JP6418225B2 - Driver physical condition detection device - Google Patents

Description

  The present invention relates to a driver physical condition detection device that detects a physical condition of a driver driving a vehicle, and belongs to the field of vehicle safety technology.

  As a part of vehicle safety technology, a technology for detecting a driver's physical condition abnormality based on a driver's driving posture or vehicle behavior is known.

  For example, in Patent Document 1, as a technique for detecting a physical condition abnormality from a driving posture of a driver, body pressure distributions of the driver's buttocks and back are detected by a body pressure distribution sensor provided in the driver's seat, and the detection result Based on the above, a technique for determining whether or not the driver's posture is in the proper posture range and detecting the presence or absence of the driver's physical condition is disclosed.

  In Patent Document 2, as a technique for detecting an abnormal condition of the driver based on the behavior of the vehicle, it is determined whether or not the vehicle deviates from the normal traveling area estimated from the traveling position of the host vehicle and the traveling lane. In the case of deviating, a technique for detecting a driver's physical condition abnormality by determining a decrease in the driver's arousal level based on the subsequent steering state is disclosed.

Japanese Patent Laid-Open No. 2015-21912 JP 2011-39735 A

  However, in the technique described in Patent Document 1, a large change appears in the driver's driving posture itself after the driver's physical condition abnormality has progressed to a certain degree or more, and the vehicle according to the technique described in Patent Document 2 The same applies to the case where a driver's physical condition abnormality is detected based on the behavior of the driver.In either case, in order to enable the driver to evacuate to a safe place, the driver's physical condition is detected earlier. It is important to detect an abnormality, and the present invention aims to solve this problem.

  By the way, the applicant of the present application is to perform a posture operation such as stepping on the foot or strengthening the grip of the steering wheel as a preliminary operation against a change in the behavior of the vehicle when turning, sudden acceleration, deceleration, etc. By finding that this preliminary movement is delayed by a slight decrease in consciousness in the initial stage of abnormal physical condition, and by relating this preliminary movement to the behavior of the vehicle as described above, only changes in the behavior of the vehicle due to the physical condition of the driver Or it was thought that abnormal physical condition could be detected earlier than the case where abnormal physical condition was estimated only from the change of driving posture.

  FIG. 11 shows the result of the experiment. The steering angle at the time of turning of the vehicle and the acceleration in the lateral direction of the vehicle (hereinafter referred to as lateral G) in the consciousness clear state when the physical condition is normal and the consciousness lowered state when the physical condition is abnormal. The time variation of the seating center of gravity position of the driver is shown. According to this, in the consciousness clear state, the change in the seating center of gravity position occurs first with respect to the change in the lateral G corresponding to the change in the steering angle, and the driver takes a preliminary action for the approach to the curve. On the other hand, in the state of reduced consciousness, as the lateral G changes, the driver's seating gravity center position changes. That is, the applicant of the present invention, in the state of reduced consciousness, even if the steering itself is normally performed, the preliminary operation for the lateral G associated therewith becomes poor, in other words, the start of the preliminary operation prior to the change in the behavior of the vehicle is delayed. New knowledge was obtained.

  In order to solve the above problems, based on the above-described knowledge, the driver physical condition detection device according to the present invention is configured as follows.

First, the driver physical condition detection device according to the invention of claim 1 is:
A center of gravity position variation detecting means for detecting a change in the seating center of gravity position of the driver;
Determination criterion acquisition means for acquiring determination criterion information for determining whether the driver's physical condition is normal or abnormal based on the variation of the seating gravity center position detected by the gravity center position variation detection means;
The change in the center of gravity position detected by the center-of-gravity position variation detection means is compared with the determination reference information acquired by the determination reference acquisition means, and the start time of the change is compared with the determination reference time included in the determination reference information. physical condition of the driver when the Te lags is closed and determining the physical condition judging means to be abnormal,
The determination criterion acquisition means detects the acceleration of the vehicle as determination criterion information,
The physical condition determination means uses the acceleration change start time detected by the determination reference acquisition means as a determination reference time, and relative to this determination reference time, the start time of the change of the seating center of gravity position detected by the gravity center position change detection means is When the preceding time is less than the predetermined time, it is determined that the driver's physical condition is abnormal,
The preceding time is a time for reaching a predetermined fluctuation ratio between the acceleration change start time and the peak and a time for reaching the predetermined fluctuation ratio between the start time of the center of gravity position change and the peak. It is characterized by the time difference between .

The invention according to claim 2 is the invention according to claim 1 ,
The acceleration detected by the criterion acquisition means is a lateral acceleration of the vehicle.

The invention according to claim 3
A center of gravity position variation detecting means for detecting a change in the seating center of gravity position of the driver;
Determination criterion acquisition means for acquiring determination criterion information for determining whether the driver's physical condition is normal or abnormal based on the variation of the seating gravity center position detected by the gravity center position variation detection means;
The change in the center of gravity position detected by the center-of-gravity position variation detection means is compared with the determination reference information acquired by the determination reference acquisition means, and the start time of the change is compared with the determination reference time included in the determination reference information. Physical condition determining means for determining that the driver's physical condition is abnormal when the driver is late,
The determination criterion acquisition means acquires the curvature of the road ahead of the vehicle as determination criterion information,
The physical condition determination means uses the curvature change start time acquired by the determination reference acquisition means as a determination reference time, and relative to the determination reference time, the start time of the change in the seated center of gravity position detected by the gravity center position change detection means is When the preceding time is less than the predetermined time, it is determined that the driver's physical condition is abnormal ,
The preceding time is a time for reaching a predetermined fluctuation ratio between the curvature change start time and the peak and a time for reaching the predetermined fluctuation ratio between the start time of the center of gravity position change and the peak. It is characterized by the time difference between .

The invention according to claim 4 is the invention according to any one of claims 1 to 3 ,
An acceleration detection means for detecting the acceleration of the vehicle;
The physical condition determination means does not determine whether the driver's physical condition is normal or abnormal when the acceleration detected by the acceleration detection means is a predetermined value or less.

  First, according to the first aspect of the present invention, attention is paid to the fact that the start time of the change in the position of the center of gravity of the driver is delayed from the normal time in the consciousness clear state when the physical condition is abnormal such as the state of consciousness decline. Thus, it is determined that the physical condition of the driver is abnormal when the start time of the change in the seating center of gravity position is delayed in light of a predetermined determination reference time. In this way, by associating the behavior of the vehicle due to the poor physical condition of the driver with the preliminary movement of the driver, it is possible to detect an abnormal condition of the occupant at an earlier stage. Safety can be improved, such as being able to evacuate to a safe place.

Further, the acceleration of the vehicle is used as the above-described determination reference information, and the change start time of acceleration included in the information is set as the determination reference time. Thus, the vehicle acceleration change start time is compared with the start time of the driver's seating center of gravity position change. Thus, the behavior of the vehicle due to the poor physical condition of the driver and the preliminary movement of the driver can be associated with each other as in the first aspect, and the abnormal condition of the occupant can be detected earlier.

  Furthermore, according to the invention described in claim 3, the acceleration detected by the determination reference acquisition means is the lateral G of the vehicle. Usually, in order to ensure the operability of the driver, there is no support member for maintaining the driver's posture in the left-right direction of the vehicle seat, and the change in the seating center of gravity position due to the lateral G is due to the longitudinal acceleration. It is easy to get. That is, it is possible to accurately detect whether the driver's physical condition is normal or abnormal by detecting the lateral variation of the driver's seating gravity center position corresponding to the lateral G of the vehicle.

According to the third aspect of the present invention, information regarding the white line on the road surface is used as the determination reference information, and the time when the curvature changes is determined as the determination reference time. Thereby, the change start time of the curvature is compared with the start time of the change of the seating center of gravity position, and the condition of the driver is abnormal when the time before the start of the change of the seating center of gravity position is less than a predetermined time. Therefore, the behavior of the vehicle due to the poor physical condition of the driver and the preliminary movement of the driver can be associated with each other as in the first aspect, and the abnormal condition of the occupant can be detected earlier. Further, the lateral G which is the criterion information of claim 3 is due to the driver's steering, and the curvature of claim 4 is due to the driver's visual recognition, both of which are aware of the driver's entry into the curve. Therefore, the same effect as in the second aspect can be obtained in the third aspect .

According to the fourth aspect of the present invention, the physical condition determination unit is configured not to determine whether the driver's physical condition is normal or abnormal when the acceleration detected by the acceleration detection unit is a predetermined value or less. Thus, it is possible to prevent erroneous determination at an acceleration of a predetermined value or less at which no preliminary movement occurs.

It is a control system figure of a driver physical condition detection device concerning an embodiment of the present invention. 1 is a schematic configuration diagram of a vehicle according to an embodiment of the present invention. It is an image figure of the detection result of a gravity center position detection means. It is a flowchart which shows the driver | operator's physical condition detection operation | movement which concerns on 1st Embodiment. It is the diagram which normalized the change of the horizontal G at the time of a physical condition, and the fluctuation | variation of the seating gravity center position. It is the diagram which normalized the change of the horizontal G at the time of a physical condition abnormality, and the fluctuation | variation of the seating gravity center position. It is a map which shows the predetermined time as a threshold value with respect to the horizontal G. It is a flowchart which shows the driver | operator's physical condition detection operation | movement which concerns on 2nd Embodiment. It is a flowchart which shows the driver | operator's physical condition detection operation | movement which concerns on a reference example . It is a figure which shows the comparative example of the fluctuation model at the time of normal of the seating gravity center position with respect to the change of the curvature which concerns on a reference example , and the fluctuation data of actual seating gravity center position. It is an example of the experimental value which compared the steering angle change at the time of a driver | operator's consciousness clearing, and a consciousness fall, the change of horizontal G, and a driver | operator's gravity center change with time.

  Hereinafter, an embodiment of a driver physical condition detection device of the present invention will be described.

  First, the configuration of the driver's physical condition detection device 1 according to the present embodiment will be described with reference to FIG. The driver's physical condition detection device 1 includes a seat weight sensor 2, a lateral G sensor 3, a control unit 4, and a notification device 5. The control unit 4 is a seating surface input from the seat weight sensor 2. A seating gravity center position detection unit 41 that detects a change in the seating gravity center position based on the pressure distribution, and a physical condition determination unit 42 that determines whether the driver's physical condition is normal or abnormal.

  As shown in FIG. 2, the seat weight sensor 2 is provided in a seat cushion of the driver's seat and includes a plurality of piezoelectric elements provided on a seating surface with which a driver's buttocks comes into contact. Each piezoelectric element constituting the seat weight sensor 2 detects a driver's seating surface pressure distribution by converting a pressure corresponding to the driving posture of the driver into a voltage signal. The driver's seat pressure distribution detected by the seat weight sensor 2 is output to the control unit 4.

  The seating center of gravity position detection unit 41 of the control unit 4 detects a change in the seating center of gravity position of the driver based on the seating surface pressure distribution detected by the seat weight sensor 2. FIG. 3 shows an example of the driver's seat pressure distribution obtained from the seat weight sensor 2. Then, as will be described later, the physical condition determination unit 42 of the control unit 4 compares the change in the driver's seating gravity center position detected by the seating gravity center position detection unit 41 with the change in the behavior of the vehicle obtained from the lateral G sensor 3. Thus, it is determined whether the driver's physical condition is normal or abnormal. In this embodiment, the lateral G sensor 3 corresponds to the determination criterion acquisition means in claim 1.

  Next, the driver's physical condition detection by the control unit 4 will be described with reference to the flowchart of FIG.

  First, in step S101 of the flowchart of FIG. 4, the seating gravity center position is detected from the driver's seating surface pressure distribution indicated by the signal from the seat weight sensor 2, and the data is stored. In step S102, the lateral G of the vehicle is detected from the signal from the lateral G sensor 3, and the data is stored. Next, in step S103, it is determined whether or not the lateral G is 0.1 or more. If the lateral G is smaller than 0.1, the process returns to step S101, and it is determined in step S103 that the lateral G is 0.1 or more. Steps S101 to S103 are repeated until

  If the lateral G is not less than 0.1 in step S103, the process proceeds to step S104, where the seating barycentric position is detected from the driver seat pressure distribution indicated by the signal from the seat weight sensor 2, and the data is stored. In step S105, the lateral G of the vehicle is detected from the signal from the lateral G sensor 3, and the data is stored. In step S106, it is determined whether or not acquisition of normalization data used for physical condition determination is complete. If acquisition of normalization data is not complete, the process returns to step S104 to acquire normalization data. Steps S104 to S106 are repeated until is completed. Note that acquisition of normalization data at this time indicates that, for example, data from the start of each change in the lateral G and the driver's seating gravity center position to the peak value has been acquired. .

  If the acquisition of normalization data is completed in step S106, the process proceeds to step S107, and normalization processing of the data of the lateral G and the driver's seating gravity center position is performed. For example, the normalization process is a process for making a quantity that cannot be compared as it is with different units such as the lateral G and the seating center of gravity position as dimensionless and can be compared with each other.

  Here, the normalization process will be described with reference to FIGS. 5 and 6. First, as shown in FIG. 5, normalization is performed so that the initial value P of the lateral G when the physical condition is normal is 0 and the peak P ′ is 1. In this example, since the peak P ′ of the lateral G is −3G, the variation of the lateral G is normalized by dividing the value of the lateral G by −3. Similarly, the fluctuation of the seating center of gravity is normalized so that the initial value Q is 0 and the peak Q 'is 1. Then, by comparing the normalized data of the lateral G and the seating center of gravity position, the fluctuation of each value becomes clearer and the difference in fluctuation start time becomes clear. Note that the time difference Δt when the normalized value of the normalized data is 0.5 is regarded as the time preceding the fluctuation start time of the center of gravity position, and the change start time of the seating center of gravity position in FIG. 450ms ahead of the start time. By comparing this preceding time with a predetermined value described later, normality or abnormality of the physical condition is determined.

  Similarly, when the physical condition is abnormal as shown in FIG. 6, the initial value S is also used for the variation of the seating center of gravity so that the initial value R of the lateral G at the time of abnormality is 0 and the peak R ′ is 1. Is normalized so that the peak S ′ is 1, and the variation timing of the lateral G and the seating center of gravity position is calculated from the normalized data. The difference from the change start time of G is Δt≈0, and it can be seen that the change of the seating gravity center position is caused by the change of the lateral G.

  Next, in step S108, a predetermined time for the horizontal G peak value serving as a threshold value is read from the map shown in FIG. 7, and a predetermined value is calculated.

  Here, a map indicating a predetermined time as a threshold for the lateral G in FIG. 7 will be described. First, the threshold map shows a predetermined time as a threshold value of a difference between the change start time of the lateral G and the change start time of the seating center of gravity position prior thereto as a predetermined time corresponding to the peak value of the horizontal G. . Specifically, the lower side of the line X of the predetermined time map as the threshold indicates lower consciousness (abnormal physical condition), and the upper side of the threshold line X indicates consciousness clearness (normal physical condition). For example, when 0.1 G, the physical condition is normal when the difference between the variation start time of the seating center of gravity position and the lateral G change start time is 0.05 s or more, and when it is less than 0.05 s, it is determined that the physical condition is abnormal. Is done. Note that the movement of the center of gravity position as a preliminary operation does not appear clearly when the lateral G is small, and may be erroneously determined in the physical condition abnormality determination. Therefore, when the lateral G is smaller than 0.1 G No physical condition is determined.

  In addition, when the lateral G is large, the threshold value of the difference between the variation start time of the seating center of gravity position and the lateral G change start time according to the lateral G is set so that the preliminary operation time is advanced. Compared with the case where the threshold is set to a constant value with respect to the lateral G, it is possible to accurately determine whether the physical condition is normal or abnormal. For example, as shown by a broken line in FIG. 7, regardless of the lateral G, when the threshold is fixed at 0.07 s, a region of 0.07 s or more is set as normal, and a region of less than 0.07 s is set as abnormal, The point x in FIG. 7 is determined to be normal because the difference between the change start time of the seating center of gravity position and the change start time of the lateral G is 0.09 s when the lateral G is 0.2 G. Since this point x is determined to be abnormal in physical condition when the threshold value X corresponding to the lateral G is used, an erroneous determination occurs when the threshold value is constant as described above.

  Returning to the flowchart of FIG. 4, it is determined in step S109 whether the preceding time with respect to the lateral G change start time of the change start time of the seating gravity center position of the driver is larger than the predetermined time of the threshold value in step S108. If larger, the process proceeds to step S110, it is determined that the physical condition is normal, and the flow ends. On the other hand, when the time is smaller than the predetermined time of the threshold value in step S108, it indicates that the preliminary operation is delayed. Therefore, the process proceeds to step S111, and it is determined that the physical condition is abnormal. finish.

  Next, a second embodiment of detecting the physical condition of the driver by the control unit 4 will be described with reference to the flowchart of FIG. In the second embodiment, in addition to the configuration of the first embodiment, a front camera 6 indicated by a virtual line in FIGS. 1 and 2 is provided. The front camera 6 has a function of acquiring an image ahead of the vehicle, and outputs the acquired image to the control unit 4.

  First, in step S201 of the flowchart of FIG. 8, the seating barycentric position is detected from the driver's seating surface pressure distribution indicated by the signal from the seat weight sensor 2, and the data is stored. In step S202, the lateral G of the vehicle is detected from the signal from the lateral G sensor 3, and the data is stored. Next, in step S203, it is determined whether or not the lateral G is 0.1 or more. If the lateral G is smaller than 0.1, the process returns to step S201 to determine that the lateral G in step S203 is 0.1. Steps S201 to S203 are repeated until.

  When the lateral G is 0.1 or more in step S203, the process proceeds to step S204, where the seating barycentric position is detected from the driver seat pressure distribution indicated by the signal from the seat weight sensor 2, and the data is stored. In step S205, the lateral G of the vehicle is detected based on the signal from the lateral G sensor 3, and the data is stored. In step S206, the front camera 6 is used to detect the white line on the road ahead of the vehicle, and in step S207, it is determined whether or not the acquisition of normalization data is complete, and the acquisition of normalization data is complete. If not, the process returns to step S204, and steps S204 to S207 are repeated until the acquisition of normalization data is completed.

  When the acquisition of the normalization data is completed in step S207, the process proceeds to step S208, and the variation of the lateral G and the driver's seating gravity center position is normalized as in the first embodiment, and in step S209, the step is performed. The curvature of the travel path is calculated from the white line of the travel path ahead of the vehicle detected in S206, and the process proceeds to step S210.

  In step 210, based on the curvature calculated in step S19, a predetermined time corresponding to the curvature of the traveling road is read from the map as a threshold, and a predetermined value is calculated. Here, the horizontal axis of the threshold value map of FIG. 7 is the curvature, and the vertical axis takes a predetermined value corresponding to the curvature to obtain the same relationship as in the first embodiment. Specifically, in order to indicate that the curve is steeper as the curvature (1 / curvature radius) increases, a relationship corresponding to the large value of the lateral G in the sharp curve is obtained, and the curvature is in accordance with the curvature. The threshold map can be used to accurately determine whether the physical condition is normal or abnormal.

  In step S211, it is determined whether or not the preceding time for the change start time of the lateral G of the change start time of the seating center of gravity position of the driver is larger than the predetermined time of the threshold value in step S210. It is determined that the physical condition is normal, and the flow ends. On the other hand, when the time is smaller than the predetermined time of the threshold value in step S210, it indicates that the preliminary operation is delayed. Therefore, the process proceeds to step S213, where it is determined that the physical condition is abnormal, and in step S214, the physical condition abnormality is notified to the driver. finish.

Next, a reference example of the driver's physical condition detection by the control unit 4 will be described with reference to the flowchart of FIG. In addition, in the reference example , in addition to the configuration of the first embodiment, the front camera 6 shown in phantom lines in FIGS. 1 and 2 is provided, and the control unit 4 further determines the seating gravity center position for each curvature when the physical condition is normal. As variation data, a seating gravity center position variation model 43 is provided.

  First, in step S301 of the flowchart of FIG. 9, the seating barycentric position is detected from the seating surface pressure distribution of the driver indicated by the signal from the seat weight sensor 2, and the data is stored. In step S302, the lateral G of the vehicle is detected based on the signal from the lateral G sensor 3, and the data is stored. Next, in step S303, it is determined whether or not the lateral G is 0.1 or more. If the lateral G is smaller than 0.1, the process returns to step S301, and steps S301 to S303 are repeated until it is determined that the lateral G in step S303 is 0.1 or more.

  If the lateral G is 0.1 or more in step S303, the process proceeds to step S304, where the seat weight sensor 2 detects the seating gravity center position based on the driver's seating surface pressure distribution, and stores the data. In step S305, the lateral G of the vehicle is detected from the signal from the lateral G sensor 3, and the data is stored. In step S306, the front camera 6 is used to detect white line information on the road ahead of the vehicle, and the data is stored. In step S307, it is determined whether acquisition of physical condition determination data is completed. If acquisition of physical condition determination data is not completed, the process returns to step S304, and step S304 is performed until acquisition of physical condition determination data is completed. Step S306 is repeated. The completion of acquisition of the physical condition determination data at this time indicates that, for example, the peak value data of the fluctuation of the driver's seating gravity center position has been acquired.

  If the acquisition of physical condition determination data is completed in step S307, the process proceeds to step S308, the curvature of the travel path is calculated from the white line on the front travel path of the vehicle detected in step S305, and the process proceeds to step S309.

  In step S309, the normal sitting center of gravity position variation model a according to the curvature as shown in FIG. The model a includes the change point t0 of the lateral G as a reference point, and the actually measured seating barycentric position data b is also associated with the change of the lateral G. In step S310, the lateral G change time t0 of the variation model a of the seating centroid position acquired in step S309 coincides with the lateral G change time t0 of the measured seating centroid position data b detected in step S304. The two are compared to determine whether the physical condition is normal or abnormal. This comparison method calculates a time difference Δt between the normal seating centroid position variation model a or the movement amount p ′ of ½ of the peak value of the actually measured seating centroid position data b, and the time difference Δt is calculated as follows. An abnormality is determined when the value is equal to or greater than a predetermined value. The predetermined value has a different value for each model.

  If the time difference Δt is smaller than the predetermined value in step S310 according to the determination method described above, it is determined in step S311 that the physical condition is normal, and the flow ends. On the other hand, when the time difference Δt is equal to or greater than the predetermined value, it indicates that the preliminary operation is delayed. Therefore, it is determined in step S312 that the physical condition is abnormal, and in step S313, the physical condition is notified to the driver, and the flow is terminated.

  In the present embodiment, the sitting centroid position variation model 43 is the variation data of the sitting centroid position for each curvature when the physical condition is normal, but the variation data of the sitting centroid position for each acceleration when the physical condition is normal may be used.

  Further, in the present embodiment, the curvature of the forward travel path of the vehicle is detected by the white line imaged by the front camera, but map information of a navigation system mounted on the vehicle may be used.

  As described above, according to the present invention, abnormalities in the physical condition of the driver are detected at an early stage. Therefore, the present invention can be suitably used in the field of vehicle safety technology.

DESCRIPTION OF SYMBOLS 1 Driver physical condition detection apparatus 2 Seat weight sensor 3 Lateral G sensor (determination reference acquisition means)
4 Control Unit 5 Notification Device 6 Front Camera 41 Seating Center of Gravity Position Detection Unit 42 Physical Condition Determination Unit 43 Seating Center of Gravity Position Variation Model

Claims (5)

A center of gravity position variation detecting means for detecting a change in the seating center of gravity position of the driver;
Determination criterion acquisition means for acquiring determination criterion information for determining whether the driver's physical condition is normal or abnormal based on the variation of the seating gravity center position detected by the gravity center position variation detection means;
The change in the center of gravity position detected by the center-of-gravity position variation detection means is compared with the determination reference information acquired by the determination reference acquisition means, and the start time of the change is compared with the determination reference time included in the determination reference information. physical condition of the driver when the Te lags is closed and determining the physical condition judging means to be abnormal,
The determination criterion acquisition means detects the acceleration of the vehicle as determination criterion information,
The physical condition determination means uses the acceleration change start time detected by the determination reference acquisition means as a determination reference time, and relative to this determination reference time, the start time of the change of the seating center of gravity position detected by the gravity center position change detection means is When the preceding time is less than the predetermined time, it is determined that the driver's physical condition is abnormal,
The preceding time is a time for reaching a predetermined fluctuation ratio between the acceleration change start time and the peak and a time for reaching the predetermined fluctuation ratio between the start time of the center of gravity position change and the peak. A driver's physical condition detecting device characterized in that the time difference is between . The driver's physical condition detection device according to claim 1 , wherein the acceleration detected by the determination criterion acquisition unit is a lateral acceleration of the vehicle. A center of gravity position variation detecting means for detecting a change in the seating center of gravity position of the driver;
Determination criterion acquisition means for acquiring determination criterion information for determining whether the driver's physical condition is normal or abnormal based on the variation of the seating gravity center position detected by the gravity center position variation detection means;
The change in the center of gravity position detected by the center-of-gravity position variation detection means is compared with the determination reference information acquired by the determination reference acquisition means, and the start time of the change is compared with the determination reference time included in the determination reference information. Physical condition determining means for determining that the driver's physical condition is abnormal when the driver is late,
The determination criterion acquisition means acquires the curvature of the road ahead of the vehicle as determination criterion information,
The physical condition determination means uses the curvature change start time acquired by the determination reference acquisition means as a determination reference time, and relative to the determination reference time, the start time of the change in the seated center of gravity position detected by the gravity center position change detection means is When the preceding time is less than the predetermined time, it is determined that the driver's physical condition is abnormal ,
The preceding time is a time for reaching a predetermined fluctuation ratio between the curvature change start time and the peak and a time for reaching the predetermined fluctuation ratio between the start time of the center of gravity position change and the peak. The driver's physical condition detection device according to claim 1, wherein a time difference between the two is detected. 4. The driver's physical condition detection device according to claim 1, wherein the predetermined time is set to a larger value as the amount of change in acceleration or the curvature is larger . 5. An acceleration detection means for detecting the acceleration of the vehicle;
The physical condition judging means, the acceleration is acceleration detected by the detection means when the predetermined value or less, claim 1, characterized in that not performed normally, the abnormality determination of the physical condition of a driver of claim 4 The driver's physical condition detection device according to item 1.

Images