JP2012128665A - Driving support device, driving support method and driving support program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a driving support device, a driving support method and a driving support program capable of reducing unnecessary notification for a driver.SOLUTION: The driving support device comprises: means for estimating a predicted relative distance between an own vehicle of the driver and a surrounding vehicle; means for comparing a measured relative distance between the own vehicle and the surrounding vehicle with the predicted relative distance; and means for notifying the driver when the comparison result indicates that the measured relative distance exceeds more than a predetermined value from the predicted relative distance.

Description

  The present invention relates to a driving support device, a driving support method, and a driving support program that support driving of a vehicle.

  2. Description of the Related Art In recent years, an apparatus that supports safe driving by notifying a driver who drives a vehicle has been proposed (see, for example, Patent Document 1). It is known that such a device may make the driver feel bothersome depending on the driver's situation and notification content. The conventional device obtains the gaze frequency at which the measurement gaze direction that the driver is gazing and the required gaze direction that the driver should gaze at, and should be notified according to the notification level determined based on the gaze frequency. Information was broadcast.

JP 7-167668 A

  However, for conventional devices that are considered to be recognized by the driver from the viewpoint of gaze frequency, it is possible to reduce the annoyance by lowering the notification level, lowering the volume, and using inconspicuous colors. It is to reduce. Conventional devices have lowered the notification level to lower the volume or use inconspicuous colors, but since there is no change in notifying the alarm, etc., there is a possibility that the driver's troublesomeness cannot be reduced. there were.

  An object of the present embodiment is to provide a driving support device, a driving support method, and a driving support program that can reduce unnecessary notifications to the driver.

  In order to solve the above-described problem, the present embodiment includes an estimation unit that estimates a predicted relative distance between a host vehicle and a surrounding vehicle by a driver, an actual measured relative distance between the host vehicle and the surrounding vehicle, and the predicted relative distance. Comparing means for comparing, and notifying means for notifying the driver when the measured relative distance deviates from the predicted relative distance by a predetermined value or more as a result of the comparison.

  In addition, what applied the arbitrary combination of the component of this embodiment, expression, or a component to a method, an apparatus, a system, a computer program, a recording medium, a data structure, etc. is also effective as an aspect of this invention.

  According to this embodiment, it is possible to provide a driving support device, a driving support method, and a driving support program that can reduce unnecessary notifications to the driver.

It is explanatory drawing of an example showing the timing of the notification to a driver | operator. It is explanatory drawing of an example showing the timing of the notification to a driver | operator in a present Example. It is a block diagram of one Example of the system containing the driving assistance device of a present Example. It is a hardware block diagram of an example of a driving assistance device. It is a block block diagram of one Example of a driving assistance device. It is a block block diagram of one Example of a visual line measuring apparatus. It is a block block diagram of one Example of the visual confirmation part of a driving assistance device. It is a processing image figure of one Example of a mirror incident image | video confirmation part. It is a block block diagram of one Example of a mirror incident image | video confirmation part. It is a block block diagram of one Example of the warning process part of a driving assistance device. It is a flowchart of an example of the process which calculates the speed and acceleration of a surrounding vehicle. It is a flowchart of an example of a vehicle visual confirmation process and a warning process. It is explanatory drawing of an example of the motion prediction method of a surrounding vehicle. It is an image figure of an example showing data required in a present Example. It is explanatory drawing showing an example of the data format and data memorize | stored in an information storage part.

  Next, modes for carrying out the present invention will be described based on the following embodiments with reference to the drawings. In this embodiment, a vehicle (automobile) will be described as an example of a vehicle.

  FIG. 1 is an explanatory diagram showing an example of the timing of notification to the driver. FIG. 1A is an example in which the driver of the vehicle feels annoyance. FIG. 1A shows a trajectory 1 of the own vehicle and a trajectory 2 of the rear vehicle. At time T1, the driver of the own vehicle visually confirms the rear and grasps the distance 3 from the rear vehicle approaching from the rear and the relative speed with the rear vehicle.

  At the time of visual observation, the driver of the own vehicle predicts the trajectory of the rear vehicle from the grasped distance 3 from the rear vehicle and the relative speed with the rear vehicle. FIG. 1 (A) also shows the locus 4 of the rear vehicle predicted by the driver. In the range 5, the trajectory 2 of the rear vehicle and the trajectory 4 of the rear vehicle predicted by the driver of the own vehicle are hardly deviated. That is, in the range 5, the trajectory 2 of the rear vehicle is close to the prediction of the driver of the own vehicle, so that even if the driver of the own vehicle is notified, the driver of the own vehicle feels bothered.

  FIG. 1B is an example of timing to be notified to the driver of the own vehicle. FIG. 1B shows a locus 6 of the own vehicle and a locus 7 of the rear vehicle. At time T2, the driver of the own vehicle visually confirms the rear, and grasps the distance 8 from the rear vehicle approaching from behind and the relative speed with the rear vehicle.

  At the time of visual observation, the driver of the own vehicle predicts the trajectory of the rear vehicle from the grasped distance 8 from the rear vehicle and the relative speed with the rear vehicle. FIG. 1 (B) also shows the locus 9 of the rear vehicle predicted by the driver. In the range 10, the locus 7 of the rear vehicle and the locus 9 of the rear vehicle predicted by the driver of the own vehicle are greatly deviated. That is, in the range 10, the locus 7 of the rear vehicle is greatly deviated from the locus 9 predicted by the driver of the own vehicle, and the distance between the own vehicle and the rear vehicle is greatly deviated from the distance predicted by the driver of the own vehicle. So you should notify your driver.

  FIG. 2 is an explanatory diagram of an example showing the timing of notification to the driver in this embodiment. FIG. 2 shows a trajectory 11 of the own vehicle and a trajectory 12 of the rear vehicle. At time T3, the driver of the own vehicle visually confirms the rear, and grasps the distance 13 from the rear vehicle approaching from the rear and the relative speed with the rear vehicle.

  At the time of visual inspection, the driver of the own vehicle predicts the locus of the rear vehicle from the grasped distance 13 from the rear vehicle and the relative speed with the rear vehicle. FIG. 2 also shows the locus 14 of the rear vehicle that is assumed to be predicted by the driver. A range 15 represented by a triangle represents a locus 14 of the rear vehicle, which is assumed to be predicted by the driver of the own vehicle, with a width. A range 15 indicated by a triangle expands with time.

  At time T4, before the rear vehicle track 12 deviates from the range 15 represented by the triangle, the driver of the own vehicle checks the rear again visually, so that the distance 16 from the rear vehicle approaching from the rear is Know the relative speed with the car behind you. In this way, at time T4, the driver of the own vehicle again confirmed the rear by visual observation. Therefore, the distance between the own vehicle and the rear vehicle coincides with the distance predicted by the driver of the own vehicle, and the driving of the own vehicle is performed. Notification to the person becomes unnecessary. Therefore, in the present embodiment, unnecessary notification to the driver of the own vehicle is suppressed at time T4, and the driver of the own vehicle is not bothered.

  At the time of visual inspection at time T4, the driver of the own vehicle predicts the locus of the rear vehicle from the grasped distance 16 from the rear vehicle and the relative speed with respect to the rear vehicle. FIG. 2 also shows a trajectory 17 of the rear vehicle that is assumed to be predicted by the driver. A range 18 represented by a triangle represents a trajectory 17 of a rear vehicle, which is assumed to be predicted by the driver of the own vehicle, with a width. A range 18 represented by a triangle expands with time.

  In the range 19, the locus 12 of the rear vehicle deviates from the range 18 represented by a triangle. As described above, in the range 19, since the driver of the own vehicle has not confirmed the rear by visual observation for a long time, the distance between the own vehicle and the rear vehicle is greatly deviated from the distance predicted by the driver of the own vehicle. ing.

  In other words, in the range 19, the trajectory 12 of the rear vehicle is greatly deviated from the trajectory 17 predicted by the driver of the own vehicle, and the distance between the own vehicle and the rear vehicle is greatly deviated from the distance predicted by the driver of the own vehicle. So you should notify your driver. Therefore, in the present embodiment, in the range 19, the driver of the own vehicle is notified of the necessity of the backward confirmation by visual observation, for example. As described above, in this embodiment, it is possible to notify the approach (movement) of the rear vehicle that the driver of the own vehicle is not aware of (not deviated from the prediction).

  In this embodiment, the driver of the own vehicle visually confirms the rear and does not notify the content estimated to be grasped, so that the driver of the own vehicle does not feel bothered. Further, in this embodiment, when there is a movement of a rear vehicle that is out of the prediction of the driver of the own vehicle, it is possible to notify the driver of the own vehicle of the timing for confirming the rear. For example, when there is an approaching vehicle that is not within the prediction of the driver of the vehicle, a warning sign is generated and the driver of the vehicle is notified, so in this embodiment, the timing at which the vehicle should be checked backwards. Can be notified. That is, in this embodiment, the driver of the own vehicle does not notify the necessity of the rear confirmation only because the driver has not confirmed the rear by visual observation for a long time or because the rear vehicle has approached.

  FIG. 3 is a configuration diagram of an embodiment of a system including the driving support apparatus of the present embodiment. 3 includes a line-of-sight measurement device 21, an ambient distance measurement detection device 22, an ambient camera 23, a driving support device 24, a navigation device 25, an audio output device 26, a display device 27, an acceleration sensor 28 and a velocity sensor 29, CAN ( Controller Area Network) bus 30. The system 20 in FIG. 3 is an example in which the driving support device 24 is realized by an in-vehicle device. The driving support device 24 in FIG. 3 uses the function of the navigation device 25.

  The driving support device 24 is connected to the line-of-sight measurement device 21, the surrounding distance measurement detection device 22, and the surrounding camera 23 so that data communication is possible. The driving support device 24 is connected to the navigation device 25 so that data communication is possible. Further, the driving support device 24 is connected to the acceleration sensor 28 and the speed sensor 29 via the CAN bus 30.

  The line-of-sight measurement device 21 measures information related to the line of sight of the driver of the vehicle as will be described later. The ambient distance measurement detection device 22 measures the distance (relative distance) from the surroundings of the vehicle with an in-vehicle millimeter wave radar or the like. The surrounding camera 23 captures the surroundings of the vehicle and acquires a video outside the vehicle. The acceleration sensor 28 measures the acceleration of the vehicle. The speed sensor 29 measures the speed of the own vehicle.

  The driving support device 24 receives information on the driver's line of sight, relative distance between the vehicle and an object such as a vehicle (neighboring vehicle) around the vehicle, an image outside the vehicle, the acceleration of the vehicle, and the speed of the vehicle. The contents to be notified to the driver of the own vehicle at the timing to be notified to the driver of the own vehicle are notified as described later.

  In addition, the driving support device 24 uses the audio output device 26 and the display device 27 of the navigation device 25 to notify the driver of the vehicle by sound or video. Note that the system 20 in FIG. 3 is an example, and may have other configurations.

  The driving support device 24, the driving support method, and the driving support program according to the present embodiment are examples, and may be devices, methods, and programs having other names, for example. For example, the driving support device 24 of this embodiment may be realized as a function of the navigation device 25. The driving support device 24 of the present embodiment is configured by hardware as shown in FIG. 4, for example. FIG. 4 is a hardware configuration diagram of an example of the driving support device.

  The driving support device 24 in FIG. 4 includes a main storage device 31, an arithmetic processing device 32, an interface device 33, a recording medium reading device 34, and an auxiliary storage device 35 that are connected to each other via a bus 37.

  The main storage device 31, the arithmetic processing device 32, the interface device 33, the recording medium reading device 34, and the auxiliary storage device 35 connected to each other via the bus 37 exchange data with each other under the control of the arithmetic processing device 32. Can do. The arithmetic processing unit 32 is a central processing unit that controls operation of the entire driving support device 24.

  The interface device 33 receives data from the line-of-sight measurement device 21, the ambient distance measurement detection device 22, the ambient camera 23, the acceleration sensor 28, the speed sensor 29, and the like, and passes the contents of the data to the arithmetic processing device 32. The interface device 33 transmits data to the navigation device 25 or the like in response to an instruction from the arithmetic processing device 32.

  The auxiliary storage device 35 stores at least a driving support program as a program for exerting the functions of the driving support device 24. Then, the arithmetic processing device 32 reads out the driving support program from the auxiliary storage device 35 to the main storage device 31 and executes it. The driving support program can be recorded in a recording medium 36 that can be read by the driving support device 24.

  Examples of the recording medium 36 include a magnetic recording medium, an optical disk, a magneto-optical recording medium, and a semiconductor memory. Magnetic recording media include HDDs, flexible disks (FD), magnetic tapes (MT) and the like. Examples of the optical disc include a DVD (Digital Versatile Disc), a DVD-RAM, a CD-ROM (Compact Disc-Read Only Memory), and a CD-R (Recordable) / RW (ReWriteable). Magneto-optical recording media include MO (Magneto-Optical disk).

  In the case where the driving support program is distributed, for example, it is possible to sell a portable recording medium 36 such as a DVD or a CD-ROM in which the driving support program is recorded. For example, the driving support device 24 that executes the driving support program reads the driving support program from the recording medium 36 on which the recording medium reading device 34 has recorded the driving support program. The arithmetic processing device 32 stores the read driving support program in the main storage device 31 or the auxiliary storage device 35.

  The driving support device 24 reads the driving support program from the main storage device 31 or the auxiliary storage device 35 which is its own storage device, and executes processing according to the driving support program. The arithmetic processing unit 32 implements various processes as described later according to the driving support program.

  FIG. 5 is a block diagram of an embodiment of the driving support device. The driving support device 24 of FIG. 5 includes a visual confirmation unit 41, a warning processing unit 42, a data collection unit 43, and an information storage unit 44. The visual confirmation unit 41 confirms whether the driver of the own vehicle visually observed the surrounding vehicle through the mirror. The warning processing unit 42 estimates the transition (trajectory) of the relative distance to the surrounding vehicle predicted by the driver at the timing when the driver of the own vehicle confirms that the surrounding vehicle is viewed through the mirror. The warning processing unit 42 notifies the driver of the vehicle by sound or video when the estimated relative distance from the neighboring vehicle deviates by a predetermined value (threshold) or more from the measured relative distance.

  The data collection unit 43 receives data from, for example, the line-of-sight measurement device 21, the ambient distance measurement detection device 22, the ambient camera 23, the acceleration sensor 28, the speed sensor 29, and the like, and stores the data in the information storage unit 44. The information storage unit 44 includes data received from the line-of-sight measurement device 21, the ambient distance measurement detection device 22, the ambient camera 23, the acceleration sensor 28, the speed sensor 29, etc., as well as mirror installation data, calculated data, and temporary data, which will be described later. Various data necessary for processing is stored.

  FIG. 6 is a block diagram of an embodiment of the visual line measuring device. The line-of-sight measurement device 21 in FIG. 6 includes a face image acquisition unit 51, a viewpoint origin calculation unit 52, a line-of-sight direction calculation unit 53, and a transmission unit 54. The face image acquisition unit 51 captures the face of the driver of the own vehicle and acquires a face image. For example, the face image acquisition unit 51 irradiates a driver with an infrared light LED and photographs the driver's face using an in-vehicle camera capable of measuring infrared light.

  The viewpoint origin calculation unit 52 performs image processing on the acquired face image and calculates, for example, between the eyes as the viewpoint origin. The line-of-sight direction calculation unit 53 performs image processing on the acquired face image to calculate the driver's line-of-sight direction. For example, the line-of-sight direction calculation unit 53 performs image processing on the acquired face image, performs three-dimensional measurement of the driver's eyeball, and calculates the driver's line-of-sight direction using an algorithm such as a corneal reflection method.

  The transmission unit 54 transmits the viewpoint origin calculated by the viewpoint origin calculation unit 52 and the line-of-sight direction calculated by the line-of-sight direction calculation unit 53 to the driving support device 24. The viewpoint origin and the line-of-sight direction are included in the information regarding the line-of-sight described above.

  FIG. 7 is a block configuration diagram of an embodiment of a visual confirmation unit of the driving support device. The visual confirmation unit 41 includes a surrounding vehicle distance measurement unit 61, a mirror visual confirmation unit 62, a mirror incident video confirmation unit 63, and a vehicle visual confirmation unit 64.

  The surrounding vehicle distance measuring unit 61 determines surrounding vehicles in the vicinity of the own vehicle using the outside image around the own vehicle received from the surrounding camera 23. Then, the surrounding vehicle distance measuring unit 61 measures the determined relative position of the surrounding vehicle using the relative distance to the surroundings of the host vehicle received from the surrounding distance measurement detection device 22.

  The mirror visual confirmation unit 62 confirms whether or not a mirror exists on the extension axis of the driver's line of sight, using the viewpoint origin and the line-of-sight direction of the driver of the own vehicle received from the line-of-sight measurement device 21. The driver's line of sight is determined from the driver's viewpoint origin and line-of-sight direction. When the mirror exists on the extension axis of the driver's line of sight, the mirror visual confirmation unit 62 determines that the driver has visually observed the mirror. The mirror visual confirmation unit 62 can determine the position where the mirror is installed using the mirror installation data 65 stored in the information storage unit 44.

  The mirror incident image checking unit 63 includes the viewpoint origin and the line-of-sight direction of the driver of the own vehicle received from the line-of-sight measuring device 21, the relative position of the surrounding vehicle measured by the surrounding vehicle distance measuring unit 61, and the mirror installation data 65. Use to check if the surrounding vehicle is reflected in the mirror. For example, if the surrounding vehicle is present on the extension axis of the driver's line of sight (virtual line of sight) turned back by the mirror, the mirror incident image checking unit 63 determines that the surrounding vehicle is reflected on the mirror. The details of the processing of the mirror incident image confirmation unit 63 will be described later. The vehicle visual confirmation unit 64 determines that the driver has visually confirmed the surrounding vehicle when it is determined that the surrounding vehicle is reflected on the mirror that the driver has determined to have visually observed.

  FIG. 8 is a processing image diagram of an embodiment of the mirror incident image confirmation unit. The viewpoint origin 71 and the line-of-sight direction 72 of the driver of the own vehicle shown in FIG. The installation position where the mirror 73 of the vehicle is installed and the installation angle of the mirror 73 can be determined from the mirror installation data 65 stored in the information storage unit 44, for example. In addition, you may make it acquire the installation position and installation angle of the mirror 73 by a measurement.

  The virtual viewpoint origin 74 of the line of sight 75 through the mirror 73 and the virtual visual field range 76 through the mirror 73 can be calculated from the viewpoint origin 71, the installation position and installation angle of the mirror 73, and the shape of the mirror 73.

  The relative distance 79 from the surrounding vehicle 78 and the relative orientation (angle) 81 in the traveling direction 80 between the host vehicle 77 and the surrounding vehicle 78 shown in FIG. Using the relative positional relationship between the host vehicle 77 and the surrounding vehicle 78, the virtual viewpoint origin 74 and the virtual field of view range 76, the mirror incident image confirmation unit 63 uses the virtual field of view range 76 in which the surrounding vehicle 78 represents the field of view through the mirror 73. Determine if it is in. The mirror incident image checking unit 63 determines that the driver of the host vehicle 77 can visually recognize the surrounding vehicle 78 when the surrounding vehicle 78 is in the virtual visual field range 76. Further, the mirror incident image confirmation unit 63 determines that the driver of the host vehicle 77 cannot visually recognize the surrounding vehicle 78 when the surrounding vehicle 78 is not in the virtual visual field range 76.

  FIG. 9 is a block diagram of an embodiment of the mirror incident image confirmation unit. The mirror incident video confirmation unit 63 includes a virtual viewpoint origin calculation unit 91, a mirror installation position acquisition unit 92, a mirror installation angle acquisition unit 93, a virtual visual line visible range calculation unit 94, and a surrounding vehicle visibility determination unit 95.

  The virtual viewpoint origin calculation unit 91 receives the viewpoint origin 71 and the line-of-sight direction 72 of the driver of the own vehicle from the surrounding vehicle distance measurement unit 61. The virtual viewpoint origin calculation unit 91 receives the installation position where the mirror 73 of the own vehicle is installed from the mirror installation position acquisition unit 92. The virtual viewpoint origin calculation unit 91 receives the installation angle of the mirror 73 of the own vehicle from the mirror installation angle acquisition unit 93.

  The virtual viewpoint origin calculation unit 91 calculates the virtual viewpoint origin 74 of the line of sight 75 through the mirror 73 from the viewpoint origin 71, the installation position and installation angle of the mirror 73, and the shape of the mirror 73 as shown in FIG.

  The mirror installation position acquisition unit 92 acquires the installation position where the mirror 73 of the own vehicle is installed, for example, by measurement. The mirror installation angle acquisition unit 93 acquires the installation angle of the mirror 73 of the host vehicle, for example, by measurement.

  Further, the virtual visual line visible range calculation unit 94 calculates the virtual visual field range 76 through the mirror 73 from the viewpoint origin 71, the installation position and installation angle of the mirror 73, and the shape of the mirror 73 as shown in FIG.

  The surrounding vehicle visually recognizable determination unit 95 receives the relative distance 79 between the surrounding vehicle 78 and the relative orientation 81 in the traveling direction 80 between the own vehicle 77 and the surrounding vehicle 78 from the surrounding vehicle distance measuring unit 61. In addition, the surrounding vehicle visual recognition determination unit 95 receives the virtual viewpoint origin 74 and the virtual visual field range 76 from the virtual visual line visual recognition range calculation unit 94.

  The surrounding vehicle visual recognition determination unit 95 uses the relative positional relationship between the host vehicle 77 and the surrounding vehicle 78, the virtual viewpoint origin 74, and the virtual field of view range 76, so that the surrounding vehicle 78 represents the field of view through the mirror 73. It is determined whether it is in 76 as shown in FIG. The surrounding vehicle visually recognizable determination unit 95 determines that the driver of the host vehicle 77 can visually recognize the surrounding vehicle 78 when the surrounding vehicle 78 is in the virtual visual field range 76. The surrounding vehicle visibility determination unit 95 determines that the driver of the host vehicle 77 cannot visually recognize the surrounding vehicle 78 when the surrounding vehicle 78 does not enter the virtual visual field range 76.

  FIG. 10 is a block diagram of an embodiment of the warning processing unit of the driving support device. The warning processing unit 42 includes a surrounding vehicle distance measurement unit 101, a surrounding vehicle speed acceleration calculation unit 102, a surrounding vehicle speed acceleration recording unit 103, a surrounding vehicle speed acceleration prediction unit 104, a host vehicle speed acceleration prediction unit 105, and a relative distance prediction unit 106. , A relative distance prediction measurement comparison unit 107 and a warning notification unit 108.

  The surrounding vehicle distance measuring unit 101 discriminates surrounding vehicles around the own vehicle using the outside image around the own vehicle received from the surrounding camera 23. Then, the surrounding vehicle distance measuring unit 101 measures the relative distance from the determined surrounding vehicle using the relative distance from the surroundings of the host vehicle received from the surrounding distance measurement detection device 22. In addition, the surrounding vehicle distance measuring unit 101 stores the measured relative distance between the host vehicle and the surrounding vehicle in the information storage unit 44 as the relative distance actual measurement result data 113.

  The surrounding vehicle speed acceleration calculation unit 102 uses the relative distance between the own vehicle and the surrounding vehicle measured every predetermined time by the surrounding vehicle distance measurement unit 101 and the speed of the own vehicle measured by the speed sensor 29 as described later. Next, the speed and acceleration of surrounding vehicles are calculated.

  The peripheral vehicle speed acceleration recording unit 103 stores the speed and acceleration of the peripheral vehicle calculated by the peripheral vehicle speed acceleration calculation unit 102 as the peripheral vehicle speed acceleration data 111, for example, in the information storage unit 44.

  The surrounding vehicle speed acceleration prediction unit 104 predicts the speed and acceleration of the surrounding vehicle using the surrounding vehicle speed acceleration data 111 stored in the information storage unit 44 as described later. The own vehicle speed acceleration prediction unit 105 predicts the speed and acceleration of the own vehicle using the own vehicle speed acceleration data 112 stored in the information storage unit 44 as described later.

  The relative distance prediction unit 106 determines the transition (trajectory) of the relative distance between the own vehicle and the surrounding vehicle at the timing when the vehicle visual confirmation unit 64 determines that the driver of the own vehicle has visually confirmed the surrounding vehicle, for example, in FIG. Predict as shown. The relative distance between the host vehicle and the surrounding vehicle predicted by the relative distance prediction unit 106 is a transition of the relative distance between the host vehicle and the surrounding vehicle that the driver seems to have predicted.

  The relative distance prediction unit 106 uses the speed and acceleration of the surrounding vehicle predicted by the surrounding vehicle speed acceleration prediction unit 104 and the speed and acceleration of the host vehicle predicted by the host vehicle speed acceleration prediction unit 105 as described later. The transition of the relative distance between the host vehicle and the surrounding vehicle that the driver thinks has been predicted is predicted. The relative distance prediction unit 106 causes the information storage unit 44 to store the predicted transition of the relative distance between the host vehicle and the surrounding vehicle as the relative distance prediction result data 114.

  The relative distance prediction measurement comparison unit 107 uses the relative distance measurement result data 113 and the relative distance prediction result data 114 stored in the information storage unit 44 to measure the relative distance between the measured (current) own vehicle and the surrounding vehicle. Compare the relative distance between the host vehicle and the surrounding vehicle that the driver thinks predicted.

  Based on the comparison result by the relative distance prediction actual measurement comparison unit 107, the warning notification unit 108 has a relative distance between the actually measured vehicle and the surrounding vehicle, and a relative distance between the own vehicle and the surrounding vehicle that the driver thinks predicted is greater than or equal to a threshold value. When there is a divergence, the driver is notified of the warning by sound or video. In the example of FIG. 2, when the rear vehicle locus 12 deviates from a triangular range 18, the measured relative distance between the own vehicle and the surrounding vehicle, and the relative relationship between the own vehicle and the surrounding vehicle that the driver seems to have predicted. The distance is more than the threshold.

  FIG. 11 is a flowchart of an example of processing for calculating the speed and acceleration of surrounding vehicles. In step S <b> 1, the surrounding vehicle distance measuring unit 101 measures the relative distance to the surrounding vehicle and stores the relative distance measurement result data 113 in the information storage unit 44.

  In step S <b> 2, the peripheral vehicle speed acceleration calculation unit 102 reads the relative distance measurement result data 113 one step before and the speed of the host vehicle measured by the speed sensor 29 from the information storage unit 44. In step S3, the peripheral vehicle speed acceleration calculation unit 102 calculates the relative distance measurement result data 113 of the previous step, the speed of the vehicle measured by the speed sensor 29, and the relative distance measurement result data 113 of the current step and the speed sensor 29. The speed of the surrounding vehicle is calculated using the measured speed of the own vehicle. The peripheral vehicle speed acceleration calculation unit 102 stores the calculated peripheral vehicle speed data 111 a in the information storage unit 44.

  In step S <b> 4, the surrounding vehicle speed acceleration calculation unit 102 reads the speed of the surrounding vehicle one step before from the information storage unit 44. In step S5, the surrounding vehicle speed acceleration calculation unit 102 calculates the acceleration of the surrounding vehicle using the speed of the surrounding vehicle one step before and the speed of the surrounding vehicle at the current step. The surrounding vehicle speed acceleration calculation unit 102 stores the calculated acceleration data 111b of the surrounding vehicle in the information storage unit 44.

  FIG. 12 is a flowchart of an example of the vehicle visual confirmation process and the warning process. In step S <b> 11, the line-of-sight measurement device 21 measures the viewpoint origin and line-of-sight direction of the driver of the host vehicle, and transmits the measurement to the driving support device 24.

  In step S12, the visual confirmation unit 41 of the driving assistance device 24 confirms whether the driver's line of sight is facing the mirror, in other words, whether the driver of the own vehicle is looking at the mirror. If the line of sight of the driver of the host vehicle does not face the mirror, the visual confirmation unit 41 again measures the viewpoint origin and the line of sight of the driver of the host vehicle and transmits it to the driving support device 24 in step S11.

  If the driver's line of sight is facing the mirror, in step S13, the visual confirmation unit 41 confirms that the surrounding vehicle is reflected on the mirror, in other words, the driver of the own vehicle visually observed the surrounding vehicle through the mirror. To check. If the surrounding vehicle is reflected in the mirror, the visual confirmation unit 41 notifies the warning processing unit 42. If the surrounding vehicle is not reflected on the mirror, the visual confirmation unit 41 again measures the viewpoint origin and the line-of-sight direction of the driver of the own vehicle in step S <b> 11, and transmits it to the driving support device 24.

  When the visual confirmation unit 41 notifies that the driver of the own vehicle has viewed the surrounding vehicle through the mirror, the warning processing unit 42 in step S14, the speed data 111a and acceleration data 111b of the surrounding vehicle reflected in the mirror. Then, the relative distance measurement result data 113 is read from the information storage unit 44 as a motion parameter of the surrounding vehicle. In step S <b> 15, the warning processing unit 42 predicts the transition (trajectory) of the relative distance between the host vehicle and the surrounding vehicle, and stores it in the information storage unit 44 as the relative distance prediction result data 114.

  In step S <b> 16, the relative distance prediction actual measurement comparison unit 107 uses the relative distance actual measurement result data 113 and the relative distance prediction result data 114 stored in the information storage unit 44 to actually measure (current) own vehicle and surrounding vehicles. Is compared with the relative distance (predicted position) between the host vehicle and the surrounding vehicle that the driver seems to have predicted, and it is determined whether the difference is equal to or greater than a threshold value.

  If the difference is greater than or equal to the threshold value, the warning notification unit 108 turns on an alarm to the driver of the vehicle in step S17. If the difference is not greater than or equal to the threshold value, in step S18, the visual confirmation unit 41 again measures the viewpoint origin and the line-of-sight direction of the driver of the host vehicle, and transmits it to the driving support device 24.

  In step S19, the visual confirmation unit 41 of the driving support device 24 confirms whether the driver's line of sight is facing the mirror, in other words, whether the driver of the own vehicle is looking at the mirror. If the line of sight of the driver of the host vehicle does not face the mirror, the warning processing unit 42 again predicts the transition (trajectory) of the relative distance between the host vehicle and the surrounding vehicle in step S15, as the relative distance prediction result data 114. After the information is stored in the information storage unit 44, the processing is continued.

  If the driver's line of sight is facing the mirror, in step S20, the visual confirmation unit 41 shows that the surrounding vehicle is reflected on the mirror, in other words, whether the driver of the own vehicle has viewed the surrounding vehicle through the mirror. Confirm. If the surrounding vehicle is not reflected in the mirror, the visual confirmation unit 41 again predicts the transition (trajectory) of the relative distance between the host vehicle and the surrounding vehicle in step S15, and stores the relative distance prediction result data 114 in the information storage unit 44. After memorizing, continue processing.

  If the surrounding vehicle is reflected in the mirror, the warning processing unit 42 receives the speed data 111a, acceleration data 111b, and relative distance measurement result data 113 of the surrounding vehicle reflected in the mirror again as information on the movement parameters of the surrounding vehicle in step S14. The processing is continued after reading from the storage unit 44.

  FIG. 13 is an explanatory diagram of an example of a motion prediction method for surrounding vehicles. FIG. 13 shows a trajectory 120 of the speed data of the own vehicle and a trajectory 121 of the speed data of the surrounding vehicles. At time T5, the driver of the own vehicle performs a backward confirmation visually.

  The relative distance prediction unit 106 refers to past speed data of the host vehicle for a predetermined time 122 from time T5 when the driver of the host vehicle visually confirms the rear. Further, the relative distance prediction unit 106 refers to the past speed data of surrounding vehicles for a predetermined time 122 from the time T5 when the driver of the own vehicle visually confirms the rear.

  The relative distance prediction unit 106 refers to the speed V1 (t5) of the host vehicle at time T5. The relative distance prediction unit 106 refers to the speed V2 (t5) of the surrounding vehicle at time T5. Further, the relative distance prediction unit 106 refers to the maximum speed fluctuation range ΔV1 of the host vehicle within the range of the predetermined time 122 and the maximum speed fluctuation range ΔV2 of the surrounding vehicles within the range of the predetermined time 122.

  That is, the relative distance prediction unit 106 uses the timing at which the driver of the own vehicle visually confirms the rear as the motion prediction start point. The speed V1 of the host vehicle and the speed V2 of the surrounding vehicles within the predetermined time 122 are obtained from the following formulas (1) and (2).

Speed of own vehicle V1 = V1 (t5) ± ΔV1 / 2 (1)
Speed of surrounding vehicle V2 = V2 (t5) ± ΔV2 / 2 (2)
Similar to the speed V1 of the host vehicle and the speed V2 of the surrounding vehicles, the relative distance prediction unit 106 obtains the acceleration α1 of the own vehicle and the acceleration α2 of the surrounding vehicles. Then, the relative distance prediction unit 106 performs motion prediction using the obtained vehicle speed V1, the speed V2 of the surrounding vehicle, the acceleration α1 of the own vehicle, and the acceleration α2 of the surrounding vehicle.

  In the example shown in FIG. 13, the driver of the own vehicle visually confirms the rear and uses the center value of the speed, acceleration, and distance (inter-vehicle distance) of the own vehicle and the surrounding vehicle at the time when the surrounding vehicle is visually recognized. Make motion predictions. In the example shown in FIG. 13, it is assumed that the speed, acceleration, and distance (inter-vehicle distance) of the surrounding vehicles at the time when the driver of the own vehicle visually recognizes the vehicle.

  FIG. 14 is an image diagram showing an example of data necessary for this embodiment. As shown in FIG. 14, in this embodiment, the speed and acceleration 131 of the own vehicle, the line of sight 132 of the driver of the own vehicle, the viewpoint origin 133 of the driver of the own vehicle, the installation position and installation angle 134 of the mirror, and the surrounding vehicles The relative distance and orientation 135 between the rear vehicle and the rear vehicle of the vehicle and the host vehicle, and the speed and acceleration 136 of the rear vehicle and the rear vehicle as surrounding vehicles are required. The data shown in FIG. 14 is acquired for every sampling time, for example, and stored in the information storage unit 44.

  FIG. 15 is an explanatory diagram showing an example of a data format and data stored in the information storage unit. Note that the data format and data shown in FIG. 15 indicate a part of the data format and data stored in the information storage unit 44.

  The data format of FIG. 15 includes, as data items, room mirror confirmation, right door mirror confirmation, left door mirror confirmation, presence / absence of vehicle in range of mirror, presence / absence of vehicle in right door mirror, presence / absence of vehicle in left door mirror, own vehicle speed, confirmation vehicle relative distance, It has confirmation vehicle relative speed, estimated relative distance, and relative distance estimation error.

  The data item “time series number” is an index number assigned every predetermined time (for example, 0.1 second). In the data item “room mirror confirmation”, 1 is input when it is determined that the driver has viewed the room mirror, and 0 is input otherwise. In the data item “confirm right door mirror”, 1 is input when it is determined that the driver has viewed the right door mirror, and 0 is input otherwise. In the data item “confirm left door mirror”, 1 is input when it is determined that the driver has viewed the left door mirror, and 0 is input otherwise.

  For the data item “presence / absence of vehicle in room mirror range”, 1 is input when it is determined that the surrounding vehicle is reflected in the room mirror and it is determined that the room mirror is viewed, and 0 is input otherwise. In the data item “presence / absence of vehicle in right door mirror”, 1 is input when it is determined that the surrounding vehicle is reflected in the right door mirror and it is determined that the right door mirror is viewed, and 0 is input otherwise. The data item “presence / absence of vehicle in left door mirror” is set to 1 when it is determined that the surrounding vehicle is reflected in the left door mirror and it is determined that the left door mirror is viewed, and 0 is input otherwise.

  In the data item “own vehicle speed”, the speed (km / h) of the own vehicle is input. In the data item “recognized vehicle relative distance”, a relative distance (m) to a confirmed vehicle (peripheral vehicle) confirmed by a driver using a mirror such as a room mirror, a right door mirror, or a left door mirror is input. Further, when the surrounding vehicle is not confirmed by the mirror, non-numeric data (NA) is input to the data item “confirmed vehicle relative distance”.

  In the data item “confirmed vehicle relative speed”, the relative speed (m / s) with respect to the surrounding vehicle confirmed by the driver using the mirror is input. When the surrounding vehicle is not confirmed by the mirror, non-numeric data (NA) is input to the data item “confirmed vehicle relative speed”.

  In the data item “estimated relative distance”, the relative distance (m) between the host vehicle and the surrounding vehicle predicted by the relative distance prediction unit 106 is input as the estimated relative distance. When the surrounding vehicle is not confirmed by the mirror, non-numeric data (NA) is input to the data item “estimated relative distance”.

  The data item “relative distance estimation error” is the relative distance (m) from the surrounding vehicle input in the data item “confirmed vehicle relative distance” and relative to the surrounding vehicle input in the data item “estimated relative distance”. A difference from the distance (m) is input. In addition, when the surrounding vehicle is not confirmed by the mirror, non-numeric data (NA) is input to the data item “relative distance estimation error”.

  The usage sequence of the data shown in FIG. 15 is as follows. Since it is not determined that the driver has looked at the mirror from time series numbers 1 to 99, data that is not a numerical value for the confirmation vehicle relative distance, the confirmation vehicle relative speed, the estimated relative distance, and the relative distance estimation error (NA) Is entered.

  In addition, it is determined that the driver has seen the rear view mirror from time series numbers 100 to 104, but since the surrounding vehicle is not reflected in the rear view mirror, the confirmation vehicle relative distance, the confirmation vehicle relative speed, the estimated relative distance, Data (NA) that is not a numerical value is input as the relative distance estimation error.

  Further, since it is not determined that the driver has looked at the mirror from time series numbers 105 to 199, data that is not a numerical value for the confirmation vehicle relative distance, the confirmation vehicle relative speed, the estimated relative distance, and the relative distance estimation error (NA). .) Is entered.

  In addition, it is determined that the driver has looked at the rearview mirror from time series numbers 200 to 204, and the surrounding vehicle is reflected in the rearview mirror, so it is determined that the surrounding vehicle is confirmed by the mirror, Input of the confirmation vehicle relative distance and the confirmation vehicle relative speed is started.

  At time series number 205, the driver's viewing of the mirror is finished. From the time series number 205, input of the estimated relative distance and the relative distance estimation error is started. In the example of FIG. 15, the confirmation vehicle relative distance is 40 m and the confirmation vehicle relative speed is 2 m / s. Note that acceleration may be used instead of speed.

  From time series numbers 205 to 209, the confirmation vehicle relative speed 2 m / s does not change. For this reason, the estimated relative distance at the time series number 210 (0.5 seconds after the estimation start) is 39 m.

  In the time series number 210, the confirmation vehicle relative speed is changed to 10 m / s. However, as the confirmation vehicle relative speed, the confirmation vehicle relative speed of 2 m / s at the time when the visual observation of the surrounding vehicle by the mirror is finished is continuously used.

  The estimated relative distance at the time series number 223 (1.8 seconds after the start of estimation) is 36.4 m. The relative distance estimation error representing the difference between the actually confirmed vehicle relative distance of 26 m and the estimated relative distance of 36.4 m is 10.4 m. When the threshold is 10 m, the relative distance estimation error exceeds the threshold at the time series number 223, so that a warning is notified to the driver of the vehicle by sound or video. The threshold value may be a value other than 10 m.

  Hereinafter, the processing procedure of the system 20 including the driving support apparatus of the present embodiment will be described. Here, a scene is assumed in which a driver is driving a car (own vehicle) on which the driving support device 24 of the present embodiment is mounted on a highway of two lanes on one side.

  For example, the surrounding distance measurement and detection device 22 of the own vehicle measures the distance from the surrounding vehicle around the own vehicle (vehicle A) at a predetermined time interval using an in-vehicle millimeter wave radar or the like. For example, if the distance to the surrounding vehicle (vehicle B) traveling diagonally behind the vehicle A is X1 in the previous measurement, X2 in the next measurement, and T is the measurement time interval, the relative speed V of the vehicle B with respect to the vehicle A is V = (X2-X1) / T. The acceleration α can be obtained in the same manner.

  The gaze measuring device 21 of the own vehicle measures the gaze of the driver of the own vehicle. For example, the driver's line of sight is measured by irradiating the driver with an infrared LED, photographing the driver's face using an in-vehicle camera that can measure infrared light, and processing the captured image to measure the eyeball three-dimensionally. And measure using an algorithm such as a corneal reflection method.

  When a surrounding vehicle exists on the extension axis of the measured driver's line of sight, the driving support device 24 defines that the driver has captured the surrounding vehicle. A peripheral vehicle on the extension axis of the line of sight (vector amount) is referred to as a directly viewable vehicle. A surrounding vehicle on the extension axis of the virtual line of sight turned back by the mirror is referred to as an indirectly visible vehicle.

  In the following description, an indirect viewable vehicle will be described. However, the present embodiment is not limited to this. It is assumed that when the vehicle B is reflected in the mirror viewed by the driver, the driver recognizes the surrounding vehicle and recognizes values close to the speed, acceleration, and distance of the vehicle B.

  The speed of the vehicle B at the moment when the driver recognizes the vehicle B is VB, the acceleration is αB, the distance between the vehicle A and the vehicle B is LAB, and each of the traveling directions of the vehicle A and the vehicle B is θAB. The driving support device 24 obtains fluctuation amounts ΔVB, ΔαB, ΔLAB, and ΔθAB of the speed, acceleration, relative distance, and relative orientation angle of the vehicle B retroactively for a predetermined time (for example, 10 seconds) from the moment when the driver recognizes the vehicle B. .

  Further, the driving support device 24 acquires the speed VA and acceleration αA of the vehicle A at the moment when the driver recognizes the vehicle B from the acceleration sensor 28 and the speed sensor 29 via the CAN bus 30. The driving support device 24 also obtains the speeds and acceleration variations ΔVA and ΔαA of the vehicle a after a predetermined time (for example, 10 seconds) from the moment when the driver recognizes the vehicle B.

  The driving support device 24 predicts that the vehicle A is moving in the speed VA ± ΔVA / 2 and acceleration αA ± ΔαA / 2, and the vehicle B is moving in the speed VB ± ΔVB / 2 and acceleration αB ± ΔαB / 2. I do.

  Motion prediction starts at the moment when the driver of the host vehicle visually recognizes the vehicle B and continues while not looking at the mirror. When indirect viewing is performed again, the motion parameters are acquired again, and motion prediction is newly started. The exercise support device 24 compares the predicted relative distance LAB with the vehicle B and the measured relative distance LAB ′ with the vehicle B.

  The exercise support device 24 can predict the approach of the vehicle B when the measured relative distance LAB ′ to the vehicle B is smaller than a predetermined value compared to the predicted relative distance LAB to the vehicle B. Notify and alert.

  For example, it is assumed that the relative speed at the moment when the driver of the own vehicle views the surrounding vehicle through the mirror is 2 m / s and the relative distance is 40 m. When the driver does not look at the 2-second mirror, the driving support device 24 assumes that the driver predicts that the relative distance is 36 m after 2 seconds from the visual inspection. However, immediately after the driver of the vehicle sees the surrounding vehicle through the mirror and accelerates 0.5 seconds later and the relative speed becomes 10 m / s, the relative distance to the vehicle B is 24 m, which is 12 m away from the prediction. Will occur.

  Here, ΔVA and ΔVB have been described without considering them. However, when prediction is performed in consideration of these motion parameters, the difference between the predicted value and the actually measured value is further increased. The system 20 of the present embodiment notifies the driver of a warning when the difference between the prediction and the actual measurement is confirmed by a predetermined value (for example, 10 m) or more. Therefore, in the system 20 of the present embodiment, it is possible to recognize the sudden approach of the vehicle B due to the driver failing to check the mirror.

  The driving support program in the present embodiment can be provided not only by package software but also by a WEB service or the like. Note that the estimation means described in the claims corresponds to the relative distance prediction unit 106, the comparison means corresponds to the relative distance prediction actual measurement comparison unit 107, and the notification means corresponds to the warning notification unit 108. The confirmation means corresponds to the visual confirmation unit 41. The mirror visual confirmation unit corresponds to the mirror visual confirmation unit 62, the mirror video confirmation unit corresponds to the mirror incident video confirmation unit 63, and the vehicle visual confirmation unit corresponds to the vehicle visual confirmation unit 64.

The present invention may have the following configurations as described below.
(Appendix 1)
An estimation means for estimating a predicted relative distance between the driver's own vehicle and the surrounding vehicle,
Comparison means for comparing the measured relative distance between the host vehicle and the surrounding vehicle and the predicted relative distance;
As a result of the comparison, the driving support device includes a notification unit that notifies the driver when the measured relative distance deviates from the predicted relative distance by a predetermined value or more.
(Appendix 2)
Further comprising confirmation means for confirming visual observation of the surrounding vehicle by the driver;
The driving support apparatus according to appendix 1, wherein the estimating means estimates a predicted relative distance between the host vehicle and a surrounding vehicle by the driver at a timing when the driver visually confirms the surrounding vehicle.
(Appendix 3)
The confirmation means is a mirror visual confirmation means for confirming the visual observation of the mirror by the driver based on the measurement result of the driver's line of sight and the position and angle of the mirror;
Mirror image confirmation means for confirming whether or not the surrounding vehicle is reflected on the mirror based on whether the surrounding vehicle exists on an extension axis of the driver's line of sight reflected by the mirror;
The driving support device according to appendix 2, further comprising: vehicle visual confirmation means for confirming visual observation of the surrounding vehicle by the driver based on visual observation of the mirror by the driver and whether the peripheral vehicle is reflected on the mirror.
(Appendix 4)
A driving support method executed by a computer,
Estimate the estimated relative distance between the driver's vehicle and the surrounding vehicle,
Compare the measured relative distance between the host vehicle and the surrounding vehicle and the predicted relative distance,
As a result of the comparison, the driver is notified when the measured relative distance deviates from the predicted relative distance by a predetermined value or more.
(Appendix 5)
Estimate the estimated relative distance between the driver's vehicle and the surrounding vehicle,
Compare the measured relative distance between the host vehicle and the surrounding vehicle and the predicted relative distance,
A driving support program for causing a computer to execute a process of notifying the driver when the measured relative distance deviates from the predicted relative distance by a predetermined value or more as a result of the comparison.

DESCRIPTION OF SYMBOLS 20 System 21 Eye-gaze measuring device 22 Ambient distance measurement detection device 23 Ambient camera 24 Driving support device 25 Navigation device 26 Audio output device 27 Display device 28 Acceleration sensor 29 Speed sensor 30 CAN (Controller Area Network) bus 31 Main memory device 32 Arithmetic processing Device 33 Interface device 34 Recording medium reading device 35 Auxiliary storage device 36 Recording medium 37 Bus 41 Visual confirmation unit 42 Warning processing unit 43 Data collection unit 44 Information storage unit 51 Face image acquisition unit 52 View point origin calculation unit 53 Gaze direction calculation unit 54 Transmission unit 61 Peripheral vehicle distance measurement unit 62 Mirror visual confirmation unit 63 Mirror incident image confirmation unit 64 Vehicle visual confirmation unit 65 Mirror installation data 91 Virtual viewpoint origin calculation unit 92 Mirror installation position acquisition unit 93 Mirror installation angle acquisition unit 94 Virtual visual line visual recognition Possible range calculation 95 Peripheral vehicle visibility determination unit 101 Peripheral vehicle distance measurement unit 102 Peripheral vehicle speed acceleration calculation unit 103 Peripheral vehicle speed acceleration recording unit 104 Peripheral vehicle speed acceleration prediction unit 105 Own vehicle speed acceleration prediction unit 106 Relative distance prediction unit 107 Relative distance Predicted actual measurement comparison unit 108 Warning notification unit

Claims (5)

An estimation means for estimating a predicted relative distance between the driver's own vehicle and the surrounding vehicle,
Comparison means for comparing the measured relative distance between the host vehicle and the surrounding vehicle and the predicted relative distance;
As a result of the comparison, the driving support device includes a notification unit that notifies the driver when the measured relative distance deviates from the predicted relative distance by a predetermined value or more. Further comprising confirmation means for confirming visual observation of the surrounding vehicle by the driver;
2. The driving support apparatus according to claim 1, wherein the estimating means estimates a predicted relative distance between the host vehicle and the surrounding vehicle by the driver at a timing when the driver visually confirms the surrounding vehicle. The confirmation means is a mirror visual confirmation means for confirming the visual observation of the mirror by the driver based on the measurement result of the driver's line of sight and the position and angle of the mirror;
Mirror image confirmation means for confirming whether or not the surrounding vehicle is reflected on the mirror based on whether the surrounding vehicle exists on an extension axis of the driver's line of sight reflected by the mirror;
The driving support device according to claim 2, further comprising: a vehicle visual confirmation unit configured to confirm visual observation of the surrounding vehicle by the driver based on visual observation of the mirror by the driver and whether the peripheral vehicle is reflected on the mirror. A driving support method executed by a computer,
Estimate the estimated relative distance between the driver's vehicle and the surrounding vehicle,
Compare the measured relative distance between the host vehicle and the surrounding vehicle and the predicted relative distance,
As a result of the comparison, the driver is notified when the measured relative distance deviates from the predicted relative distance by a predetermined value or more. Estimate the estimated relative distance between the driver's vehicle and the surrounding vehicle,
Compare the measured relative distance between the host vehicle and the surrounding vehicle and the predicted relative distance,
A driving support program for causing a computer to execute a process of notifying the driver when the measured relative distance deviates from the predicted relative distance by a predetermined value or more as a result of the comparison.

Images