JP7302275B2 - Drowsiness estimation device - Google Patents

Description

The present disclosure relates to a drowsiness estimation device.

Patent Literature 1 describes an awakening degree processing device. The arousal level processing device acquires the arousal level from the physiological data using the model. Patent Literature 2 describes an arousal level estimation device. The wakefulness estimation device recognizes the first wakefulness from the driving information. Further, the wakefulness estimation device recognizes the second wakefulness from the biological information of the driver by the second wakefulness recognizing unit. The device for estimating arousal level causes the second arousal level recognizing unit to perform learning such that the second arousal level matches the first arousal level.

JP 2018-82931 A JP 2018-127112 A

A model included in the awakening level processing device described in Patent Document 1 is created for each individual. Creating a model for each individual requires a great deal of labor and cost. In the technique described in Patent Literature 2, the entire second awakening level recognition unit is updated through learning. Therefore, it becomes difficult for the developer of the arousal level estimation device to handle the second arousal level recognition unit after the device is put on the market.

One aspect of the present disclosure is to provide a drowsiness estimation device that can reduce the labor and cost required to create a unit that estimates drowsiness, and that can handle the unit that estimates drowsiness even after the product is put on the market. preferable.

One aspect of the present disclosure is a basic information acquisition unit (11) configured to acquire basic information including at least one of physiological information representing a physiological condition of an occupant of a vehicle and vehicle information representing a condition of the vehicle. an estimating unit (17) configured to estimate a first drowsiness that is the degree of drowsiness of the occupant based on the basic information; and a second drowsiness is calculated by correcting the first drowsiness. a notification unit (21) configured to notify the occupant of the second drowsiness; and a third drowsiness level of the occupant input by the occupant. a third drowsiness acquisition unit (23) configured to acquire and an update unit (25) configured to learn based on said second drowsiness and said third drowsiness and update said correction unit is a drowsiness estimation device (1) comprising:

A drowsiness estimation device that is one aspect of the present disclosure performs learning based on the second drowsiness and the third drowsiness to update the correction unit. The third drowsiness is close to true drowsiness. Therefore, the drowsiness estimation device that is one aspect of the present disclosure can bring the second drowsiness closer to true drowsiness.

A drowsiness estimation device that is one aspect of the present disclosure may not necessarily update the estimation unit. Therefore, the developer of the drowsiness estimation device, which is one aspect of the present disclosure, can handle the estimation unit after the drowsiness estimation device is put on the market.

The estimation unit and correction unit included in the drowsiness estimation device, which is one aspect of the present disclosure, need not be created in advance for each individual. Therefore, it is possible to reduce the labor and cost of creating the estimation unit and the correction unit.

It is a block diagram showing the structure of a drowsiness estimation apparatus. It is a block diagram showing a functional configuration of a drowsiness estimation device. It is an explanatory view showing composition of an instrument panel. FIG. 4 is an explanatory diagram showing the configuration of a steering wheel and input switches; 4 is a flowchart showing processing executed by the drowsiness estimation device; FIG. 4 is an explanatory diagram showing a display mode of a display unit;

Exemplary embodiments of the present disclosure are described with reference to the drawings.
<First embodiment>
1. Configuration of Drowsiness Estimating Device 1 The configuration of the drowsiness estimating device 1 will be described with reference to FIGS. 1 to 4. FIG. The drowsiness estimation device 1 is mounted on a vehicle. The vehicle may be a vehicle capable of automatic operation or a vehicle capable of only manual operation.

As shown in FIG. 1, the drowsiness estimation device 1 includes a microcomputer having a CPU 3 and a semiconductor memory such as RAM or ROM (hereinafter referred to as memory 5).
Each function of the drowsiness estimation device 1 is realized by executing a program stored in a non-transitional substantive recording medium by the CPU 3 . In this example, the memory 5 corresponds to a non-transitional substantive recording medium storing programs. Also, by executing this program, a method corresponding to the program is executed. Note that the drowsiness estimation device 1 may be provided with one microcomputer, or may be provided with a plurality of microcomputers.

As shown in FIG. 2, the drowsiness estimation device 1 includes an occupant identification unit 7, a calling unit 9, a basic information acquisition unit 11, a motion information acquisition unit 13, a driving state information acquisition unit 15, and an estimation unit 17. , a correction unit 19 , a notification unit 21 , a third drowsiness acquisition unit 23 and an update unit 25 .

The drowsiness estimation device 1 includes a plurality of correction units 19 . Each correction unit 19 is associated with one passenger. That is, there is a correction unit 19 associated with each of the multiple occupants.

As shown in FIG. 1, the drowsiness estimation device 1 is connected to other components mounted on the vehicle. Other configurations include a camera 27, a physiological condition sensor 29, a vehicle condition sensor 31, a car navigation system 33, a communication device 35, an in-vehicle device 37, a display unit 39, an input switch 41, and an identification information input unit 43.

The camera 27 captures the face of the vehicle occupant and generates an image (hereinafter referred to as a face image). Camera 27 sends the face image to drowsiness estimation device 1 . A facial image corresponds to physiological information. The occupant is, for example, the driver of the vehicle. Also, the occupant is, for example, an occupant of a vehicle that automatically drives.

The physiological state sensor 29 measures the physiological state of the occupant and generates physiological information. The physiological condition sensor 29 sends physiological information to the drowsiness estimation device 1 . Physiological information is information representing a physiological state. Physiological conditions include, for example, electrocardiogram, respiration, electrooculography, skin conductance, skin temperature, brain waves, pulse waves, and body motion. The physiological condition sensor 29 may generate one type of physiological information, or may generate multiple types of physiological information.

The vehicle state sensor 31 measures the state of the vehicle and generates vehicle information. The vehicle state sensor 31 sends vehicle information to the drowsiness estimation device 1 . The vehicle information is information representing the state of the vehicle. Vehicle conditions include, for example, vehicle speed, acceleration, steering angle, and brake depression amount. The vehicle state sensor 31 may generate one type of vehicle information, or may generate multiple types of vehicle information.

The car navigation system 33 has a function of acquiring vehicle position information, a function of setting a planned travel route of the vehicle, a function of determining the type of road on which the vehicle is traveling, and the like. When the passenger operates the car navigation system 33 , the car navigation system 33 generates motion information and sends it to the drowsiness estimation device 1 . The action information sent by the car navigation system 33 is information indicating that the occupant performed an action to operate the car navigation system 33 .

The communication device 35 can communicate with the outside of the vehicle and acquire various information. Information to be acquired includes, for example, information representing the weather, the current time, and the like.
Examples of the in-vehicle equipment 37 include air conditioners, power windows, car audio systems, lights, wipers, and the like. When the occupant operates the vehicle-mounted device 37 , the vehicle-mounted device 37 generates motion information and sends it to the drowsiness estimation device 1 . The action information sent by the vehicle-mounted device 37 is information indicating that the occupant performed a motion to operate the vehicle-mounted device 37 .

As shown in FIG. 3, the display unit 39 is provided on an instrument panel 45 of the vehicle. A passenger can visually recognize the display unit 39 . The display unit 39 can display second drowsiness and third drowsiness, which will be described later. Functions of the display unit 39 will be described later.

The instrument panel 45 includes a tachometer 47 , a fuel gauge 49 , a speedometer 51 , a name display section 53 , a date/time and temperature display section 55 and a rangefinder 57 . The name display section 53 displays the name of the passenger identified in step 1, which will be described later. The date/time and temperature display section 55 displays the date, time, and temperature.

As shown in FIG. 4, the input switch 41 is provided on the steering wheel 59 of the vehicle. The input switch 41 has a knob 61 . The passenger can operate the knob 61 in the A direction and in the B direction. The input switch 41 sends an A output to the drowsiness estimation device 1 when the knob 61 is operated in the A direction. The input switch 41 sends a B output to the drowsiness estimation device 1 when the knob 61 is operated in the B direction.

The identification information input unit 43 receives the identification information of the passenger. The identification information includes, for example, a password entered by the passenger, a facial image of the passenger, biological information of the passenger, and the like.
2. Processes Executed by Drowsiness Estimation Device 1 Processings executed by the drowsiness estimation device 1 will be described with reference to FIGS. 5 and 6. FIG. In step 1 of FIG. 5, the occupant identification unit 7 acquires occupant identification information using the identification information input section 43 . The occupant identification unit 7 identifies the occupants currently in the vehicle based on the identification information.

In step 2, the calling unit 9 calls the correction unit 19 associated with the occupant identified in step 1 above.
In step 3, the basic information acquisition unit 11 acquires basic information. Basic information includes physiological information and vehicle information. The basic information acquisition unit 11 acquires physiological information using the camera 27 and physiological condition sensor 29 . The basic information acquisition unit 11 acquires vehicle information using the vehicle state sensor 31 .

In step 4 , the motion information acquisition unit 13 performs processing for acquiring motion information from the car navigation system 33 and the in-vehicle device 37 .
In step 5, the operating state information acquisition unit 15 performs processing for acquiring operating state information. The driving state information is information representing the driving state of the vehicle. As the driving state of the vehicle, for example, the driving duration from the start of driving to the present time, whether the road on which the vehicle is traveling is an expressway or a general road, the current time, whether the vehicle is on the outbound or Which of the return trips the vehicle is traveling on, the weather in the area where the vehicle is traveling, the percentage of the time the vehicle was stopped in the past unit time, the steering angle of the vehicle, and the like.

The driving state information acquisition unit 15 can acquire driving state information using, for example, the car navigation system 33, the communication device 35, the vehicle state sensor 31, and the like.
In step 6, the estimation unit 17 estimates the first drowsiness based on the basic information acquired in step 3 above. The first drowsiness is the degree of drowsiness of the passenger. The method by which the estimation unit 17 estimates the first drowsiness is as follows.

The estimating unit 17 extracts features that are correlated with occupant drowsiness from the basic information. Features include, for example, features that can be extracted from a face image. Features that can be extracted from the facial image include, for example, lowered upper eyelids, closed eyes, slow blinking, non-blinking, and loose cheeks or around the mouth. be done.

Features include, for example, features that can be extracted from vehicle information. Features that can be extracted from vehicle information include, for example, that the frequency of occurrence of steering angle fluctuations is low.
The estimation unit 17 estimates the first drowsiness higher when the feature is extracted than when the feature is not extracted. The estimation unit 17 estimates the first drowsiness higher as the features are more prominent. The estimation unit 17 estimates the first drowsiness higher as the number of extracted features increases.

In step 7, the correction unit 19 called in step 2 corrects the first drowsiness to calculate the second drowsiness. The second drowsiness is the degree of drowsiness of the passenger. The correction unit 19 calculates a second drowsiness based on the first drowsiness, the motion information acquired in step 4 above, and the driving state information acquired in step 5 above.

The correction unit 19 is a function that outputs the second drowsiness when the first drowsiness, motion information, and driving state information are input. The contents of the correction unit 19 are initially standard contents. The initial stage is, for example, when the drowsiness estimation device 1 is shipped from the factory. The contents of the correction unit 19 are updated by the process of step 11, which will be described later.

If motion information could not be obtained in said step 4, the correction unit 19 uses a predetermined standard motion information. If the driving state information could not be obtained in step 5, the correction unit 19 uses a predetermined standard driving state information.

In step 8, the notification unit 21 uses the display section 39 to notify the passenger of the second drowsiness calculated in step 7 above. As shown in I of FIG. 6 , the display unit 39 displays the first area 65 within the frame 63 . The first area 65 differs in color or brightness from other areas in the frame 63 . The first area 65 extends rightward from the left end of the frame 63 . A length L2 in the width direction of the first region 65 corresponds to the second drowsiness. The greater the length L2, the higher the second drowsiness. The passenger can know the second drowsiness by looking at the display unit 39 .

In step 9, the third drowsiness acquisition unit 23 determines whether or not there is an input of the third drowsiness. The third drowsiness is the drowsiness of the occupant, which the occupant inputs using the input switch 41 . The occupant subjectively determines and inputs the third drowsiness.

The method by which the occupant inputs the third drowsiness is as follows. When the occupant determines that the third drowsiness is lower than the second drowsiness displayed on the display unit 39, the occupant operates the knob 61 in the A direction shown in FIG.

At this time, as shown in IIA of FIG. 6, the second area 67 is displayed. The second area 67 is an area that differs in color or luminance from other areas. The second area 67 extends rightward from the left end of the frame 63 . A length L3 in the width direction of the second region 67 represents the degree of drowsiness. The length L2 decreases as the number of operations in the A direction increases. The occupant repeats the operation in the A direction until it is determined that the length L3 matches the third drowsiness. The third drowsiness obtaining unit 23 obtains the third drowsiness corresponding to the final length L3.

When the passenger determines that the third drowsiness is higher than the second drowsiness displayed on the display unit 39, the occupant operates the knob 61 in the B direction shown in FIG. At this time, as shown in IIB of FIG. 6, the second area 67 is displayed. The length L3 increases as the number of operations in the B direction increases. The occupant repeats the operation in the B direction until it is determined that the length L3 matches the third drowsiness. The third drowsiness obtaining unit 23 obtains the third drowsiness corresponding to the final length L3.

If it is determined that the third drowsiness has been input, the process proceeds to step 10 . If it is determined that the third drowsiness has not been input, the process proceeds to step 3 .
In step 10, the update unit 25 updates the first drowsiness estimated in step 6, the second drowsiness calculated in step 7, the third drowsiness determined to be input in step 9, and the motion information obtained in step 4. , and based on the operating state information obtained in step 5, the correction unit 19 is learned.

If the motion information could not be obtained in said step 4, the update unit 25 uses the predefined standard motion information. If the driving state information could not be obtained in said step 5, the updating unit 25 uses the predetermined standard driving state information.

The content of the learning is such that the second drowsiness calculated by the correction unit 19 approaches the third drowsiness when a situation in which the first drowsiness, the motion information, and the driving state information are the same or similar occurs in the future. It is to do.

In step 11, the update unit 25 updates the correction unit 19 based on the learning result in step 10 above. The update unit 25 stores the updated correction unit 19 .
In step 12, the notification unit 21 updates the second drowsiness value to the third drowsiness value. That is, the notification unit 21 updates the length of L2 to the length of L3, as indicated by IIIA and IIIB in FIG. Also, the notification unit 21 erases the second area 67 .

3. Effects of Sleepiness Estimation Device 1 (1A) The sleepiness estimation device 1 performs learning based on the second sleepiness and the third sleepiness, and updates the correction unit 19 . The third drowsiness is close to true drowsiness. Therefore, the drowsiness estimation device 1 can bring the second drowsiness closer to true drowsiness.
Here, true drowsiness is a closed sensation within an individual and is difficult to measure objectively. In recent years, a hypothesis has been proposed that the cause of drowsiness is the phosphorylation of proteins in the brain. However, measuring protein changes in the human brain is not easy. Moreover, even if protein changes in the human brain can be measured, the relationship between the amount of measurement and true sleepiness, which is the individual's sense of sleepiness, is thought to differ from person to person.

(1B) Estimation unit 17, for example, is not updated. Therefore, for example, the developer of the drowsiness estimation device 1 can handle the estimation unit 17 after the drowsiness estimation device 1 is put on the market.

(1C) The estimation unit 17 and the correction unit 19 do not have to be created in advance for each individual. Therefore, the labor and cost of creating the estimation unit 17 and the correction unit 19 can be reduced.

(1D) The correction unit 19 calculates a second drowsiness based on the first drowsiness, the motion information, and the driving state information. The update unit 25 learns and updates the correction unit 19 based on the first drowsiness, the second drowsiness, the third drowsiness, the motion information and the driving state information.

Therefore, the drowsiness estimation device 1 can more accurately calculate the second drowsiness. In addition, the drowsiness estimation device 1 can update the correction unit 19 more appropriately based on the motion information and the driving state information.
<Second embodiment>
1. Differences from First Embodiment Since the basic configuration of the second embodiment is the same as that of the first embodiment, the differences will be described below. Note that the same reference numerals as in the first embodiment indicate the same configurations, and refer to the preceding description.

In the second embodiment, the drowsiness estimation device 1 does not acquire driving state information. In the second embodiment, the correction unit 19 calculates the second drowsiness based on the first drowsiness and the motion information acquired in step 4 above. That is, the correction unit 19 is a function that outputs the second drowsiness when the first drowsiness and motion information are input.

In the second embodiment, the update unit 25 determines the first drowsiness estimated in step 6, the second drowsiness calculated in step 7, the third drowsiness determined to be input in step 9, and the drowsiness obtained in step 4. Based on the obtained operation information, the correction unit 19 is learned.

The content of the learning is to learn so that the second drowsiness calculated by the correction unit 19 approaches the third drowsiness when a situation in which the first drowsiness and the motion information are the same or similar occurs in the future. .

2. Effects of Drowsiness Estimation Device 1 According to the second embodiment described above, the effects (1A) to (1C) of the first embodiment described above are obtained, and the following effects are obtained.

(2A) The correction unit 19 calculates a second drowsiness based on the first drowsiness and the motion information. The update unit 25 learns and updates the correction unit 19 based on the first drowsiness, the second drowsiness, the third drowsiness and the motion information.

Therefore, the drowsiness estimation device 1 can more accurately calculate the second drowsiness. In addition, the drowsiness estimation device 1 can update the correction unit 19 more appropriately based on the motion information.
<Third Embodiment>
1. Differences from First Embodiment Since the basic configuration of the third embodiment is the same as that of the first embodiment, the differences will be described below. Note that the same reference numerals as in the first embodiment indicate the same configurations, and refer to the preceding description.

In the third embodiment, the drowsiness estimation device 1 does not acquire motion information. In the third embodiment, the correction unit 19 calculates the second drowsiness based on the first drowsiness and the driving state information acquired in step 5 above. That is, the correction unit 19 is a function that outputs the second drowsiness when the first drowsiness and driving state information are input.

In the third embodiment, the update unit 25 determines the first drowsiness estimated in step 6, the second drowsiness calculated in step 7, the third drowsiness determined to be input in step 9, and the drowsiness obtained in step 5. Based on the obtained operating state information, the correction unit 19 is learned.

The content of the learning is to learn so that the second drowsiness calculated by the correction unit 19 approaches the third drowsiness when a situation in which the first drowsiness and the driving state information are the same or similar occurs in the future. be.

2. Effects of Drowsiness Estimation Device 1 According to the third embodiment described above, the effects (1A) to (1C) of the first embodiment described above are obtained, and the following effects are obtained.

(3A) The correction unit 19 calculates the second drowsiness based on the first drowsiness and the driving state information. The update unit 25 learns and updates the correction unit 19 based on the first drowsiness, the second drowsiness, the third drowsiness and the driving state information.

Therefore, the drowsiness estimation device 1 can more accurately calculate the second drowsiness. In addition, the drowsiness estimation device 1 can update the correction unit 19 more appropriately based on the driving state information.
<Fourth Embodiment>
1. Differences from First Embodiment Since the basic configuration of the fourth embodiment is the same as that of the first embodiment, the differences will be described below. Note that the same reference numerals as in the first embodiment indicate the same configurations, and refer to the preceding description.

In the fourth embodiment, the drowsiness estimation device 1 does not acquire motion information and driving state information. In the fourth embodiment, the correction unit 19 calculates the second drowsiness based on the first drowsiness. That is, the correction unit 19 is a function that outputs the second drowsiness when the first drowsiness is input.

In the fourth embodiment, the update unit 25 learns the correction unit 19 based on the second drowsiness calculated in step 7 and the third drowsiness determined to be input in step 9 above.

The content of the learning is to learn so that the second drowsiness calculated by the correction unit 19 approaches the third drowsiness when a situation similar to or the same as the first drowsiness occurs in the future.
2. Effects of Drowsiness Estimating Device 1 According to the fourth embodiment described above, the effects (1A) to (1C) of the first embodiment described above are achieved.
<Other embodiments>
Although the embodiments of the present disclosure have been described above, the present disclosure is not limited to the above-described embodiments, and various modifications can be made.

(1) The position of the display unit 39 may be, for example, above the A/C panel, the car navigation system 33, or the like. Moreover, the same display as that of the display unit 39 may be performed by a head-up display. Moreover, you may notify a passenger|crew of the display content of the display part 39 by a sound.

(2) Instead of the input switch 41, a touch panel, voice recognition system, remote touch controller, or the like may be used.
(3) The drowsiness estimation device 1 may acquire motion information by other methods. For example, the drowsiness estimation device 1 may detect the motion of the passenger based on the image of the camera 27 and acquire motion information.

(4) The basic information may be physiological information only. The basic information may be only vehicle information.
(5) The drowsiness estimation device 1 and method thereof described in the present disclosure were provided by configuring a processor and memory programmed to perform one or more functions embodied by a computer program. It may also be implemented by a dedicated computer. Alternatively, the drowsiness estimation device 1 and its technique described in the present disclosure may be implemented by a dedicated computer provided by configuring a processor with one or more dedicated hardware logic circuits. Alternatively, the drowsiness estimation device 1 and its method described in the present disclosure are a processor and memory programmed to perform one or more functions, and a processor configured by one or more hardware logic circuits. It may also be implemented by one or more dedicated computers configured in combination. Computer programs may also be stored as computer-executable instructions on a computer-readable non-transitional tangible storage medium. The method for realizing the function of each part included in the drowsiness estimation device 1 does not necessarily include software, and all the functions may be realized using one or more pieces of hardware.

(6) A plurality of functions possessed by one component in the above embodiment may be realized by a plurality of components, or a function possessed by one component may be realized by a plurality of components. . Also, a plurality of functions possessed by a plurality of components may be realized by a single component, or a function realized by a plurality of components may be realized by a single component. Also, part of the configuration of the above embodiment may be omitted. Moreover, at least part of the configuration of the above embodiment may be added or replaced with respect to the configuration of the other above embodiment.

(7) In addition to the drowsiness estimation device 1 described above, a system having the drowsiness estimation device 1 as a component, a program for causing a computer to function as the drowsiness estimation device 1, and a non-transitional device such as a semiconductor memory recording this program The present disclosure can also be implemented in various forms such as a physical recording medium, wakefulness estimation device, drowsiness estimation method, wakefulness estimation method, and the like.

DESCRIPTION OF SYMBOLS 1... Drowsiness estimation apparatus, 11... Basic information acquisition unit, 13... Motion information acquisition unit, 15... Driving state information acquisition unit, 17... Estimation unit, 19... Correction unit, 21... Notification unit, 23... Third drowsiness acquisition unit , 25... update unit, 39... display unit, 41... input switch

Claims (7)

a basic information acquisition unit (11) configured to acquire basic information including at least one of physiological information representing the physiological condition of an occupant of the vehicle and vehicle information representing the condition of the vehicle;
an estimating unit (17) configured to estimate a first drowsiness measure of the occupant's drowsiness based on the basic information;
a correction unit (19) configured to correct said first drowsiness to calculate a second drowsiness;
a notification unit (21) configured to notify the occupant of the second drowsiness;
a third drowsiness acquisition unit (23) configured to acquire a third drowsiness level of the occupant input by the occupant;
an updating unit (25) configured to learn based on said second drowsiness and said third drowsiness and update said correction unit;
a motion information acquisition unit (13) for acquiring motion information representing motion of the occupant ;
with
the correction unit is configured to calculate the second drowsiness based on the first drowsiness and the motion information;
A drowsiness estimation device (1) , wherein said updating unit is adapted to learn based on said first drowsiness, said second drowsiness, said third drowsiness and said motion information to update said correction unit. a basic information acquisition unit (11) configured to acquire basic information including at least one of physiological information representing the physiological condition of an occupant of the vehicle and vehicle information representing the condition of the vehicle;
an estimating unit (17) configured to estimate a first drowsiness measure of the occupant's drowsiness based on the basic information;
a correction unit (19) configured to correct said first drowsiness to calculate a second drowsiness;
a notification unit (21) configured to notify the occupant of the second drowsiness;
a third drowsiness acquisition unit (23) configured to acquire a third drowsiness level of the occupant input by the occupant;
an updating unit (25) configured to learn based on said second drowsiness and said third drowsiness and update said correction unit;
a driving state information acquisition unit (15) for acquiring driving state information representing the driving state of the vehicle ;
with
the correction unit is configured to calculate the second drowsiness based on the first drowsiness and the driving state information;
The drowsiness estimation device (1) , wherein the updating unit is configured to learn based on the first drowsiness, the second drowsiness, the third drowsiness and the driving state information to update the correction unit. a basic information acquisition unit (11) configured to acquire basic information including at least one of physiological information representing the physiological condition of an occupant of the vehicle and vehicle information representing the condition of the vehicle;
an estimating unit (17) configured to estimate a first drowsiness measure of the occupant's drowsiness based on the basic information;
a correction unit (19) configured to correct said first drowsiness to calculate a second drowsiness;
a notification unit (21) configured to notify the occupant of the second drowsiness;
a third drowsiness acquisition unit (23) configured to acquire a third drowsiness level of the occupant input by the occupant;
an updating unit (25) configured to learn based on said second drowsiness and said third drowsiness and update said correction unit;
a motion information acquisition unit (13) for acquiring motion information representing motion of the occupant;
A driving state acquisition unit (15) for acquiring driving state information representing the driving state of the vehicle;
with
the correction unit is configured to calculate the second drowsiness based on the first drowsiness, the motion information and the driving state information;
The updating unit is configured to learn based on the first drowsiness, the second drowsiness, the third drowsiness, the motion information, and the driving state information, and update the correction unit ( 1). The drowsiness estimation device according to claim 1 or 3,
The motion information is information indicating that the occupant has performed a motion to operate a car navigation system or an in-vehicle device.
Drowsiness estimator. The drowsiness estimation device according to claim 2 or 3,
The driving state information includes the driving duration from the start of driving to the current time, the current time, which route the vehicle is traveling on, the outward route or the return route, the weather in the area where the vehicle is traveling, Alternatively, information representing the steering angle of the vehicle,
Drowsiness estimator. The drowsiness estimation device according to any one of claims 1 to 5,
the basic information includes the physiological information;
The physiological information is an image of the occupant's face,
Drowsiness estimator. The drowsiness estimation device according to any one of claims 1 to 6 ,
The drowsiness estimation device, wherein the updating unit is configured to update the correction unit such that the second drowsiness approaches the third drowsiness based on the result of the learning.

Images