JP2015032005A - Driver's condition determination device, vehicle controller, and driver's condition determination method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a driver's condition determination device capable of determining with high accuracy whether a driver is recognizing dangers.SOLUTION: A driver's condition determination device 10 comprises: a cardiotachometer 2 for measuring heart rate of a vehicle driver; an eyeblink detector 3 for detecting the driver's eyeblink; and an ECU 4. The ECU 4 includes: a heart rate's continuous rise detection part 6 for detecting whether the driver's heart rate rises continuously on the basis of a measured value by the cardiotachometer 2; an eyeblink interval detection part 7 for detecting whether the driver's eyeblink intervals are longer than necessary on the basis of a detection result by the eyeblink detector 3; and a danger recognition determination part 8 for determining whether the driver is recognizing dangers on the basis of detection results by the heart rate's continuous rise detection part 6 and the eyeblink interval detection part 7.

Description

  The present invention relates to a driver state determination device and method for determining the state of a driver of a vehicle, and a vehicle control device that controls a vehicle driven by the driver.

  As a conventional driver state determination device, for example, as described in Patent Document 1, a driver's pupil is photographed by a CCD camera, and a pupil region is extracted from an image of the CCD camera by image processing. It is known to calculate the diameter and estimate whether the driver recognizes a warning such as a danger based on the change in the pupil diameter.

Japanese Patent Laid-Open No. 11-245683

  However, the following problems exist in the prior art. That is, the extraction accuracy of the pupil contour in image processing can be greatly affected by the brightness of the surrounding environment. Further, when the driver closes his eyes or blinks frequently in front of danger, it may be difficult to determine whether or not the pupil diameter has changed. Therefore, in the method of detecting the change in the pupil diameter of the driver, it is not possible to determine with high accuracy whether or not the driver recognizes the alarm (sense of danger).

  An object of the present invention is to provide a driver state determination device, a vehicle control device, and a driver state determination method that can determine with high accuracy whether or not the driver recognizes a sense of danger.

  The present invention relates to a driver state determination device that determines the state of a driver of a vehicle, and includes a blink detection unit that detects a blink of the driver's eyes, and a driver's blink based on the driver's blinks detected by the blink detection unit. It is characterized by comprising a danger determination means for determining a duration of the eye open state and determining whether or not the driver recognizes the danger based on the duration of the eye open state of the driver. .

  Thus, in the driver state determination device of the present invention, the driver's eye blink is detected, the duration of the driver's eye open state is obtained based on the driver's eye blink, and the driver is based on the duration of the eye open state. It is determined whether or not is in a state of recognizing a sense of danger. For example, when the duration time of the driver's eye open state is longer than a predetermined time, it is determined that the driver recognizes a sense of danger. At this time, since the eyelid size is larger than the pupil, detection of the driver's eye open state due to the brightness of the surrounding environment, eye opening when the driver closes his eyes or blinks frequently in front of danger・ Easy detection of closed eyes. As a result, it is possible to determine with high accuracy whether or not the driver recognizes a sense of danger.

  Preferably, the danger determination means calculates the driver's eye open state reference value based on the driver's past open eye state duration, and compares the driver's current open eye state duration with the eye open state reference value. Then, it is determined whether or not the driver recognizes the danger. The duration of the open eye state in the blink of an eye varies greatly among individuals. Therefore, by calculating the driver's eye open state reference value based on the driver's past eye open state duration, and comparing the driver's current eye open state duration with the eye open state reference value, Regardless of the individual difference, it is possible to determine with high accuracy whether or not the driver recognizes the danger.

  At this time, it is preferable that the danger determination unit suppresses the execution of the calculation of the eye open state reference value when it is determined that the driver has recognized the danger last time. In this case, as the eye-opening state reference value, almost normal data in which the driver hardly recognizes the danger is obtained, so whether or not the driver recognizes the danger is determined. It can be determined with higher accuracy.

  Further, the vehicle stop detection means for detecting whether or not the vehicle is stopped is provided, and the danger determination means detects the eye open state reference value when the vehicle stop detection means detects that the vehicle is stopped. It is preferable to suppress the execution of the calculation and determine that the driver does not recognize the danger. When the vehicle stops, the duration of the eye-open state in the blink of an eye is likely to change due to various emotions caused by various actions of the driver. Therefore, by suppressing the execution of the calculation of the eye-opening state reference value when the stop of the vehicle is detected, the influence on the determination of whether or not the driver is aware of the danger when driving the vehicle Can be reduced.

  Preferably, the apparatus further comprises a consciousness detection means for detecting whether or not the driver's consciousness is lowered, and when the consciousness reduction detection means detects that the driver's consciousness is not lowered, the danger determination means. To implement. When the driver's awareness is reduced, the driver's eyes blink less and the duration of the driver's eye-closed state becomes longer. Therefore, it is preferable to implement the danger determination means when it is detected that the driver's awareness is not lowered.

  The present invention is also a vehicle control device for controlling a vehicle driven by a driver, the blink detecting means for detecting the blink of the driver's eyes, and the driver based on the blink of the eyes of the driver detected by the blink detecting means. The determination result by the danger determination means and the determination result by the danger determination means for determining whether the driver is aware of the danger based on the duration of the eye open state of the driver And a support means for supporting control of the vehicle in response.

  In such a vehicle control device of the present invention, the driver's eye blink is detected, the duration of the driver's eye open state is obtained based on the driver's eye blink, and the driver is determined based on the duration of the eye open state. It is determined whether or not the state is aware of danger. For example, when the duration time of the driver's eye open state is longer than a predetermined time, it is determined that the driver recognizes a sense of danger. At this time, since the eyelid size is larger than the pupil, detection of the driver's eye open state due to the brightness of the surrounding environment, eye opening when the driver closes his eyes or blinks frequently in front of danger・ Easy detection of closed eyes. As a result, it is possible to determine with high accuracy whether or not the driver recognizes a sense of danger. Then, by providing vehicle control support according to the determination result, it is possible to prevent the driver from feeling uncomfortable.

  Furthermore, the present invention is a driver state determination method for determining a state of a driver of a vehicle, the step of detecting blinking of the driver's eyes, and obtaining a duration of the driver's eye open state based on the blinking of the driver's eyes. And determining whether or not the driver recognizes a sense of danger based on the duration of the eye open state of the driver.

  As described above, in the driver state determination method of the present invention, for example, when the duration of the eye open state of the driver is longer than a predetermined time, it is determined that the driver recognizes a sense of danger. As a result, as described above, it can be determined with high accuracy whether or not the driver recognizes a sense of danger.

  According to the present invention, it is possible to determine with high accuracy whether or not the driver recognizes a sense of danger. This can prevent the driver from feeling uncomfortable when performing vehicle control support.

It is a block diagram which shows the vehicle control apparatus provided with 1st Embodiment of the driver state determination apparatus which concerns on this invention. It is a flowchart which shows the detail of the process sequence performed by the heart rate continuation rise detection part shown in FIG. It is a flowchart which shows the detail of the process sequence performed by the long blink interval detection part shown in FIG. It is a flowchart which shows the detail of the process sequence performed by the danger determination part shown in FIG. It is a block diagram which shows the vehicle control apparatus provided with 2nd Embodiment of the driver state determination apparatus which concerns on this invention. It is a flowchart which shows the detail of the process sequence performed by the heart rate continuation rise detection part shown in FIG. It is a flowchart which shows the detail of the process sequence performed by the long blink interval detection part shown in FIG. It is a flowchart which shows the detail of the modification of the process sequence performed by the long blink interval detection part as 3rd Embodiment of the driver state determination apparatus which concerns on this invention. It is a graph which shows an example of the recognition judgment result of a driver | operator's danger when the process shown in FIG. 8 is implemented. It is a block diagram which shows the vehicle control apparatus provided with 4th Embodiment of the driver state determination apparatus which concerns on this invention. It is a flowchart which shows the detail of the process sequence performed by the heart rate continuation rise detection part shown in FIG. It is a flowchart which shows the detail of the process sequence performed by the long blink interval detection part shown in FIG. It is a block diagram which shows the vehicle control apparatus provided with 5th Embodiment of the driver state determination apparatus which concerns on this invention. It is a flowchart which shows the detail of the process sequence performed by the avoidance action detection part shown in FIG. It is a block diagram which shows the vehicle control apparatus provided with 6th Embodiment of the driver state determination apparatus which concerns on this invention. It is a flowchart which shows the detail of the process sequence performed by the danger magnitude estimation part shown in FIG. It is a block diagram which shows the vehicle control apparatus provided with 7th Embodiment of the driver state determination apparatus which concerns on this invention. It is a flowchart which shows the detail of the process sequence performed by the heart rate continuation rise detection part shown in FIG. It is a flowchart which shows the detail of the process sequence performed by the long blink interval detection part shown in FIG.

  Hereinafter, preferred embodiments of a driver state determination device, a vehicle control device, and a driver state determination method will be described in detail with reference to the drawings.

  FIG. 1 is a configuration diagram illustrating a vehicle control device including a first embodiment of a driver state determination device according to the present invention. In the figure, a vehicle control device 1 of the present embodiment is a device that performs driving support control of a vehicle.

  The vehicle control device 1 includes a heart rate measuring device 2, a blink detector 3, an ECU (Electronic Control Unit) 4, and a driving support unit 5.

  The heart rate measuring device 2 is a device that measures the heart rate of the driver of the vehicle, and is composed of, for example, electrodes provided on the steering. Note that the pulse rate may be measured instead of the heart rate. The blink detector 3 is a device (blink detection means) that detects blinks (blinks) of the driver's eyes, and is configured by a camera, for example. The driving support unit 5 is an accelerator, a brake, and a steering.

  The ECU 4 includes a CPU, a memory such as a ROM and a RAM, an input / output circuit, and the like. The ECU 4 includes a heartbeat continuation rise detection unit 6, a long blink interval detection unit 7, a danger determination unit 8, and a driving support control unit 9.

  The heart rate continuation rise detection unit 6 detects whether or not the driver's heart rate continues to rise based on the measurement value of the heart rate measuring device 2. The long blink interval detection unit 7 detects whether the blink interval of the driver is longer than necessary based on the detection result of the blink detector 3. The blink interval is a time interval between blinks. The danger determination unit 8 determines whether or not the driver recognizes the danger based on the detection result of the heartbeat continuation rise detection unit 6 and the detection result of the long blink interval detection unit 7. The driving support control unit 9 controls the driving support unit 5 according to the determination result of the danger determination unit 8.

  FIG. 2 is a flowchart showing details of the processing procedure executed by the heart rate continuation rise detection unit 6. In FIG. 2, first, the latest heart rate HR (n) measured by the heart rate meter 2 is larger than the heart rate HR (n−1) previously measured by the heart rate meter 2, that is, HR ( It is determined whether n)> HR (n−1) is satisfied (step S101).

  When HR (n)> HR (n-1) is satisfied, an elapsed time H_pt after HR (n)> HR (n-1) is satisfied is measured (step S102). Subsequently, it is determined whether the elapsed time H_pt is longer than a preset threshold time Tth (step S103). When the elapsed time H_pt is longer than the threshold time Tth, the heart rate increase flag is set to 1 (procedure S104), and when the elapsed time H_pt is not longer than the threshold time Tth, the heart rate increase flag is set to 0 (procedure). S105).

  FIG. 3 is a flowchart showing details of a processing procedure executed by the long blink interval detection unit 7. In FIG. 3, first, the elapsed time B_pt of the eye open state from the immediately preceding blink timing detected by the blink detector 3 is measured (step S111). The open eye state refers to a state where the distance between the upper and lower eyelids is a predetermined value or more. The predetermined value here is a value larger than the pupil diameter.

  Subsequently, it is determined whether or not the elapsed time B_pt in the open eye state is longer than a preset threshold time Bth (step S112). When the elapsed time B_pt in the open eye state is longer than the threshold time Bth, the long blink interval flag is set to 1 (step S113), and when the elapsed time B_pt in the open eye state is not longer than the threshold time Bth, The eye interval flag is set to 0 (step S114).

  FIG. 4 is a flowchart showing details of a processing procedure executed by the danger determination unit 8. In FIG. 4, it is first determined whether or not the heartbeat increase flag is set to 1 in the heartbeat continuation rise detection unit 6 or whether the long blink interval flag is set to 1 in the long blink interval detection unit 7 (procedure 121). .

  When the heart rate raising flag is set to 1 or when the long blink interval flag is set to 1, it is determined that the driver is in a state of perceiving danger (step S122). When the heart rate raising flag is set to 0 and the long blink interval flag is set to 0, it is determined that the driver is not aware of the danger (step S123).

  Returning to FIG. 1, the driving support control unit 9, although not particularly illustrated, avoids a vehicle collision when the danger determination unit 8 determines that the driver recognizes the danger. When the driving support unit 5 is controlled and it is determined by the danger determination unit 8 that the driver does not recognize the danger, the driving support unit 5 is not controlled.

  In the above, the heart rate measuring device 2, the blink detector 3, the heartbeat continuation rise detection unit 6, the long blink interval detection unit 7 and the danger determination unit 8 of the ECU 4 constitute the driver state determination device 10. ing. At this time, the long blink interval detection unit 7 and the danger determination unit 8 obtain the duration of the eye open state of the driver based on the blink of the eyes of the driver detected by the blink detection unit 3, and continue the eye open state of the driver. A danger determination means for determining whether or not the driver recognizes the danger based on the time is configured. The driving support control unit 9 and the driving support unit 5 of the ECU 4 constitute support means for supporting control of the vehicle according to the determination result by the danger determination means.

  As described above, in the present embodiment, the blink detector 3 detects the blink (blink) of the driver's eyes, and the elapsed time B_pt of the eye open state from the immediately preceding blink timing is longer than the threshold time Bth. When the driver is long, that is, when the driver keeps his eyes wide open for a long time, it is determined that the driver recognizes the danger and the driving support unit 5 is controlled. At this time, the elapsed time B_pt of the eye-opening state is measured without being greatly affected by the lightness and darkness of the surrounding environment due to the fact that the size of the eyelid is larger than the pupil and that the three-dimensional boundary is clear. It becomes possible. Thereby, it can be determined with high accuracy whether or not the driver recognizes a sense of danger. As a result, it is not necessary to give the driver a sense of incongruity when carrying out vehicle driving control support.

  FIG. 5 is a configuration diagram illustrating a vehicle control device including a second embodiment of the driver state determination device according to the present invention. In the figure, the same or equivalent elements as those in the first embodiment are denoted by the same reference numerals, and the description thereof is omitted.

  In the figure, the ECU 4 of the vehicle control device 1 according to the present embodiment includes a blink interval in addition to the heart rate continuation rise detection unit 6, the long blink interval detection unit 7, the danger determination unit 8, and the driving support control unit 9. A calculation unit 11 is included. The blink interval calculation unit 11 calculates the above-described blink interval of the driver based on the detection result of the blink detector 3. The long blink interval detection unit 7 detects whether the blink interval of the driver is longer than necessary based on the detection result of the blink detector 3 and the calculation result of the blink interval calculation unit 11.

  FIG. 6 is a flowchart showing details of a processing procedure executed by the heart rate continuation rise detection unit 6. In addition, the same code | symbol is attached | subjected to the same procedure as the procedure shown in FIG.

  In FIG. 6, first, based on the heart rate HR measured by the heart rate measuring device 2, the heart rate moving average HR_ave from the latest heart rate HR to the heart rate HR a predetermined time (T seconds) ago is calculated (procedure). S107). Subsequently, it is determined whether or not the latest heart rate HR measured by the heart rate measuring device 2 is larger than the heart rate moving average HR_ave, that is, whether HR> HR_ave is satisfied (step S108).

  When HR> HR_ave is not established, the process returns to step S107. When HR> HR_ave is satisfied, an elapsed time H_pt after HR> HR_ave is satisfied is measured (step S109). Then, the same steps S103 to S105 as the process shown in FIG. 2 are executed.

  FIG. 7 is a flowchart showing details of a processing procedure executed by the long blink interval detecting unit 7. In addition, the same code | symbol is attached | subjected to the same procedure as the procedure shown in FIG.

  In FIG. 7, in parallel with step S111 shown in FIG. 3, based on the driver's blink interval Bint calculated by the blink interval calculation unit 11, a certain number of times (N times) before the latest blink interval Bint. The blink interval moving average Bint_ave up to the blink interval Bint is calculated (step S116).

  Subsequently, based on the blink interval Bint of the driver calculated by the blink interval calculation unit 11, a blink interval variation index Bint_v representing the magnitude of variation of the blink interval Bint is calculated (step S117). The blink interval variation index Bint_v is expressed as, for example, n × standard deviation σ.

  Subsequently, it is determined whether or not the elapsed time B_pt in the open eye state is longer than the value obtained by adding the blink interval variation index Bint_v to the blink interval moving average Bint_ave (open eye state reference value), that is, whether B_pt> Bint_ave + Bint_v is satisfied. (Procedure S118). When B_pt> Bint_ave + Bint_v is satisfied, the long blink interval flag is set to 1 (step S113), and when B_pt> Bint_ave + Bint_v is not satisfied, the long blink interval flag is set to 0 (step S114).

  In the above, the heart rate measuring device 2, the blink detector 3, the heartbeat continuation rise detection unit 6, the blink interval calculation unit 11, the long blink interval detection unit 7 and the danger determination unit 8 of the ECU 4 are drivers. The state determination device 10 is configured. At this time, the blink interval calculation unit 11, the long blink interval detection unit 7, and the danger determination unit 8 obtain the duration of the driver's eye open state based on the blink of the driver's eyes detected by the blink detection unit 3. The risk determination means for determining whether or not the driver recognizes the danger based on the duration of the eye open state of the driver is configured.

  As described above, in the present embodiment, the driver is in danger by comparing the elapsed time B_pt of the eye open state from the previous blink timing with the value obtained by adding the blink interval variation index Bint_v to the blink interval moving average Bint_ave. It is determined whether or not the feeling is recognized. In other words, since the elapsed time B_pt of the eye-opening state is compared with an appropriate value according to the individual state of the driver, is the state in which the driver recognizes a sense of danger despite a large individual difference in the blink interval? Whether or not can be determined with high accuracy.

  FIG. 8 is a flowchart showing details of a modification of the processing procedure executed by the long blink interval detection unit 7 as the third embodiment of the driver state determination device according to the present invention. In addition, the same code | symbol is attached | subjected to the same procedure as the procedure shown in FIG.

  In FIG. 8, in parallel with the procedure S111 shown in FIG. 2, it is determined whether or not the previously set long blink interval flag is 0 (procedure S119). When the long blink interval flag set last time is 0, the above steps S116 and S117 are sequentially executed. On the other hand, when the long blink interval flag set last time is 1 instead of 0, the above steps S116 and S117 are not executed.

  By performing such processing, the comparison target of the elapsed time B_pt in the open eye state is limited to the data of the normal blink interval of the driver. As a result, it can be determined with higher accuracy whether or not the driver recognizes the danger.

FIG. 9 is a graph showing an example of a driver's risk recognition determination result when the above processing is performed. In FIG. 9 (a), the solid line _{P 1} indicates the heart rate of the driver, the broken line _{Q 1} is shows the heart rate moving average HR_ave. The time interval T of the heart rate moving average HR_ave is 60 seconds, and the threshold time Tth is 10 seconds. In FIG. 9 (b), the solid line P ₂ shows a blink interval drivers, dashed Q ₂ is shows the blink interval moving average Bint_ave + blink interval distribution index Bint_v. The number N of blink interval moving average Bint_ave is 50, and the blink interval variation index Bint_v is 3σ.

In FIG. 9, an area X is a time zone in which the driver has traveled while paying attention to the bicycle, and an area Y is a time zone in which the driver has traveled at a point where the driver has poor visibility. In the vicinity of the regions X and Y, the heart rate increase flag F ₁ and the long blink interval flag F ₂ are both 1, and it is determined that the driver recognizes a sense of danger.

  In the third embodiment, when the previously set long blink interval flag is 1, the above steps S116 and S117 are not executed, that is, the blink interval moving average Bint_ave and the blink interval variation index Bint_v Although the calculation is not performed, the present invention is not limited to this, and it is also possible to reduce only the usage amount of the blink interval moving average Bint_ave and the blink interval variation index Bint_v.

  FIG. 10 is a configuration diagram illustrating a vehicle control device including a fourth embodiment of the driver state determination device according to the present invention. In the figure, elements that are the same as or equivalent to those in the second embodiment are given the same reference numerals, and descriptions thereof are omitted.

  In the figure, the vehicle control device 1 of this embodiment further includes a vehicle speed sensor 12 for measuring the vehicle speed of the vehicle. Further, the ECU 4 of the vehicle control device 1 includes the heartbeat continuation rise detection unit 6, the blink interval calculation unit 11, the long blink interval detection unit 7, the danger determination unit 8, and the driving support control unit 9. .

  FIG. 11 is a flowchart showing details of a processing procedure executed by the heart rate continuation rise detection unit 6. In addition, the same code | symbol is attached | subjected to the same procedure as the procedure shown in FIG.

  In FIG. 11, first, based on the vehicle speed measured by the vehicle speed sensor 12, it is determined whether or not the vehicle is stopped (step S131). When the vehicle is stopped, the elapsed time H_pt is set to 0 (step S132), and the process proceeds to the above step S103. When the vehicle is not stopped, the above steps S107 to S109 are sequentially executed, and the process proceeds to step S103. With this process, the heart rate raising flag is set to 0 when the vehicle is stopped.

  FIG. 12 is a flowchart showing details of a processing procedure executed by the long blink interval detection unit 7. In addition, the same code | symbol is attached | subjected to the same procedure as the procedure shown in FIG.

  In FIG. 12, first, based on the vehicle speed measured by the vehicle speed sensor 12, it is determined whether or not the vehicle is stopped (step S133). When the vehicle is stopped, the above steps S116 and S117 are not executed. On the other hand, when the vehicle is not stopped, the above steps S116 and S117 are sequentially executed.

  Further, in parallel with the procedure S133, it is determined whether the vehicle is stopped as in the procedure S133 (procedure S134). When the vehicle is stopped, the elapsed time B_pt in the open eye state is set to 0 (procedure S135), and the process proceeds to the above-described procedure S118. On the other hand, when the vehicle is not stopped, the above-described procedure S111 is executed, and the procedure proceeds to procedure S118. With this process, the long blink interval flag is set to 0 when the vehicle is stopped.

  In the above, the heart rate measuring device 2, the blink detector 3, the vehicle speed sensor 12, the heart rate continuation rise detection unit 6, the blink interval calculation unit 11, the long blink interval detection unit 7 and the danger determination unit of the ECU 4 8 constitutes the driver state determination device 10. At this time, the vehicle speed sensor 12 constitutes vehicle stop detection means for detecting whether or not the vehicle is stopped.

  When the vehicle is stopped, the biological signal changes due to various emotions caused by various driver actions, and thus erroneous determination of the driver's perception of danger is likely to occur. However, in the present embodiment, when the vehicle is stopped, the heart rate increase flag and the long blink interval flag are 0, so that it is determined that the driver is not aware of danger. . In addition, when the vehicle is stopped, the calculation of the heart rate moving average HR_ave and the blink interval variation index Bint_v is not performed. It is possible to reduce the influence on the determination of the recognition of the driver's danger at the time. Thereby, it can be determined with higher accuracy whether or not the driver recognizes the danger.

  In the fourth embodiment, when the vehicle is stopped, the steps S116 and S117 are not executed, that is, the blink interval moving average Bint_ave and the blink interval variation index Bint_v are not calculated. However, the present invention is not limited to this, and it is also possible to reduce only the usage amount of the blink interval moving average Bint_ave and the blink interval variation index Bint_v.

  FIG. 13: is a block diagram which shows the vehicle control apparatus provided with 5th Embodiment of the driver state determination apparatus based on this invention. In the figure, elements that are the same as or equivalent to those in the second embodiment are given the same reference numerals, and descriptions thereof are omitted.

  In the figure, the vehicle control device 1 of the present embodiment further includes the vehicle speed sensor 12, the steering angle sensor 13, the acceleration sensor 14, and the brake sensor 15. The steering angle sensor 13 is a sensor that measures the steering angle of the steering of the vehicle. The acceleration sensor 14 is a sensor that measures the acceleration of the vehicle. The brake sensor 15 is a sensor that detects whether the brake is turned on.

  Further, the ECU 4 of the vehicle control device 1 avoids the heartbeat continuation rise detection unit 6, the blink interval calculation unit 11, the long blink interval detection unit 7, the danger determination unit 8, and the driving support control unit 9. A behavior detection unit 16 is included.

  The avoidance action detection unit 16 avoids the collision of the vehicle by the driver based on the determination result of the danger determination unit 8, the measured values of the vehicle speed sensor 12, the steering angle sensor 13 and the acceleration sensor 14, and the detection result of the brake sensor 15. It is detected whether or not an action is taken, and the degree of danger recognized by the driver is determined.

  FIG. 14 is a flowchart showing details of a processing procedure executed by the avoidance action detection unit 16. In FIG. 14, first, it is determined whether or not the danger determination unit 8 determines that the driver recognizes the danger (step S141).

  When it is determined that the driver recognizes a sense of danger, it is determined whether or not the vehicle speed reduction rate measured by the vehicle speed sensor 12 is lower than the threshold value P (step S142). When the reduction rate of the vehicle speed is lower than the threshold value P, it is determined that the risk of danger recognized by the driver is high (step S143).

  When the vehicle speed reduction rate is not lower than the threshold value P, it is determined whether or not the steering angle measured by the steering angle sensor 13 is larger than the threshold value θ (step S144). When the steering angle is larger than the threshold value θ, it is determined that the risk of danger recognized by the driver is high (step S143).

  If the steering angle is not larger than the threshold value θ, it is determined whether or not the vehicle deceleration measured by the acceleration sensor 14 is larger than the threshold value G (step S145). When the deceleration of the vehicle is larger than the threshold value G, it is determined that there is a high risk of danger recognized by the driver (step S143).

  When the deceleration of the vehicle is not greater than the threshold value G, it is determined whether or not the brake ON is detected by the brake sensor 15 (step S146). When the brake ON is detected, it is determined that the risk of danger recognized by the driver is high (step S143). When the brake OFF is detected, it is determined that the danger perceived by the driver is low (step S147).

  Returning to FIG. 13, the driving support control unit 9 controls the driving support unit 5 to perform driving support at a level corresponding to the degree of danger determined by the avoidance behavior detection unit 16. Specifically, the driving support control unit 9 controls the driving support unit 5 to perform high-level driving support when it is determined by the avoidance action detection unit 16 that the risk of danger recognized by the driver is high. When it is determined by the avoidance behavior detection unit 16 that the risk of danger recognized by the driver is low, the driving support unit 5 is controlled so as to perform low level driving support.

  In the above, the heart rate measuring device 2, the blink detector 3, the vehicle speed sensor 12, the steering angle sensor 13, the acceleration sensor 14, the brake sensor 15, the heart rate continuation rise detection unit 6 of the ECU 4, and the blink interval. The calculation unit 11, the long blink interval detection unit 7, the danger determination unit 8, and the avoidance action detection unit 16 constitute a driver state determination device 10.

  As described above, in this embodiment, when the avoidance action by the driver's operation is performed, it is determined that there is a high risk of danger recognized by the driver, so it is possible to accurately determine the high risk of danger accompanying the driver's operation. Can do.

  FIG. 15: is a block diagram which shows the vehicle control apparatus provided with 6th Embodiment of the driver state determination apparatus based on this invention. In the figure, elements that are the same as or equivalent to those in the second embodiment are given the same reference numerals, and descriptions thereof are omitted.

  In the figure, the ECU 4 of the vehicle control device 1 of the present embodiment includes a heartbeat continuation rise detection unit 6, a blink interval calculation unit 11, a long blink interval detection unit 7, a danger determination unit 8, and a driving support control unit. In addition to 9, a danger magnitude estimation unit 17 is provided. The danger magnitude estimation unit 17 estimates the magnitude of danger recognized by the driver based on the determination result of the danger determination unit 8 and the measurement value of the heart rate measuring device 2.

  FIG. 16 is a flowchart showing details of a processing procedure executed by the danger magnitude estimation unit 17. In FIG. 16, first, it is determined whether or not the danger determination unit 8 determines that the driver recognizes the danger (step S151).

If it is determined that the driver recognizes a sense of danger, it is determined whether the heart rate at the time of danger recognition determination is increasing based on the measurement value of the heart rate meter 2 (step S152). . When the heart rate at the time of risk recognition determination is increasing, the normalized heart rate is calculated by the following equation (step S153).
Normalized heart rate = heart rate / average heart rate for a certain period (N seconds) immediately before judgment of danger

Subsequently, the magnitude of danger is calculated by the following equation (step S154). As a result, the magnitude of the danger perceived by the driver can be estimated quantitatively.

  On the other hand, when it is determined in step S152 that the heart rate at the time of risk recognition determination is not increasing, it is determined that the driver recognizes a risk that the heart rate response is small (step S155).

  Returning to FIG. 15, the driving support control unit 9 controls the driving support unit 5 to perform driving support at a level corresponding to the magnitude of the danger estimated by the danger magnitude estimation unit 17.

  In the present embodiment, the heart rate is normalized by the average heart rate for each individual, but the heart rate may not be particularly normalized. Further, when normalizing the heart rate, another method may be used, for example, recording the difference between the minimum value and the maximum value of the heart rate fluctuation in a certain period immediately before the risk recognition determination, and averaging the values. You may divide your heart rate by

  In the above, the heart rate measuring device 2, the blink detector 3, the heart rate continuation rise detection unit 6, the blink interval calculation unit 11, the long blink interval detection unit 7, the danger determination unit 8, and the danger feeling magnitude of the ECU 4 The height estimation unit 17 constitutes the driver state determination device 10.

  FIG. 17: is a block diagram which shows the vehicle control apparatus provided with 7th Embodiment of the driver state determination apparatus based on this invention. In the figure, elements that are the same as or equivalent to those in the second embodiment are given the same reference numerals, and descriptions thereof are omitted.

  In the figure, the vehicle control apparatus 1 of the present embodiment further includes a consciousness decrease detector (consciousness decrease detection means) 18 for detecting whether or not the driver's consciousness is decreased. As the consciousness decrease detector 18, for example, a camera is used. At this time, the consciousness decrease detector 18 may be used also as the blink detector 3.

  Further, the ECU 4 of the vehicle control device 1 includes the heartbeat continuation rise detection unit 6, the blink interval calculation unit 11, the long blink interval detection unit 7, the danger determination unit 8, and the driving support control unit 9. .

  FIG. 18 is a flowchart showing details of the processing procedure executed by the heart rate continuation rise detection unit 6. In addition, the same code | symbol is attached | subjected to the same procedure as the procedure shown in FIG.

  In FIG. 18, it is first determined whether or not the driver's consciousness detector 18 detects that the driver's consciousness has been reduced (step S161). When it is detected that the driver's consciousness is lowered, this process is forcibly terminated. When it is detected that the driver's awareness is not lowered, the above steps S107 to S109 are sequentially executed.

  FIG. 19 is a flowchart showing details of a processing procedure executed by the long blink interval detection unit 7. In addition, the same code | symbol is attached | subjected to the same procedure as the procedure shown in FIG.

  In FIG. 19, it is first determined whether or not the driver's consciousness detector 18 detects that the driver's consciousness has decreased (step S162). When it is detected that the driver's consciousness is lowered, this process is forcibly terminated. When it is detected that the driver's awareness is not lowered, the above-described procedure S111 is executed, and the above-described procedures S116 and S117 are executed sequentially.

  In the above, the heart rate measuring device 2, the blink detector 3, the consciousness drop detector 18, the heart rate continuation rise detection unit 6, the blink interval calculation unit 11, the long blink interval detection unit 7 and the danger feeling of the ECU 4 The determination unit 8 constitutes a driver state determination device 10.

  As described above, in the present embodiment, it is possible to determine with high accuracy whether or not the driver recognizes a sense of danger in a normal state where the driver's consciousness has not deteriorated.

  As described above, several preferred embodiments of the driver state determination device, the vehicle control device, and the driver state determination method according to the present invention have been described, but the present invention is not limited to the above-described embodiment.

  For example, in the above-described embodiment, the driver's heart rate measurement value and the driver's blink detection result are used to determine whether or not the driver recognizes a sense of danger. It is possible to determine whether or not the driver recognizes a sense of danger using only the detection result of the blink of the eyes of the driver without using the measured value of the heart rate of the driver. . In place of the heart rate, at least one of a blood pressure value, a sweating amount, skin conductance, and skin temperature may be measured.

  In the above embodiment, when it is determined that the driver recognizes the danger, the vehicle driving control support is performed, and the driver is determined not to recognize the danger. When it is determined that the driver recognizes the danger, the driver does not recognize the danger. You may make it raise the level of driving control assistance of a vehicle compared with when it determines with it being in a state.

  Further, in the above-described embodiment, the brake operation and the steering operation for avoiding the collision are performed as the vehicle driving control support. However, for example, information notification to the driver may be performed.

  DESCRIPTION OF SYMBOLS 1 ... Vehicle control apparatus, 2 ... Heart rate measuring device, 3 ... Blink detector (blink detection means), 4 ... ECU, 5 ... Driving support unit (support means), 6 ... Heart rate continuation rise detection part, 7 ... Long Blink interval detection unit (danger sensation determination unit), 8 ... Danger determination unit (danger determination unit), 9 ... Driving support control unit (support unit), 10 ... Driver state determination device, 11 ... Blink interval calculation unit (Dangerousness determining means), 12 ... Vehicle speed sensor (vehicle stop detecting means), 18 ... Decrease in consciousness detector (consciousness decrease detecting means).

Claims (7)

A driver state determination device for determining a state of a driver of a vehicle,
Blink detection means for detecting blinks in the eyes of the driver;
A state in which the driver's eye open state duration is determined based on the driver's eye blink detected by the blink detection means, and the driver recognizes a sense of danger based on the driver's eye open state duration A driver state determination device comprising: a risk determination unit that determines whether or not   The danger determination means calculates the driver's eye open state reference value based on the driver's past eye open state duration, and compares the driver's current eye open state duration with the eye open state reference value. Then, it is determined whether or not the driver recognizes a sense of danger.   3. The risk determination unit according to claim 2, wherein the risk determination unit suppresses execution of the calculation of the eye-opening state reference value when it is determined that the driver has recognized the risk of danger last time. Driver status determination device. Vehicle stop detection means for detecting whether or not the vehicle is stopped;
When the vehicle stop detection means detects that the vehicle is stopped, the danger determination means suppresses the calculation of the eye-opening state reference value, and the driver recognizes the danger. 4. The driver state determination device according to claim 2, wherein the driver state determination device determines that the state is not in a state. Further comprising a consciousness decrease detection means for detecting whether or not the driver's consciousness is reduced,
The driver state according to any one of claims 1 to 4, wherein the danger determination unit is implemented when the driver's consciousness is not decreased by the consciousness decrease detection unit. Judgment device. A vehicle control device for controlling a vehicle driven by a driver,
Blink detection means for detecting blinks in the eyes of the driver;
A state in which the driver's eye open state duration is determined based on the driver's eye blink detected by the blink detection means, and the driver recognizes a sense of danger based on the driver's eye open state duration A danger determination means for determining whether or not
A vehicle control apparatus comprising: a support unit that performs control support of the vehicle according to a determination result by the danger determination unit. A driver state determination method for determining a state of a driver of a vehicle,
Detecting blinking eyes of the driver;
Determining a duration of the driver's eye open state based on the driver's eye blink, and determining whether the driver recognizes a danger based on the driver's eye open state duration A driver state determination method comprising:

Images