WO2019215811A1

WO2019215811A1 - Drive assist device and drive assist method

Info

Publication number: WO2019215811A1
Application number: PCT/JP2018/017767
Authority: WO
Inventors: 中川　幸彦
Original assignee: 三菱電機株式会社
Priority date: 2018-05-08
Filing date: 2018-05-08
Publication date: 2019-11-14

Abstract

A passenger shaking detection unit (30) detects shaking of a first passenger using image data of an image in which the interior of a vehicle is captured. A vehicle shaking detection unit (31) detects shaking of the vehicle using vehicle information. A shaking relationship calculation unit (32) calculates a first relationship value indicating the relationship between the shaking of the first passenger as detected by the passenger shaking detection unit (30) and the shaking of the vehicle as detected by the vehicle shaking detection unit (31). A reference relationship value acquisition unit acquires a reference relationship value indicating the relationship between the shaking of a passenger and the shaking of the vehicle when the passenger is not carsick. When the first relationship value calculated by the shaking relationship calculation unit (32) has deviated from the reference relationship value by a set amount or greater, a determination unit (33) determines that the first passenger is carsick.

Description

Driving support device and driving support method

The present invention relates to a driving support device for determining occupant sickness.

In the driving support device of Patent Document 1, the occupant head position detection unit detects the position of the occupant's head, and the occupant influence calculation unit calculates the behavior of the occupant's head. Then, when the head of the occupant is greatly shaken, it is determined that the occupant may have a car sickness.

Japanese Patent Laid-Open No. 2005-326962

However, just because the occupant is shaking does not mean that car sickness will occur. This is because there may be a case where the occupant is swaying due to factors other than car sickness such as a personal habit of the occupant. Therefore, the driving support device of Patent Document 1 that makes a determination based on whether the occupant is simply shaking has a problem that the determination accuracy of car sickness is low.

The present invention has been made to solve the above-described problems, and an object of the present invention is to obtain a driving support device that improves the accuracy of determination of car sickness.

The driving support device according to the present invention includes an occupant swing detection unit that detects a swing of the first occupant using image data of an image taken inside the vehicle, and a vehicle swing that detects the swing of the vehicle using vehicle information. A detecting unit, and a swing correlation calculating unit that calculates a first correlation value indicating a correlation between the first passenger's shake detected by the passenger's swing detecting unit and the vehicle shake detected by the vehicle swing detecting unit; The first correlation value calculated by the reference correlation value acquisition unit for acquiring the reference correlation value indicating the correlation between the shake of the passenger and the shake of the vehicle when the occupant is not sick is the first correlation value calculated by the swing correlation calculation unit. And a determination unit that determines that the first occupant is drunken when the reference correlation value deviates from the setting by more than a setting.

According to this invention, the accuracy of determination of car sickness is improved.

It is a block diagram which shows the structure of the driving assistance device which concerns on Embodiment 1, and its periphery. 3 is a graph showing the amount of occupant swing and the amount of vehicle swing. 3A and 3B are diagrams illustrating a hardware configuration example of the driving support apparatus according to Embodiment 1. FIG. 4 is a flowchart illustrating an example of processing by the driving support device according to the first embodiment. It is a block diagram which shows the structure of the driving assistance device which concerns on Embodiment 2, and its periphery. 6 is a flowchart illustrating an example of processing performed by a driving support apparatus according to Embodiment 2.

Hereinafter, in order to explain the present invention in more detail, modes for carrying out the present invention will be described with reference to the accompanying drawings.
Embodiment 1 FIG.
FIG. 1 is a block diagram illustrating a configuration of the driving support device 3 according to the first embodiment and its surroundings.
The driving support device 3 is mounted on a vehicle. The driving support device 3 is connected to the occupant detection unit 1, the vehicle state detection unit 2, the vehicle unit 4, and the notification unit 5.

The occupant detection unit 1 is mounted on a vehicle and outputs image data of an image taken inside the vehicle to the driving support device 3. The occupant detection unit 1 is realized by a camera 10, a stereo camera 11, a TOF (Time of Flight) camera, or the like.

The vehicle state detection unit 2 outputs vehicle information indicating the state of the vehicle to the driving support device 3. For example, the vehicle state detection unit 2 acquires a CAN (Controller Area Network) signal 20 and outputs it to the driving support device 3 as vehicle information. The CAN signal 20 indicates information such as the vehicle speed, the steering angle of the steering wheel, or the brake amount. Moreover, the vehicle state detection part 2 is implement | achieved by the acceleration sensor 21, for example, and outputs the acceleration of a vehicle to the driving assistance apparatus 3 as vehicle information.

The driving support device 3 includes an occupant swing detection unit 30, a vehicle swing detection unit 31, a swing correlation calculation unit 32, a determination unit 33, and a control unit 34.
The occupant swing detection unit 30 acquires the image data output from the occupant detection unit 1, and detects the sway of the occupant using the image data. In addition, it is possible to detect the swaying of an occupant appearing in an image by a known image processing technique. The occupant swing detection unit 30 outputs the detected occupant swing to the swing correlation calculation unit 32. Here, the swing specifically refers to the swing amount.

The vehicle swing detection unit 31 acquires the vehicle information output by the vehicle state detection unit 2, and detects the swing of the vehicle using the vehicle information. Note that it is possible to detect a vehicle shake by a known calculation method using the CAN signal 20 or the acceleration of the vehicle. The vehicle swing detection unit 31 outputs the detected vehicle swing to the swing correlation calculation unit 32.

The swing correlation calculation unit 32 calculates a correlation value indicating a correlation between the passenger swing detected by the passenger swing detection unit 30 and the vehicle swing detected by the vehicle swing detection unit 31. The swing correlation calculation unit 32 outputs the calculated correlation value to the determination unit 33. The correlation value calculated by the swing correlation calculating unit 32 may be any value as long as it indicates the correlation between the passenger's shake and the vehicle shake. For example, the correlation value indicating the correlation between passenger sway and vehicle sway is a non-patent document “Shizushima Shimada,“ Vibration Analysis of Structures by Correlation Analysis Method ”, JSCE Proceedings, February 1970, 174th. No., p. It can be calculated using the equations (4)-(8) disclosed in “11-23”.

The determination unit 33 uses the correlation value calculated by the rocking correlation calculation unit 32 to determine whether the occupant is intoxicated. The determination unit 33 outputs the determination result to the control unit 34.

When the determination unit 33 determines that the occupant is intoxicated, the control unit 34 notifies that the occupant is intoxicated or controls the vehicle. Specifically, the control unit 34 outputs a control signal to the vehicle unit 4 to control the vehicle. In addition, the control unit 34 outputs a control signal to the notification unit 5 and notifies the passenger that the occupant is intoxicated through the notification unit 5.

The vehicle unit 4 indicates an accelerator 40, a brake 41, a window 42, a suspension 43, a handle 44, or the like. When the control signal is output from the control unit 34, the vehicle unit 4 operates according to the control signal.

The notification unit 5 is realized by an indicator 50 or a speaker 51 mounted on the vehicle. When the control signal is output from the control unit 34, the notification unit 5 performs a warning display or a warning sound output indicating that the occupant is intoxicated according to the control signal.

Here, the concept of car sickness determination performed by the driving assistance device 3 will be described using the graph shown in FIG. FIG. 2 shows an occupant swing amount as the occupant swing and a vehicle swing amount as the vehicle swing. The swaying of the occupant indicates the relative swaying when viewed from the vehicle, and the swaying of the vehicle indicates the relative swaying when viewed from the ground.
When the occupant is not intoxicated, the occupant's swing amount and the vehicle's swing amount show a correlation of approximately the same degree from time 0 (sec) to time 100 (sec).

On the other hand, assuming that a car sickness occurs in the occupant around the time 60 (sec), the degree of correlation between the occupant swing amount and the vehicle swing amount after the time 60 (sec) is 0 (sec) before that time. From the degree of correlation between the occupant's rocking amount and the vehicle's rocking amount from time to time 60 (sec). This is because when the vehicle shakes due to the operation of the accelerator 40, the brake 41 or the handle 44, the occupant shakes the vehicle with inertial force unless he / she is intoxicated. When this occurs, the occupant stops suspending himself / herself from the inertial force and prevents the vehicle from shaking relative to the worsening of car sickness.

The driving support device 3 utilizes the fact that the correlation value indicating the correlation between the swaying of the occupant and the swaying of the vehicle shows different values when the occupant is not sick and when the occupant is sick. , Make a determination of car sickness.

Next, a hardware configuration example of the driving support device 3 will be described with reference to FIGS. 3A and 3B.
The functions of the occupant swing detection unit 30, the vehicle swing detection unit 31, the swing correlation calculation unit 32, the determination unit 33, and the control unit 34 of the driving support device 3 are realized by a processing circuit. The processing circuit may be dedicated hardware or a CPU (Central Processing Unit) that executes a program stored in a memory. The CPU is also called a central processing unit, a processing unit, an arithmetic unit, a microprocessor, a microcomputer, a processor, or a DSP (Digital Signal Processor).

3A, the functions of the occupant sway detection unit 30, the vehicle sway detection unit 31, the sway correlation calculation unit 32, the determination unit 33, and the control unit 34 are realized by a processing circuit 100 that is dedicated hardware. It is a figure which shows the hardware structural example in a case. The processing circuit 100 is, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array), or a combination thereof. To do. The functions of the occupant swing detection unit 30, the vehicle swing detection unit 31, the swing correlation calculation unit 32, the determination unit 33, and the control unit 34 may be realized by combining separate processing circuits 100. These functions may be realized by one processing circuit 100.

FIG. 3B shows the functions of the occupant sway detection unit 30, the vehicle sway detection unit 31, the sway correlation calculation unit 32, the determination unit 33, and the control unit 34. The CPU 102 executes a program stored in the memory 101. It is a figure which shows the hardware structural example at the time of implement | achieving. In this case, the functions of the occupant sway detection unit 30, the vehicle sway detection unit 31, the sway correlation calculation unit 32, the determination unit 33, and the control unit 34 are based on software, firmware, or a combination of software and firmware. Realized. Software and firmware are described as programs and stored in the memory 101. The CPU 102 reads out and executes the program stored in the memory 101 to thereby execute functions of the occupant swing detection unit 30, the vehicle swing detection unit 31, the swing correlation calculation unit 32, the determination unit 33, and the control unit 34. Is realized. That is, the driving support device 3 includes a memory 101 for storing a program or the like that results in the processing of steps ST1 to ST11 shown in the flowchart of FIG. 4 to be described later. In addition, these programs may cause a computer to execute the procedures or methods of the occupant swing detection unit 30, the vehicle swing detection unit 31, the swing correlation calculation unit 32, the determination unit 33, and the control unit 34. I can say that. Here, the memory 101 is, for example, a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable ROM) or an EEPROM (Electrically Erasable Programmable Semiconductor ROM). This corresponds to a memory or a disk-shaped recording medium such as a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, or a DVD (Digital Versatile Disc).

Note that some of the functions of the occupant sway detection unit 30, the vehicle sway detection unit 31, the sway correlation calculation unit 32, the determination unit 33, and the control unit 34 are realized by dedicated hardware, and part of them. It may be realized by software or firmware. For example, the functions of the occupant swing detection unit 30 and the vehicle swing detection unit 31 are realized by a processing circuit as dedicated hardware, and the swing correlation calculation unit 32, the determination unit 33, and the control unit 34 are processed by a processing circuit. The function can be realized by reading and executing the program stored in the memory.

As described above, the processing circuit is configured by the above-described occupant swing detection unit 30, vehicle swing detection unit 31, swing correlation calculation unit 32, determination unit 33, and control unit 34 by hardware, software, firmware, or a combination thereof. The function of each part can be realized.

Next, an example of processing by the driving support device 3 configured as described above will be described with reference to the flowchart shown in FIG. The process shown in FIG. 4 is started when the driving of the vehicle is started, for example.
The occupant swing detection unit 30 acquires image data from the occupant detection unit 1 (step ST1). The occupant swing detection unit 30 determines whether an occupant other than the driver is on board using the image data (step ST2). Hereinafter, passengers other than the driver are simply referred to as “occupants”.

When an occupant other than the driver is not on board (step ST2; NO), the process of step ST1 is performed again. Note that, when no passenger other than the driver is on board, the process of step ST1 is not performed again, and the driving support device 3 may end the process as it is.
On the other hand, when an occupant other than the driver is on board (step ST2; YES), the vehicle swing detection unit 31 acquires vehicle information from the vehicle state detection unit 2 (step ST3).

Subsequently, the occupant swing detection unit 30 detects the swaying of the occupant using the image data from the occupant detection unit 1 (step ST4). In the following description, for simplification of description, it is assumed that there is only one occupant and that the first occupant swing detection unit 30 has detected the swing of the first occupant. The occupant swing detection unit 30 outputs the detected first occupant swing to the swing correlation calculation unit 32.
When there are a plurality of occupants, the occupant swing detection unit 30 detects the swing for each occupant.
Further, the vehicle swing detection unit 31 detects vehicle swing using the vehicle information from the vehicle state detection unit 2 (step ST5). The vehicle swing detection unit 31 outputs the detected vehicle swing to the swing correlation calculation unit 32.

The occupant swing detection unit 30 and the vehicle swing detection unit 31 detect the swing of the first occupant and the swing of the vehicle during the first set time from the start of operation, for example, for 10 minutes from the start of the drive, and the swing correlation calculation unit Continue to output to 32. The swing correlation calculation unit 32 uses the first occupant's shake and the vehicle's shake detected during the 10 minutes from the start of driving to correlate the first occupant's shake and the vehicle's shake during the 10 minutes from the start of driving. Is calculated (step ST6). In addition, the 1st setting time should just be set to the time considered that normal car sickness does not generate | occur | produce, and is not restricted to 10 minutes. That is, the reference correlation value calculated in step ST6 is a value indicating the correlation between the shake of the first occupant and the shake of the vehicle when the first occupant is not sick. The swing correlation calculation unit 32 outputs the calculated reference correlation value to the determination unit 33.
When there are a plurality of passengers, the reference correlation value is calculated for each passenger.

Even after the reference correlation value is calculated, the acquisition of the image data and the detection of the first occupant's shake are continuously performed by the occupant fluctuation detection unit 30, and the detected first occupant fluctuation is calculated as the fluctuation correlation. Is output to the unit 32. Similarly, after the reference correlation value is calculated, vehicle information acquisition and vehicle shake detection by the vehicle swing detection unit 31 are continued, and the detected vehicle swing is detected by the swing correlation calculation unit. 32.
The swing correlation calculation unit 32 uses the first occupant shake and the vehicle shake detected after the reference correlation value is calculated, and the first occupant shake and the vehicle shake after the calculation of the reference correlation value. The first correlation value indicating the correlation with is calculated (step ST7). The first correlation value does not use only one instantaneous value for each of the first occupant's shake and the vehicle's shake, but rather the first occupant's shake and the vehicle's shake for a certain amount of time, or a certain number of times. It is calculated using the first passenger's shake of the minute and the shake of the vehicle. The rocking correlation calculation unit 32 outputs the calculated first correlation value to the determination unit 33.

Subsequently, the determination unit 33 determines whether or not the first correlation value calculated by the swing correlation calculation unit 32 deviates from the reference correlation value by a setting or more (step ST8). When the first correlation value deviates from the reference correlation value by more than the setting (step ST8; YES), the determination unit 33 outputs a determination result that the first occupant is intoxicated to the control unit 34. On the other hand, if the first correlation value has not deviated from the reference correlation value by more than the setting (step ST8; NO), the determination unit 33 determines that the first occupant is not sick, and the process returns to step ST7. A first correlation value calculation process using the newly detected shake and the shake of the vehicle for the first occupant is performed.

When the determination unit 33 determines that the first occupant is intoxicated, the control unit 34 outputs a control signal to the notification unit 5 and notifies that the occupant is intoxicated through the notification unit 5. (Step ST9).

Subsequently, the control unit 34 determines whether to control the vehicle (step ST10). For example, when the driver uses the input unit (not shown) such as a touch panel to select the control unit 34 to control the vehicle, the control unit 34 determines to control the vehicle.
When it is determined that the control unit 34 does not control the vehicle (step ST10; NO), the process returns to step ST7, and the first correlation using the newly detected shake and the shake of the vehicle for the first occupant. A value calculation process is performed.

On the other hand, when the control unit 34 determines to control the vehicle (step ST10; YES), the control unit 34 outputs a control signal to the vehicle unit 4, and the accelerator 40, the brake 41, the handle 44, the suspension 43, or the window. The vehicle such as 42 is controlled (step ST11). For example, the control unit 34 adjusts at least one of the effect of the accelerator 40, the effect of the brake 41, the effect of the handle 44, the effect of the suspension 43, and the opening degree of the window 42. For example, if the accelerator 40 is effective, an adjustment is made so that the accelerator 40 becomes less effective with respect to the amount of depression of the driver's accelerator pedal in order to suppress sudden acceleration of the vehicle.
After step ST11, the process returns to step ST7, and a first correlation value calculation process using the newly detected shake and vehicle shake for the first occupant is performed.

In this way, the driving support device 3 determines whether the first occupant is swinging by using the relationship between the swing of the first occupant and the swing of the vehicle other than the first occupant. Thus, the determination accuracy can be improved.

The vehicle is not limited to manual driving, but may be one that can be driven automatically. When the determination unit 33 determines that the first occupant is drunken while the vehicle is traveling by automatic driving, the control unit 34 may perform control to change the mode of automatic driving. Good. For example, when there are a plurality of modes depending on how gently acceleration and deceleration are performed as the mode of automatic operation, the control unit 34 changes to a mode in which acceleration and deceleration are performed more gently. . Whether or not to change the mode of the automatic driving may be selectable by the occupant.
Further, the control unit 34 may select the mode desired by the occupant manually instead of automatically changing the mode of automatic driving.
In addition, the control unit 34 may propose switching from automatic operation to manual operation via the notification unit 5.
Further, in the case of automatic operation, the above-described process of step ST2 can be omitted.

In addition, when the determination unit 33 determines that the occupant is drunken, the control unit 34 causes the navigation device (not illustrated) to search for a guide route with a small curve, or to perform a process of changing the guide route. May be.

Moreover, in the above, the case where the driving assistance apparatus 3 was mounted in the vehicle was shown. However, the driving support device 3 may be built in a server (not shown). In this case, the server receives image data and vehicle information of an image taken inside the vehicle from the vehicle by wireless communication. And the said server transmits the control signal which the control part 34 produces | generates toward a vehicle. In this way, the server may function as the driving support device 3.
Similarly, the driving support device 3 may be built in a smartphone brought into the vehicle.

Further, for example, the occupant swing detection unit 30 and the vehicle swing detection unit 31 may be built in a server (not shown), and the server may detect the passenger swing and the vehicle swing. In this case, the server receives image data and vehicle information of an image taken inside the vehicle from the vehicle by wireless communication, and transmits the detected passenger shake and vehicle shake to the vehicle. The swing correlation calculation unit 32, the determination unit 33, and the control unit 34 are provided in the vehicle, and perform processing using the passenger's shake and the vehicle shake received from the server. Thus, each part which comprises the driving assistance apparatus 3 may be disperse | distributed to a different place called a vehicle and a server.

In the above description, the driving support device 3 includes the control unit 34. However, the driving support device 3 does not include the control unit 34, and may be a device that simply performs processing up to determining occupant sickness. Good. For example, when the driving support device 3 uploads the location where the car sickness occurs to the server, information from a plurality of vehicles is accumulated in the server, and it becomes possible to find a bad road where the car sickness is likely to occur. .

In the above description, the reference correlation value is calculated using the occupant shake and the vehicle shake detected during the first set time from the start of driving of the vehicle. However, the reference correlation value is not calculated every time the vehicle is driven, but the reference correlation value calculated between the start of driving and the first set time when the vehicle first rides is not individual. The reference correlation value of the corresponding individual may be read out from the storage unit according to the personal authentication result of the occupant that is stored in the illustrated storage unit and then boarded on the vehicle. Further, the reference correlation value calculated for the first ride on the vehicle stored for each individual in the storage unit may be updated using the reference correlation value calculated during subsequent driving.
In addition, the reference correlation value is not calculated using the occupant swing detection unit 30, the vehicle swing detection unit 31, and the swing correlation calculation unit 32 of the driving support device 3, but already when the driving support device 3 is shipped. The reference correlation value may be stored in a storage unit (not shown) of the driving support device 3. For example, by generating statistical data for several people for values indicating the correlation between passenger sway and vehicle sway when there is no car sickness, an average reference correlation value for a human can be calculated. The driving assistance device 3 is shipped in a state where such an average reference correlation value is stored in the driving assistance device 3.
In addition, a test run mode for acquiring a reference correlation value is provided in the driving support device 3 so that the reference correlation value calculated for each individual during traveling in the test run mode is stored in a storage unit (not shown). Also good. For example, after confirming that the driver is not intoxicated, the driver instructs the driving support device 3 to execute the test run mode.
In short, the reference correlation value only needs to indicate the correlation between the swaying of the occupant and the swaying of the vehicle when the occupant is not sick.

When the reference correlation value stored in the storage unit (not shown) as described above is used in the process, the driving support device 3 acquires the reference correlation value from the storage unit and outputs the reference correlation value to the determination unit 33 (not shown). The reference correlation value acquisition unit is provided. In the description using the flowchart of FIG. 4, as can be seen from the fact that the rocking correlation calculation unit 32 calculates the reference correlation value and outputs the reference correlation value to the determination unit 33 in step ST <b> 6, the rocking correlation calculation unit 32 It functions as a reference correlation value acquisition unit that acquires correlation values.

As described above, the driving support apparatus 3 according to Embodiment 1 calculates the first correlation value indicating the correlation between the passenger's shake and the vehicle's shake, and the first correlation value indicates that the passenger is not sick. It is determined that the occupant is intoxicated when the reference correlation value indicating the correlation between the occupant's sway and the vehicle's sway is more than the setting. The determination accuracy can be improved by determining the car sickness by utilizing the relationship between the vibration of the occupant and the vibration of the vehicle other than the occupant.

In addition, when the determination unit 33 determines that the first occupant is drunken, the driving support device 3 notifies that the occupant is drunken or controls the vehicle. Is provided. Therefore, it is possible to notify the driver of the occurrence of car sickness or to control the vehicle so that the car sickness does not deteriorate.

Further, the control unit 34 adjusts at least one of the effect of the accelerator 40, the effect of the brake 41, the effect of the handle 44, the effect of the suspension 43, and the opening degree of the window 42. Therefore, it is possible to adjust factors that affect car sickness.

Further, when the determination unit 33 determines that the first occupant is drunken while the vehicle is traveling by automatic driving, the control unit 34 changes the mode of automatic driving. Therefore, automatic driving considering car sickness is possible.

Embodiment 2. FIG.
When a plurality of occupants are present in the vehicle and one of them begins to get intoxicated, the occupant who is intoxicated tends to swing differently from other occupants. In the second embodiment, a description will be given of a mode in which an occupant who is shaking differently from other occupants is determined to be intoxicated.
FIG. 5 is a block diagram showing the configuration of the driving support device 3A according to the second embodiment and its surroundings. Components having the same or corresponding functions as those already described in the first embodiment are denoted by the same reference numerals, and description thereof is omitted or simplified as appropriate.

The driving support device 3A is capable of executing processing for determining car sickness when there are at least three passengers.
The occupant swing detection unit 30 acquires the image data output by the occupant detection unit 1, and detects occupant swings for a plurality of occupants using the image data. The occupant swing detection unit 30 outputs each detected swing of the occupants to the swing correlation calculation unit 32.

The rocking correlation calculation unit 32 calculates a second correlation value that is a correlation value indicating the correlation between the shaking of the occupant detected by the occupant rocking detection unit 30 and the shaking of a plurality of occupants other than the occupant. Similar to the first correlation value and the reference correlation value described in the first embodiment, the second correlation value can be calculated using, for example, equations (4) to (8) disclosed in the non-patent document. it can. The swing correlation calculation unit 32 outputs the calculated second correlation value to the determination unit 33.

The determination unit 33 uses the second correlation value calculated by the rocking correlation calculation unit 32 to determine whether the occupant is intoxicated.

Next, an example of processing performed by the driving support apparatus 3A configured as described above will be described with reference to the flowchart shown in FIG. The process shown in FIG. 6 is started when the driving of the vehicle is started, for example.
Following the processing of step ST1 described in the first embodiment, the occupant swing detection unit 30 determines whether three or more occupants are on board using the image data from the occupant detection unit 1 (step ST20). .

When the number of passengers is less than three (step ST20; NO), the process of step ST1 is performed again. When the number of passengers is less than three, the process of step ST1 is not performed again, and the driving support device 3A may end the process as it is.
On the other hand, when three or more occupants are on board (step ST20; YES), the occupant swing detection unit 30 detects the sway of the occupant using the image data from the occupant detection unit 1 (step ST21). In the following description, it is assumed that there are three occupants in the vehicle and that the occupant swing detection unit 30 detects the swings of the first occupant, the second occupant, and the third occupant. The occupant swing detection unit 30 outputs the detected first occupant swing, second occupant swing, and third occupant swing to the swing correlation calculation unit 32, respectively.

The swing correlation calculation unit 32 calculates a second correlation value indicating the correlation between the swing of one passenger and the swings of a plurality of passengers other than the passenger (step ST22). For example, in calculating the second correlation value indicating the correlation between the shaking of the first occupant and the shaking of the second and third occupants other than the first occupant, the average of the shaking of the second occupant and the shaking of the third occupant, And the shaking of the first occupant is used. That is, the second correlation value indicating the correlation between the shaking of the first occupant and the shaking of the second and third occupants other than the first occupant is that the shaking of the first occupant is the shaking of the second occupant and the shaking of the third occupant. It shows how much it correlates with the average. The second correlation value indicating the correlation between the shaking of the second occupant and the shaking of the first and third occupants other than the second occupant, the shaking of the third occupant and the shaking of the first and second occupants other than the third occupant The same applies to the second correlation value indicating the correlation.
As described above, the swing correlation calculation unit 32 calculates, for each occupant, the second correlation value indicating the correlation between the swing of one occupant and the swings of a plurality of occupants other than the occupant. Then, the rocking correlation calculation unit 32 outputs the lowest second correlation value, that is, the second correlation value indicating the lowest correlation to the determination unit 33. Here, the second correlation value indicating the correlation between the shake of the first occupant and the shake of the second and third occupants other than the first occupant is the lowest among the three second correlation values calculated for each occupant. Suppose that

Subsequently, the determination unit 33 sets the second correlation value, which is calculated by the swing correlation calculation unit 32, and indicates the correlation between the swing of the first occupant and the swings of the second and third occupants other than the first occupant. It is determined whether the value is greater than or equal to the value (step ST23). When the second correlation value is less than the set value, that is, when the correlation greater than the set value is not indicated (step ST23; NO), the determination unit 33 notifies the control unit 34 of the determination result that the first occupant is intoxicated. Output. On the other hand, when the second correlation value is greater than or equal to the set value, that is, when the correlation is greater than or equal to the set value (step ST23; YES), the determination unit 33 determines that the first occupant is not sick, and the process proceeds to step ST21. And the second correlation value calculation process using the newly detected shake for each occupant is performed.
The processes in steps ST9 to ST11 in FIG. 6 are as described in the first embodiment.

In this way, the driving assistance device 3A does not simply determine whether the first occupant is shaking, but uses the relationship between the shaking of the first occupant and the shaking of the second and third occupants other than the first occupant. The determination accuracy can be improved by determining the sickness.

Note that the motion sickness determination method shown in the second embodiment may be used in combination with the motion sickness determination method shown in the first embodiment. That is, the determination unit 33 is a case where the first correlation value indicating the correlation between the shake of the first occupant and the shake of the vehicle deviates from the reference correlation value by more than a set value, and the first occupant shake and the first occupant It is determined that the first occupant is intoxicated when the second correlation value indicating the correlation with the shaking of the second and third occupants other than the above indicates no correlation greater than the set value.

As described above, the driving assistance apparatus 3A according to Embodiment 2 calculates the second correlation value indicating the correlation between the swaying of the occupant and the swaying of a plurality of occupants other than the occupant, and the second correlation value is set. When the above correlation is not shown, it is determined that the occupant is drunken. The determination accuracy can be improved by determining the car sickness using the relationship between the vibration of the occupant and the vibration other than the occupant.

Further, within the scope of the invention, the invention of the present application can be freely combined with each embodiment, modified with any component in each embodiment, or omitted with any component in each embodiment. is there.

As described above, the driving assistance apparatus according to the present invention has improved accuracy of determination of car sickness, and thus can be used, for example, to be mounted on a vehicle and assist driving of the driver.

1 occupant detection unit, 2 vehicle state detection unit, 3, 3A driving support device, 4 vehicle unit, 5 notification unit, 10 camera, 11 stereo camera, 12 TOF camera, 20 CAN signal, 21 acceleration sensor, 30 occupant swing detection Unit, 31 vehicle swing detection unit, 32 swing correlation calculation unit, 33 determination unit, 34 control unit, 40 accelerator, 41 brake, 42 window, 43 suspension, 44 handle, 50 indicator, 51 speaker, 100 processing circuit, 101 Memory, 102 CPU.

Claims

An occupant swing detection unit that detects the swing of the first occupant using image data of an image taken inside the vehicle;
A vehicle sway detector that detects vehicle sway using vehicle information;
A rocking correlation calculating unit that calculates a first correlation value indicating a correlation between the swing of the first occupant detected by the occupant rocking detection unit and the vibration of the vehicle detected by the vehicle rocking detection unit;
A reference correlation value acquisition unit that acquires a reference correlation value indicating a correlation between the vibration of the passenger and the vibration of the vehicle when the passenger is not intoxicated;
And a determination unit that determines that the first occupant is intoxicated when the first correlation value calculated by the swing correlation calculation unit deviates from the reference correlation value by a setting or more. A driving support device.
An occupant swing detection unit that detects occupant swing for a plurality of occupants using image data of an image taken inside the vehicle;
A rocking correlation calculating unit that calculates a second correlation value indicating a correlation between the shaking of the first occupant detected by the occupant rocking detection unit and the shaking of a plurality of occupants other than the first occupant;
And a determination unit that determines that the first occupant is drunken when the second correlation value calculated by the swing correlation calculation unit does not indicate a correlation greater than or equal to the setting. Support device.
The control unit for notifying that the occupant is intoxicated or controlling the vehicle when the determining unit determines that the first occupant is intoxicated. The driving support device according to claim 1 or 2.
The driving support device according to claim 3, wherein the control unit adjusts at least one of an accelerator effect, a brake effect, a handle effect, a suspension effect, and a window opening condition.
The swing correlation calculating unit calculates a second correlation value indicating a correlation between the swing of the first occupant detected by the occupant swing detection unit and the swings of a plurality of occupants other than the first occupant;
The determination unit determines that the first occupant is intoxicated when the first correlation value deviates from the reference correlation value by a setting or more and the second correlation value does not indicate a setting or more. The driving support device according to claim 1, wherein it is determined that the vehicle is operating.
The control unit changes the mode of automatic driving when the determination unit determines that the first occupant is drunken while the vehicle is traveling by automatic driving. The driving assistance apparatus as described.
An occupant swing detection step in which the occupant swing detection unit detects the swing of the first occupant using image data of an image taken inside the vehicle;
A vehicle swing detection step in which the vehicle swing detection unit detects the swing of the vehicle using the vehicle information;
A swing for calculating a first correlation value indicating a correlation between the swing of the first occupant detected by the occupant swing detection step and the swing of the vehicle detected by the vehicle swing detection step. A correlation calculation step;
A reference correlation value acquisition unit for acquiring a reference correlation value indicating a correlation between the shaking of the occupant and the shaking of the vehicle when the occupant is not sick;
A determination step in which the determination unit determines that the first occupant is intoxicated when the first correlation value calculated in the swing correlation calculation step deviates from the reference correlation value by a setting or more; A driving support method comprising the steps of: