JP5541145B2 - Driver fatigue estimation device - Google Patents

Description

  The present invention relates to a technique for estimating a driver's fatigue level.

  In the technology of Patent Document 1, the control unit estimates the physique of the driver based on the driver information in the IC card, the seat slide position, and the like, and estimates the mechanical impedance based on the estimated physique. Alternatively, the control unit forcibly rotates the steering wheel by a predetermined amount for measurement, and estimates the mechanical impedance based on the operation amount of the driver's steering operation performed in response thereto. Then, the steering reaction force of the steering is controlled based on the estimated mechanical impedance.

JP 2007-245904 A

However, since the hardness (impedance) of the arm varies depending on the position of the driver's hand, the method described in Patent Document 1 cannot accurately detect the driver's impedance. As a result, there is a possibility that the driver's fatigue degree cannot be accurately estimated only from the mechanical impedance estimated in Patent Document 1.
The present invention has been made paying attention to the above points, and an object of the present invention is to estimate a driver's fatigue level with higher accuracy.

  In order to solve the above problems, the present invention estimates the impedance of a driver who operates the operating element with respect to disturbance torque input to the operating element, and the difference between the estimated impedance and a pre-set or measured prior impedance. Based on the above, the driver's fatigue level is estimated.

  The driver's fatigue level is estimated based on the difference from the reference prior impedance. Thereby, the estimation accuracy of the driver's impedance is improved, and the driver's fatigue level can be estimated more accurately.

1 is a conceptual diagram of a vehicle including a fatigue estimation device according to an embodiment of the present invention. It is explanatory drawing of the driver | operator's fatigue estimation apparatus which concerns on embodiment based on this invention. 1 is a system configuration diagram of a driver fatigue estimation apparatus according to an embodiment of the present invention. It is a figure explaining detection of a driver's posture. It is a figure explaining disturbance torque A. It is a figure explaining the disturbance torque B. FIG. It is a block diagram of impedance measurement. It is a figure which shows the frequency characteristic of an impedance. It is a figure which shows the relationship between a muscular strength estimated value and a difference. It is explanatory drawing which estimates a muscular strength from the difference value of an impedance. It is explanatory drawing of the change of the impedance by the position of a hand, and the setting of a judgment threshold value. It is explanatory drawing of the change of the impedance by the position of a hand, and the change of a judgment threshold value. It is explanatory drawing of the change of the judgment threshold value by the position of an elbow. It is explanatory drawing of the method of changing a sensitivity according to the joint angle of an elbow, so that an elbow is in a position far from a trunk. It is a figure explaining the detection of a fatigue state. It is a figure explaining the detection of a fatigue state. It is a flowchart figure explaining the process example of a driver | operator's fatigue estimation apparatus. It is a figure explaining the operation example of this embodiment.

Next, embodiments of the present invention will be described with reference to the drawings.
In the following embodiments, a steering wheel 7 will be described as an example of an operator operated by a driver. The target operator is not limited to the steering wheel 7, but may be a brake pedal, a shift lever, or the like. In short, it is applicable to any operator that is operated when the driver is driving and generates an operation reaction force on the driver.

FIG. 1 is an overall schematic diagram of a vehicle equipped with the fatigue estimation device of the present embodiment.
Reference numeral 7 denotes a steering wheel as an operator operated by the driver. The steering wheel 7 is connected to the steering gear mechanism 1 via steering. The steering gear mechanism 1 is a device that converts steering rotation into movement in the vehicle width direction (lateral direction) by a conversion mechanism such as a rack / pinion mechanism.

Reference numerals 2 and 3 are steered wheels. The steered wheels 2 and 3 are steered according to the displacement in the vehicle width direction by the steering gear mechanism 1. As the tires of the steered wheels 2 and 3, for example, tires attached to a general sedan type can be exemplified, and the type (flatness, tire diameter, radial / studless, etc.) is not limited.
Reference numeral 5 denotes a steering reaction force actuator. The steering reaction force actuator 5 is a motor installed on the steering column. The steering reaction force actuator 5 is a device that can be used to amplify and assist the driver's force required to rotate the steering, or to reduce unnecessary disturbance that enters from the tire side. The steering reaction force actuator 5 generates a desired reaction force by inputting a control current in accordance with a command from the reaction force device controller 6.

Reference numeral 6 denotes a reaction force device controller. The reaction force device controller 6 outputs a control current corresponding to the target steering reaction force and steering angle. The reaction device controller 6 is a control device for driving the steering reaction force actuator 5.
Reference numeral 4 denotes a reaction force device motor angle sensor. The reaction force device motor angle sensor 4 is a motor angle sensor installed between the steering column shaft and the steering reaction force actuator 5, and detects and outputs the angle of the rotating steering wheel.

Reference numeral 8 denotes a torque sensor. In this embodiment, a torque sensor with a built-in steering link portion is illustrated. The torque sensor 8 measures the force generated when the driver operates (steers) the steering wheel 7.
Reference numeral 9 denotes a steering information transmission harness. The steering information transmission harness 9 includes a cable for transmitting a steering force, a steering angle, and a disturbance torque signal from the steering reaction force actuator 5 to the muscle force calculation / instruction device use determination device 10.

Reference numeral 10 denotes a muscular strength calculation / instruction device use determination device. The muscular strength calculation / instruction device use determination device 10 is an arithmetic device for estimating the muscular strength of the upper limb body (target musculoskeletal portion) of the driver and estimating the degree of fatigue. Details of the muscular strength calculation / instruction device utilization determination device 10 will be described later.
Reference numeral 11 denotes a driver camera. The driver camera 11 is an imaging device for detecting the posture of the driver. Examples of the camera include a CCD camera and a CMOS camera.

Reference numeral 12 denotes a tire lateral force detection sensor. The tire lateral force detection sensor 12 is a sensor that measures the lateral force acting on the tire.
Reference numeral 13 denotes a pointing device. When the driver inputs a signal indicating that the driver is in a fatigued state or a state that is likely to lead to fatigue in the future, for example, the instruction device 13 prompts the driver to take a break, or a driving support device such as a lane keeping support device Prompt to use. The prompt is performed by voice or display.

Next, the muscular strength calculation / instruction device use determination device 10 will be described with reference to FIGS.
The muscle strength calculation / instruction device utilization determination device 10 is configured by a computer and functionally configured as shown in FIGS.
That is, the muscular strength calculation / instruction device use determination device 10 includes a disturbance torque signal detection unit 14, an arm state measurement unit 15, an muscular impedance measurement unit 16, a muscular strength estimation unit 17, a fatigue level estimation unit 18, and an indication device utilization. A determination unit 19 is provided.

  The disturbance torque signal detection unit 14 is a program that detects a predetermined disturbance torque that can be used for measuring the driver's impedance from the steering reaction force generated in a past fixed time and the steering reaction force generated by the suspension state. . That is, the disturbance torque signal detection unit 14 is based on a measured torque signal of the past certain time input from the torque sensor 8 and a signal of a force acting on the tire due to self-aligning torque, road surface disturbance, and side wind disturbance (tire generation torque signal). That is, the disturbance torque that can be used to measure the driver's impedance is detected from the history of the tire generation torque signal and the measured torque.

The tire generation torque signal used for measurement and the detection conditions of the actually measured torque are, for example, as follows. That is, the detection condition is that the coherence with the actually measured torque is 0.9 or more and the phase of the self-aligning torque is earlier at a frequency of 1 Hz to 100 Hz.
When this detection condition is not satisfied, the disturbance torque signal detection unit 14 uses a predetermined disturbance torque signal prepared in advance. That is, the disturbance torque signal detection unit 14 drives the steering reaction force actuator 5 via the reaction force device controller 6 and transmits the disturbance torque under a preset torque condition to the steering wheel 7.

The self-aligning torque is estimated by calculating the self-aligning torque generated in the front wheels 2 and 3 which are steered wheels by the product of the tire lateral force and the caster rail based on the signal input from the tire lateral force detection sensor 12. presume. For example, the estimated self-aligning torque is used as the tire generation torque signal.
The arm state measurement unit 15 is a program that detects the state of the body (mainly the arm) from the image captured by the driver camera and calculates the position of the hand / elbow and the angle of the hand / elbow joint. For example, as shown in FIG. 4, the joint position and joint angle of the driver are detected from the image taken by the camera 11. The arm state measurement unit 15 also detects the grip position of the driver.

The whole muscle impedance measuring unit 16 is a program for measuring the impedance of the whole muscle. For example, the impedance measuring unit 16 for the entire muscle employs a technique for obtaining the frequency characteristics of the arm using disturbance torque. That is, the impedance measuring unit 16 for the entire muscle inputs the disturbance torque, the actually measured torque, and the steering angle, calculates the impedance based on the following formula, and outputs the impedance Hnms as an impedance characteristic.
Impedance calculation formula:
Hnms = −Swf / Swx [(Nm · Nm) / (Nm · deg)]

This equation is an equation in which a value obtained by dividing the cross spectrum Swf of the disturbance torque Ta and the actually measured torque Tb by the disturbance torque Ta and the cross spectrum Swx of the steering angle is an impedance value. Note that other known methods may be adopted as the impedance measurement method.
Here, impedance calculation methods are generally classified into the following three types 1) to 3).
In the following 1) and 2), after obtaining the viscosity, elasticity, and inertia, the impedance is determined by substituting it into a second-order transfer function and obtaining its frequency characteristics. For this reason, only how to obtain viscosity, elasticity and inertia will be described. Further, 3) is a method for directly obtaining frequency characteristics without obtaining viscosity, elasticity, and inertia.

(Example of impedance calculation method)
1) Calculation method using simultaneous equations Impedances are calculated by solving simultaneous equations with viscosity, elasticity, and inertia as variables from measured torque measured at predetermined time intervals set in advance.
2) Method of calculating impedance by individually measuring viscosity, elasticity, and inertia In other words, after obtaining an elastic value by applying stepwise disturbance torque to the steering wheel 7, the steering wheel 7 is rotated at a constant speed to obtain a viscosity term. Ask for. The inertia term is obtained using the weight of the arm as a reference value.
3) Method of calculating impedance by measuring actual torque and steering angle while applying disturbance torque to steering wheel 7

This method is calculated by the following processes 3-1) to 3-3).
3-1) Add disturbance torque to the operator operated by the driver.
3-2) Measure the actual torque generated in the steering wheel 7 and the shift amount of the steering angle.
The impedance of the driver's body is measured from the comparison of the frequency characteristics of 3-1) and 3-2).
Of the above three impedance calculation methods, 3) can measure impedance most accurately. Based on this, in this embodiment, the impedance is measured (calculated) by the calculation method of 3). This impedance calculation method uses disturbance torque transmitted to the steering wheel 7.

Here, as the target disturbance torque, the following disturbance torque A and disturbance torque B may be employed.
A) Predetermined predetermined signal: Disturbance torque A due to pseudo M series, sweep waveform, white noise (see FIG. 5) convoluted by shifting the phase of a sine wave in a frequency band of 0 to 5 Hz
B) Disturbance torque B obtained from vehicle travel data
As the disturbance torque B, for example, time series data of the force transmitted from the tire to the steering or the force transmitted from the steering to the tire is recorded in the storage unit. When the waveform of the time series data has a predetermined shape and a predetermined spectrum set in advance, it is used as a disturbance torque (see FIG. 6).

The disturbance torque has a lower limit value at which the steering wheel 7 starts moving, and an upper limit value that can be overridden by the driver. Within this range, it is estimated that the driver's impedance can be measured.
FIG. 7 is a schematic diagram of impedance measurement by the method 3) with the highest accuracy. That is, FIG. 9 is a schematic diagram when disturbance torque is applied by the actuator 5 connected to the steering wheel 7 in a state where the driver is holding the steering wheel 7 as the operation unit.
In this schematic diagram, Ta: disturbance torque Tb: measured torque Tc: difference value between disturbance torque and measured torque (= Ta-Tb)
x: Steering angle.

  At this time, the driver's impedance can be obtained from “−STaTb / STax (a value obtained by dividing the cross spectrum of the disturbance torque and the actual torque by the cross spectrum of the disturbance torque and the steering angle)” as described above. FIG. 8 shows an example of the impedance measurement result. From this lower left figure, it is possible to estimate the state of force entering the arm, that is, the state of hardness of the musculoskeletal part. For example, it can be confirmed that when the force for gripping the steering wheel 7 increases, the impedance increases in a frequency band of 1 Hz or less.

The muscle strength estimation unit 17 is a program that estimates muscle strength based on the impedance obtained by the impedance measurement unit 16 of the entire muscle.
First, the muscular strength estimation unit 17 obtains a difference value that is a difference between the impedance of the arm measured in advance (preliminary impedance) and the current impedance.
Next, the muscular strength estimation unit 17 corrects the obtained difference value based on the measurement information (driver's posture) of the arm state measurement unit 15. Specifically, when the driver's gripping position with respect to the steering wheel 7 is at the lower end of the steering wheel 7, and when the elbow is in contact with the armrest or the elbow is in contact with the trunk, Impedance is high even if it is removed. In view of this, when the position of the driver's hand with respect to the steering wheel 7 is at the lower end of the steering wheel 7, the elbow is in contact with the armrest, or the elbow is in contact with the trunk, The absolute value of the difference value is corrected to be small. Instead of correcting the absolute value of the difference value, the sensitivity when estimating the muscle strength of the next arm may be corrected. In the following description, an example in the case of correcting the sensitivity when estimating the strength of the arm will be described.

  Next, the muscular strength estimation unit 17 estimates the muscular strength of the arm with reference to a preset “impedance-muscle strength” map based on the corrected difference value. The muscular strength estimation unit 17 estimates the muscular strength of the arm at a preset control cycle, and records the muscular strength in a recording unit (not shown) each time the estimation is performed. As shown in FIG. 9, the “impedance-muscle strength” map can be estimated that the greater the difference value, the greater the muscle strength. What is shown in FIG. 9 is a case where the impedance in a state where the driver is relaxed and gripped is adopted as the prior impedance. For this reason, when the difference value is zero, the muscle strength estimation value is zero (y intercept is 0). In the case where the predetermined impedance shows a predetermined muscular strength, the muscular strength estimated value when the difference value is zero is set to a value corresponding thereto.

Here, the arm impedance (preliminary impedance) at the time of relaxation measured in advance will be described.
In this embodiment, “relax” refers to “a straight running or a state where the steering wheel 7 is gripped but not operated” in a stopped state or a state estimated as the state. Since the state (impedance) of the arm gripping the steering wheel 7 by applying a disturbance is measured, measurement is not possible when the steering wheel 7 is not gripped.

Moreover, what is necessary is just to implement the measurement method and measurement timing of the impedance at the time of relaxation based on AB as follows.
A. The average driver's relaxed impedance is obtained in advance by experiments or the like. With respect to the driver's impedance obtained using at least a vehicle signal of several tens of seconds, if the waveform of 1 Hz or less in the horizontal axis in FIG. The impedance at that time is obtained as a prior impedance.

  B. The driver measures the impedance when relaxed for the first time. First, a voice is instructed to “hold the steering wheel 7 with a force sufficient to touch the steering wheel 7”. Next, disturbance torque for impedance is input to the steering wheel 7, and the actual torque and steering angle at that time are measured. And a driver | operator's impedance is calculated | required as a prior impedance from the relationship of the said disturbance torque, an actual torque, and a steering angle.

C. Measure the impedance when relaxing from the running history. From the history of steering in a low speed range (40 km / h or more) where there is no road disturbance at all and the driving load is considered to be small, the steering wheel 7 The state of not operating is extracted, and the driver's relaxation impedance is measured from the vehicle signal at that time. It has been known from past experiments that the driver becomes more nervous and puts force on the arm as the speed increases, and the impedance at the time of high speed driving is not used as the impedance at the time of relaxation.
Here, as the prior impedance, it is preferable that the impedance at the time of relaxation is adopted because a larger difference can be obtained, but the prior impedance may not be the impedance at the time of relaxation. The prior impedance may be an impedance that can be specified as a reference.

Further, a supplementary explanation will be given of the processing of the muscle strength estimation unit 17.
When the driver grasps the steering wheel 7 (see FIG. 10A), the prior impedance in the relaxed and maintained state and the impedance in the firmly maintained state are obtained by experiments as shown in FIG. ) Results were obtained. As can be seen from FIG. 10B, it can be confirmed that the impedance increases in a frequency band (impedance low frequency component) of 1 Hz or less when the force for gripping the steering wheel 7 increases.

Then, as shown in FIG. 10 (c), the difference between the impedance characteristic value (physical value) in advance in a relaxed and maintained state and the impedance characteristic value (physical value) in the current maintained state. By obtaining the value, a difference value from the previous impedance as a reference can be obtained. The muscle strength can be specified by using the “impedance-muscle strength” map as shown in FIG.
Here, as the characteristic value (physical value) of the impedance, for example, an average value or area of an impedance low frequency component, or an elastic term obtained assuming that the arm is a quadratic model is employed. That is, a value proportional to the impedance low frequency component may be used as the impedance characteristic value (physical value).

Next, correction of the difference value based on the gripping position of the steering wheel 7 and the position of the driver's elbow will be described.
When the relationship between the case where the steering wheel 7 was held in a relaxed manner and the case where the steering wheel 7 was firmly held was determined, the result shown in FIG. 11 was obtained. As can be seen from FIG. 11, the impedance characteristic value increases as the driver grips the lower side of the steering wheel 7 in spite of the same gripping force. Further, the lower the steering wheel 7 is gripped, the smaller the difference between the impedance characteristic value in the relaxed grip and the impedance characteristic value in the firm grip.

  Further, from the above result (FIG. 11), it is easy to estimate the muscular strength when the gripping position is on the upper side of the steering wheel 7, but when the gripping position is on the lower side of the steering wheel 7, the gripping state is relaxed and firmly. Therefore, it is difficult to estimate because there is little difference in muscle strength from the gripping state. Based on this, a determination threshold value for determining whether or not to estimate muscle strength based on the gripping position may be set. For example, when the gripping position is below the position of 07:25, muscle strength estimation, that is, fatigue estimation may not be performed.

  Further, as described above, when the gripping position is on the upper side of the steering wheel 7, it is easy to estimate the muscular strength, but when the gripping position is on the lower side of the steering wheel 7, the difference in muscular strength is small and difficult to estimate. Based on this, the sensitivity for estimating muscle strength is changed according to the grip position. Specifically, as the lower side of the steering wheel 7 is gripped, correction is performed so as to increase the sensitivity of the muscle strength estimation value with respect to the difference value between the relaxed grip and the firm grip as shown in FIG.

  Further, when the driver puts his elbow on the armrest and grips the steering wheel 7 in a state where the steering wheel 7 is firmly gripped, the impedance is increased even if the driver does not put force on the arm. When the impedance is high in this way, the muscle strength estimation value tends to be larger than the actual value. On the other hand, in this embodiment, when the driver puts his elbow on the armrest and grips the steering wheel 7, as shown in FIG. 13, the estimated muscle strength value for the difference value between the relaxed grip and the firm grip Reduce the sensitivity. At this time, when the driver puts his elbow on the armrest, muscle strength estimation may not be performed.

  The relationship between the position of the elbow and the impedance is as shown in FIG. 14B (except when the elbow is supported on the vehicle body as described above). That is, the farther the elbow is from the trunk, such as when the arm is extended, the lower the impedance is even when force is applied (closer to the impedance when relaxed). In other words, the impedance becomes closer to the impedance in the relaxed grasping state and the difference value becomes smaller as the elbow moves away from the trunk even though force is applied. In view of this, as shown in FIG. 14C, the sensitivity decreases as the elbow moves away from the trunk.

The fatigue level estimation unit 18 is a program that estimates the fatigue level based on the history of muscle strength estimated by the muscle strength estimation unit 17.
For example, the fatigue level is estimated based on the muscle strength duration curve shown in FIG. 15, and when the muscle strength history is located above the muscle strength duration curve, the fatigue status is determined. Moreover, it determines with a fatigue degree being so large that the duration above the muscular strength duration curve is long.

Further details.
Here, a processing method for estimating the driver's fatigue based on the muscle strength estimation value that is an index of the degree of fatigue will be described.
In general, the duration of muscle strength (instantaneous value) has a relationship as shown in FIG. Therefore, when the muscular strength is large, the duration in which the muscle continues to exert force is short. Conversely, if the muscle strength is small, the duration that the muscle can continue to exert force is long. As shown in FIG. 15, the duration decreases exponentially as the muscular strength increases.
Based on this, as a fatigue estimation method based on FIG. 15, for example, the following policy 1 and policy 2 can be exemplified.

(Measure 1 for fatigue estimation)
Fatigue is determined from the frequency of instantaneous muscle strength. That is, based on the duration of the instantaneous value from the relationship of FIG. 15 (fatigue curve), the degree of fatigue state and the fatigue state to be notified to the driver are estimated.
For the above estimation, recognition processing based on statistical knowledge may be performed. Examples of statistical processing include Bayesian estimation, fuzzy estimation, support vector machine, genetic algorithm, neural network, and the like.

(Measure 2 for fatigue estimation)
Depending on whether or not the instantaneous value of the muscular strength exceeds a preset threshold value, it is estimated how much the fatigue state is and whether the fatigue state should be notified to the driver. There may be one threshold value or a plurality of threshold values.
Here, when a muscle exerts a force of 100% or more, a phenomenon called a crane occurs. Therefore, the threshold is set based on the extreme state.
Next, the fatigue estimation process that employs the above policy 1 will be described.
First, as preprocessing, muscle strength calculation and a histogram of muscle strength and time are calculated.
Next, as the feature extraction process, the feature amount is obtained based on the area and size of the histogram.

Next, as a post-treatment, an average, median, and mode value of feature amounts are calculated, or a feature space is created based on a plurality of different feature amounts.
Next, as a determination process, it is determined whether or not a preset numerical value or a value within a predetermined range of the feature space (a value exceeding a threshold value). If this determination is not satisfied, the process returns to the preprocess and the process is repeated. On the other hand, when this determination is satisfied, it is determined that the driver is required to be notified of fatigue, and the process ends.

Further, when the muscle strength estimation value is acquired as a power spectrum, as shown in FIG. 16, when the high frequency component (for example, a band of 1 Hz or more) of the power spectrum of the muscle strength estimation value is equal to or higher than a preset threshold, The degree of fatigue is set to be larger as the state in which the high frequency component of the power spectrum of the muscle strength estimation value is equal to or higher than a preset threshold value continues.
Based on the fatigue level estimated by the fatigue level estimation unit 18, the determination unit 19 using the pointing device is a program that determines whether or not the user is fatigued or is likely to get tired and outputs a notification signal to that effect to the pointing device. For example, when the fatigue level is equal to or higher than a preset fatigue level threshold, the determination unit 19 using the pointing device determines that the fatigue state is present.

Next, an example of processing of the muscle strength calculation / instruction device usage determination device 10 will be described with reference to the flowchart of FIG.
The muscular strength calculation / instruction device utilization determination device 10 operates at a preset control cycle. When the operation is started, disturbance torque is first measured in step S1. That is, the disturbance torque generated when the tire is taken on a bad road is recorded in the storage unit. Thereafter, the process proceeds to step S2.
In step S2, the disturbance torque recorded in step S1 is subjected to FFT processing to obtain a power spectrum of the disturbance torque. Then, the power spectrum of the obtained disturbance torque is compared with the power spectrum when traveling on a good road measured in advance. Alternatively, it is calculated whether the impedance spectrum calculation judgment line is exceeded for the obtained disturbance torque power spectrum. Thereafter, the process proceeds to step S3.

In step S3, when it is determined that the impedance can be calculated by the process in step S2, the data is cut out from the time series data of the disturbance torque in the storage unit for a predetermined time interval, and at the same time. The driver's impedance is measured from the actual torque and steering angle. Whether or not the impedance can be calculated may be determined based on whether or not the disturbance torque condition described above is satisfied.
If it is determined that the impedance cannot be calculated for a preset time, the preset disturbance torque is input to the steering, and the driver's impedance is measured from the actual torque and the steering angle.

Thereafter, the process proceeds to step S4.
In step S4, the difference value between the impedance measured in step S3 and the preset prior impedance is calculated. The prior impedance of this embodiment is obtained by measuring in advance the impedance in a relaxed state (disarming where the driver is holding the steering wheel without applying any force to the arm). The difference value is a value proportional to the strength of the arm. Thereafter, the process proceeds to step S5. Here, the difference value is a difference value for the characteristic value of the low frequency component of the impedance as described above.

In step S5, when the difference value is smaller than a preset threshold value, the process is terminated and returned, and step S1 is operated in the next control cycle. If the difference value is greater than or equal to a preset threshold value, the process proceeds to step S6.
Here, when the difference value is small, the resolution is low, so that the muscle strength may not be estimated. In addition, when the difference value is small, it is estimated that it is close to a relaxed state (a state where no force is applied to the driver's arm). And when the said difference value is more than the preset threshold value, it is considered that the arm has force and it transfers to step S6.

In step S6, an estimated muscle strength value is obtained from the difference value of the obtained low frequency component. Thereafter, the process proceeds to step S7.
In step S <b> 7, the driver's gripping position with respect to the steering wheel 7, the elbow position with respect to the driver's trunk, and the degree of elbow extension are detected based on image data captured by one or more cameras 11. Thereafter, the process proceeds to step S8.

In step S8, based on the detection in step S7, it is determined whether or not the gripping position with respect to the steering wheel 7 is located on the lower side. If it is located on the lower side, the process proceeds to step S9. If it is not located on the lower side, the process proceeds to step S10.
In step S9, the sensitivity of the muscular strength estimation value with respect to the difference value is increased as it is determined that the lower side of the steering wheel 7 is gripped. Thereafter, the process proceeds to step S10.

In step S10, it is determined whether or not the elbow is in contact with a part of the vehicle body such as an armrest. If it is determined that the elbow is in contact with a part of the vehicle body such as an armrest, the process proceeds to step S11. Otherwise, the process proceeds to step S12.
When the process proceeds to step S11, the measured impedance is increased even if the driver is not putting force on the arm even when the steering wheel 7 is gripped, such as when the elbow is placed on the armrest. It is in a state. For this reason, in step S11, the sensitivity of the muscle strength estimation value with respect to the difference value is lowered. Thereafter, the process proceeds to step S12.

In step S12, it is determined whether or not the elbow angle is greater than or equal to a predetermined angle set in advance. If the elbow angle is greater than or equal to a predetermined angle set in advance, the process proceeds to step S13. If the elbow angle is less than the predetermined angle, the process proceeds to step S14.
In step S13, it is difficult to put force on the arm and the impedance is low when the side is open and the elbow joint is extended. Decrease sensitivity when arm is extended by sitting. Thereafter, the process proceeds to step S14.

In step S14, the maximum muscle strength data is read. Thereafter, the process proceeds to step S15. The maximum muscle strength refers to the instantaneous value (peak value) of the obtained muscle strength.
In step S15, an estimated value of muscle strength in the maximum muscle strength is calculated. Thereafter, the process proceeds to step S16.
In step S16, a deviation amount between the estimated muscle strength value and the muscle strength duration curve in the maximum muscle strength is calculated. Here, a positive value is set when each muscle strength estimation value is in the upper right of the muscle strength sustaining curve, and a negative value is set when the muscle strength estimation value is in the lower left. The process proceeds to step S17.

In step S17, when the deviation calculated in step S16 is equal to or larger than a predetermined value set in advance, the process proceeds to step S18. Otherwise, the process proceeds to step S19.
In step S18, since it is a case where deviation | shift amount is more than predetermined, it is estimated as a fatigue state. Thereafter, the process proceeds to step S19. The amount of deviation corresponds to the degree of fatigue.
If it is determined in step S19 that the accumulated value of the fatigue state exceeds a predetermined value set in advance, the process proceeds to step S20. Otherwise, the process ends and returns, and step S1 is activated in the next control cycle.

In step S20, it is determined whether or not the setting of the pointing device is the switch ON state. If it is ON, the process proceeds to 21. Otherwise, the process ends and returns, and step S1 is activated in the next control cycle.
In step S21, the instruction device is turned on to notify the driver that the user is in a fatigued state. Thereafter, the process is terminated and the process returns, and step S1 is operated in the next control cycle.

(About operation and others)
Here, conventional techniques for determining the muscle strength of a driver are roughly classified into the following two (a) and (b).
(A) Muscle strength estimation based on electromyogram (b) Muscle strength estimation based on musculoskeletal model However, muscle strength estimation based on an electromyogram is troublesome because it is necessary to attach an electromyographic sensor to the skin. In addition, the muscle strength estimation by the musculoskeletal model has a large error because the force cancelled internally cannot be predicted for estimating the muscle strength using a huge storage unit.

On the other hand, in this embodiment, the driver's fatigue can be easily estimated without using the myoelectric sensor or the musculoskeletal model.
Also, I don't know if the reference and impedance are high, whether the force is applied or not. On the other hand, in the present embodiment, by using the difference from the assumed impedance as a reference, it is possible to estimate with higher accuracy.

Next, an operation example according to the present embodiment will be described with reference to FIG.
Disturbance torque is detected during traveling, and the disturbance torque data is recorded in the storage unit. At the same time, the measured torque and the steering angle are also detected and stored in the storage unit.
Then, when the power spectrum obtained by performing the FFT process on the disturbance torque data recorded in the storage unit exceeds the impedance measurement determination line for a preset duration, the disturbance torque data for the predetermined section is cut out.

If no situation occurs in which the power spectrum exceeds the impedance measurement determination line for a preset duration at a preset time, the steering reaction force actuator 5 is driven to set a preset disturbance. Torque is input to the steering.
Then, the impedance is calculated from the disturbance torque, the actual torque, and the steering angle of the cut-out predetermined section at the same time. Subsequently, a difference value which is a difference between the low frequency component of the calculated impedance and a low frequency component of the previous impedance which is the impedance in a relaxed state set in advance is obtained. The muscle strength is estimated from the difference value.

At this time, if it is determined that the driver is holding the lower side of the steering wheel 7, the sensitivity at the time of obtaining the muscle strength estimation value from the difference value is increased. Also, when the elbow is away from the driver's trunk, the sensitivity is lowered.
The estimated muscle strength is stored in the storage unit.
Then, based on the estimated muscular strength, the time during which the arm is in force is counted, and when the estimated muscular strength value is on the upper right of the muscular strength duration curve, it is determined that the user is fatigued.

When the fatigue determination value is accumulated and exceeds a predetermined value, the indicating device is turned on, for example, a sound for prompting the driver to take a break is generated.
Here, the steering wheel 7 constitutes an operator. The disturbance torque signal detector 14 constitutes a disturbance torque detector. The impedance measurement unit 16 constitutes impedance estimation means. The fatigue level estimation unit 18 constitutes fatigue level estimation means. The disturbance torque signal detector 14 constitutes a disturbance torque applying unit. The arm state measurement unit 18 constitutes a grip position detection unit. The camera 11 constitutes driving posture detection means.

(Effect of this embodiment)
(1) The disturbance torque signal detection unit 14 detects disturbance torque input to an operation element for giving a driving instruction to the vehicle. The impedance measurement unit 16 estimates the impedance of the driver who operates the operation element with respect to the disturbance torque detected by the disturbance torque signal detection unit 14. The fatigue level estimation unit 18 estimates the driver's fatigue level based on the difference between the previously set or measured prior impedance and the impedance estimated by the impedance measurement unit 16.
The driver's fatigue level is estimated based on the difference from the reference prior impedance. Thereby, the estimation accuracy of the driver's impedance is improved, and the driver's fatigue level can be estimated more accurately.
The disturbance torque input to the operator may be a weak load reaction force, so that it is possible not to disturb the driver's operation.

(2) The prior impedance is a value measured when the driver is estimated to be relaxed.
By using the impedance when it is estimated that the driver is relaxed as the reference prior impedance, the difference can be increased by that amount (the resolution can be increased), so that the driver's fatigue level is estimated. Accuracy is improved.
(3) The disturbance torque applying means applies a disturbance torque used for the estimation of the impedance to the operation element.
As a result, the driver's impedance can be estimated even when no disturbance torque is input to the operator.

(4) The grip position detection means detects the grip position of the steering wheel 7 by the driver. The fatigue level estimation unit 18 corrects the estimated fatigue level based on the grip position detected by the grip position detection means.
The correction is performed by correcting the determination threshold of the fatigue level or changing the sensitivity for obtaining fatigue from the difference value.
Even if the difference value is the same, the fatigue state differs depending on the gripping position. On the other hand, the accuracy of fatigue estimation is improved by performing correction based on the gripping position.
It should be noted that it is preferable not to provide unnecessary information to the driver so as not to perform fatigue estimation on the impedance measurement result that is difficult to detect whether force is applied or removed depending on the position of the hand.

(5) The driving posture detection means detects the driving posture of the driver. The fatigue level estimation unit 18 corrects the estimated fatigue level based on the elbow position based on the driving posture detected by the driving posture detection means.
Even if the difference value is the same, the fatigue state varies depending on the elbow position of the driver. On the other hand, the accuracy of fatigue estimation is improved by correcting the position based on the elbow position of the driving vehicle.

(6) In the correction based on the elbow position, the correction amount is changed according to the position of the elbow with respect to the trunk of the driver. For example, as the elbow is located farther from the trunk (depending on the joint angle of the elbow), the sensitivity of fatigue level determination is lowered.
Depending on whether the arm is extended or bent, that is, depending on the position of the elbow, the actual impedance differs even if the difference value is the same. Based on this, the accuracy of fatigue estimation is improved by changing the correction amount according to the position of the elbow with respect to the trunk of the driver.

5 Steering reaction force actuator 6 Reaction force controller 7 Steering wheel (operator)
8 Torque Sensor 10 Muscle Strength Calculation / Indication Device Utilization Determination Device 11 Driver Camera 13 Indicator Device 14 Disturbance Torque Signal Detection Unit 15 Muscle State Measurement Unit 16 Impedance Measurement Unit 17 Muscle Strength Estimation Unit 18 Fatigue Level Estimation Unit 19 Judgment part

Claims (6)

Disturbance torque detecting means for detecting disturbance torque input to an operator for giving a driving instruction to the vehicle;
Impedance estimation means for estimating an impedance of a driver who operates the operation element with respect to disturbance torque detected by the disturbance torque detection means;
Fatigue degree estimation means for estimating the driver's fatigue level based on the difference between the previously set or measured prior impedance and the impedance estimated by the impedance estimation means .   2. The driver fatigue estimation apparatus according to claim 1, wherein the prior impedance is a value measured when the driver is estimated to be relaxed.   The driver's fatigue estimation device according to claim 1, further comprising disturbance torque applying means for applying a disturbance torque used for the impedance estimation to the operation element. The operating element is a steering wheel,
A gripping position detecting means for detecting a gripping position of the steering wheel by the driver;
The driver fatigue estimation according to any one of claims 1 to 3, wherein the fatigue level estimating means corrects the estimated fatigue level by a gripping position detected by the gripping position detecting means. apparatus. Driving posture detection means for detecting the driving posture of the driver,
5. The driver fatigue estimation apparatus according to claim 4, wherein the fatigue level estimation unit corrects the estimated fatigue level by an elbow position based on the driving posture detected by the driving posture detection unit.   6. The driver fatigue estimation apparatus according to claim 5, wherein the correction based on the elbow position changes a correction amount according to the position of the elbow with respect to the trunk of the driver.

Images