JP3435413B2 - Human machine system for vehicles - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a human-machine system, for example, a human-machine system which is a man-machine system that acts on a driver's sensitivity in the relationship between an automobile and a driver.

[0002]

2. Description of the Related Art Humans cause a decrease in arousal (attention) when they continue monotonous work for a long time, and after a while, they recover to awake state. Regarding the rhythm of arousal that repeats such arousal and low arousal state, a method for quantitatively measuring it even if it is possible to perceive vaguely is not clarified.

Conventionally, a circadian rhythm that repeats sleep and activity in a cycle of about 24 hours is well known as a long-term human rhythm, which is a long-term rhythm related to human life activity. However, it does not indicate a rhythm relating to arousal on a time scale of about 10 minutes to several hours from a tense state to a decreased awakening state. Moreover, in general, the duration of human attention is psychologically 2
It is said that it takes about 0 to 30 minutes, but it has not been clarified what happens to humans when viewed as a phenomenon of several hours over time.

Conventionally, for example, as a drowsy driving detection device during driving, as disclosed in Japanese Patent Laid-Open No. 164529/1984, drowsiness is detected based on the fact that the steering state of the steering wheel is disturbed and the response speed is delayed. There is known a drowsy driving detection device which gives an alarm to a driver.
The man-machine system requires a system design that allows the user to enjoy using the machine, as represented by the relationship between the car and the driver, or the computer and the user.

In recent years, in the automobile industry, there has been devised a technique of making full use of electronic control technology to adaptively control a machine function according to a driver's preference, skill and feeling. For example, in U.S. Pat. No. 4,829,434, the driver's behavior such as personal qualities, mood, taste, physiology, psychology, etc.
It is disclosed that the vehicle characteristics are changed by adaptive control, judging from the running state and environment of the vehicle. In this U.S. patent, as a specific example, the distance between vehicles, the vehicle speed, and the amount of rainfall are detected, and the results are compared with pre-stored data. Based on the determination results, a driver who prefers light driving can use automatic transmission. A mechanism that puts the shift pattern in power mode and switches to economy mode is given to drivers who prefer comfortable driving.

[0006]

In recent years, there has been an increasing need for a technique for quantitatively knowing the time characteristics of such awakening deterioration to prevent awakening deterioration. For example, a long-term test drive of a vehicle at the development stage of an automobile manufacturer. Test drivers are required to operate in harsh monotonous operation. In such a situation, it is considered that an awakened state (a state in which the judgment is slow even if the person is not dozing) attacking the driver causes an accident caused by the driver.

The drowsy driving detection device for driving shown in the above-mentioned Japanese Patent Laid-Open No. 55-164529 detects a decrease in arousal level and issues an alarm only when the driver is in a drowsy state or immediately before that. Therefore, it may be too late. Further, in the vehicle based on the conventional electronic control technique described above, when adaptive control adapted to the driver's sensitivity is achieved, the following problems occur from the driver's point of view.

(1) Since the intellectual ability of the machine is superior to that of the driver and the control is too adapted to the driver characteristics, the machine takes over the work that should be judged and processed by the driver. Accidents tend to occur because the driver's willingness to control the vehicle is lost and the driving skill of the driver is reduced as a result. In other words, in a vehicle that performs adaptive control, the adaptive control changes the vehicle characteristics to such an extent that the driver's driving operation is affected. Therefore, the driver himself feels that the characteristics have changed, and as a driver, the characteristics have changed. You are forced to drive the vehicle.

(2) The driver feels coldness and boringness with respect to the high degree of intelligent adaptability of the vehicle, and the vehicle seems to be an inorganic thing for the driver, and the joy of driving and the affinity for the vehicle are lost. The present invention has been made in view of the above points, and an object thereof is to detect detected dry
Depending on the state of the bar, it operates to change its state
By providing a human machine system for a vehicle
The driver and
It is to guide to the driving state .

[0010]

SUMMARY OF THE INVENTION The present invention has been made for the purpose of solving the above-mentioned problems, and as a means for solving the above-mentioned problems, a human-machine system for vehicles is used.
The stem has the following configuration. That is, the driver of the vehicle
Vibrating means (28) for applying vibration to the seat on which the
Fingertip volume pulse wave detection for detecting the fingertip volume pulse wave of the driver
The exit means (13) and the vehicle are traveling on a rough road
Rough road detection means (32, 33, S1 in FIG. 9)
0, S11) and a white noise (W) generating
Noise generator (25) and the fingertip volume pulse wave
The detection result of the detection means and the detection result of the bad road traveling detection means
Based on the result, control the state of vibration by the vibrating means
Control means (26,30,34,29) and provided with the front
The control means is based on the detection result of the fingertip volume pulse wave detection means.
Based on the state quantity related to the psychological state of the driver
(H) is obtained, and the vehicle is detected by the rough road traveling detection means.
If it is detected that the vehicle is not driving on a rough road,
The smaller the state quantity, the sensory stimulation generation control parameter
While setting the value of (I) to a large value, driving on the rough road
It is possible that the vehicle is traveling on a bad road due to the detection means.
When it is detected, the smaller the state quantity, the more the feeling.
When the value of the stimulus generation control parameter is set to a small value
Both, the white noise (W), the sensory stimulus generation
The larger the parameter value, the more
Filter characteristics that reduce the amount of dynamic power reduction to a small value
By using the variable filter (26) of FIG. 12
And the state of vibration by the vibrating means
The higher the engine speed (RPM) increases
When controlling according to this frequency (ω),
By adding vibration according to the white noise generated,
It is characterized in that fluctuation is applied to the vibration state .

In a preferred embodiment, the control means
Indicates that the vehicle is traveling on a rough road by the rough road traveling detection means.
If the state that it is running continues for a predetermined time
In this case, even if the detection state continues,
It is not detected that both are driving on a bad road.
The vibration state caused by the vibrating means.
Good to change .

Further , another means for solving the above-mentioned problems
As a step, the human machine system for vehicles is
With configuration. That is, the fuel of the engine installed in the vehicle
An injector (36) and a fingertip volume pulse wave of the driver are detected.
The fingertip volume pulse wave detection means (13) and the vehicle are congested
Traffic congestion detection means to detect that you are traveling on a road
(FIG. 17) and a white noise (W) generating
Noise generator (25) and the fingertip volume pulse wave detection
The detection result of the exit means and the detection result of the traffic jam traveling detection means
Based on the control of the fuel injection device (36).
And a control means (26,30,34,35), said control
Means, based on the detection result of the fingertip volume pulse wave detection means
The state quantity (H) related to the psychological state of the driver
The vehicle is in a traffic jam due to the traffic jam detection unit.
If it is detected that you are not driving on
The smaller the state quantity, the sensory stimulation generation control parameter
While setting the value of (I) to a large value, the traffic congestion
The vehicle is traveling on a congested road due to the detection means.
If it is detected that the state quantity is small,
How to reduce the value of the sensory stimulus generation control parameter
While setting the white noise (W),
The frequency increases as the value of the stimulus generation parameter increases.
The amount of decrease in vibration power due to
A variable filter (26) having a filter characteristic (FIG. 12) is used.
Processed by the fuel injection device
Increase the engine speed (RPM) of the vehicle
To control according to the fundamental frequency (ω) that increases with
At this time, the processed whiteno
By applying vibration according to the fuel
It is characterized by adding fluctuation to the state .

In a preferred embodiment, the control means
Indicates that the vehicle is in a traffic jam due to the traffic jam detection means.
If the vehicle is detected to be traveling on the road, it will continue for a certain period of time.
If the detection status continues,
Also detects that the vehicle is driving on a congested road.
The fuel injection device
It is advisable to change the fuel injection state depending on the position .

[0014]

[0015]

[0016]

Human machine for vehicles having the above-described configurations
According to the system, the machine function is
Depending on the state of, it operates to adapt or de-adapt to human sensibilities in order to change that state .
Driver, the interaction between human and machine
It is possible to guide all driving conditions .

[0018]

DESCRIPTION OF THE PREFERRED EMBODIMENTS [Outline of the Invention] <Basic Functions of the Invention> The human machine system of the present invention is different from the conventional adaptive system in that It does not directly affect the driving behavior of the driver. (2) In order to promote the affinity between the driver and the vehicle, the vehicle is being simulated. It is based on the guideline.

Therefore, the human-machine system of the present invention has the following basic functions as a man-machine system. (1) Adaptive function A function that detects the sensitivity of the driver and adapts the functions of the machine such as physical elements, control characteristics, environment, etc. to the sensitivity so that the driver is comfortable with respect to the state.

The detection of kansei here means to make an index that can quantitatively measure the human state from the physiological amount, psychological amount, and behavior of the human, and to estimate the human state by using the result. Means In addition, the measurement of comfort is
The evaluation index that matches the psychological state is created from the physiological amount and the action amount (the amount that represents the action), and based on this, the human physiological and psychological states are estimated. In order to realize comfort, a method is used in which a stimulus is presented to a human, the effect on the stimulus is evaluated using an index, and the stimulus amount is controlled so that the human is in the most comfortable state. The so-called idea of biofeedback of arousal level is included in this realization method. (2) Non-adaptive function A function of detecting the sensitivity of the driver and making the functions of the machine such as physical elements, control characteristics, environment, etc. non-adaptive to the sensitivity so that the driver is uncomfortable with the state. (3) Detecting the sensibility of the pseudo-human-type functional driver, detecting the physical element for that state,
A function that makes the driver feel the illusion that the machine has the will or emotion by adapting or not adapting the machine functions such as control characteristics and environment to the driver's sensitivity. <Control Target> The control target of the human machine system of the present invention includes human senses such as visual sense, hearing sense, tactile sense, vestibular sense such as sense of acceleration and gravity, mood of arousal, excitement, and tension, and further pleasure. Feelings of anger and fatigue, fatigue due to driving, attachment to things, tiredness, and a high level of affection, hate, values, ethics, aesthetics, the driver's driving skill and habit. This is the behavior as a feature of.

The machine function to be controlled does not control the function which directly affects the driving which is the work of the driver. That is, the function to be controlled is an indirect mechanical function with respect to the target work and does not directly affect the work. For example, changing the steering characteristics of the vehicle according to the skill of the driver can be said to be direct control because the change greatly affects the driving of the driver. However, even if the steering characteristic is changed and the driver can sense the change, if it is within a range that does not affect the driving of the driver, it is indirect control.

The mechanical functions that indirectly affect are:
There are environmental stimuli such as sound, vibration, heat from air conditioning, air flow rate, and illuminance that directly or sensally act on humans. <Basic Structure of the Invention> The human machine system of the present invention (hereinafter, “Human Mimetek Machine”)
tic Machine, abbreviated as HMM)) is configured to make the driver feel that the reaction is a human intention through the reaction of the machine. Therefore, in order to express the interaction between a person and a machine, the pattern of communication of intentions that is performed between human beings is realized between the person and the machine.

In the man-machine system, a relationship similar to the relationship between close human beings exchanging messages with each other, that is, a human being obtains the most psychological satisfaction through interaction with a close person, in the man-machine system. Realization will solve the above-mentioned subject. FIG. 1 shows the pattern of interaction obtained by observing the interactions between close people. In the figure, the attentive reaction means always paying the utmost attention to the state and reaction of the other person, and the personal reaction means to respond with attentiveness and assertion while paying attention. means. The emotional reaction is a selfish reaction that sometimes gives priority to emotions, and the obedience reaction is a reaction that a respected person tries to obey.

Based on these observations, the HMM of the present invention returns a human-like reaction to the driver by sometimes returning an emotional and personal reaction while the machine responds to the driver's action. It makes you feel like it. FIG. 2 is a block diagram showing the configuration of the HMM of the present invention. The HMM shown in the same figure detects the road environment and detects the states of the vehicle control unit 1 and the driver 11 which control the vehicle characteristics to the optimum running state accordingly, and generates a message according to the state. The pseudo human reaction generation unit 5 is included.

The vehicle control unit 1 realizes an optimal vehicle behavior according to the driver's operation and the road environment, and controls for adapting the vehicle to the external environment, such as ABS and four-wheel drive (4WS). ) Is equivalent to this. The pseudo-human reaction generation unit 5 controls the vehicle environment in order to return a human-like reaction to the driver. Here, in order to control the physiological state to the optimal one and to make the user feel the psychological satisfaction by acting on the physiological state and the psychological state of the driver by the reaction of the vehicle, only the message transmission to the driver for the state of the driver is performed. To do. The pseudo human reaction generation unit 5 is a human state detection unit 9 that estimates the driver's physiology and psychological state in real time, a stimulus control unit 6 that determines a human reaction presentation pattern for the driver, and a command from the stimulus control unit 6. The sensory stimulus generator 10 generates a stimulus for the driver 11 based on

The stimulus control unit 6 is composed of an adaptive control unit 7 and a non-adaptive control unit 8. Upon receiving a signal from the vehicle behavior unit 3, the adaptive control unit 7 optimizes the driver's physiology and psychology in real time when the stimulus is generated. A command is issued to the sensory stimulus generation unit 10 for adaptive control so as to keep the state. In addition, the non-adaptive control unit 8 interrupts a command to the sensory stimulus generation unit 10 of the adaptive control unit 7 irregularly based on the accumulation of the vehicle behavior motion history information for a long period of time. Then, the stimulus is selected based on the stimulus generation rule defined in a normative manner, and the sensory stimulus generation unit 10 is instructed to generate a non-adaptive stimulus for the driver. However, the stimulus does not directly control the steering characteristics of the vehicle, but is a stimulus such as sound, vibration, scent, or air-conditioning environment in the vehicle that is not directly involved in vehicle traveling.

The operation of the adaptive control unit 7 expresses an attentive reaction to the driver 11, and the adaptive control unit 7 operates as a master to the non-adaptive control unit 8 to express a compliant reaction. Then, the driver 11 of the non-adaptive control unit 8
The personality and emotional reactions are expressed by adjusting the rules of the reaction to. The non-adaptive control unit 8 accumulates the long-term vehicle behavior motion history information because the memory of past events due to a long-term relationship is
This is to express that it affects the psychological relationship in the current interaction with the other person. For example, when driving at the upper limit of performance continues for a long time, such as driving on a highway, it is possible to make the driver feel a message that prompts the driver to stop driving by controlling the vehicle environment.

In the HMM of the present invention, the pseudo-human reaction generation unit 5 does not directly affect the vehicle control unit 1, and the vehicle control unit 1 concentrates on vehicle control. The control of the vehicle behavior is not directly controlled according to the physiology and psychological state of the driver, but is left to the driver's will. This is based on the idea that the joy of driving a vehicle lies in the independent act of controlling the vehicle by the driver himself. In other words, in this HMM, the physiology and psychology of the driver are affected only by the environmental stimulus that directly affects humans, the driver is taught a driving operation that matches the vehicle behavior, and at the same time, an environment suitable for the teaching is created.

Next, preferred embodiments of the present invention will be described in detail. <First Embodiment> FIG. 3 shows the H according to the first embodiment of the present invention.
It is a block diagram which shows the whole structure of MM. In FIG. 1, the vehicle control unit 2 controls the vehicle behavior unit 3 so as to realize an optimum vehicle behavior according to the operation of the driver 11 and the road environment from the environment 20 detected by the detection unit 4. is there. A vehicle body distortion R, which will be described later, is output from the vehicle behavior unit 3 to the central control unit (CPU) 15 of the stimulation control unit 6 as the vehicle behavior.

The human state detecting section 9 estimates the physiological and psychological states of the driver in real time.
The smoothing circuit 21 for smoothing the pulse wave from the fingertip volume pulse wave detector 13 provided for detecting the human state of the human body, and a calculation for obtaining the human state amount described later based on the result. The calculation unit 22 is included. The stimulus control unit 6 includes an adaptive control unit 7, a non-adaptive control unit 8, and a central control unit (CPU) 15 that controls them. The adaptive control unit 7 optimizes the driver's physiology and psychology in real time during stimulus generation. A command is issued to the sensory stimulation vehicle control unit 12 for adaptive control so as to keep the state. Further, the non-adaptive control unit 8 interrupts the command to the sensory stimulation vehicle control unit 12 of the adaptive control unit 7 irregularly based on the accumulation of the vehicle behavior motion history information for a long period of time.

The sensory stimulation vehicle control unit 12 receives a command from the stimulation control unit 6 and generates a stimulation for the driver 11 based on the human reaction presentation pattern determined by the stimulation control unit 6. FIG. 4 is a block diagram showing a detailed configuration around the CPU 15 that constitutes the stimulation control unit 6. In the figure, the CPU 15 controls the vehicle behavior unit 3 to A
The vehicle body distortion R is received through the / D conversion unit 18, and at the same time, the human state amount H is received from the human state detection unit 9 through the A / D conversion unit 19. The CPU 15 generates a sensory stimulus generation control parameter I described later based on these inputs.

FIG. 5 is a diagram for explaining a method of calculating the vehicle body strain R in the vehicle behavior unit 3. In the figure, the strain gauge 32 installed at the bottom of the vehicle body 31 is
The strain S _{x in} the direction and the strain S _{y in the} y direction are detected. The strain calculation circuit 33 obtains the root mean square of S _x and S _y from the strain detection result of the strain gauge 32, and outputs the value as the vehicle body strain R to the stimulation control unit 6.

FIG. 6 is a diagram showing a method of detecting the human state of the driver in the human state detecting section 9. Here, the pulse wave ((a) in FIG. 6) as the voltage waveform obtained by the fingertip volume pulse wave detector 13 is processed to calculate the index H of the driver 11's frustration and the degree of vignetting. FIG. 6B shows a smoothing circuit 21 for the pulse wave shown in FIG.
Shows a smoothed waveform, and the triangular portion formed by decreasing its amplitude is extracted due to frustration and biting which are characteristic changes of the pulse wave.

The human state quantity H is obtained by performing an operation according to the following equation (1) on the smoothed waveform. As a result of the calculation, the waveform shown in FIG.
The larger the value of, the more irritated or sick the human is, and the smaller the value, the calmer.

[0035]

[Equation 1]

FIG. 7 is a map showing the relationship between the human state quantity H and the sensory stimulus generation control parameter I. In FIG. 7A, the human state quantity H and the adaptive control unit 7 are calculated. And the sensory stimulus generation control parameter I, which has a characteristic that the value of the parameter I decreases as the value of the state quantity H increases. Also, FIG. 7B
Represents the relationship between the human state quantity H and the sensory stimulus generation control parameter I ′ calculated by the non-adaptive control unit 8,
Contrary to (a) of FIG. 7, it has a characteristic that the larger the value of the state quantity H, the larger the value of the parameter I ′.

Next, the procedure for calculating the sensory stimulus generation control parameters I and I'in this embodiment will be described. FIG. 8 is a flow chart showing a main routine for calculating the sensory stimulus generation control parameters I and I ′. In step S1a, the adaptive control unit 7 causes the human state amount H obtained by the human state detection unit 9 to be H. Then, in step S1b, the corresponding sensory stimulus generation control parameter I is calculated with reference to the map shown in FIG. In step S2, it is determined whether or not an interrupt, which will be described later, has occurred. If there is no interrupt, the process proceeds to step S3 to output I to the sensory stimulation vehicle control unit 12 so that comfortable driving can be performed.

However, if it is determined in step S2 that an interrupt has occurred, the sensory stimulus generation control parameter I'is calculated in step S5. This calculation of I ′ is shown in FIG.
Refer to the map shown in FIG. Then, after replacing the parameter I with I'in step S6, the process proceeds to step S3 for outputting I. In step S4, a waiting time T _W is provided to see the effect of outputting the parameter I,
After T _W elapses, the process returns to step S1 again.

The flow chart shown in FIG. 9 shows the interrupt generation routine in step S2 of FIG. In the figure, the CPU 15 determines the vehicle behavior unit 3 in step S10.
Then, the vehicle body strain R is input from the above, and it is determined in the next step S11 whether or not the vehicle is traveling on a rough road. Here, FIG.
As shown in, when the value of the vehicle body strain R calculated by the vehicle behavior unit 3 exceeds the predetermined value R _r for a predetermined time t _r seconds, it is determined that the vehicle is traveling on a rough road.

If the determination in step S11 is YES, the process proceeds to step S12, and it is determined whether or not the parameter I'has been output n times or more consecutively. If this determination result is YES, it is determined that the non-adaptive control has continued for a certain period of time or longer, and it is determined that the driver is in a state where the driver cannot rest for some reason, and the process proceeds to step S13 and step S11.
Cancel the bad road judgment in and perform adaptive control. But,
If the determination in step S12 is NO, it means that adaptive control is being performed even though the vehicle has been running on a rough road for a certain period of time to the extent that it burdens the vehicle body. Therefore, an interrupt is issued in step S14. Generate. Therefore, the determination in step S2 in FIG. 8 described above becomes YES, and non-adaptive control is performed, so that the driver 11 is controlled in a direction that increases frustration and the degree of shock, and the driver 11 is discouraged from driving, resulting in a break. Will be prompted.

FIG. 11 is a block diagram showing the internal structure of the sensory stimulation vehicle control unit 12. In the figure, the controller 34 receives the sensory stimulation generation control parameter I from the stimulation control section 6 and calculates a filter control signal F for selecting the characteristic of the variable filter 26 according to the value. The white noise W from the white noise generator 25 is constantly input to the variable filter 26, and a signal W ′ filtered by a filter characteristic according to the value of F is output. This signal W ′ is converted into a digital signal by the A / D conversion unit 30 and then converted into a signal W ″ by the controller 34.
Entered in.

The variable filter 26 has a 1 / f characteristic as shown in FIG. 12, and as the value of the parameter I increases, the signal W'changes from white noise to 1 / f ⁿ noise (n
Is a real number greater than or equal to 1 and it is preferable to use a value between 1 and 2). The controller 34 further calculates a vibration fluctuation control signal V described later, and the vibration control unit 29.
Is the vibration fluctuation control signal V and the engine rotation signal RP.
According to M, a signal is sent to the vibrator 28 provided under the seat 27 on which the driver 11 sits. As a result, sheet 2
7 is vibrated at a predetermined frequency described later.

FIG. 13 is a map showing the frequency characteristic of the vibration exciter 28 for vibrating the sheet 27. FIG. 13A shows the relationship between the engine rotation signal RPM and the basic vibration frequency ω. Further, FIG. 11B shows the relationship between the vibration fluctuation control signal V and the fluctuation Δω of the vibration frequency. The seat 27 is vibrated at a frequency that changes according to the sum of ω and Δω corresponding to the vibration fluctuation control signal V and the engine rotation signal RPM output from the vibration control unit 29 as described above. It

After all, since the sensory stimulation vehicle control unit 12 drives the shaker 28 which supports the driver's seat in accordance with the sensory stimulation generation control parameter I from the stimulation control unit 6, the driver is provided with vibration stimulation. It is possible to control the irritability and the degree of risk. FIG. 14 is a flow chart showing a procedure for calculating the vibration fluctuation control signal V and the filter control signal F in the controller 34. In step S15 of the figure, the controller 15 determines the filter characteristic from the 1 / f characteristic map shown in FIG. 12 based on the sensory stimulation generation control parameter I from the stimulation control section 6, and corresponds to the characteristic. The filter control signal F is calculated, and the signal F is output in step S16.

The controller 34 is moved up at step S17.
Excitation based on the output signal W ″ from the A / D conversion unit 30 described above.
The fluctuation control signal V is calculated, and the vibration is applied in the subsequent step S18.
The signal V is output to the control unit 29. In step S19,
Time T has passed since the filter control signal F was calculated in the previous processing.
_W It is determined whether has passed. Where time T_W Has passed
If not, return to step S17
Although the vibration fluctuation control signal V is calculated, it takes a predetermined time.
If the error is confirmed, return to step S15 and restart
The ilter control signal F is calculated.

FIG. 15 shows the relationship between the human state and the vibration frequency of the seat in the HMM according to the embodiment. In the figure, when the human state detection unit 9 of the HMM performs adaptive control based on the driver's irritability and the degree of bite, 1 / f ⁿ type (n is a real number of 1 or more, between 1 and 2) It is preferable to use the value of (1)) to provide the fluctuation of the seat vibration frequency, and when the driver is in a quiescent state, the fluctuation of the white noise type is provided. In the case of non-adaptive control,
The fluctuation of the seat vibration frequency opposite to that of the adaptive control is given.

As described above, according to this embodiment,
The driver's annoyance and biteness are detected using the fingertip plethysmogram, and vehicle distortion is detected as the vehicle behavior.
The seat fluctuation vibration corresponding to these values is used as a stimulus to perform adaptive control so that the driver can drive comfortably, and when it is determined that the vehicle is traveling on a rough road, the predetermined seat fluctuation vibration is used. Is used as a stimulus to perform non-adaptive control, which has the effect of prompting the driver to rest. <Second Embodiment> A second embodiment according to the present invention will be described. Note that the overall configuration of the HMM according to this embodiment is the same as that of the HMM according to the first embodiment shown in FIG. 3, so description thereof will be omitted.

FIG. 16 is a block diagram showing a detailed configuration around the CPU 15 which constitutes the stimulation control unit 6 of the HMM.
In the figure, the CPU 15 receives the vehicle speed V from the vehicle behavior unit 3 via the A / D conversion unit 18, and at the same time receives the human state amount H from the human state detection unit 9 via the A / D conversion unit 19. The CPU 15 generates the sensory stimulus generation control parameter I based on these inputs.

The method of detecting the human state of the driver by the human state detecting unit 9 and the human state amount H in this embodiment.
And the sensory stimulus generation control parameter I are the same as the detection method (FIG. 6) and the relationship between the state quantity H and the parameters I and I ′ (FIG. 7) in the first embodiment. Next, a procedure for calculating the sensory stimulus generation control parameters I and I ′ in this embodiment will be described.

Since the main routine for parameter calculation in this embodiment is the same as the parameter calculation procedure in the first embodiment shown in FIG. 8, only the interrupt generation routine will be described here. FIG. 17 shows an interrupt generation routine according to the present embodiment. In step S21,
The CPU 15 determines whether or not the vehicle is in a traffic jam. Here, as shown in FIG. 18, when the vehicle behavior unit 3 keeps the vehicle speed V at or below a certain constant vehicle speed V _J for a predetermined time t _J seconds or more, it is determined that there is congestion.

If the determination in step S21 is YES, the process proceeds to step S22, and it is determined whether or not the parameter I'has been output n times or more consecutively. If this determination result is YES, it is determined that the non-adaptive control has continued for a certain period of time or longer, and it is determined that the driver is in a state where the driver cannot rest for some reason, and the process proceeds to step S23 and step S21.
Cancel the traffic congestion judgment at and perform adaptive control. But,
If the determination in step S22 is NO, it means that the adaptive control is being performed even though the traffic jam continues for a certain time or longer, so that an interrupt is generated in step S24.

By this processing, the determination in step S2 in FIG. 8 becomes YES, and non-adaptive control is performed, so that the driver 1 is controlled in the direction of increasing the annoyance and the degree of vibration.
For No. 1, the drive will be discouraged and a break will be prompted. FIG. 19 is a block diagram showing the internal configuration of the vehicle control unit 12 for sensory stimulation of the HMM of this embodiment. In the figure, the controller 34 receives the sensory stimulation generation control parameter I from the stimulation control section 6 and calculates a filter control signal F for selecting the characteristic of the variable filter 26 according to the value. The white noise W from the white noise generator 25 is always input to the variable filter 26,
The signal W ′ filtered by the filter characteristic according to the value of F is output. This signal W ′ is supplied to the A / D conversion unit 30.
After being converted into a digital signal at, the signal W ″ is input to the controller 34.

The variable filter 26 is similar to that of the first embodiment.
The signal W ′ has 1 / f characteristics as shown in FIG. 12, and as the value of the parameter I increases, the signal W ′ changes from white noise to 1 / f ⁿ noise (n is a real number of 1 or more and between 1 and 2). It is preferable to use the value). Controller 3
4 calculates the fuel injection control signal E based on the signal W ″ as described later, and sends the calculation result to the EGI control unit 35. Then, the fuel injection device 36.
Performs fuel injection according to the control signal from the EGI control unit 35.

FIG. 20 shows the fuel injection control signal E and the EGI control unit 35 to the fuel injection device 36.
Is a map showing the relationship with the fuel control signal sent to the.
(A) of the figure shows the relationship between the fuel injection control signal E and the fuel injection amount, and (b) shows the relationship between the fuel injection control signal E and the injection interval. The safe allowable range in the figure is provided in order to prevent the fuel injection amount and the fluctuation amount of the injection interval from extremely changing with respect to an appropriate value. It does not take a large value.

The HMM sensory stimulation vehicle control unit 12 controls the fuel injection amount or the fluctuation of the injection interval according to the sensory stimulation generation control parameter I from the stimulation control unit 6 to control the engine output, and the driver. Since not only the engine output but also the engine sound and vibration are given to, the irritability and the degree of shock can be controlled.

FIG. 21 is a flow chart showing a procedure for calculating the fuel injection control signal E and the filter control signal F in the controller 34. In step S25 of the figure, the controller 15 uses the sensory stimulation generation control parameter I from the stimulation controller 6 as shown in FIG.
A filter characteristic is determined from the / f characteristic map, a filter control signal F corresponding to the characteristic is calculated, and step S2 is performed.
The signal F is output at 6.

The controller 34 is moved up at step S27.
Based on the output signal W ″ from the A / D conversion unit 30 described above, the fuel is
The injection control signal E is calculated, and the subsequent step S28 is performed.
Then, the signal E is output to the vibration control unit 29. Step S29
Now, after calculating the filter control signal F in the previous processing,
Time T _W It is determined whether has passed. Where time T_W But
If it is determined that the time has not elapsed, go to step S27.
Returning to this, the fuel injection control signal E is calculated,
If it is confirmed that the predetermined time has passed, go to step S25.
Then, the filter control signal F is calculated again.

FIG. 22 shows the relationship between the human state and the fluctuation of fuel injection in the HMM according to this embodiment. In the figure, when the human state detection unit 9 of the HMM performs adaptive control based on the driver's irritability and the degree of bite, 1 / f ⁿ type (n is a real number of 1 or more, between 1 and 2) It is desirable to use the value of)
Gives white noise type fluctuations when the driver is in a quiescent state. In the case of non-adaptive control, the fluctuation of fuel injection opposite to that in adaptive control is given.

As described above, according to this embodiment,
The driver's annoyance and biteness are detected using the fingertip plethysmogram, the vehicle behavior is judged based on the vehicle speed, and the congestion is judged, and the fuel injection fluctuations corresponding to these values determine the engine output or engine. Sound and vibration are used as stimuli to perform adaptive control so that the driver can drive comfortably, and when it is determined that the traffic congestion has continued for a certain period of time, a stimulus due to a predetermined fuel injection fluctuation is used. This has the effect of performing non-adaptive control and prompting the driver to rest. <Reference Example> The following describes Reference Example according to the present invention.

In this reference example , the driver's arousal level is detected, and a stimulus corresponding to the arousal level is presented to the driver to keep the driver in an optimal arousal state. Figure 23 is San
It is a schematic block diagram showing the overall structure of an HMM according to Reference Example. In the figure, the vehicle control unit 2 controls the vehicle behavior unit 3 to realize an optimum vehicle behavior according to the operation of the driver 11 and the road environment from the environment 20 detected by the detection unit 4.

The human state detecting section 9 detects the awake state of the driver 11 based on the brain waves from the driver 11. The adaptive control unit 7 issues a command to the sensory stimulation vehicle control unit 12 in order to perform adaptive control so as to maintain the driver in an optimum physiological and psychological state in real time when the stimulus is generated. Further, the sensory stimulation vehicle control unit 12 receives a command from the adaptive control unit 7 and generates a stimulus to the driver 11 based on a presentation pattern of a human reaction to the driver.

FIG. 24 is a detailed block diagram of the HMM of this reference example . In this case , the electrode 2a is attached to the head of the driver 11 and the potential difference is measured to measure the brain waves of the driver. . The electroencephalogram of the driver 11 is guided to the physiological measurement amplifier 43 to undergo signal amplification, and the A / D
After being converted into a digital signal by the converter 44, the CPUb
Sent to 42. The CPUb42 calculates an awakening degree estimated value X, which will be described later, and delivers the estimated value X at that moment to the CPUa41 in accordance with a command from the CPUa41 for controlling the entire system and for awakening degree control.

The CPUa 41 constantly monitors the awakening degree estimated value X, and at the same time, measures the engine sound of the engine 45 by the microphone 46 provided in the engine room 54. The measured engine sound is converted into a digital signal by the A / D converter 47 and taken into the CPUa 41.
When the CPUa41 determines that the driver is in the low awakening state based on the value of the awakening degree estimated value X, the CPUa41 sends the taken engine sound to the D / A converter 51 to convert it into an analog signal and to the amplifier 50. It sends a gain control signal to control its amplification. And the amplifier 50
The engine sound amplified according to the control signal from the CPU 1a is presented to the driver 11 as a stimulus via the speaker 53. The amplification degree of the amplifier 50 is, as described later,
The CPUa 41 determines in accordance with the monitoring result of the engine sound output level from the speaker 53 picked up by the ear microphone 52 installed near the ear of the driver 11.

On the other hand, when the CPUa 41 determines that the driver 11 is in an overstrained state, the scent presenting device 48
Start and present a calming scent to the driver. It should be noted that the driver 11 can operate the end switch 49 by itself to stop the driving of the system when the driving of the present system takes a long time and the physical fatigue is accumulated.

Next, the wakefulness continuation processing procedure in the HMM of this reference example will be described. FIG. 25 is a flow chart showing the procedure of the optimum arousal level sustaining process according to the reference example . In the figure, in step S31, the following values and parameters are set as initial values. (1) Optimal arousal level estimation value Rbest (2) Tension level limit value R1 (3) Awakening reduction level value R2 (4) Allowable control arousal level estimate value range Rerr1 (for tension level) (5) Allowable control arousal level estimate value range Rerr2 (for low arousal level) (6) Engine sound capture time T1 (7) Speaker volume E0 (8) Safety limit engine volume Esafe (9) Engine-on playback sound rate Δ0 (10) Safety limit counter Nsafe-lim (11) Engine sound reproduction time Tsafe at the safety limit (12) Limit number of scent presentation Ns-lim (13) Individual arousal level estimation parameters a1, a2, a3,
a4, a5 (14) HMM drive time counter Nl (initial value is set to 1) (15) HMM drive time upper limit counter Nlset Here, the relationship between the awakening degree and the awakening degree estimated value will be described.

FIG. 26 illustrates the relationship between the arousal level and the estimated arousal level, and the two have a monotonically increasing relationship with each other. When the estimated arousal value is Rbest, it is assumed that the person is in an optimal arousal state, and Rrr1 is placed on the hypertension side around Rbest,
When the range of Rerr2 (hatched portion in the figure) is defined on the side of lowering awakening, the normal awakening state is set when the estimated awakening level is within this range.

Further, when the awakening degree estimated value is larger than R1, the state is overstrained, and when it is smaller than R2, the state is low awakened. In the next step S32, the awakening degree estimated value X1 is fetched, and in the following step S33, it is determined whether or not X1 is larger than R1, that is, whether or not there is an excessive tension.
If the determination here is YES, it is determined that the robot is in an overstrained state, and the process proceeds to step S36 to execute an overstress handling process described later.

However, if the determination in step S33 is NO, whether X1 is smaller than R2 in step S34,
That is, it is determined whether or not the person is in a low awakening state. If it is determined that the vehicle is not in the low awakening state, the process returns to step S32 and the awakening degree estimated value X1 is fetched. On the contrary, if it is determined in step S34 that the vehicle is in the low awakening state, the process proceeds to step S35 to execute a low awakening coping process described later.

In step S37, the value of the HMM drive time counter Nl is set to the HMM drive time upper limit counter Nlset.
Is larger than the value of. The determination here is NO
If so, it means that the optimum wakefulness duration process has not been executed the set number of times.
After the value of is incremented by 1, the process returns to step S32. However, the value of the counter Nl is equal to the value of the counter Nlset.
If it exceeds, the determination in step S37 is YES.
Then, in the next step S39, the pressing of the end switch 49 by the driver is detected. If the switch is pressed, this process is terminated because the driver intends to stop driving the HMM.

Next, a method for estimating the arousal level as a numerical value from the electroencephalogram in the HMM of this reference example will be described. FIG. 27 is a block diagram of the awakening degree estimation unit in the CPUb 42 that constitutes the HMM of this reference example . Here, the degree of human arousal is quantified absolutely, and the vibration,
By giving the driver a stimulus such as sound and scent, the driver's arousal state is maintained at a certain level in a biofeedback manner, and it is possible to prevent falling asleep or excessive excitement.

In FIG. 27, the electroencephalogram detecting section 51a is not tested.
From the electrode 2 mounted on the head of the driver 11 who is a person
Is detected, and the part that amplifies the electroencephalogram by an amplifier (not shown)
It is composed of Is the electroencephalogram processing unit 52a the electroencephalogram detection unit 51a?
Into a physical quantity that is easy to estimate arousal by processing the brain wave signals
Signal processing. That is, it was amplified by the electroencephalogram detection section 51a.
The EEG signal is taken in via the A / D converter described later and
Delta wave of the electroencephalogram (1-3H
_Z ), Theta wave (3-6H_Z ), Alpha wave (8-13H _Z ), Β
Wave (13H_Z -30H_Z ) Frequency band
Process to the physical quantity required for evaluation.

The arousal level estimation unit 53a calculates an estimated value deeply related to arousal based on the arousal level estimation parameters and the electroencephalogram band data described later. In addition, the awakening degree determination unit 54a determines to what extent the stimulus is presented based on the awakening degree estimated by the awakening degree estimation unit 53a. Then, the stimulus presentation unit 50
The a presents an appropriate stimulus to the driver or stops the driver in response to an instruction from the awakening degree determination unit 54a.

The arousal level estimation unit of this reference example estimates the arousal level based on the electroencephalogram and the reaction time, and uses the information to provide the driver with an optimal stimulus in a biofeedback manner. Therefore, the concept of the biofeedback applied to this reference example will be described with reference to FIG. Figure 28 is a diagram often used in psychology and ergonomics, the vertical axis is work efficiency,
It also represents processing capacity, and if limited to driving, it corresponds to the target steering wheel operation and judgment. In addition, the horizontal axis indicates the amount of stimulation, and visual stimulation, auditory stimulation, tactile stimulation, olfactory stimulation, etc. can be considered. These are similar to the scenery flow in front, noise and vibration in the car,
Or it will be a smell.

The hatched portion B in FIG. 28 means that the driver whose stimulus amount increases more than a certain amount cannot perform an accurate operation, and eventually falls into a panic state (hyperawakening). ing. As an example, when you increase the vehicle speed at high speed, you will eventually be unable to keep up with changes in the road environment and you will be unable to operate the steering wheel, or you will be unable to avoid when a spin occurs. be able to.

The hatched portion A in the figure means that the stimulus is small in spite of the fact that attention is required, so that the subject falls into a monotonous state and causes a decrease in arousal. An example of this is a state where awakening is lowered during monotonous driving such as on an expressway. By the way, what the conventional drowsiness alarm device aims at is to give an alarm to the driver when this area is reached.

As can be seen from the above description, human beings cannot cope with a certain amount of information and environmental changes, and similarly, awakening is lowered for a stimulating amount of a certain amount or less. From this fact, the most desirable safety device is to clarify the optimal stimulation amount for each person, though it varies depending on the individual person, and adjust and give an effective stimulation according to the physiological state in the vicinity of the stimulation amount. That is. The situation is shown in the center of FIG. 28 as the optimum stimulus amount.

In this reference example, it is possible in principle to achieve the optimal awake state shown in FIG. 28 by detecting the human awake state and applying a stimulus corresponding thereto. Therefore, first, a wakefulness estimation parameter determination device that determines parameters for calculating the wakefulness by the wakefulness estimation unit 53a will be described. The arousal level estimation parameter determination device determines the arousal level estimation parameter by simultaneously measuring the reaction time and the EEG that have a high correlation with the arousal level and performing multiple regression analysis based on the obtained EEG data and the reaction time data. To do.

FIG. 29 is a schematic block diagram of the awakening degree estimation parameter determination device. This figure shows the situation in which this device simultaneously measures human reaction time and EEG over a long period of time, and it is possible to obtain estimated parameters by analyzing the correlation between the obtained reaction time and EEG. it can. Figure 2
9, the electroencephalogram processing unit 70 uses the subject (driver) 1
The electroencephalogram from the electrode 2 attached to 1 is measured, and the reaction time measuring unit 71 measures the time during which the subject 11 reacts to the given stimulus. Based on these measurement results, the wakefulness estimation parameter determination unit 72 estimates the wakefulness unique to the subject.

FIG. 30 is a block diagram showing the structure of the reaction time measuring unit 71 for measuring the reaction time (selective reaction time). In the figure, the reaction stimulus presentation unit 30A generates a reaction measurement stimulus to the subject 11 under the control of the selective reaction time measurement unit 35A, and the reaction from the subject 11 is selected reaction via the reaction input unit 31A. It is input to the time measuring unit 35A.

The selective reaction time measuring section 35A sets a main control section 36A including a CPU 36a for controlling the entire apparatus and a storage section for storing various data, and sets a time for presenting a stimulus for reaction and measures a reaction time. Timer 37A for
It is composed of As the data storage unit, a stimulus presentation time data storage unit 36a for storing data relating to the time at which the reaction stimulus is presented, a stimulus pattern data storage unit 36c for storing the reaction stimulus pattern, and the measured value. There is a reaction time data storage unit 36d for storing the reaction time. A disk device or a semiconductor memory is used as a storage medium for these storage units.

The reaction stimulus presentation section 30A can generate a CRT 30a for generating a stimulus as an image, a speaker 30b for generating a stimulus sound, and a tactile stimulus for compressing a human body by incorporating a vibrating body. It is composed of a seating portion 30c on which the subject can easily sit so as to respond to the stimulus. On the other hand, the reaction input unit 31A has a button 31a for the subject 11 to respond to the stimulus with the finger 1c and a microphone 31b for reacting with the voice. The reaction from the reaction input unit 31A is the selected reaction time measuring unit 35.
It is stored in the reaction time data storage unit 36d as reaction time data via the CPU 36a of A.

FIG. 31 shows an example in which a CRT is used in the reaction stimulus presentation section as a reaction time measurement situation. In the figure, the subject 11 sits on the seat 30c and holds the bar 40A having a plurality of reaction buttons (not shown) attached thereto. The reaction stimulus is displayed on the CRT 30a in front of the subject 11 and is set to disappear momentarily. When the stimulus is displayed on the screen of the CRT 30a, the test subject 11
You will be tasked with selecting and pressing the buttons as soon as possible according to established rules.

On the CRT 30a, the colored circles set according to the stimulus presentation time and the stimulus pattern data set in the selective reaction time measuring section 35A are momentarily large and bright enough to be seen by the subject 11. To be presented There are three types of colors, and the order in which these colors are presented is set randomly. Also, the displayed time interval is set randomly. In addition, the type of color at this time may be any color.

[0084] Next, the procedure of measuring definitive reaction time in this reference example in accordance with the flow chart shown in FIG. 32. At step S30a in the figure, TC is set as the stimulus presentation time t1, the measurement end time, and the missed time when the awakening level decreases and the reaction is delayed. Stimulus presentation time t1
Is obtained by generating a random number, and the presentation time interval may be uniformly selected at random from 5 seconds to 30 seconds, but this time interval and distribution may be set arbitrarily.
Further, it is desirable to set the measurement time to 30 minutes to several hours.

Next, in step S31a, a stimulus presentation pattern indicating which color is selected as a stimulus is randomly set. Here, the circle marks with colors (red, blue, yellow) are set. However, the randomness of the stimulus generation may be biased, and may be presented in a certain order in extreme situations. Depending on the level of attention, it is recommended to set the stimulus pattern so that it can be predicted easily when increasing monotonousness.

At step S32a, the timer 37A is set to C.
The PU 36a is notified of the current time, and the subsequent step S33a
Then, the coincidence between the stimulus presentation time t ₁ set in step S30a and the current time ta is checked. If the result of the determination here is YES, that is, when the set stimulus presentation time is reached,
In step S34a, the subject is presented with a predetermined stimulation pattern. By doing this, it is possible to put weight on the stimulus.

In step S35a, the timer 37A notifies the CPU 36a of the current time tb after the stimulus is presented. Then, the reaction signal from the subject 11 is detected in step S36a, and if there is a reaction, tb-ta is calculated as the reaction time in the subsequent step S37a. However, step S3
If the reaction signal from the subject 11 cannot be detected in 6a, the process proceeds to step S38, where the awakening degree of the subject decreases,
As a missed time when the reaction is delayed, Tc set in advance and the waiting time tb-ta are compared. If this waiting time does not exceed Tc, the process returns to step S35a, the current time tb is input, and the next step S36a waits for a reaction signal again. If the result of the determination in step S38a is YES, that is, if the waiting time exceeds Tc, it is determined to be overlooked, and the reaction time is set to Tc for convenience in step S39a.

In step S40a, the reaction time data for the obtained stimulus presentation time is stored in the reaction time data storage unit 3
In step S41a, the CPU 36a determines whether the measurement end time has come, and the determination result is N.
If it is O, the process returns to step S30a to repeat the above process until the set time comes. However, if the measurement end time can be detected in step S41a, this processing ends.

FIG. 33 is a block diagram showing the configuration of the electroencephalogram processing section 70. In the electroencephalogram detecting section 70a shown in FIG.
The brain wave electrode is connected to the human head with an adhesive having a dedicated conductivity, and the resulting brain wave is taken as a noise countermeasure by, for example, a head amplifier (corresponding to the physiological measurement amplifier 43 in FIG. 24). And amplify the signal. Brain wave detector 7
The electroencephalogram signal from 0a is the A / D converter 61 (A in FIG. 24).
Corresponding to the / D converter 44) and stored in the brain wave data storage unit 62 as brain wave data. This data is separated by a digital filter (63a to 63d) into a δ wave band, a θ wave band, an α wave band, and a β wave band, and each band data is stored in a filtering data storage section (64a to 64d).

Next, the power amount calculator (65a to 65d)
The average power amount is calculated from the electroencephalogram data of each band. The average time T _P (about 1 second) at this time is determined in a timely manner. In order to use the average power data thus obtained as the average power amount having a high correlation with the arousal level, the smoothing processing units (66a to 66)
Smoothing in d). A method of setting the magnitude of the smoothing time at this time will be described later. The average power data is the power data storage section (67a to 67d) for each band.
Stored in.

Regarding the electroencephalogram smoothing process, in order to correspond to the reaction time, the power amount from the stimulus presentation time to the electroencephalographic smoothing time before is averaged to obtain the electroencephalogram power amount at that time. Here, the EEG smoothing time (T
δ, Tθ, Tα, Tβ) must be determined. These parameters are the electroencephalogram after smoothing (4 variables).
It is necessary to set the one that has the highest correlation between and the reaction time. Therefore, the reaction time is used as the objective variable, and the four variables of the δ wave power amount, the θ wave power amount, the α power amount, and the β wave power amount are used as explanatory variables. I do. Of the multiple correlation coefficients obtained for various combinations of smoothing parameters in this way, the maximum multiple correlation coefficient is clarified, and the combination of smoothing parameters at that time is determined as the optimum smoothing time for the electroencephalogram. At the same time, the parameters a1, a2, a3, a4 obtained from the multiple correlation analysis and the multiple regression equation using the parameters are defined as the arousal level estimation equation. This is because the reaction time is predicted, but it can be considered that the reaction time and the arousal level correspond to each other with high correlation.

Next, although the arousal level can be quantitatively calculated, it is necessary to determine to what threshold value the stimulus presentation should be started. The procedure of determination at this time is shown in the flow chart of FIG. At step S80a in the figure, the reaction time data is fetched, and at the next step S81a, the reaction time data is logarithmically converted. The average value A and the standard value σ are calculated in step S82a. This is because it is generally known that the distribution of reaction time is a Gaussian distribution. In order to know that the awakening has deviated significantly from the average, step S8
In 3a, A + 2σ is set from the threshold reaction time Tth = A + σ, which makes it possible to judge that when the reaction time becomes late, the awakening is approaching. Also,
Similarly, when the reaction time Tth = A−σ becomes smaller than A−2σ, it is possible to judge that the person is considerably awake. Based on this idea, Tth is used as the value of the threshold arousal level at which the arousal stimulus is started to be applied.

Obtained Tth and Tδ, Tθ, Tα, Tβ
And a1, a2, a3, a4 and the multiple regression equation using the same are used as the arousal level estimation parameters. Next, each component of the awakening degree estimation unit in the CPU b42 that constitutes the HMM of this reference example will be described in detail. <Electroencephalogram detection unit> An electrode for picking up the electroencephalogram is connected to the head of the subject (driver) with a dedicated conductive adhesive, and the obtained electroencephalogram is amplified by a head amplifier (not shown) to prevent noise. After the application, it is amplified by an electroencephalogram amplifier (not shown). <Electroencephalogram Processing Section> FIG. 35 shows a detailed block diagram of the electroencephalogram processing section 52a.
Since the electroencephalogram processing unit in the figure has the same function as the electroencephalogram processing unit shown in FIG. 33 except for the method of calculating the power amount of the electroencephalogram, the same components are designated by the same reference numerals, and detailed description thereof is omitted here. The description is omitted. <Arousal level estimation unit> The arousal level estimation unit 53a uses the arousal level estimation parameter obtained by the arousal level estimation parameter determination device shown in FIG. 29 and the multiple regression equation to process the electroencephalogram processed by the electroencephalogram processing unit 52a. Awakening degree is calculated from each band power data of. Then, the value is delivered to the awakening degree determination unit 54a.

The arousal level is expressed by the following estimation formula.
That is, arousal level = a1 * P [delta] + a2 * P [theta] + a3 * P [alpha] + a4 * P [beta] + a5 (2-1) Further, the arousal level estimated value is calculated by the following awakening level estimated value = (-1) x awakening level (2-2).

An example of estimation of arousal level by the above method is shown in FIG.
And shown in FIG. FIG. 36 shows the maximum multiple correlation coefficient, the reaction time, and the electroencephalogram smoothing time obtained for the vibration conditions of each subject and the vibration exciter. In the figure, numerical values represent, in order from the left, the multiple correlation coefficient, the smoothing time (second) of the reaction time and the smoothing time (second) of the electroencephalogram when the coefficient was obtained.
The multiple correlation coefficient was rounded off to two decimal places.

FIG. 37 shows the result of FIG. 36 as a change with time of the reaction time and the predicted value of the reaction time by the multiple regression equation. In the figure, the vertical axis represents reaction time and the horizontal axis represents elapsed time. In addition, in the figure, the dotted line indicates the reaction time, and the solid line indicates the estimated value from the electroencephalogram. The breaks between the data represent breaks. In this example, the multiple correlation coefficient between the reaction time and the electroencephalogram is 0.89, which means that the method described above is effective for estimating the arousal level, and that the arousal level can be estimated in detail. is doing. The magnitude of the multiple correlation coefficient varies from 0.7 to 0.9, depending on the subject.
It can take between eight.

<Awakening Level Determining Unit> The processing procedure in the awakening level determining unit 54a will be described with reference to the flow chart shown in FIG. In step S81 of the figure, the awakening degree estimated by the awakening degree estimating unit 53a is fetched, and step S82
The reaction time Rt corresponding to the arousal level is estimated at. here,
Since Rt once exceeded Tth, it is dangerous to immediately determine that the wakefulness has changed. This is because it is possible that there are variations due to some factor.

Therefore, in step S83, the threshold frequency N _S for signal output is set. Then in step S84, Rt
The value of the frequency counter N indicating the frequency at which T exceeds Tth is set to 0.
And That is, Rt is Tt more than a certain constant frequency N _S.
When the value exceeds h, it is determined that the awakening is decreased. However, the sampling time of Rt can be set at a proper time. The Rt thus determined is used as a criterion for lowering awakening.

After N _S is determined in step S83 and the frequency counter value is set to 0, in step S85 the short-term EEG data is acquired. Then, in step S86, the reaction time estimated value Reeg (awakening degree) is calculated using the estimated parameter and the multiple regression equation. In step S87, Reeg and Rt
Are compared with each other, and if the reaction time estimated value Reeg does not exceed Rt, the process returns to step S85, the brain wave is captured again, and the process of calculating the reaction time estimated value Reeg is repeated. However, in step S87, the estimated reaction time value Reeg is R
If t is exceeded, 1 is added to the frequency counter N in step S88, and the value N of the frequency counter is compared with the set frequency Nt in step S89. If the determination here is NO, the process returns to step S85, but YES
If so, that is, the frequency N set by the value N of the frequency counter
If s is exceeded, it is determined that the awakening is abnormal, and the subsequent step S9
When the value is 0, an awakening lowering signal which means lowering of awakening is output. This awakening lowering signal is sent to the stimulus presentation unit.

At step S91, it is determined whether the processing is completed. If Reeg exceeds Rt and does not exceed Rt for a while, a mechanism for resetting the counter N as noise may be attached. <Stimulation presentation unit> The stimulation presentation unit 50a is an awakening degree determination unit 54a.
Driver 11 who is the subject after receiving the awakening lowering signal from
The sound, vibration, scent, and other stimuli with awakening effect are output for a certain period of time until the stop signal comes. At this time, the type of the stimulus and the presentation method may be those obtained by the awakening stimulus setting device described later.

FIG. 39 is a block diagram showing the structure of the awakening stimulus setting device. The present arousal stimulus setting device determines the type of stimulus most suitable for the individual and the method of presenting the stimulus in advance, because the stimulus that enhances the arousal effect may vary depending on the individual. In this device, the electroencephalogram processing unit 102, the reaction time measurement unit 103, and the awakening degree estimation parameter determination unit 101, excluding the stimulation presentation unit 104 and the stimulation parameter setting unit 110, have the same functions as the awakening state determination device shown in FIG. 29. Therefore, the description thereof will be omitted.

In FIG. 39, the subject 11 is caused to perform a selective reaction work, and brain waves at that time are simultaneously recorded. Then, parameters are determined by the arousal level estimation parameter determination unit from the obtained reaction time and EEG, and the arousal level is estimated. If the subject 11 is evaluated by the awakening degree estimation parameter determination unit 72 of the awakening state determination device shown in FIG. 29 before being evaluated by this device, the awakening degree can be estimated immediately.

In the stimulation parameter setting unit 110, the kind of stimulation and the presentation method for determining the optimum stimulation state are provisionally set in advance. Then, the stimulation presentation unit 104
When it is determined that the state of awakening is low, the stimulation parameter setting unit 11
In accordance with the data set to 0, a sound is emitted from the speaker 105a via the speaker driving unit 105, the seat 106a including the vibrating body is vibrated at a high frequency via the vibration driving unit 106, or the vibration driving unit 107 is operated. The vibrating device 107a is vibrated at a low frequency through the vibration to present the stimulus to the subject 11. Then, the arousal level of the subject is estimated from the electroencephalogram when the stimulus is given, and the arousal level estimation values before and after the stimulus presentation are compared to see whether or not there is an effect.

By changing the stimulus state and the type of stimulus, the stimulus most effective for awakening control can be selected,
Furthermore, based on the obtained results, the awakening of the individual can be maintained and the effect of awakening control can be improved. Next, FIG.
The low alertness coping process corresponding to step S35 of the flowchart of FIG. 25 will be described in detail with reference to the flowchart shown in FIG. In the figure, step S10
1 resets the safety limit sound level counter, that is, NΔ
To 1. In the next step S102, the speaker generated sound effective value Ea indicating the power of the noise amount (amplified engine sound) is calculated as described later.

In step S103, the effective value E of the speaker generated sound obtained in step S102 is used as a safety volume check.
a and the safety limit engine volume Esa set as the initial value
Compare the level with fe. If it is determined that Ea is greater than Esafe, the sound generated by the speaker exceeds the safety limit and is dangerous, so the process proceeds to step S104, where the sound generation process at the safety limit described later is executed and the process returns to step S102. .

However, in step S103, Ea changes to Esa.
When it is determined that the value is smaller than fe, step S10
At 5, the speaker sound described later is generated, and at the subsequent step S106, the awakening degree estimated value X2 is fetched according to the awakening degree estimating method described above. In step S107, it is determined whether or not the driver's arousal level has reached the optimum arousal level due to the generation of the speaker sound, that is, the arousal level estimated value X2 and Rbest-.
A comparison with the value of Rerr2 is made. Where X2 is Rbe
If it is smaller than st-Rerr2, it is determined that the driver has not returned to the normal awake state, and the process returns to step S102 again. However, if it is determined that X2 is larger than Rbest-Rerr2, the driver determines that the low wakefulness state has returned to the normal wakeful state, and ends the present process.

FIG. 41 is a flow chart showing the procedure for calculating the speaker effective sound value. In the same figure, in step S111, the CPUa 41 captures the engine sound e _in (t) via the microphone 46. Here, the acquisition of the engine sound is changed from e _in (t ₀ ) to e _in (t ₀ + T).
Go to ₁ ). In the following step S112, the engine sound emitted from the speaker 53 through the ear microphone 52 installed near the driver's ear (this is the ear sound S
(T). This ear sound S (t) is taken in from S (t ₀ ) to S (t ₀ + T ₁ ).

In step S113, the effective value E ₀ of the engine sound captured in step S111 is calculated according to equation (3), and in step S114, the ear sound captured in step S112 according to equation (4). Effective value Se
Calculate ar.

[0109]

[Equation 3]

[0110]

[Equation 4]

In step S115, the engine sound reproduction increase rate Δe is calculated according to the following equation. .DELTA.e = .DELTA..multidot. (X-Rbest) .multidot.N.DELTA. (5) Then, in the next step S116, the reproduction engine sound effective value (speaker generated sound effective value) Ea is calculated according to the equation (6). Ea = E ₀ · Δe (6) In step S117, the sound pressure Eb during reproduction, which is the effective value of the ear sound during engine sound reproduction, is calculated according to equation (7) for the purpose of protecting the driver's ears. .

[0112]

[Equation 7]

FIG. 42 is a flow chart showing the speaker sound generation processing. CPU in step S121 of FIG.
a41 is the step S of the float chart shown in FIG.
113, and the effective value E _{0 of} the engine sound obtained in S116
And playback engine sound effective value (speaker generated sound effective value) Ea
From the above, the gain G of the amplifier 50 is calculated by the following formula. G = Ea / E ₀ (8) At the next step S122, the gain control signal e _out
Calculate (t). This signal is calculated by the following formula.

E _out (t) = G · e (t) (9) FIG. 43 is a flow chart showing the procedure for processing the sound generated at the safety limit. In the figure, in step S125,
Reset the safe sound reproduction time counter (Nsafe =
1). Next, in step S126, the safety limit sound is reproduced.
The safety limit engine sound volume Esafe is reproduced with, and in the following step S127, the reproduction is continued for the engine sound reproduction time Tsafe seconds at the safety limit.

At step S128, the speaker generated sound Ea is calculated by the same procedure as that of the flow chart shown in FIG. 41, and at step S129, as a safety sound check,
The level of the sound pressure during reproduction Eb is compared with the safety limit engine volume Esafe set as the initial value. If it is determined in this step S129 that Eb is smaller than Esafe, the processing is terminated as it is, but if it is determined that Eb is larger than Esafe, the process proceeds to step S130, where the safe sound reproduction time counter Nsafe is set to the safety limit counter. It is determined whether or not it is larger than Nsafe-lim.

If the determination result in step S130 is NO, it is determined that the number of times the safe limit engine volume Esafe has been reproduced has not reached the predetermined number, and the process returns to step S126. However, if the determination in step S130 is YES, even if the safe limit engine volume Esafe is presented a predetermined number of times, there is no effect in improving the arousal level, and it is determined that the driver's tolerance is exceeded, and a warning is issued in step S131.

Next, the excessive tension coping process (step S36 in FIG. 25) according to the present reference example will be described. Figure 44 shows
It is a flow chart which shows the procedure of a hypertension coping process. In the figure, in step S141, the scent presentation counter Ns is reset (Ns = 1), and in step S142,
Present a calming scent (here, a diasmine scent). Then, in step S143, the awakening degree estimated value X3 is fetched according to the awakening degree estimating method described above.

In step S144, step S142 is executed.
Whether or not the driver's arousal level has reached the optimum arousal level based on the presentation of the scent, that is, the arousal level estimation value X3 and Rbe
The value of st + Rerr2 is compared. If X3 is larger than Rbest + Rerr2, it is determined that the driver is still in an overstrained state and has not returned to the normal awakening state, and the process proceeds to step S146. In step S146, it is determined whether the scent presentation counter Ns is larger than the scent presentation limit number Ns-lim. If the result of the determination is NO, the value of the scent presentation counter Ns is set to 1 in step S147.
After only incrementing, the process returns to step S142 again.

However, the determination result of step S146 is Y.
If it is ES, it is determined that the fragrance presentation limit has been exceeded, and a warning is issued in step S148 and the processing ends. Step S1
If it is determined at 44 that X3 is smaller than Rbest + Rerr2, it is determined that the driver has returned from the hypertonic state to the normal awakening state, and the presentation of the scent is stopped at the next step S145.

As described above, according to the present reference example , the degree of arousal can be detected quantitatively and accurately by the electroencephalogram, and the driver is stimulated in accordance with the degree of arousal to prevent a decrease in arousal. There is an effect that the driver can be kept in the optimum awakening state. Further, not only when the wakefulness is lowered, the wakefulness of the driver who has fallen into an excessively tense state can be returned to the normal state, and the driver can be guided to a safe driving state.

In the above-mentioned reference example , the arousal level was quantitatively obtained by the electroencephalogram, but the awakening state of the driver can be easily maintained by the following method. FIG. 45 is a block diagram showing the overall configuration of the awakening state determination device for quantitatively detecting the awakening rhythm. The apparatus shown in the figure measures the electroencephalogram from the subject 11, processes it into a physical quantity that is closely related to the arousal level, and detects the arousal rhythm cycle based on the frequency analysis of the processed electroencephalogram data.

In FIG. 45, the arousal rhythm measuring section 20c
Then, the electroencephalogram data from the subject 11 is measured and stored, and the arousal rhythm detection section 21c detects the arousal rhythm cycle based on the obtained data. Then, in the stimulation presentation unit 22c,
A stimulus such as a sound described later is generated based on the awakening rhythm cycle. The arousal rhythm measurement unit 20c outputs signals from the electrode 2 that detects the brain waves of the subject 11 and the head amplifier unit 3c and the head amplifier unit 3c that pre-amplify to improve the S / N ratio of the weak current from the electrode 2. A / D converter 5c at the next stage
It is composed of a main amplifier section 4c for amplifying a signal for obtaining a sufficient level for conversion into digital data and a main control section 6c for controlling the entire apparatus. The main control unit 6c includes a CPU 6a that functions as a calculation unit and an electroencephalogram data storage unit 6b that stores the digital signal converted by the A / D conversion unit 5c. An anti-aliasing filter may be provided before the A / D converter, or the sampling frequency may be set sufficiently high.

The awakening rhythm detecting section 21c takes in the electroencephalogram data from the electroencephalogram data storing section 6b, and outputs the electroencephalogram to the δ wave (1 to 3 Hz), θ by the band filter 7c which is a digital filter.
Wave (4 to 7 Hz), α wave (8 to 13 Hz), β wave (18
Separated into 4 bands of ˜30 Hz. Then, the power amount calculator 8c calculates the power amount of each electroencephalogram band data.

In order to further remove noise from the obtained power amount, the smoothing time setting unit 10c sets an appropriate smoothing time T _S , and the smoothing processing unit 9c performs smoothing processing to detect an electroencephalogram rhythm. Convert to easy data. The smoothed data is stored in the smoothed data storage unit 11c, and the frequency analysis unit 12c performs frequency analysis to read the peak of the level on the frequency to detect the arousal rhythm cycle.

The arousal rhythm cycle storage section 13c of the stimulus presentation section 22c stores the arousal rhythm cycle obtained as a result of the frequency analysis, and the timer 14c and the stimulus generation section 15c are interlocked with each other to make a predetermined time and a predetermined time. Present stimuli at intervals.
With reference to the flow chart shown in FIG. 46, an awakening rhythm detection processing procedure in the awakening rhythm detecting unit 21c will be described.

The electroencephalogram data is stored in step S1c of FIG.
The electroencephalogram data is acquired from the section 6b, and in step S2c,
As mentioned above, the EEG data is delta wave,
Divide into four bands of θ wave, α wave and β wave. And continued
In step S3c, the power of the electroencephalogram data for each band above
Calculate the quantity. Here, according to the filtered band
Take the root mean square of the time series data of the power spectrum.
Average time T at this time _PIs about 1 second, the sampling cycle
Is about 30 seconds. This allows the power spectrum for each band
It is possible to follow the time series change of the ram and at the same time
It can be reduced.

In step S4c, an appropriate smoothing time T _S is set to smooth the power amount obtained as a result of the calculation in step S3c in order to further remove noise and obtain a tendency for the power amount to change for a long time. Processing is performed to convert the electroencephalogram rhythm into data that is easy to detect. It has been clarified by the study of parameters that the smoothing time T _S is set to about 300 seconds to easily detect the electroencephalogram rhythm.
In step S5c, the smoothed data is stored in the smoothed data storage unit 11 to prepare for the next frequency analysis.

In step S6c, frequency analysis is performed on the smoothed data for each band, and the peak of the level on the frequency is read to detect the cycle of the awakening rhythm. For frequency analysis here, Fourier transform or maximum entropy method (MEM) is used. In particular, since Fourier transform is not effective for data having a strong randomness, rhythm cycle detection by MEM, which is easy to detect peaks, is effective.

FIG. 47 shows a comparison of the results of the frequency analysis obtained from the time when a certain stimulus is presented and the human reaction to it, and the results of the rhythm analysis of the electroencephalogram (α wave in this case). The solid line shows the response. Time, wavy lines represent brain waves. If there is a peak in terms of frequency, it may be interpreted that there is an arousal rhythm at that peak frequency. As can be seen from FIG. 47, the reaction time and the EEG peak are in good agreement,
Further, since the electroencephalograms, particularly the α-waves and the θ-waves, appear well in response to the awakened state, the electroencephalogram rhythm can be directly interpreted as the arousal rhythm. The existence of such a rhythm means that the arousal state is regularly repeated in a plurality of combined cycles, and the awake state and the low awake state are repeated.

FIG. 48 shows how a driver of a vehicle is stimulated in accordance with the rhythm cycle of awakening obtained by the awakening state determination device. In the figure, a certain kind of sound is generated from the stimulation presentation unit 22c of the awakening state determination device 100c, and the sound from the speaker 25c stimulates the driver to maintain the awakening. As a stimulus,
In addition to sound, for example, vibration, scent, heat, or the like that raises the alertness may be used. As will be described later, this device continues to give a stimulus to the driver after chasing a rhythm whose arousal level is peculiar to the individual, for example, a phenomenon in which the arousal is increased or decreased in about 30 minutes in a substantially regular cycle.

FIG. 49 schematically shows changes in the arousal level with time. Next, the awakening continuation processing procedure will be described with reference to the flow chart shown in FIG. At step S10c of FIG. 50, the awakening rhythm cycle Tr unique to the individual detected by the awakening rhythm detecting section 21c is set in the awakening rhythm cycle storing section 13c, and at the next step S11c, when the awakening lowering time comes to the timer 14c. , A time interval Δt (stimulation presentation time width) that determines how much stimulation is performed. Then, in step S12c, t = 0 is set in the timer 14c as the operation start time.

Time measurement is started from the time indicated by t = 0, and it is determined in step S13c whether or not half the time of the awakening rhythm cycle Tr has elapsed. If the determination here is YES, t = 0 is set again as the driving start time in the timer 14c in the subsequent step S14c, and as the first stimulus for the driver, in step S15c, the stimulus generator 15c stimulates for a certain period of time. give. This is because it is considered that the driver is in an awake state at the start of driving, and the first state of awakening is visited in half the awakening rhythm cycle Tr. Therefore, after that, the awakened state comes every awakening rhythm cycle Tr, so that the driver is stimulated in accordance with the cycle.

At step S16c, the awakening rhythm cycle Tr
And the stimulus presentation time width Δt from the arithmetic difference, the timer setting time t
Has reached the stimulus presentation time, and if the decision is YES, it is determined in the next step S17c whether or not the stimulus presentation time has been exceeded. While the determination in step S17c is NO, the stimulus presentation in step S19c is continued. However, if it is determined in step S17c that the stimulus presentation time has been exceeded, the stimulus is stopped in step S18c.

At step S20c, as an operation related to the end of the operation, for example, an operation such as pulling out the ignition key is detected. If the operation is not completed, step S
21c, t = 0 is set to the timer 14c. But,
If it is determined that the operation has ended, this processing ends. By doing this, the electroencephalogram can be detected and the arousal rhythm can be quantitatively detected from the frequency analysis of the time-series data of the electroencephalogram, and the driver can be kept awake by giving a stimulus according to the rhythm. It becomes possible to do.

Next, a method of detecting the arousal rhythm based on the reaction time to the stimulus will be described. FIG. 51 is a block diagram showing the overall configuration of the awake state determination device for detecting the awake rhythm based on the reaction time to the stimulus. In the figure, the awakening state determination device of the reference example (see FIG.
0), and the same components as those of the awakening state determination device (FIG. 45) for quantitatively detecting the awakening rhythm, the same reference numerals are given, and the detailed description thereof is omitted here.

The reaction time measurement situation and the reaction time measurement procedure here are also the measurement situation in the reference example (see FIG. 3).
Since 1) and the measurement procedure (FIG. 32) are the same, description thereof will be omitted. The flow chart shown in FIG. 52 corresponds to that shown in FIG.
7 shows an awakening rhythm detection processing procedure based on the reaction time in the awakening rhythm detecting unit 21c.

At step S50c of FIG. 52, the reaction time data is fetched from the reaction time data storage section 36d, and at the next step S51c, the reaction time data is logarithmically converted by the logarithmic conversion section 38c. Next, in step S52c, the smoothing time width T _S is set, and in step S53c, smoothing for noise removal and averaging for frequency analysis are performed. This T _S is set to about T _S = 300 seconds because the cycle of the awakening rhythm is 5 minutes or more. Further, the average of the reaction times when the stimulus is presented during a time T _S ahead of the measurement start time is calculated. The average may be obtained by any of an arithmetic average, a weighted average, and an exponential average. With such a method, the time is shifted by the sampling period T _samp , and the same operation is repeated to obtain the reaction time until the final measurement time. The setting time of T _samp should be determined by the characteristic of the random number of the stimulus presentation time width. For example, when the stimulus is uniformly generated in about 20 seconds, T _samp is set.
Should be set to about 20 seconds.

In step S54c, the reaction time data subjected to the above processing is stored in the smoothed data storage section 11c,
In the next step S55c, the reaction time is subjected to frequency analysis, and the period of the awakening rhythm is obtained by detecting the peak of the level on the frequency. Freie's transformation method and maximum entropy method (ME
M) is used.

FIG. 53 shows an example of frequency analysis by applying the maximum entropy method to the reaction time data. In this example, there are peaks at 8 minutes and 3 minutes, and it can be seen that the subject who is the subject of the measurement has a characteristic that the awakening rhythm of the 8-minute cycle and the 3-minute cycle comes. Based on the wake-up rhythm cycle thus obtained, the driver of the vehicle is stimulated by the stimulus presentation unit 22c, so that the driver's awake state can be maintained. The state of presenting the stimulus to the driver using the stimulus presenting unit 22c is the same as that shown in FIG. 48, and therefore the illustration and description thereof will be omitted. Also,
The awakening continuation processing procedure in the stimulus presentation unit 22c is also the same as the above-described processing (FIG. 50) in the stimulus presentation unit 22c, and thus the description thereof will be omitted.

As described above, since the wakefulness rhythm cycle can be quantitatively measured from the reaction time to the stimulus, it is possible to perform a monotonous and long-time highway traveling according to the detected wakefulness rhythm cycle. A certain driver can be stimulated to return to the awake state before the driver falls into the awake state, and the awakening can be achieved. <Modification> In the HMM of the above reference example, the arousal level is estimated as a numerical value from the electroencephalogram, but the arousal level can be estimated by the heart rate or blink frequency in addition to the electroencephalogram.

First, a method of estimating the arousal level based on the heart rate will be described. In FIG. 54, the cardiac potential detecting section 51
Reference numeral aa includes a portion for detecting a cardiac potential from the electrode 2 attached to a region between the heart of the driver 11 as a subject and amplifying the cardiac potential signal by an amplifier (not shown). The cardiac potential processing unit 52aa processes the cardiac potential signal from the cardiac potential detecting unit 51aa to process the signal into a physical quantity that makes it easy to estimate awakening. That is, the cardiac potential signal amplified by the cardiac potential detecting unit 51aa is taken in via an A / D converter described later, and processed into a heart rate required for arousal level evaluation.

The arousal level estimation unit 53a calculates an estimated value deeply related to arousal based on a later-described arousal level estimation parameter and heart rate data. In addition, the awakening degree determination unit 54a determines to what extent the stimulus is presented based on the awakening degree estimated by the awakening degree estimation unit 53a. Then, the stimulus presentation unit 50a
The driver presents an appropriate stimulus to the driver or stops the driver in accordance with an instruction from the awakening degree determination unit 54a.

The arousal level estimation unit of the present modification estimates the arousal level based on the heart rate and reaction time, and uses the information to provide the driver with an optimal stimulus in a biofeedback manner. The concept of the biofeedback applied to this modified example is the same as that described in the reference example with reference to FIG. By detecting the awake state and giving a stimulus corresponding thereto, it is possible in principle to achieve the continuation of the optimum awake state shown in FIG.

First, a wakefulness estimation parameter determining device for determining the parameters for calculating the wakefulness in the wakefulness estimating unit 53a will be described. The arousal level estimation parameter determination device simultaneously measures a reaction time and an electrocardiographic signal that are highly correlated with the degree of arousal, and calculates a heart rate from the cardiac potential. Then, a wakefulness estimation parameter is determined by performing regression analysis based on the obtained heart rate data and reaction time data.

FIG. 55 is a schematic block diagram of the awakening degree estimation parameter determination device. This figure shows the situation where this device measures human reaction time and heart rate at the same time for a long time, and the estimated parameters are obtained by analyzing the correlation between the obtained reaction time and heart rate. be able to. In FIG. 55, the cardiac potential processing unit 70aa measures the cardiac potential from the electrode 2 attached to the subject (driver) 11, and the reaction time measuring unit 71 measures the time during which the subject 11 responds to the given stimulus. measure. Based on these measurement results, the wakefulness estimation parameter determination unit 72 estimates the wakefulness unique to the subject.

Since the method of measuring the reaction time (selective reaction time) in this modification is the same as that of the reference example , its explanation is omitted (see FIG. 30). FIG. 56 shows an electrocardiographic processing unit 7.
It is a block diagram which shows the structure of 0aa. In the electrocardiographic detection unit 149 shown in the figure, an electrocardiographic electrode is used to connect to a region sandwiching the human heart with a conductive adhesive for exclusive use, and the electrocardiographic signal obtained therefrom is A head amplifier (corresponding to the physiological measurement amplifier 43 in FIG. 24) takes noise countermeasures and amplifies the signal.

The cardiac potential signal from the cardiac potential detector 149 is
A / D converter 61 (corresponding to A / D converter 44 in FIG. 24)
Electrocardiographic data storage unit 62a as electrocardiographic data via
It is stored in a. The heart rate is the electrocardiographic data storage unit 62a.
It is calculated by the original heart rate calculation unit 150 based on the cardiac potential data stored in a. The obtained original heart rate data is smoothed by the averaging unit 152 in order to obtain the average heart rate data having a high correlation with the arousal level. The time setting for this smoothing will be described later. Further, the heart rate data after the averaging process is stored in the heart rate storage unit 153.

Regarding the smoothing processing of the heart rate data, in order to correspond to the reaction time, the heart rate from the stimulus presentation time to the smoothing time before the heart rate is averaged, and this is taken as the heart rate at that time. . Here, it is necessary to determine the smoothing time of the heart rate. It is necessary to set these parameters such that the correlation between the heart rate after smoothing and the reaction time is highest. Therefore, using the reaction time as the objective variable and the heart rate as the explanatory variable, the smoothing times are changed and the regression analysis is performed by the regression analysis unit 72aa. Of the correlation coefficients thus obtained for various combinations of smoothing parameters,
The maximum correlation coefficient is clarified, and the combination of smoothing parameters at that time is determined as the optimum smoothing time for the heart rate. At the same time, the parameters a1 and a2 obtained from the correlation analysis and the regression equation using the parameters a1 and a2 are used as the arousal level estimation equation. This is because the reaction time is predicted, but it can be considered that the reaction time and the arousal level correspond to each other with high correlation.

Next, although the arousal level can be calculated quantitatively, it is necessary to decide to what threshold value the stimulus presentation should be started. The procedure of determination at this time is the same as that of the flow chart ( reference example ) of FIG. Here, the obtained Tth, a1, and a2 and the regression equation using the obtained Tth, a1 and a2 are used as the arousal level estimation parameters. Next, each component of the awakening degree estimation unit of this modification will be described in detail. <Electrical potential detection unit> An electrode for picking up the cardiac potential is connected to a region between the heart of the subject (driver) with a special conductive adhesive, and the obtained cardiac potential signal is output by a head amplifier (not shown). Amplify and take measures against noise, and then amplify with a sanitary amplifier (not shown). <Electrocardiographic Processing Section> FIG. 57 shows a detailed block diagram of the electrocardiographic processing section 52aa. Since the electrocardiographic processing unit in the figure has the same function as that shown in FIG. 56, the same components are designated by the same reference numerals,
The detailed description is omitted here. <Arousal level estimation unit> The arousal level estimation unit 53a uses the arousal level estimation parameter and the regression equation obtained by the arousal level estimation parameter determination device shown in FIG. 55 to process the heartbeat processed by the electrocardiographic potential processing unit 52aa. Calculates arousal level from numerical data. Then, the value is delivered to the awakening degree determination unit 54a.

The arousal level is expressed by the following estimation formula.
That is, arousal level = a1 * H + a2 (10) However, H is an example in which the arousal level is estimated by a method equal to or higher than the heart rate in FIG.
8 shows. FIG. 58 shows the maximum correlation coefficient obtained for the vibration conditions of each subject and the vibration exciter. The correlation coefficient is
Rounded off to two decimal places. The smoothing time is 300 seconds for both the reaction time and the heart rate.

From FIG. 58, there is a high correlation between the heart rate and the reaction time, which means that it is possible to finely estimate the arousal level. However, this method is not as accurate as the estimation of the arousal level by the electroencephalogram shown in the reference example . <Arousal Level Determining Unit> For the processing procedure in the awakening level determining unit 54a, see FIG.
Since it is the same as the flow chart shown in No. 8 ( reference example ), the description thereof is omitted. <Stimulation presenting unit> The stimulus presenting unit 50a receives the wakefulness lowering signal from the wakefulness determining unit 54a and outputs a stimulus having a wakefulness effect such as sound, vibration, and scent to the driver 11 who is the subject for a certain period of time until a stop signal comes. To do.

FIG. 59 is a block diagram showing the structure of the awakening stimulus setting device. The present arousal stimulus setting device determines the type of stimulus most suitable for the individual and the method of presenting the stimulus in advance, because the stimulus that enhances the arousal effect may vary depending on the individual. In this device, the electrocardiographic potential processing unit 161, the reaction time measurement unit 103, and the awakening degree estimation parameter determination unit 101, excluding the stimulation presentation unit 104 and the stimulation parameter setting unit 110, have the same functions as the awakening state determination device shown in FIG. 55. Therefore, the description thereof is omitted.

In FIG. 59, the subject 11 is caused to perform a selective reaction work, and the electrocardiogram at that time is simultaneously recorded. And
Parameters are determined by the arousal level estimation parameter determination unit from the obtained reaction time and heart rate, and the arousal level is estimated. If the subject 11 is evaluated by the awakening degree estimation parameter determination unit 72 of the awakening state determination device shown in FIG. 55 before being evaluated by this device, the awakening level can be estimated immediately.

In the stimulation parameter setting unit 110, the kind of stimulation and the presentation method for determining the optimum stimulation state are provisionally set in advance. Then, the stimulation presentation unit 104
When it is determined that the state of awakening is low, the stimulation parameter setting unit 11
In accordance with the data set to 0, a sound is emitted from the speaker 105a via the speaker driving unit 105, the seat 106a including the vibrating body is vibrated at a high frequency via the vibration driving unit 106, or the vibration driving unit 107 is operated. The vibrating device 107a is vibrated at a low frequency through the vibration to present the stimulus to the subject 11. Then, the wakefulness of the subject is estimated from the heart rate when the stimulus is given, and the wakefulness estimated value before and after the stimulus is presented is compared to see whether or not the effect is obtained.

By changing the stimulus state and the type of stimulus, it is possible to select the stimulus most effective for arousal control,
Furthermore, based on the obtained results, the awakening of the individual can be maintained and the effect of awakening control can be improved. As described above, according to the reference example , the arousal level can be detected quantitatively and accurately by the heart rate, and by providing the driver with a stimulus according to the arousal level, the awakening lowering is prevented and the driver's individual There is an effect that can be maintained in the optimal awake state.

Further, not only when the wakefulness is lowered, there is an effect that the wakefulness of the driver who is in an excessively tense state can be returned to a normal state and can be guided to a safe driving state.
Next, a method of estimating the awakening level based on the blink frequency will be described. In FIG. 60, a blink detection section 51bb
Is composed of a portion that detects the eye movement from the electrodes 2 attached to the upper and lower sides of the eyeball of the driver 11, which is the subject, or left and right, and amplifies the eye movement potential by an amplifier (not shown).
The blink processing unit 52bb processes the eye movement signal from the blink detection unit 51bb to perform signal processing into a physical quantity that facilitates estimation of arousal. That is, the eye movement signal amplified by the blink detection section 51bb is taken in via an A / D converter described later, and processed to a blink frequency necessary for awakening degree evaluation by calculation.

The arousal level estimation unit 53a calculates an estimated value deeply related to arousal based on the awakening level estimation parameters and blink frequency data described later. In addition, the awakening degree determination unit 54a
Determines how much stimulus should be presented from the arousal level estimated by the arousal level estimation unit 53a. Then, the stimulus presentation unit 50a presents an appropriate stimulus to the driver or stops the stimulus according to an instruction from the awakening degree determination unit 54a.

In the arousal level estimation unit of this reference example , blinking,
And the arousal level is estimated based on the reaction time, and the optimal stimulus is applied to the driver in a biofeedback manner by using the information. The concept of the biofeedback applied to this modified example is the same as that described in the reference example with reference to FIG. By detecting the awake state and giving a stimulus corresponding thereto, it is possible in principle to achieve the continuation of the optimum awake state shown in FIG.

Next, a wakefulness estimation parameter determining device for determining the parameters for calculating the wakefulness in the wakefulness estimating unit 53a will be described. The arousal level estimation parameter determination device measures the reaction time and the eye movement signal that are highly correlated with the degree of awakening at the same time, and performs regression analysis based on the blink frequency data and the reaction time data obtained from the eye movement. The awakening degree estimation parameter is determined.

FIG. 61 is a schematic block diagram of the awakening degree estimation parameter determination device. This figure shows the situation in which this device measures human reaction time and eye movement simultaneously for a long time, and the estimated parameters are obtained by analyzing the correlation between the obtained reaction time and blinking frequency. be able to. The awakening degree estimation parameter determination device measures the eye movement from the electrode 2 attached to the subject (driver) 11 in the blink processing unit 70bb, and the reaction time measurement unit 71 measures the time during which the subject 11 responds to the given stimulus. Is measured, and the arousal level estimation parameter determination unit 72 estimates the arousal level peculiar to the subject based on these measurement results.

Since the method of measuring the reaction time (selective reaction time) in this modification is the same as that of the reference example , its explanation is omitted (see FIG. 30). FIG. 62 is a block diagram showing the configuration of the blink processing unit 70bb. In the eye movement detecting section 155 shown in the figure, the eye movement signals obtained by connecting the human eye to the upper and lower sides or the right and left sides of the eye with a dedicated conductive adhesive using electrodes are used as, for example, a head amplifier ( A noise countermeasure is taken by the physiological measurement amplifier 43 in FIG. 24) to amplify the signal.

The eye movement signal from the eye movement detecting section 155 is stored in the eye movement signal data storage section 62bb as eye movement signal data via the A / D conversion section 61 (corresponding to the A / D converter 44 in FIG. 24). Is stored. The eye movement signal data is based on the eye movement signal data stored in the eye movement signal data storage unit 62bb, and is based on the original blink frequency calculation unit 15
Calculated as 6.

The obtained original blink frequency data is smoothed by the averaging unit 158 in order to obtain average blink frequency data having a high correlation with the arousal level. The average blink frequency data after the averaging process is stored in the blink frequency data storage unit 159. The time setting for this smoothing and the procedure for determining the threshold value when starting the stimulus presentation are the same as the method in the arousal level estimation based on the above heart rate, and therefore the description thereof will be given. Is omitted.

Next, each constituent element of the awakening degree estimation unit of this modification will be described in detail. <Eye movement detector> The electrodes for picking up eye movement signals are connected to the upper and lower sides of the eye of the subject (driver) with a dedicated electrically conductive adhesive, and the obtained signals are connected to a head amplifier (not shown). Amplify and take measures against noise, and then amplify with a sanitary amplifier (not shown). <Eye movement signal processing unit> FIG. 63 shows a blink processing unit 52.
The detailed block diagram of bb is shown. Since the blink processing section in the figure has the same function as that shown in FIG. 62,
The same components are designated by the same reference numerals, and detailed description thereof will be omitted here. <Arousal level estimation unit> The arousal level estimation unit 53a uses the arousal level estimation parameter and the regression equation obtained by the arousal level estimation parameter determination device shown in FIG. 61 to blink the processing unit 52bb.
The arousal level is calculated from the blink frequency data processed in. Then, the value is delivered to the awakening degree determination unit 54a.

The arousal level is expressed by the following estimation formula.
That is, arousal level = a1 * E + a2 (11) where E is the blink frequency. An example in which the awakening degree is estimated by the above method is shown in FIGS. 64 and 65. FIG. 64 shows the maximum correlation coefficient obtained for the vibration conditions of each subject and the vibration device. The correlation coefficient was rounded off to two decimal places. The smoothing time is 30 for both reaction time and heart rate.
0 seconds.

FIG. 65 shows the results of FIG. 64 with respect to the reaction time and the change over time in the predicted value from the blinking frequency.
In the figure, the vertical axis represents the logarithmic reaction time and the horizontal axis represents the elapsed time.
Also, in the figure, the dotted line indicates the reaction time, and the solid line indicates
The predicted value from the blink frequency is shown. The data breaks represent breaks between experimental sessions. From these figures, it can be seen that the above-described method is effective for estimating the arousal level and is capable of finely estimating the arousal level. However, this method is not as accurate as the estimation of the arousal level by the electroencephalogram shown in the reference example .

<Awakening Level Determining Unit> The processing procedure in the awakening level determining unit 54a is shown in FIG.
Since it is the same as the flow chart shown in No. 8 ( reference example ), the description thereof is omitted. <Stimulation presenting unit> The stimulus presenting unit 50a receives the wakefulness lowering signal from the wakefulness determining unit 54a and outputs a stimulus having a wakefulness effect such as sound, vibration, and scent to the driver 11 who is the subject for a certain period of time until a stop signal comes. To do.

FIG. 66 is a block diagram showing the structure of the awakening stimulus setting device. The present arousal stimulus setting device determines the type of stimulus most suitable for the individual and the method of presenting the stimulus in advance, because the stimulus that enhances the arousal effect may vary depending on the individual. The reaction time measuring unit 103 and the awakening degree estimation parameter determining unit 101 in this device are
Since the function is the same as that of the awakening state determination device shown in FIG. 61, the description thereof will be omitted.

In FIG. 66, the subject 11 is caused to perform a selective reaction work, and the eye movement at that time is simultaneously recorded. Then, the parameter is determined by the arousal level estimation parameter determination unit from the obtained reaction time and blink frequency, and the arousal level is estimated. In addition, before the subject 11 is evaluated by this device,
When the wakefulness estimation parameter determination unit 72 of the wakefulness determination device shown in FIG. 61 is evaluating, the wakefulness can be estimated immediately.

In the stimulation parameter setting unit 110, the kind of stimulation and the presentation method for determining the optimum stimulation state are provisionally set in advance. Then, the stimulation presentation unit 104
When it is determined that the state of awakening is low, the stimulation parameter setting unit 11
In accordance with the data set to 0, a sound is emitted from the speaker 105a via the speaker driving unit 105, the seat 106a including the vibrating body is vibrated at a high frequency via the vibration driving unit 106, or the vibration driving unit 107 is operated. The vibrating device 107a is vibrated at a low frequency through the vibration to present the stimulus to the subject 11. Then, the arousal level of the subject is estimated from the blink frequency when the stimulus is given, and the arousal level estimation values before and after the stimulus presentation are compared to see whether or not the effect has been achieved.

By changing the stimulus state and the type of stimulus, the stimulus most effective for awakening control can be selected,
Furthermore, based on the obtained results, the awakening of the individual can be maintained and the effect of awakening control can be improved. As described above, according to the present reference example, it is possible to detect the arousal level quantitatively and accurately based on the blink frequency, and by applying a stimulus to the driver according to the degree of the awakening, the awakening lowering is prevented, and the driver There is an effect that the optimum awakening state of the individual can be maintained.

Further, not only when the wakefulness is lowered, there is an effect that the wakefulness of the driver who is in an excessively tense state can be returned to a normal state and can be guided to a safe driving state.
Although the target is limited to the driver in the above embodiment,
For example, a user such as a computer can also be applied. Also, the scope of application is not limited to the automobile industry, and there is a human-machine relationship, such as home appliances,
It can also be applied to the life-related industry called architecture.

[0173]

As described above, according to the present invention,
Depending on the detected driver status, change the status
Providing a human-machine system for vehicles that operates as much as possible
Can be driven by human-machine interaction.
There is an effect that the iva can be guided to a safe driving state .

[Brief description of drawings]

[Figure 1]

FIG. 2 is a diagram for explaining the basic configuration of the present invention,

[Figure 3]

FIG. 15 is a view for explaining the first embodiment according to the present invention,

16]

FIG. 22 is a view for explaining a second embodiment according to the present invention,

FIG. 23

FIG. 53 is a view for explaining a reference example according to the present invention,

FIG. 54

FIG. 66 is a diagram for explaining a modified example of the reference example according to the present invention .

─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Yoshihisa Okamoto 3-1, Shinchi Fuchu-cho, Aki-gun, Hiroshima Prefecture Mazda Co., Ltd. (72) Inventor Nami Hirahata 3-1-1 Shinchu, Fuchu-cho, Hiroshima Prefecture Mazda Incorporated (72) Inventor Akihaku 3-1, Shinchu, Fuchu-cho, Aki-gun, Hiroshima Prefecture Mazda Co., Ltd. (72) Inventor Tomoyuki Abe 3-1-1 Shinchu, Fuchu-cho, Aki-gun, Hiroshima Prefecture Mazda Corporation (72) Inventor, Neiichi Nagamura 1-34-17 Komatsu, Tsuchiura-shi, Ibaraki (72) Inventor Ahei Sadoyama 4-926-102, Namiki, Tsukuba-shi, Ibaraki (56) Reference JP-A 64-72759 ( JP, A) JP 55-101270 (JP, A) JP 60-592 (JP, A) JP 1-131648 (JP, A) JP 56-67632 (JP, A) JP Sho 62-15129 (JP, A) JP-A-63-222711 JP, A) (58) investigated the field (Int.Cl. ^7, DB name) A61M 21/00 A61B 5/16 300 G06F 17/00

Claims (4)

(57) [Claims] 1. A vibrating means for vibrating a seat on which a driver of a vehicle is seated, a fingertip volume pulse wave detecting means for detecting a fingertip volume pulse wave of the driver, and the vehicle traveling on a rough road. Based on the detection result of the fingertip volume pulse wave detection means and the detection result of the rough road traveling detection means, the vibrating means And a control means for controlling the state of vibration due to, the control means, based on the detection result of the fingertip volume pulse wave detection means, obtains a state quantity related to the psychological state of the driver, the bad road running detection When it is detected by the means that the vehicle is not traveling on a rough road, the smaller the state quantity, the larger the value of the sensory stimulus generation control parameter is set, while When it is detected by the road running detection means that the vehicle is running on a bad road, the smaller the state quantity, the smaller the value of the sensory stimulus generation control parameter is set, and the white noise is set. Is processed by using a variable filter having a filter characteristic in which the decrease amount of the vibration power with the increase in frequency is adjusted to a smaller value as the value of the sensory stimulus generation parameter is larger, and the state of vibration by the vibrating means is processed. Is controlled in accordance with a fundamental frequency that increases with an increase in the engine speed of the vehicle, vibration is applied to the vibration state according to the processed white noise, thereby imparting fluctuations to the vibration state. Human machine system for vehicles. 2. The control means, when the state in which the vehicle is traveling on a rough road is detected by the rough road traveling detection means for a predetermined time, the detected state continues. The human-machine system for a vehicle according to claim 1, wherein the state of vibration by the vibrating means is changed by treating that the vehicle is not detected to be traveling on a rough road. 3. A fuel injection device for an engine mounted on a vehicle, fingertip volume pulse wave detecting means for detecting a fingertip volume pulse wave of a driver, and detecting that the vehicle is traveling on a road that is congested. Congestion traveling detection means, white noise generator for generating white noise, control means for controlling the fuel injection device based on the detection result of the fingertip volume pulse wave detection means and the detection result of the traffic congestion traveling detection means The control means, based on the detection result of the fingertip volume pulse wave detection means, obtains a state quantity related to the psychological state of the driver, the traffic congestion detection means by the traffic congestion detection means to the road in a traffic jam. When it is detected that the vehicle is not traveling, the smaller the state quantity is, the larger the value of the sensory stimulus generation control parameter is set, while the congestion traveling detection hand is set. When it is detected that the vehicle is traveling on a road that is congested, by setting the value of the sensory stimulus generation control parameter to a smaller value as the state amount is smaller, the white noise, When the value of the sensory stimulus generation parameter is larger, the processing is performed by using a variable filter having a filter characteristic in which the decrease amount of the vibration power with the increase of the frequency is adjusted to a smaller value, and the fuel injection state by the fuel injection device is changed. When controlling according to a fundamental frequency that increases with an increase in the engine speed of the vehicle, by adding a vibration corresponding to the processed white noise to the injection amount of the fuel,
A human-machine system for a vehicle, characterized by imparting fluctuation to the fuel injection state. 4. When the control means detects that the vehicle is traveling on a congested road by the traffic congestion detection means for a predetermined period of time, the detected state continues. 4. The vehicle according to claim 3, wherein the fuel injection state of the fuel injection device is changed by treating that the vehicle is not detected to be traveling on a congested road. Human machine system.