JP2008217482A - Visibility enhancement support device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a visibility enhancement support device for changing a noise intensity, based on a condition of a driver including, for example, low wakefulness and dozing, so as to enhance visibility. <P>SOLUTION: This visibility enhancement support device imparts at first a noise of optimum noise intensity as an initial value, in step 100, when starting driving. The device determines a proper driving range (proper driving range determination processing), in step 110. Then, the device monitors a vehicle behavior and changes a calculation method of a driver condition (vehicle behavior acquisition processing) in response to the vehicle behavior, in step 120. Then, the device changes the noise intensity (driver condition acquisition processing), based on a standard deviation of a load distribution, in step 130. Then, the device imparts the noise of noise intensity set as above, in step 140. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a visibility enhancement support apparatus.

  Conventionally, there has been proposed a visual power improvement support device that prevents an accident due to carelessness or oversight by improving the driver's visual recognition ability to the situation around the host vehicle and making the reaction to the surrounding situation sensitive. (See Patent Document 1).

According to the visibility improving device disclosed in Patent Document 1, noise having a strength corresponding to the surrounding situation and the driver state is generated and applied to the driver. Therefore, the driver's visibility can be improved.
JP 2006-172315 A

  However, the above-described conventional visual acuity improving device changes the noise intensity to the driver at the time when the driver is in a state where it is difficult to perceive the noise itself such as a decrease in arousal level or snoozing. When the driver's state changes below an expected level, there is a problem that it is difficult to support improvement in visibility by applying noise.

  The present invention has been made in view of such a problem. For example, the visibility improving device can improve the visibility by changing the noise intensity based on the state of the driver, not limited to a decrease in arousal level or snoozing. The purpose is to provide.

  The present invention focuses on the results of research on stochastic resonance (SR, Stochastic Resonance) (experimental verification of stochastic resonance in the human brain, Hidaka et al., Biological and Physiological Symposium Vol.15th, P.261-264). It is a thing.

  The stochastic resonance phenomenon in a living body is a phenomenon in which the ability to detect an input signal below a threshold value in a sensory nerve cell is increased by applying appropriate noise to the sensory nerve cell. Experimental verification and the like have revealed that this stochastic resonance phenomenon improves human biological functions (for example, macro functions such as perception, regulation, and action).

  By the way, looking at examples where the effects of stochastic resonance have appeared so far, there are research results that improve the function by adding noise to the reduced function (for example, I. Hidaka, D. Nozaki, Y Yamamoto: Functional stochastic resonance in the human brain: Noise induced sensitization of baroreflex system. Phys. Rev. Lett. 85: 3740-3743, 2000.).

  In other words, the stochastic resonance phenomenon has an effect of lowering the threshold of detection ability, but since the effect is to bring out the ability inherent to human beings, the effect is considered to be small when the maximum ability is being exhibited. On the other hand, in the state where human ability is reduced, the effect as a stochastic resonance phenomenon is considered to be large because the allowance for extracting ability is large.

  However, a situation in which human ability has been reduced is a situation in which driving itself is difficult such as falling asleep. In the conventional technology, when a nap is detected, the driver is awakened by awakening means such as a seat vibration, but the effect of stochastic resonance is detected by detecting a decline in human ability and giving noise before reaching a nap. You can get the most out of it.

The present invention has been obtained based on the above-described findings. Hereinafter, each claim will be described.
(1) The invention of claim 1 generates visual noise having a noise intensity (for example, optimum noise intensity) set based on a threshold value of the visually observable noise intensity, and the generated visual noise is generated during vehicle driving. The present invention relates to a visibility enhancement support device that assists the driver in improving the visibility by giving to the driver.

  In particular, in the present invention, the driver state (for example, using a pressure sensor) acquires the driver state (indicating the driver's operation ability) before reaching the driver's arousal level or falling asleep. The relationship storage means stores the correspondence between the driver state and the visual noise. The noise application mode control unit controls the visual noise application mode from the driver state acquired by the driver state acquisition unit based on the correspondence stored in the correspondence storage unit.

  In order to obtain the maximum effect of stochastic resonance, when the driver's ability to operate the vehicle is reduced, it is preferable that visual noise having an optimal noise intensity according to the state can be given. Even if the noise intensity is increased after the awakening level is actually lowered or the state of dozing is increased, the effect is low.

  Therefore, in the present invention, before falling into a state where the arousal level is actually lowered or falling into a dozing state, it is detected by a change in the load distribution at the seat back, for example, that the driver's operation ability has decreased than usual, The visual noise application mode is controlled in accordance with the driver state. For example, visual noise having a noise intensity corresponding to the driver state is given to the driver.

  By doing in this way, even if a driver's operation capability falls, visual noise can be given with respect to the driver of a vehicle before falling into a state impossible for driving, such as dozing. Further, as described above, the effect of the stochastic resonance phenomenon can be maximized by applying noise aiming at the time when the capacity is lowered. Thereby, the improvement in visibility can be supported more effectively than before.

Note that the correspondence storage means stores data such as the intensity of an electrical signal (noise signal) for generating visual noise that is actually given to the driver (the same applies hereinafter).
(2) The invention of claim 2 generates a visual noise having a noise intensity (for example, an optimum noise intensity) set based on a threshold value of a visually observable noise intensity, and uses the generated visual noise during driving of the vehicle. The present invention relates to a visibility enhancement support device that assists the driver in improving the visibility by giving to the driver.

  In particular, in the present invention, the driver state acquisition unit acquires the driver state based on the load distribution variation in the seat back portion generated by the driver, and the correspondence relationship storage unit determines the driver state and the visual noise. The correspondence is memorized. Then, the noise application mode control unit controls the visual noise application mode from the driver state acquired by the driver state acquisition unit based on the correspondence stored in the correspondence storage unit.

  As described above, even if the noise intensity is increased after the awakening level is actually lowered or the dozing state is reached, the effect is low. Therefore, in the present invention, it is detected based on the load distribution fluctuation in the seat back portion generated by the driver that the driver's operation ability (before reaching a decrease in arousal level or falling asleep) is lower than usual. The visual noise application mode is controlled in accordance with the driver state. For example, visual noise having a noise intensity corresponding to the driver state is given to the driver.

Thereby, it is possible to support the improvement of the visibility more effectively.
(3) In the invention of claim 3, the vehicle state is acquired by the vehicle state acquisition means, and the correspondence relationship between the driver state and the vehicle state and the visual noise is stored by the correspondence relationship storage means. Then, by applying the noise from the driver state and the vehicle state acquired by the driver state acquisition unit and the vehicle state acquisition unit by the noise addition mode control unit, the visual noise is applied based on the correspondence stored in the correspondence storage unit. Control aspects.

  In measuring the state of a driver in a car, the behavior of the car itself due to a rough road, a sharp curve, or the like becomes an artifact. Therefore, in the present invention, the vehicle state acquisition means acquires the vehicle behavior in the vertical direction and the horizontal direction, for example, and if there is a vehicle behavior that affects the acquisition of the driver state by the driver state acquisition means, the influence is obtained. Change the measurement method and so on. As a result, it is possible to accurately perform control that supports improvement in the visibility.

(4) In the invention of claim 4, the driver state acquisition means includes a plurality of pressure sensors arranged on the seat back portion.
In the present invention, in order to calculate the position of the center of gravity of the pressure applied to the seat back portion, a plurality (preferably three or more) of pressure sensors are arranged in the seat back portion (for example, inside).

  (5) In the invention according to claim 5, the pressure sensor of the seat back portion is within the range from the driver's waist to both shoulders in the vertical direction at a predetermined distance from the seat surface, and the driver in the horizontal direction. At least three or more sensors are disposed within the shoulder width range.

  In the present invention, the pressure sensor is disposed in a portion where the load of the driver is applied in the seat back portion. As the part where the driver's load is applied, the vertical direction may be the height from the seating surface to both shoulders of the driver, but if the pressure sensor is placed close to the seating surface, no load is applied to the pressure sensor. Since it is considered that there are drivers who drive in a posture that does not apply and drivers who always drive in a posture in contact with the pressure sensor, they are arranged at a position away from the seating surface. In the horizontal direction, the pressure sensor is arranged in a range narrower than the driver's shoulder width. It should be noted that a range in which the load can be captured even with an average shoulder width of a woman is desirable.

(6) In the invention according to claim 6, two or more pressure sensors are arranged in the seat back portion at least in the horizontal direction and in the vertical direction.
In the present invention, in order to enable determination based on the positions of the center of gravity in the horizontal direction and the vertical direction, at least two pressure sensors are arranged in the horizontal direction and two or more pressure sensors are arranged in the vertical direction.

  It is desirable that this arrangement can capture the driver's load distribution and is as wide as possible. Accordingly, in order to obtain pressure in the vertical direction from a position at a certain distance from the seat surface to both shoulders of the driver, and in the horizontal direction to the shoulder width of the driver, three or more sensors are required. In addition, when four or more sensors are used, it is desirable to arrange them symmetrically within the aforementioned range.

(7) In the invention of claim 7, the driver state acquisition means acquires the state of the driver based on the load distribution obtained from the plurality of pressure sensors.
Since the load distribution obtained from the pressure sensor changes according to the driver's state, the driver's state can be detected from this load distribution.

(8) In the invention of claim 8, the driver's state is acquired based on the standard deviation of the load center position obtained from a plurality of pressure sensors at a predetermined time.
As the driver's state changes, the load distribution on the seat back portion changes. Therefore, for example, the center of gravity of the load distribution (referred to as the load center position) can be obtained, and the state of the driver can be evaluated based on the amount of movement. Specifically, for example, in order to determine the variation of the center of gravity with respect to the seat back portion of the driver within a certain time, the load center position is obtained for a certain time, and the standard deviations of the load center positions in the horizontal and vertical directions are obtained.

As will be described later, when the driver's operating ability decreases, the standard deviation of the load distribution tends to decrease.
(9) The invention according to claim 9 is characterized in that the load center position is obtained from a relative value between the pressure sensors.

  For example, the pressure distribution is normalized by dividing the output from the pressure sensor by the sum of the outputs of all the sensors. As a result, it is possible to acquire the driver state without taking into account differences between individuals such as a difference in weight and a difference in posture with respect to the seat back portion.

  (10) In the invention according to claim 10, when the standard deviation of the load center position falls below the predetermined first threshold, the noise application mode control means has a noise intensity higher than the visual noise applied so far. It is characterized in that visual noise with relatively large is added.

  In the present invention, for example, the standard deviation of the load center position is acquired a plurality of times in a certain period from the start of operation, and whether or not the standard deviation of the load center position after the certain period is included for this acquired set. Determine the driver status.

  It is considered that the driver's state gradually decreases from the start of driving, and the driver's state is good at the start of driving and is considered to be suitable for driving. Therefore, it is determined that outliers not included in the set are not suitable for driving. Specifically, the state suitable for operation is considered to be a range in which the standard deviation of the load center position at an arbitrarily set time is included in the set of standard deviations of the load center position in an arbitrary period from the start of operation. . Among those not included in this range, for example, it is determined that the time when the driving ability of the driver is lowered is below the lower limit of the range.

  At this time, visual noise having a relatively higher noise intensity than the visual noise that has been applied so far is applied by the noise application mode control means. Thereby, visibility is improved.

(11) The invention of claim 11 is characterized in that the first threshold value is the lower limit of the proper operation range.
The present invention exemplifies the first threshold value. The appropriate driving range is a range of a state suitable for driving, and can be set based on, for example, a deviation (deviation amount) from the lane center line during driving. For example, when the deviation amount is smaller than a predetermined value, it can be considered that the state is suitable for driving.

  (12) In the invention of claim 12, the driving proper range is 25% to 50% centering on the median value in the range obtained by the standard deviation of the load center position in a predetermined period after the driver starts driving. It is characterized by being in the range.

  In the present invention, for example, the standard deviation of the load center position is acquired a plurality of times in an arbitrary period from the start of operation, and the range corresponding to 25% to 50% centering on the median value of the range is set as the proper operation range.

  (13) In the invention of claim 13, the noise application mode control means equalizes the noise intensity of the visual noise to the initial set value (default value) when the standard deviation of the load center position falls within the appropriate operating range. Noise level.

  Here, the state suitable for driving is considered to be a range in which the standard deviation of the load center position at a time interval that can be arbitrarily set is included in the proper driving range. Therefore, when the standard deviation of the load center position is within the proper operation range, the noise intensity is changed to the initial setting value. This is because the noise application mode control means changes the noise intensity in accordance with the driver's condition, whereas the driver condition is suitable for driving. It means to return to. Normally, when the determination that the driver's condition is suitable for driving continues, the noise intensity is not changed.

  (14) In the invention of claim 14, when the standard deviation of the load center position exceeds a predetermined second threshold, the noise application mode control means has the same noise intensity as the visual noise that has been applied so far. Of noise. However, the second threshold is larger than the first threshold.

  In the present invention, for example, the standard deviation of the load center position is acquired a plurality of times in an arbitrary period from the start of operation, for example, the second threshold value is lower by 25% from the median value that is the first threshold value in the range. A range of 50% centering on the median value, which is 25% higher than the median value, is set as an appropriate operating range. The state suitable for driving is considered to be a range in which the standard deviation of the load center position at an arbitrarily set time interval is included in the driving proper range. If an outlier that is not statistically included in the set exceeds a second threshold (for example, an upper limit), it is considered that a large body movement of the driver such as re-sitting has influenced.

Since these are artifacts in the evaluation of the driver's condition, the noise application mode control means does not change the noise intensity that has been given so far without being controlled.
Even when the driver does not sit back, the standard deviation of the load center position may be influenced by the vehicle behavior, as will be described later. In this case as well, although the standard deviation value becomes large, the noise intensity can be changed based only on the change in the driver's state by taking the above measures.

(15) In the invention of claim 15, the second threshold is the upper limit of the proper driving range.
The present invention exemplifies the second threshold value.
(16) In the invention of claim 16, the driver state acquisition means calculates the number of samples at the center of the load distribution for obtaining the standard deviation from the load center position according to the vertical vehicle behavior obtained from the vehicle state acquisition means. Increase or decrease.

  The present invention deals with a case where the vehicle vibrates in the vertical direction, such as when traveling on a rough road. The driver state is obtained from the standard deviation of the load center position for a fixed time, but when the vehicle vibrates in the vertical direction, the value increases regardless of the driver state. At this time, the vehicle state acquisition means increases the number of samples of the load center position for obtaining the standard deviation, and averages the influence of the load center position including the influence of the vehicle behavior, so that the driver even when there is a behavior. The state can be detected.

  (17) In the invention of claim 17, the driver state acquisition means calculates the number of samples at the center of the load distribution for obtaining the standard deviation from the load center position in accordance with the vehicle behavior in the left-right direction obtained from the vehicle state acquisition means. Increase or decrease.

  The present invention deals with a case where a lateral G is applied to the vehicle, such as when turning left and right. When the lateral G is applied, a biased load center position is obtained regardless of the driver's condition. At this time, when the vehicle state obtaining means increases the number of samples of the load center position for obtaining the standard deviation and averages the influence of the load center position including the influence of the lateral G, there is a vehicle behavior in the left-right direction. Also made it possible to detect the driver status.

Next, the best mode (embodiment) of the present invention will be described.
a) First, a schematic configuration of the visibility enhancement support device of the present embodiment will be described.
As shown in FIG. 1 (a), the visibility enhancement support apparatus 1 of the present embodiment is mounted on a vehicle, and generates a noise (electric noise) and a noise generation apparatus 3 that generates noise (electric noise). A noise generation device 5 that outputs noise, a driver state acquisition device 7 that acquires the state of the driver, a vehicle state acquisition device 9 that acquires the state of the vehicle, and an operation switch 11 are provided.

  The noise generation device 3 outputs a noise intensity storage unit 3a that stores a threshold value of noise intensity capable of perceiving predetermined noise, which will be described later, and a noise intensity threshold value stored in the noise intensity storage unit 3a. Correspondence between the optimum noise intensity setting unit 3b for setting the optimum noise intensity (optimum noise intensity), the driver state and vehicle state acquired by the driver state acquisition device 7 and the vehicle state acquisition device 9, and the optimum noise strength A correspondence relationship storage unit 3c that stores the relationship, and a control unit 3d that is connected to the noise intensity storage unit 3a, the noise optimum strength setting unit 3b, and the correspondence relationship storage unit 3c and controls the entire noise generation device 3 are provided. .

  The control unit 3d is connected to the driver state acquisition device 7, determines the optimum noise intensity corresponding to the acquired driver state based on the correspondence stored in the correspondence storage 3c, and the determination is made. Generate noise with optimal noise strength. Then, a control signal (drive request signal) corresponding to the generated noise is output to the noise generator 5.

  The noise generator 5 is, for example, an indoor lamp, and light (noise light) on which visual noise is superimposed is output (irradiated) from the indoor lamp or the like. In addition, when functioning as a normal room light, a relatively high luminance is required, but in the case of noise light, a lower luminance is sufficient.

  As a specific example of the interior lamp used as the noise generating device 5 of the present embodiment, as shown in FIG. 1 (b), it is conceivable that the interior lamp is mounted on the ceiling of the passenger compartment ahead of the driver's head. This interior light plays a role as a so-called room lamp, and if it is mounted on the ceiling of the passenger compartment ahead of the driver's head in this way, in terms of giving visual noise to the driver. preferable.

b) Here, visual noise will be described.
The visual noise used to induce stochastic resonance (SR) in visual recognition must be random broadband noise that does not exhibit strong intensity in a specific frequency band. Therefore, random noise with a uniform frequency distribution of the noise intensity of visual noise in a visually recognizable frequency band, or the frequency distribution of the noise intensity of visual noise is inversely proportional to the frequency of visual noise. By using f noise or the like, a stochastic resonance phenomenon (SR) in visual recognition can be induced.

  As described above, random wideband noise may reduce the signal-to-noise ratio of the output signal if the noise intensity is too small or too large to exceed the threshold regardless of the input signal intensity. There is a known noise intensity that maximizes the signal to noise ratio of the signal output.

  Therefore, in order to generate noise light having a noise intensity suitable for inducing the stochastic resonance phenomenon (SR), for example, a noise intensity suitable for increasing the signal-to-noise ratio of the signal output is obtained in advance by experiments or the like. The noise intensity obtained by this experiment or the like may be stored as the optimum noise intensity in the noise intensity storage unit 3a.

  In addition, since there is an individual difference in the optimum noise intensity, in the present embodiment, the following measures are taken to supplement the individual difference. That is, when the control unit 3d gradually increases the intensity of the noise light output from the noise generating device 5 and operates the operation switch 11 at the timing when the driver first perceives the noise, the intensity of the noise light at the operation timing Is determined as a noise intensity threshold value and stored in the noise intensity storage unit 3a.

  Then, the noise optimum intensity setting unit 3b has a predetermined ratio of noise intensity (for example, the random noise) suitable for increasing the signal-to-noise ratio of the signal output with respect to the threshold value stored in the noise intensity storage unit 3a. In the case of (1), the noise intensity is set to about 100% of the threshold value, and in the case of the 1 / f noise, the noise intensity (about 69% of the threshold value) is set as the optimum noise intensity. Note that the ratio of the noise intensity to the threshold suitable for increasing the signal-to-noise ratio of the signal output may be obtained in advance by experiments or the like. Moreover, you may comprise so that the ratio of the noise intensity with respect to a threshold value can be changed.

c) Next, the driver state acquisition device 7 will be described.
As shown in FIG. 2, the driver state acquisition device 7 includes a plurality of pressure sensors 15 arranged inside the seat back portion 13 of the driver's seat, and detects the driver's seating situation.

  The pressure sensor 15 is disposed at a portion where the driver reliably contacts the seat back portion 13. Specifically, as shown in FIG. 3 (a), the vertical direction is a range from the driver's waist (specifically, a position above the seat surface to a predetermined value (for example, about 24 cm)) to the height of the driver's shoulder, As shown in FIG. 3 (b), the horizontal direction is a range within the shoulder width of the driver.

  For example, as shown in FIG. 2A, the pressure sensor 15 is arranged in a cross shape near the center of the seat, and the shoulder is taken out as shown in FIG. 2B. A method of arranging in a T-shape, a method of arranging in a rectangular shape as shown in FIG.

  Further, in order to reliably acquire the driver state for a driver of any body type, it is desirable to dispose many pressure sensors 15. As the arrangement method, for example, a grid-like arrangement as shown in FIG. 2 (d) is mentioned, but in order to further increase the density, the arrangement as shown in FIG. 2 (f) is desirable.

  Furthermore, in this embodiment, as will be described later, in order not to depend on the height, body shape, etc. of the driver, the standard deviation of the pressure distribution with respect to the seat back portion 13 is adopted as the seating state acquisition method, The value obtained from each pressure sensor 15 is replaced with a relative value.

d) Next, processing in this embodiment will be described.
(1) First, FIG. 4 shows the overall processing flow.
When driving is started, first, in step (S) 100, the visual noise having the above-mentioned (initial value) optimum noise intensity is added to the driver.

  Thereafter, in step 110, as will be described later, a process for determining an appropriate driving range (an appropriate driving range determining process) is performed. In the present embodiment, the noise specification is changed depending on whether or not the driver state is included in the range.

Next, in step 120, as will be described later, the vehicle behavior is monitored, and a process (vehicle behavior acquisition process) for changing the driver state calculation method according to the vehicle behavior is performed.
Next, in step 130, as will be described later, processing for changing the noise intensity from the standard deviation of the load distribution (driver state acquisition processing) is performed.

Next, in step 140, a process of applying visual noise having the noise intensity set as described above is performed, and the process returns to step 120.
(2) Next, FIG. 5 shows the appropriate range determination process.

In Step 200, the appropriate state of motion is set by the vehicle state acquisition device 9 from the load center position of the initial acquisition time T0 (for example, 10 minutes) that can be arbitrarily set from the start of driving.
Here, the output from each pressure sensor 15 is divided by the sum of the outputs of all the sensors, thereby making the output of the pressure sensor 15 a relative value, normalizing the pressure distribution, and obtaining from the normalized pressure distribution. Find the center of gravity (load center position) of the load distribution.

  That is, based on the relative value of each pressure sensor 15, the load center position of the seat back portion 13 of the driver is calculated at an arbitrary interval (for example, at an interval of 0.1 second), and then a time that can be arbitrarily set ( For example, the standard deviation D of 100 points of the center of gravity (load center position at each sample point) for the past 10 seconds is calculated.

  Then, the load center position of an initial acquisition time T0 (for example, 10 minutes) that can be arbitrarily set is acquired from the start of operation, and 60 points of the standard deviation D of the load center position (that is, one obtained by data for 10 seconds). The standard deviation D of 10 minutes = 60 points) is taken as the appropriate range of motion.

In the subsequent step 210, the lower limit value of the appropriate exercise range is set to dmin, the upper limit value is set to dmax, and the present process is temporarily terminated.
(3) Next, FIG. 6 shows the vehicle state acquisition process.

In step 300, the vehicle state acquisition device 9 detects vertical vibrations and lateral G in the horizontal direction (detects vehicle behavior).
In the subsequent step 310, it is determined whether there is vibration and lateral G, that is, whether there is a vehicle behavior. If an affirmative determination is made here, the process proceeds to step 320, while if a negative determination is made, the process proceeds to step 330.

  In step 320, since there is a vehicle behavior, in order to reduce the influence of the vehicle behavior, the calculation time of the standard deviation D of the load center position (acquisition time for determination used for determining the driver state) T1 is lengthened (to Tlong). Setting), increasing the number of samples, and once ending this processing.

  On the other hand, in step 330, since there is no vehicle behavior, the calculation time (acquisition time for determination) T1 of the standard deviation D of the load center position is set to an initial value (set to Tdef), and this process is temporarily terminated. Note that Tlong> Tdef.

(4) Next, FIG. 7 shows the driver state acquisition process.
In step 400, based on the time (determination acquisition time T1) for calculating the standard deviation D of the load center position set by the vehicle state acquisition process of FIG. A standard deviation D is calculated.

  In the following step 410, it is determined whether or not the standard deviation D of the load center position obtained at this time is below the lower limit value dmin of the proper operation range. If an affirmative determination is made here, the process proceeds to step 420, while if a negative determination is made, the process proceeds to step 430.

  In step 420, since the standard deviation D of the load center position is below the lower limit value dmin of the appropriate driving range, it is determined that the driver state is not suitable for driving, and processing for increasing the noise intensity is performed and this processing is terminated once. To do.

  On the other hand, in step 430, it is determined whether or not the standard deviation D of the load center position is equal to or less than the upper limit value dmax of the proper operation range. If an affirmative determination is made here, the process proceeds to step 440, whereas if a negative determination is made, the present process is temporarily terminated.

  In step 440, since the standard deviation D of the load center position is within the proper operating range, the noise intensity specification is changed to the initial value (noise intensity smaller than the value set in step 420), and this process is performed once. finish.

  In addition, when a negative determination is made in step 430, the standard deviation D of the load center position exceeds the appropriate driving range, but this is determined to have been affected by the driver's re-sitting or vehicle behavior, and as it is. This process ends.

And the process mentioned above is repeated only for the period when a driver | operator continues a driving action.
e) Next, the above-described method for setting the proper operation range will be described in more detail.
The proper operation range of the present embodiment is set based on the experiment shown below.

This experiment examined driver status and lane departure.
Specifically, the standard deviation of the load center position was obtained by the pressure sensor, and at the same time, the deviation amount (staggered amount) from the lane (specifically, the lane center line) was measured. Here, the sampling interval for calculating the standard deviation was 0.1 second, and the standard deviation was calculated from the data for 10 seconds. The result is shown in FIG.

  In FIG. 8, the thin line indicates the deviation amount, and the thick line indicates the driver state (standard deviation of the load center position). In FIG. 8, the vertical axis indicates the deviation (deg (steering angle)) and standard deviation, and the horizontal axis indicates time (seconds).

  As is clear from FIG. 11, it can be seen that the value of the standard deviation is decreased from around 1200 seconds after the start of operation in which the deviation amount increases (operation capacity decrease section). For this reason, it is necessary to set the appropriate operating range so that at least this section can be detected.

  The appropriate operating range is obtained from the following formulas (1) and (2) from the maximum value and the minimum value of the range of the load center position obtained in a time (for example, 10 minutes) that can be arbitrarily set from the start of the experiment. Here, x is a parameter that determines an appropriate operating range.

Upper limit: (1/2) (1 + x) (maximum value-minimum value) (1)
Lower limit: (1/2) (1-x) (maximum value-minimum value) (2)
FIG. 9 shows a proper operation range (range between upper and lower broken lines) of x = 0.75 (x = 0.5 in FIG. 10, x = 0.25 in FIG. 11) with respect to the standard deviation of 10 minutes. The driver state (standard deviation of the load center position) obtained by the pressure sensor is shown. In addition, the vertical axis | shaft of FIGS. 9-11 has shown the standard deviation, and the horizontal axis has shown time (second).

  As shown in FIG. 9, when x = 0.75, the section shown in FIG. 8 (operation ability decrease section) cannot be detected, whereas x = 0.5, x = 0. In FIG. 25, it can be seen that it is possible to detect the operation capability decrease section.

However, if the proper driving range is narrower than when x = 0.25, the value exceeding the upper limit of the proper driving range increases, and the possibility of erroneous determination of sitting again increases.
For the above reasons, the appropriate driving range is 25% to 50% of the range from the minimum value to the maximum value in the standard deviation set from the start of operation to an arbitrary time (that is, 25% centering on the median standard deviation). To 50%).

  d) As described above, in the present embodiment, the driver state is acquired based on the fluctuation of the load distribution in the seat back 13 generated by the driver, and the correspondence data between the driver state and the visual noise is used. The noise intensity of the visual noise given to the driver is changed.

  In other words, since the effect is low even if the noise intensity is increased after the state of awakening is actually lowered or the state of dozing is reached, in the present embodiment, based on the variation of the load distribution in the seat backrest portion 13 ( It detects that the driver's operation ability (before reaching a reduction in arousal level, falling asleep, etc.) is lower than normal, and gives the driver visual noise with noise intensity corresponding to the driver state. Thereby, it is possible to effectively support the improvement of the visibility.

  Further, in measuring the state of the driver in the automobile, the behavior of the automobile itself due to a rough road, a sharp curve, etc. becomes an artifact, so in this embodiment, the vehicle behavior in the vertical direction and the horizontal direction is acquired, When there is a vehicle behavior that affects the acquisition of the driver state, the measurement method and the like are changed so as to suppress the influence. As a result, it is possible to accurately perform control that supports improvement in the visibility.

Needless to say, the present invention is not limited to the above-described embodiments, and can be implemented in various modes without departing from the scope of the present invention.
For example, the noise intensity is adjusted by adjusting at least one of the luminance of light irradiating the driver, the light emission area, the amount of change in luminance, the amount of change in light emission area, the lighting time, the amount of change in wavelength, and the amount of change in chromaticity. Can be changed. Note that the chromaticity changes as the luminance increases, the emission area increases, the luminance change amount increases, the emission area change amount increases, the lighting time increases, the wavelength change amount increases. The higher the amount, the greater the noise intensity.

(A) is a block diagram which shows schematic structure of the visibility improvement assistance apparatus of this embodiment, (b) is explanatory drawing of the specific example of a noise generator. It is explanatory drawing which shows the example of arrangement | positioning of the pressure sensor in a sheet | seat. It is explanatory drawing which shows the arrangement | positioning range of the vertical direction and horizontal direction of the pressure sensor in a sheet | seat. It is a flowchart which shows the main routine of the process which the visibility improvement assistance apparatus of this embodiment implements. It is a flowchart which shows the process which sets a driving | operation appropriate range. It is a flowchart which shows the process which acquires a vehicle behavior. It is a flowchart which shows the process which acquires a driver | operator state. It is a graph which shows the experimental result of the experiment example which investigated the relationship between a driver | operator state and the deviation | shift amount from a lane. It is a graph which shows the driving | operation appropriate range at the time of setting x = 0.75. It is a graph which shows the driving | operation appropriate range at the time of setting x = 0.50. It is a graph which shows the driving | operation appropriate range at the time of setting x = 0.25.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Visibility improvement assistance apparatus 3 ... Noise generating apparatus 3a ... Noise intensity memory | storage part 3b ... Noise optimal intensity | strength setting part 3c ... Correspondence relation memory | storage part 3d ... Control part 5 ... Noise generating apparatus 7 ... Driver state acquisition apparatus 9 ... Vehicle State acquisition device 11 ... operation switch 13 ... seat backrest 15 ... pressure sensor

Claims (17)

Visibility of the driver is improved by generating visual noise of the noise intensity set based on the threshold of noise intensity that can be visually recognized, and giving the generated visual noise to the driver while driving the vehicle. A device for improving visibility,
Driver state acquisition means for acquiring a driver state before reaching a driver's arousal level or falling asleep;
Correspondence storage means for storing the correspondence between the driver state and the visual noise;
From the driver state acquired by the driver state acquisition unit, a noise application mode control unit that controls the visual noise application mode based on the correspondence stored in the correspondence storage unit;
A visual power improvement assisting device characterized by comprising: Visibility of the driver is improved by generating visual noise of the noise intensity set based on the threshold of noise intensity that can be visually recognized, and giving the generated visual noise to the driver while driving the vehicle. A device for improving visibility,
Driver state acquisition means for acquiring a driver state based on a load distribution variation in a seat back portion generated by the driver;
Correspondence storage means for storing the correspondence between the driver state and the visual noise;
From the driver state acquired by the driver state acquisition unit, a noise application mode control unit that controls the visual noise application mode based on the correspondence stored in the correspondence storage unit;
A visual power improvement assisting device characterized by comprising: In the visual enhancement improvement support device according to claim 1 or 2,
Furthermore, a vehicle state acquisition means for acquiring the vehicle state is provided,
Correspondence storage means for storing the correspondence between the driver state and vehicle state and visual noise;
Noise application for controlling the visual noise application mode based on the correspondence stored in the correspondence storage unit from the driver state and the vehicle state acquired by the driver state acquisition unit and the vehicle state acquisition unit. Aspect control means;
A visual power improvement assisting device characterized by comprising: In the visibility improvement assistance apparatus in any one of the said Claims 1-3,
The driver state acquisition means includes a plurality of pressure sensors arranged on a seat back portion, and the visibility improvement assisting device. In the visibility improvement assistance apparatus of the said Claim 4,
The seat backrest pressure sensor is at least within a range from the waist of the driver to both shoulders at a predetermined distance from the seat surface in the vertical direction, and within a range of the shoulder width of the driver in the horizontal direction. A visual enhancement improvement support device, wherein three or more sensors are arranged. In the visual enhancement improvement support device according to claim 4 or 5,
Two or more pressure sensors are arranged in the seat back portion at least in the horizontal direction and in the vertical direction, respectively. In the visibility improvement assistance apparatus in any one of the said Claims 4-6,
The driver state acquisition means acquires the driver state based on a load distribution obtained from the plurality of pressure sensors. In the visual enhancement improvement support device according to claim 7,
The visibility improving support apparatus, wherein the driver state is acquired based on a standard deviation of a load center position obtained from the plurality of pressure sensors at a predetermined time. In the visual enhancement improvement support device according to claim 8,
The load center support device is characterized in that the load center position is obtained from a relative value between the pressure sensors. In the visual enhancement improvement support device according to claim 8 or 9,
When the standard deviation of the load center position falls below a predetermined first threshold, the noise application mode control means is configured to visually increase the noise intensity relative to the visual noise that has been applied so far. Visibility improvement support device characterized by adding noise. In the visual acuity improvement support device according to claim 10,
Said 1st threshold value is made into the minimum of the driving | operation appropriate range, The visibility improvement assistance apparatus characterized by the above-mentioned. In the visual enhancement improvement support device according to claim 11,
The proper driving range is a range obtained from the standard deviation of the load center position in a predetermined period after the driver starts driving, and the range is 25% to 50% centering on the median value. Visibility improvement support device. In the visual enhancement improvement support device according to claim 11 or 12,
When the standard deviation of the load center position falls within the appropriate operating range, the noise application mode control means sets the noise intensity of the visual noise to a noise intensity equivalent to an initial set value. Visibility improvement support device. In the visibility improvement assistance apparatus in any one of the said Claims 8-13,
When the standard deviation of the load center position exceeds a predetermined second threshold, the noise application mode control unit applies noise having the same noise intensity as the visual noise that has been applied. Visibility improvement support device. In the visibility improvement assistance apparatus of the said Claim 14,
Said 2nd threshold value is made into the upper limit of the said driving | operation appropriate range, The visibility improvement assistance apparatus characterized by the above-mentioned. In the visibility improvement assistance apparatus in any one of the said Claims 8-15,
The driver state acquisition unit increases or decreases the number of samples at the center of the load distribution for obtaining a standard deviation from the load center position according to the vehicle behavior in the vertical direction obtained from the vehicle state acquisition unit. Visibility improvement support device. In the visibility improvement assistance apparatus in any one of the said Claims 8-15,
The driver state acquisition means increases or decreases the number of samples at the center of the load distribution for obtaining a standard deviation from the load center position according to the vehicle behavior in the left-right direction obtained from the vehicle state acquisition means. Visibility improvement support device.