. therefore It is possible to calculate the periodic linkage between the steering angle and the head roll angle in a frequency band where the synchronism between the cortical activity and the brainstem activity appears significantly, and the accuracy of drowsiness level determination can be improved.

[0013] Claim 2 The invention of one In the invention, in the coherence calculation step, when the steering angle is equal to or greater than a predetermined threshold value, the coherence is calculated, which is a feature. With this configuration, it is possible to calculate the periodic linkage between the steering angle and head roll angle movement in a driving environment where synchrony between cortical activity and brainstem activity is prominent, thereby further improving the accuracy of drowsiness level judgment.

[0014] Claim 3 The invention is as follows: one or two In the invention, the preparation process is characterized in that the driver is in an awake state, the first and second thresholds are preset based on coherence calculated using the steering angle and head roll angle in a frequency band of 0.06 Hz or more and 0.15 Hz or less, and the preset first and second thresholds are stored in a storage medium. According to this configuration, it is possible to execute the wakefulness maintenance assistance when the driver is in a state of drowsiness or light drowsiness before the driver is aware of it, thereby improving the effectiveness of the wakefulness maintenance assistance. Effect of the Invention

[0015] According to the driver drowsiness estimation method of the present invention, by evaluating the periodic linkage between cortical activity and brainstem activity, drowsiness or a light drowsiness state before the driver is aware of it can be accurately detected. [Brief description of the drawings]

[0016]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Mode for Carrying Out the Invention

[0017] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. The following description exemplifies the application of the present invention to a driver riding in a vehicle (driving simulator), and does not limit the present invention, its application, or its use.

Example

[0018] Hereinafter, Example 1 of the present invention will be described with reference to FIGS. 1 to 8. When the vehicle is in operation, the sleepiness estimation device 1 determines the sleepiness level at the stage before the driver becomes aware of it or a very shallow sleep state through a sign (coherence) related to the brain function. When this sleepiness estimation device 1 determines a very shallow sleep state, it executes arousal maintenance control to restore the driver's wakefulness, and when a high sleepiness state is determined, abnormal time control is executed for danger avoidance.

[0019] As shown in FIG. 1, the sleepiness estimation device 1 mainly includes an in-vehicle camera 2 (head movement detection means), a steering angle sensor 3 (steering angle detection means), one or more displays 4, an audio output device 5, a steering control device 6, a brake control device 7, an accelerator control device 8, and an ECU (Electronic Control Unit) 10.

[0020] The in-vehicle camera 2 is installed, for example, inside the windshield (not shown) in order to obtain time-series data of the driver's head movement every 100 msec. This in-vehicle camera 2 is an image sensor such as a CCD (Charge Coupled Device) or CMOS (Complementary Metal Oxide Semiconductor) that can capture the interior situation including the driver who operates the steering wheel (not shown). The image captured by the in-vehicle camera 2 is transmitted to the ECU 10 via the in-vehicle network (not shown). The steering angle sensor 3 detects the steering angle of the steering wheel, which is one of the driving operation behaviors, every 100 msec and outputs the detected steering angle to the ECU 10.

[0021] Each of the devices 4 to 8 receives a command signal from the ECU 10 and performs a prescribed operation. The display 4 is a display visible to the driver and includes a center display, a head-up display, etc. arranged on the instrument panel (not shown). When executing the wakefulness maintenance control in which a mild drowsy state (for example, drowsiness level 2) is determined, the display 4 gives a visual stimulus to the driver within a range where the driving operation is not hindered. For example, the brightness or lightness of the traveling lane image in front of the host vehicle may be increased (or decreased), or an alarm may be displayed to such an extent that it does not interfere with driving.

[0022] The voice output device 5 gives an auditory stimulus to the driver within a range where the driving operation is not hindered when executing the wakefulness maintenance control. For example, it plays a preset BGM or the like at a low volume. The display 4 and the voice output device 5 mainly operate when executing the wakefulness maintenance control. When executing the abnormal time control in which a high drowsy state (for example, drowsiness level 3) is determined, the display 4 warns the driver more strongly via visual information than when executing the wakefulness maintenance control, and the voice output device 5 warns the driver more strongly via voice than when executing the wakefulness maintenance control.

[0023] The steering control device 6 controls the steering angle for the vehicle to turn a curve. The brake control device 7 generates a braking force for decelerating the vehicle. The accelerator control device 8 generates a driving force for the vehicle to travel. For example, when a high drowsy state is estimated and determined as an abnormal situation, an abnormal output is given, and abnormal control is performed to reduce the vehicle speed under vehicle control and make an emergency stop on the road shoulder to avoid danger.

[0024] Next, the ECU 10 will be described. The ECU 10 is composed of a central processing unit that executes various programs, memories (RAM, ROM), an input / output bus, etc. As shown in FIG. 1, the ECU 10 includes a head roll angle calculation unit 11, a coherence calculation unit 12, a drowsiness level determination unit 13, etc. Note that the following description includes an explanation of a drowsiness estimation method using the drowsiness estimation device 1.

[0025] The head roll angle calculation unit 11 extracts the driver's head behavior from the images continuously captured in time series by the in-vehicle camera 2 during driving, and calculates the driver's head roll angle based on the head roll angle movement among this head behavior. Note that the head roll angle is defined as the inclination angle in the left-right direction of the head around the rotation center axis when the vehicle body front-rear direction axis is the rotation center axis. The coherence calculation unit 12 calculates the coherence value between the steering angle detected by the steering angle sensor 3 and the head roll angle.

[0026] Here, in order to clarify the correlation between the steering angle and the driver's head roll angle during vehicle driving, a verification experiment using a driving simulator (hereinafter abbreviated as DS) was conducted. As shown in FIG. 2, the DS is a course in which 12 straight lines and 12 curves are alternately and continuously formed. The experimental conditions are a target vehicle speed of 80 km / h and a driving time of about 60 minutes. The curve has a curvature radius that requires the steering angle to be equal to or greater than a determination threshold value (for example, about 30°). During the verification experiment, the vehicle behavior (vehicle speed, accelerator pedal depression amount, steering angle) corresponding to the driving operation behavior by the driver, the head behavior of the driver (head yaw angle, head pitch angle, head roll angle), and the driver's drowsiness level based on a preset evaluation criterion (see Fig. 8) were detected in time series.

[0027] As shown in Figs. 3(a) to 3(d), in the region where the drowsiness level is 4 or more, large disturbances occur in all of the vehicle speed (a), accelerator pedal depression amount (b), and steering angle (c). In the region near the drowsiness level of 2, there is almost no disturbance in the vehicle speed (a) and the accelerator pedal depression amount (b), and only the steering angle (c) has a disturbance. Thus, it was found that near the drowsiness level of 2, during driving, among the cortical activities coordinated by the cerebral cortex, the change in the steering angle (c) is more visible than other cortical activities.

[0028] As shown in Figs. 4(a) to 4(d), in the region where the drowsiness level is 4 or more, large disturbances occur in all of the head yaw angle (a), head pitch angle (b), and head roll angle (c). In the region near the drowsiness level of 2, there are no significant changes in the head yaw angle (a) and the head pitch angle (b), and only the head roll angle (c) has a significant change. Thus, it was found that near the drowsiness level of 2, during driving, among the brainstem activities coordinated by the brainstem, the change in the head roll angle (c) is more visible than other brainstem activities.

[0029] Based on the above analysis results, since the coherence value by frequency corresponds to the periodic linkage, for the time series data of the steering angle and the head roll angle, the coherence value by frequency was calculated using the coherence (correlation degree function). Note that for the calculation of coherence, a fast Fourier transform (FFT) or a frequency analysis method was used.

[0030] Fig. 5 shows the coherence analysis results of the steering angle and the head roll angle. In the analysis results, the coherence value (correlation intensity) between the steering angle and the head roll angle is represented by a value of 0 to 1 for each frequency band. The higher the brightness of the region, the higher the coherence value, and the maximum value of the coherence value is set to 1.0.

[0031] As shown in FIG. 5, regardless of the operation status such as a straight line or a curve in the DS, the higher the frequency band, the lower (darker) the coherence value tends to be. On the other hand, in a frequency band of a predetermined minute frequency (for example, 0.2 Hz) or less, there are regions where the coherence value is high (bright) corresponding to each curve region. As a result, it was found that when the driver is in a waking state (normal), in the minute frequency band, the steering angle which is the result of cortical system activity and the head roll angle which is the result of brainstem system activity have periodic interlocking.

[0032] Also, as shown in FIG. 5, even in the minute frequency band, the higher the sleepiness level, that is, the longer the driving time, the mutual synchronism between the steering angle and the head roll angle is lost, and the coherence values of both tend to be low. As a result, it was confirmed by experimental data that the periodic interlocking between cortical system activity and brainstem system activity is impaired by the occurrence of sleepiness, that is, that sleepiness can be regarded as a kind of minor brain operation abnormality as a function of the whole brain.

[0033] Next, As shown in FIG. 6 Regarding four frequency bands of 0.06 Hz, 0.12 Hz, 0.18 Hz, and 0.23 Hz among the minute frequency bands in which periodic interlocking occurs, the coherence values at the first to fourth, sixth, ninth, and eleventh curve points P1 to P7 were obtained respectively, and a graph showing the correlation was created. Except for the sleepiness level 4 in the frequency bands of 0.06 Hz and 0.12 Hz, all frequency bands have the characteristic that the higher the sleepiness level, the lower the coherence value.

[0034] As shown in FIG. 6, at the sleepiness level 1 (point P1) corresponding to the waking state, the coherence value in the minute frequency band of 0.2 Hz or less is 0.9 (second threshold value) or more. At the sleepiness level 2 (point P3), the coherence value in any frequency band is approximately 0 .It is 7 (the first threshold) or more, and as long as it is the coherence value in the micro frequency band of 0.2 Hz or less, all coherence values are 0.7 or more. At drowsiness level 3 (point P5), the coherence value in the micro frequency band of 0.2 Hz or less is 0.7 or more. If it slightly exceeds drowsiness level 3 (point P6), all coherence values become less than 0.7.

[0035] Based on the above, in this embodiment, when the coherence value is 0.9 or more, it is determined as drowsiness level 1, when it is 0.7 or more and less than 0.9, it is determined as drowsiness level 2, and when it is less than 0.7, it is defined as drowsiness level 3. In other words, the first threshold of 0.7 and the second threshold of 0.9 can be said to be the coherence values calculated using the steering angle and the head roll angle in the frequency band of 0.2 Hz or less, particularly 0.06 Hz or more and 0.15 Hz or less, and are stored in advance in the storage medium (for example, hard disk, etc.) of the vehicle. That is, the first and second thresholds are set using the coherence value in the frequency band of 0.2 Hz or less, and the drowsiness level 3 is also determined using the coherence value in the frequency band of 0.2 Hz or less.

[0036] Return to the description of ECU10. When the coherence value calculated by the coherence calculation unit 12 is 0.7 or more and less than 0.9, the drowsiness level determination unit 13 determines it as the preset drowsiness level 2, and when the coherence value is less than 0.7, it determines it as drowsiness level 3, which is more drowsy than drowsiness level 2. When the drowsiness level determination unit 13 determines drowsiness level 2, ECU10 executes wakefulness maintenance control, and when it determines drowsiness level 3, it executes abnormal time control. When drowsiness level 1 is determined, neither control is executed.

[0037] Next, based on the flowchart of FIG. 7, the drowsiness estimation process will be described. Here, Si (i = 1, 2,...) indicates the steps for each process. First, as shown in FIG. 7, in the drowsiness estimation device 1, various information such as the photographed data acquired by the in-vehicle camera 2, the steering angle detected by the steering angle sensor 3, and the coherence values corresponding to the first and second threshold values are read (S1), and the process proceeds to S2. Also, S1 also serves as a preparation step for preparing the in-vehicle camera 2 and the steering angle sensor 3.

[0038] In S2, it is determined whether the steering angle of the steering wheel operated by the driver is equal to or greater than the determination threshold value. As a result of the determination in S2, if the steering angle is equal to or greater than the determination threshold value, it means that the vehicle is traveling on a curve, and since the periodic linkage between the cortical activity and the brainstem activity can be used for drowsiness level determination, the process proceeds to S3. As a result of the determination in S2, if the steering angle is less than the determination threshold value, it means that the vehicle is traveling straight, and the periodic linkage between the cortical activity and the brainstem activity is low, and it is difficult to use the periodic linkage in this driving scene for drowsiness level determination, so the process returns. In S3, after the head roll angle calculation unit 11 calculates the head roll angle of the driver, the coherence calculation unit 12 calculates the coherence value (S4).

[0039] In S5, it is determined whether the coherence value is less than 0.7, which is the first threshold value. As a result of the determination in S5, if the coherence value is less than 0.7, it means that the driver is strongly drowsy, so drowsiness level 3 is determined (S6), and the process proceeds to S7. In S7, after the ECU 10 executes the abnormal time control, the process returns. As a result of the determination in S5, if the coherence value is 0.7 or more, the process proceeds to S8.

[0040] In S8, it is determined whether the coherence value is less than 0.9, which is the second threshold value. As a result of the determination in S8, if the coherence value is less than 0.9, it means that the driver is less drowsy than at drowsiness level 3, so drowsiness level 2 is determined (S9), and the process proceeds to S10. In S10, after the ECU 10 executes the wakefulness maintenance control, the process returns. As a result of the determination in S8, when the coherence value is 0.9 or more, the driver has no drowsiness at all and is at drowsiness level 1 corresponding to the normal state, so simply return.

[0041] Next, the operation and effect of the above driver drowsiness estimation method will be described. According to this driver drowsiness estimation method, since it has a preparation step (S1) of preparing an in-vehicle camera 2 for detecting the head movement of the driver and a steering angle sensor 3 for detecting the steering angle by the steering operation of the driver, it is possible to surely extract the head movement of the driver and the steering angle reflecting the steering angle operation of the steering wheel by the driver which is a higher-order voluntary movement. Since it has a head roll angle calculation step (S3) of calculating the head roll angle based on the head movement detected by the in-vehicle camera 2, it is possible to calculate the head roll angle of the driver reflecting the lower-order involuntary movement that is involuntarily executed as a brainstem activity. Since it has a coherence calculation step (S4) of calculating the coherence value between the steering angle detected by the steering angle sensor 3 and the head roll angle, it is possible to calculate the periodic linkage between the steering angle which is a cortical activity result and the head roll angle which is a brainstem activity result in a micro frequency band. Since it has a drowsiness level determination step (S5, S6, S8, S9) of determining that it is the preset drowsiness level 2 when the coherence value is 0.7 or more which is the first threshold value and less than 0.9 which is the second threshold value, and determining that it is the drowsiness level 3 with stronger drowsiness than the drowsiness level 2 when it is less than the first threshold value, it is possible to determine the drowsiness before the driver becomes aware or the light drowsiness state based on the periodic linkage between the cortical activity and the brainstem activity.

[0042] In the coherence calculation step (S4), since the coherence is calculated using the steering angle and the head roll angle in a frequency band of 0.2 Hz or less, it is possible to calculate the periodic linkage between the steering angle and the head roll angle in a frequency band where the synchronism between the cortical activity and the brainstem activity is significantly manifested, and the accuracy of the drowsiness level determination can be improved.

[0043] In the coherence calculation process (S4), a coherence value is calculated when the steering angle is equal to or greater than a predetermined threshold value. This makes it possible to calculate the periodic linkage between the steering angle and head roll angle movement in a driving environment where synchrony between cortical activity and brainstem activity is prominent, thereby further improving the accuracy of drowsiness level judgment.

[0044] In the preparation process (S1), when the driver is in an awake state, first and second threshold values ​​are preset based on a coherence value calculated using a steering angle and a head roll angle in a frequency band of 0.06 Hz or more and 0.15 Hz or less, and the preset first and second threshold values ​​are stored in a storage medium. Therefore, wakefulness maintenance assistance can be performed when the driver is in a drowsy state or a light drowsy state before he or she is aware of it, thereby enhancing the effectiveness of the wakefulness maintenance assistance.

[0045] Next, a modified example in which the above embodiment is partially changed will be described. 1) In the above embodiment, an example was described in which drowsiness levels were divided into five stages. However, it may also be possible to use a two-stage evaluation based on the presence or absence of drowsiness, and perform wakefulness maintenance control when drowsiness is determined to be equivalent to drowsiness level 2.

[0046] 2) In the above embodiment, an example was described in which wakefulness maintenance control for drowsiness level 2 is performed when the level is equal to or greater than the first threshold and less than the second threshold, and abnormal condition control for drowsiness level 3 is performed when the level is less than the first threshold. However, it is also possible to set three thresholds and perform first wakefulness maintenance control for drowsiness level 2 when the level is equal to or greater than the first threshold and less than the second threshold, perform second wakefulness maintenance control, which is more stimulating than the first wakefulness maintenance control for drowsiness level 3, when the level is equal to or greater than the second threshold and less than the third threshold, and perform abnormal condition control for drowsiness level 4 when the level is equal to or greater than the third threshold.

[0047] 3) In addition, a person skilled in the art can implement the present invention in a form in which various modifications are added to the above-mentioned embodiment without departing from the spirit of the present invention, and the present invention also includes such modifications. [Explanation of symbols]

[0048] 1. Drowsiness estimation device S3 Head roll angle calculation step S4 Coherence calculation step

Claims

1. A method for estimating a driver's drowsiness, which estimates the drowsiness of a driver of a vehicle, a preparation step of preparing a head movement detection means for detecting the head movement of the driver and a steering angle detection means for detecting the steering angle by the steering angle operation of the driver; a head roll angle calculation step of calculating a head roll angle operation based on the head movement detected by the head movement detection means; a coherence calculation step of calculating the coherence between the steering angle detected by the steering angle detection means and the head roll angle, the coherence calculation step of calculating the coherence using the steering angle and the head roll angle in a frequency band of 0.2 Hz or less; a drowsiness level determination step of determining a first drowsiness level preset when the coherence is equal to or greater than a first threshold value and less than a second threshold value, and determining a second drowsiness level with stronger drowsiness than the first drowsiness level when it is less than the first threshold value; A method for estimating a driver's drowsiness, characterized by comprising the above.

2. The method for estimating a driver's drowsiness according to claim 1, wherein in the coherence calculation step, the coherence is calculated when the steering angle is equal to or greater than a predetermined threshold value.

3. In the preparation step, the first and second threshold values are preset based on the coherence calculated using the steering angle and the head roll angle movement in a frequency band of 0.06 Hz or more and 0.15 Hz or less, with the driver being in an awake state, and the preset first and second threshold values are stored in a storage medium. The method for estimating a driver's drowsiness according to claim 1 or 2, characterized by this.

Images