JP4986135B2 - Database creation device and database creation program - Google Patents

Description

  The present invention relates to a database creation apparatus and a database creation program for creating a database related to an event that occurs while a vehicle is running, for example.

There is a technique for detecting a driver's dangerous reaction based on vehicle operation information such as brake, accelerator, and steering wheel operation amounts, and driver's biological information such as heartbeat and sweating.
For example, Patent Document 1 discloses a technique for recording information on a point where a driver feels a dangerous reaction (a near miss) such as “a pedestrian jumps out” or “an oversight of another vehicle in a blind spot” in a database. Has been proposed.
JP 2007-47914 A

In the recording system described in Patent Document 1, for example, character information such as “30 km / h pedestrian jumping out of a pedestrian crossing straight” is recorded in the database as situation information when the driver feels a dangerous reaction. .
However, even if such character information is referred to, the viewer of the database cannot easily grasp the situation when the driver feels a dangerous reaction.
On the other hand, a driver who has performed a dangerous driving operation does not recognize that he has performed a dangerous driving operation.

Accordingly , an object of the present invention is to make a driver who has performed a dangerous driving operation easily and objectively recognize a dangerous driving operation situation .
It is another object of the present invention to create information that can easily grasp the situation when a change in the driver's biological information is detected.

(1) In the first aspect of the invention, from the vehicle in which the change in the biological information of the driver due to the event based on the specific cause vehicle is detected, the detection time at which the change in the biological information is detected, and the vehicle position Detection information acquisition means for acquiring detection information including: related vehicle specification means for specifying related vehicles related to each event based on the acquired detection information; and acquiring surrounding images of the specified related vehicles Based on the acquired image, reproduced image creating means for creating a reproduced image of each event, causal vehicle identifying means for identifying the cause vehicle, and the created reproduction on the identified cause vehicle There is provided a database creating apparatus comprising access information transmitting means for transmitting access information for browsing images .
(2) In the invention according to claim 2, the cause vehicle specifying means specifies the cause vehicle from the acquired surrounding image of each related vehicle. provide.
(3) In the invention according to claim 3, the image acquisition means is configured to obtain a vehicle image before and after the detection of a change in a driver's biological information from a vehicle that has acquired the detection information as a surrounding image of the related vehicle. The database creation device according to claim 1, wherein a surrounding moving image is acquired, and the reproduced image creating unit creates a reproduced image of each event based on the acquired moving image. I will provide a.
(4) In the invention described in claim 4, the access information transmitting means further transmits access information for viewing the created reproduction image to at least one of the specified related vehicles. A database creation device according to claim 1, 2, or 3 is provided.
(5) In the invention according to claim 5, the reproduction image creating means creates a reproduction image of the behavior of the cause vehicle as the reproduction image of the event. A database creation device according to item 3 or claim 4 is provided.
(6) In the invention described in claim 6, a database including detection information acquisition means, related vehicle identification means, image acquisition means, reproduction image creation means, cause vehicle identification means, and access information transmission means In the creation device, the detection information acquisition means detects the detection time and vehicle position at which the change in the biological information is detected from the vehicle in which the change in the biological information of the driver due to the event based on the specific cause vehicle is detected. A detection information acquisition function for acquiring detection information including the related vehicle specifying means, a related vehicle specifying function for specifying a related vehicle related to each event based on the acquired detection information, and the image acquisition means, An image acquisition function for acquiring the surrounding image of each identified related vehicle, and a reproduction image in which the reproduction image creation means creates a reproduction image of each event based on the acquired image And forming function, the cause vehicle identification means, causing the vehicle specifying function for specifying the cause vehicle, the access information transmitting means, the specified reason vehicle, the access information for browsing reproduced image above prepared A computer-readable database creation program for realizing an access information transmission function for transmission is provided.

According to the present invention, a reproduction image of an event that causes a change in the driver's biological information is created, and access information for viewing the created reproduction image is transmitted to the cause vehicle. It is possible to make a person easily recognize a dangerous driving operation situation objectively.
In addition, by creating a reproduction image of the event that caused the cause, it is possible to easily grasp the situation when a change in the biological information of the driver is detected.

Hereinafter, preferred embodiments of a database creation apparatus and a database creation program according to the present invention will be described in detail with reference to FIGS.
(1) Outline of Embodiment When the driver feels some danger during traveling, the driver's biometric information (heart rate and skin impedance) changes from the sense of tension. Using this characteristic, the probe car detects a driver's dangerous reaction (near-hat reaction) from changes in the driver's biological information.
The probe car is used for driving information such as vehicle position information, time information, vehicle operation information, and image information captured around the vehicle before and after detection of a driver's dangerous reaction. Information about the event that the person showed a dangerous reaction is transmitted to the data center as near-miss information.
In the data center, the near-miss information collected from the probe cars is distributed (grouped) into related information groups in units of events based on the position information and time information.

Then, the data center analyzes the content of the near-miss information (image information, vehicle information, etc.) collected in the same information group, and causes the near-miss event that the vehicle from which the near-hat information was transmitted encounters. Identify the vehicle. That is, the cause vehicle of the event that triggered the driver's dangerous reaction is specified.
The data center creates (produces) a reproduced moving image of the near-miss event based on the information on the identified cause vehicle, the image information on the vehicle related to the near-miss event, and the like. Then, the reproduction video of the near-miss event is stored in the database.
A user can view a reproduced moving image of a near-miss event later on a terminal device such as a mobile phone or a PC by accessing the database via the network.

(2) Details of Embodiment FIG. 1 is a diagram showing the configuration of a database creation system according to this embodiment.
As shown in FIG. 1, the database creation system includes a probe car 1 equipped with an information processing device and a data center 2 having a database creation device. The probe car 1 (information processing device) and the data center 2 are Connected via a wireless network.
In addition, various data can be transmitted and received between the probe cars 1 via the wireless network.
The data center 2 is provided via a network 6 so as to be connectable to a PC (personal computer) 3, a mobile phone 4 and a vehicle 5 having a network connection function. That is, the PC 3, the mobile phone 4, and the vehicle 5 can browse the information stored in the data center 2 by accessing the data center 2 via the network 6.

FIG. 2 is a diagram showing a schematic configuration of the data center 2.
The data center 2 has a function as a central server that plays a central role in the database creation system, collects (accumulates) near-miss information transmitted from a plurality of probe cars 1, and analyzes the information. .
As shown in FIG. 2, the data center 2 includes a CPU 21, a ROM 22, a RAM 23, an input device 24, an output device 25, a communication device 26, a storage unit 27, and a database 28.
The CPU 21 is a central processing unit that controls the data center 2 according to various programs and data, and performs various arithmetic processes related to database creation.

The ROM 22 stores various programs related to database creation, such as an image analysis program, a cause vehicle identification program, a reproduction moving image creation program, and the like.
The RAM 23 is a storage area that functions as a work area for the CPU 21 and temporarily stores data being processed.
The input device 24 is a device for inputting commands and data to the data center 2, and includes a keyboard and a mouse.
The output device 25 is a device that outputs the result processed by the data center 2 to the outside, and includes a display, a printer, and the like.
The communication device 26 is connected to a wired network or a wireless network, and transmits and receives various data to and from the probe car 1 and an external terminal device (PC3, mobile phone 4).

The storage unit 27 includes various information storage units such as a data table storage unit 271, a near-miss information storage unit 272, and a cause vehicle information storage unit 273.
The data table storage unit 271 stores a data table storing the identification information of the probe car 1 registered in the data center 2 and the address information of the information processing apparatus mounted on the probe car 1.
The near-miss information storage unit 272 stores near-miss information received from the probe car 1.
The cause vehicle information storage unit 273 stores information about the cause vehicle, for example, the vehicle number, the vehicle type / format, the color information, and the like specified from the result of the image analysis process.
The database 28 stores a reproduced moving image of a near-miss event.

Next, the configuration of the information processing apparatus mounted on the probe car 1 in this embodiment will be described. Note that the information processing apparatus mounted on the probe car 1 functions as an information collecting apparatus.
As shown in FIG. 1, the information processing device includes an ECU (electronic control device) 10 that controls the entire information processing device according to various programs and data. The ECU 10 includes a current position detection device 11 and a biological information acquisition device 12. The environmental information acquisition device 13, the vehicle information acquisition device 14, the storage device 15, the image input device 16, the display device 17, the audio output device 18, the input device 19, and the communication device 20 are connected.

The current position detection device 11 is for detecting the current position (absolute coordinate value composed of latitude and longitude) of the vehicle on which the driving support device is mounted, and is a GPS that measures the position of the vehicle using an artificial satellite. A receiving device is provided.
Note that the current position detection device 11 may include a geomagnetic sensor, a gyro sensor, a vehicle speed sensor, and the like that detect the geomagnetism and determine the direction of the vehicle as a device that supplements the current position detection by the GPS receiver. About a gyro sensor and a vehicle speed sensor, you may make it share with the vehicle speed sensor 144 and the gyro sensor 145 of the vehicle information acquisition apparatus 14 mentioned later, or you may make it prepare separately.
The biological information acquisition device 12 includes a heart rate sensor 121 and a skin impedance sensor 122 as sensors for acquiring the driver's biological information. Note that the sensors that acquire the driver's biological information are not limited to these, and other sensors such as a sensor that measures brain waves or a sensor that measures blood pressure may be used.
When the vehicle starts running, the heart rate and the amount of sweat are detected at predetermined time intervals and supplied to the ECU 10.

The heart rate sensor 121 is a sensor that detects the heart rate of the driver, and detects the heart rate from the pulse rate of the driver. The heart rate sensor 121 according to the present embodiment detects a heart rate by collecting a heart rate signal from the hand of a driver who is driving by an electrode disposed on a steering wheel (handle). The heart rate sensor 121 may be provided with a dedicated sensor on the driver's body such as a wrist.
Skin impedance sensor 122 is disposed on the steering and measures the skin impedance (resistance value) of the driver. Since the skin impedance value changes according to the change in the sweating state, the driver's sweating state can be detected from the change in the skin impedance.

The environmental information acquisition device 13 includes an inter-vehicle distance / relative speed measurement device 131 and an image processing device 132.
The inter-vehicle distance / relative speed measuring device 131 includes millimeter wave radars, laser radars, and the like arranged in front and rear of the vehicle, and detects inter-vehicle distances and relative speeds with the preceding vehicle and inter-vehicle distances and relative speeds with the rear vehicle. At the same time, the presence or absence of an oncoming vehicle is determined.
The image processing device 132 performs image processing of an image outside the vehicle captured by an image input device 16 (stereo camera 161), which will be described later, and detects a forward field of view (weather, rain, snow, fog, and other obstacles). Is supposed to do.

The environment information acquisition device 13 includes a time sensor as a sensor for determining the current time.
In addition to (or instead of) the image processing of the image processing device 132, the environment information acquisition device 13 is a sensor for determining the weather, a wiper sensor for detecting the operating state of the wiper, A vehicle width light sensor for detecting the on state and a head lamp sensor for detecting the on state of the headlamp are provided.

The vehicle information acquisition device 14 includes a steering wheel steering angle sensor 141, a brake pedal force sensor 142, an accelerator pedal force sensor 143, a vehicle speed sensor 144, a gyro sensor 145, and other sensors. To detect.
The steering angle sensor 141 detects the steering angle of the steering wheel, the steering wheel operation amount, the steering wheel operation torque, and the steering wheel operation status.
The brake pedal force sensor 142 detects the speed at which the brake is depressed, the brake pedal force, and the brake operation status.

Based on the value detected by the brake pedal force sensor 142, the ECU 10 determines a sudden brake, that is, determines whether or not a sudden brake operation has been performed.
FIG. 3 is a diagram illustrating an example of a detection result of the brake pedal force sensor 142.
For example, the ECU 10 determines that the sudden braking operation has been performed when the brake pedal force satisfies the condition of the average value ± 3Σ or more or the brake operation state average value ± 3Σ or more.
The average value indicates the average during driving by the current driver.

The accelerator pedal force sensor 143 detects the speed at which the accelerator is depressed, the accelerator pedal force, and the like.
The vehicle speed sensor 144 detects the vehicle speed.
The gyro sensor 145 detects the acceleration of the vehicle.

The storage device 15 includes a ROM, a RAM, a magnetic recording medium such as a flexible disk, a hard disk, and a magnetic tape, a semiconductor recording medium such as a memory chip and an IC card, a CD-ROM, an MO, and a PD (phase change rewritable optical disk). Recording media on which information is optically read, and recording media on which data and computer programs are recorded by various methods.
As the recording medium, a different medium is used depending on the recording contents. For example, when storing information whose contents are updated (added) as needed, a recording medium capable of rewriting data is used.

  The storage device 15 includes a biological information processing program 151, a vehicle information processing program 152, an environmental information processing program 153, a near / miss determination program 154, a near / hat information storage memory 155, a vehicle information storage memory 156, and an environment information storage memory 157. Various data used in this embodiment, such as the image storage memory 158, the map database 159, the history information storage memory 160, and other data are stored.

The biological information processing program 151 is a program that obtains detection values obtained by the heart rate sensor 121 and the skin impedance sensor 122 as biological information and stores them in the RAM.
The RAM stores, as biological information, the detection results of the heart rate sensor 121 and the skin impedance sensor 122, the presence / absence of changes in the biological information determined from these detection results, and the determination result of the degree of tension (Level). Is done.
The vehicle information processing program 152 is a program that acquires detection values of the steering wheel angle sensor 141, the brake pedal force sensor 142, the accelerator pedal force sensor 143, the vehicle speed sensor 144, and the gyro sensor 145 as vehicle information and stores them in the vehicle information storage memory 156. is there.

The environmental information processing program 153 is a program that always acquires the measurement results by the inter-vehicle distance / relative speed measurement device 131 and the information of the processing results by the image processing device 132 as vehicle information and stores them in the environmental information storage memory 157.
The near-miss / hit determination program 154 was surprised or frightened by a situation where the driver felt “near” or “rapid” based on biological information, vehicle information, environmental information, etc., that is, a sudden unexpected event. This is a program for determining (detecting) the state. In other words, the near-miss determination program 154 is an unexpected event (hereinafter referred to as a near-miss event) for a driver who detects an unusual danger reaction (a near-miss reaction) such as an abnormality in biological information or an abnormality in driving operation. It is a program that detects that it has encountered.

The near-miss information storage memory 155 stores information related to a near-miss event in which a near-miss reaction (abnormality of biological information) is detected by the near-miss determination program 154. Information on such near-miss reactions and near-miss events is referred to as near-miss information.
FIG. 4 is a diagram showing an example of information stored in the near-miss information storage memory 155.
As shown in FIG. 4, the near-miss information storage memory 155 images the result of the near-miss reaction, vehicle information when encountering a near-hat event, environmental information, position information, and surroundings of the vehicle. Image data, information about the own vehicle (vehicle identification information), and the like are stored.

In the near-miss information storage memory 155, a flag indicating whether or not a near-miss reaction has been detected is stored as a determination result. As vehicle information, vehicle speed, vehicle acceleration, vehicle lateral acceleration, steering operation torque, steering operation Situation, brake pedal force, and brake operation status are saved, environmental information includes the following vehicle distance and forward vehicle relative speed, and location information includes road type, intersection determination, time, east longitude, and north latitude data. .
The near-miss information storage memory 155 stores image data of 30 seconds (15 seconds before detection + 15 seconds after detection) before and after the detection of the near-miss reaction as image information, and the vehicle identification information The vehicle number, vehicle type / format, and color data are stored.

Returning to the description of FIG. 1, the vehicle information storage memory 156 stores vehicle speed (km / h), vehicle acceleration (G), steering wheel operation torque (N · m), steering wheel operation status (N · m / ms), accelerator. The pedal effort (N · m), the brake effort (N · m), and the brake operation status (N · m / ms) are stored.
The environmental information storage memory 157 always stores detection data in the inter-vehicle distance / relative speed measuring device 131 and the image processing device 132 based on the environmental information processing program 153.

Based on the environmental information processing program 153, the image storage memory 158 always stores a surrounding image (peripheral image) of the vehicle captured by the image input device 16 (stereo camera 161).
Image storage processing in the image storage memory 158 is performed by a FIFO (first-in first-out) method, and is configured so that old image data is deleted in order. In the present embodiment, the image storage memory 158 is configured to always store (save) image data for 30 minutes.
When the ECU 10 detects a near-miss reaction, that is, an encounter with a near-miss event, the ECU 10 obtains image data for 30 seconds before and after the detection of the near-hat reaction from the image data stored in the image storage memory 158. The information is read and stored in the near-miss information storage memory 155 separately.
The map database 159 is a database in which information related to various maps such as map information, road information, and curve information is stored.
The history information storage memory 160 stores not only the processing result in the information processing apparatus but also various history information received from the data center 2 and information processing apparatuses of other vehicles. The history information storage memory 160 stores, for example, information received from the data center 2 that the vehicle is determined to be the cause vehicle of the near-miss event.

The image input device 16 includes a stereo camera 161 disposed in front of the vehicle.
The image outside the vehicle picked up by the image input device 16 is used for analysis of a near-miss event or creation of a reproduction image of a near-miss event.
The captured image is supplied to the image processing device 132 and used for visibility determination, inter-vehicle distance determination, and weather (weather) determination.
As the display device 17, various display devices such as a liquid crystal display device, a CRT, and a head-up display are used, and a warning image that alerts that a near-miss reaction (a near-miss event) in this embodiment has been detected. Warning characters (text) are displayed as part of driving assistance.
The display device 17 displays a near-miss or hat on the display screen when a map around the vehicle or a route searched for the route based on the navigation function, a searched route, a nearby facility guidance screen, or the like is displayed. A warning at a point where a reaction (near hat event) is detected may be displayed.

The audio output device 18 includes a plurality of speakers and an audio control device arranged in the vehicle. The voice output device 18 outputs a voice controlled by the voice control unit, for example, a warning voice that alerts that a near-miss reaction (a near-miss event) has been detected and alerts the driver. It has become so. The audio output device 18 may also be used as an audio speaker.
As the input device 19, various devices such as a touch panel (functioning as a switch), a keyboard, a mouse, a light pen, a joystick, an infrared remote controller, a touch panel attached to the display screen of the display device, a remote controller, and a voice recognition device can be used. It constitutes an input means for inputting various information.
The communication device 20 functions as a communication unit, connects to a wireless network, and transmits and receives various data to and from the external data center 2. In addition, transmission and reception of various data with other probe cars 1 are performed via the communication device 20.

Next, the operation of database creation processing in the database creation system configured as described above will be described.
First, a method for acquiring near-miss information in the probe car 1 will be described.
FIG. 5 is a flowchart showing an operation procedure of acquisition processing of near-miss information in the probe car 1.
This near-miss information acquisition process is executed by the information processing apparatus of the probe car 1 when the vehicle starts to run and continues to be executed while the vehicle is running, but is executed when the ignition is turned on and ends when the ignition is turned off. Good.

ECU10 detects a driver | operator's biometric information by the biometric information acquisition apparatus 12 (step 11).
And ECU10 judges the presence or absence of a change of biometric information about the detected biometric information according to predetermined biometric information determination conditions (step 12). Here, it is determined whether or not an abnormal reaction of the driver's biological information has occurred.
In this embodiment, in order to transmit (provide) near-miss information with higher accuracy to the data center 2, the presence / absence of a change in biological information is determined (detected) based on detection results of a plurality of types of biological information. . In addition, this process means judging a driver | operator's mental state (tension state) from the change of a driver | operator's biometric information.

FIG. 6 is a diagram illustrating a determination example of changes in biological information.
6A and 6B, the upper part shows the detection (measurement) result of the skin impedance sensor 122, and the lower part shows the detection (measurement) result of the heart rate sensor 121.
In the present embodiment, for example, when the skin impedance and the heart rate suddenly increase as shown in FIG. 6 (a), or when the skin impedance suddenly increases as shown in FIG. 6 (b). When the heart rate suddenly decreases, it is determined that the biological information has changed.

Returning to the description of FIG. 5, when a change in the biological information is not detected (step 12; N), the ECU 10 returns to step 11 and continues detecting the biological information.
When the change of the biological information is detected (step 12; Y), the ECU 10 refers to the vehicle information storage memory 156, acquires the vehicle information at the timing (time) when the change of the biological information occurs, and temporarily stores it in the RAM. Store (step 13).
Here, as vehicle information, vehicle speed, vehicle acceleration, steering wheel operation torque, steering wheel operation status, accelerator pedal force, brake pedal force, brake operation status, and the like are detected.

Next, the ECU 10 refers to the environment information storage memory 157, acquires the environment information at the timing (time) when the biometric information changes, and temporarily stores it in the RAM (step 14).
Here, measurement data and the like in the inter-vehicle distance / relative speed measuring device 131 and the image processing device 132 are acquired as the environmental information.

Next, based on the detected change information of the biological information, vehicle information, and environmental information, the ECU 10 should provide the data center 2 with the event that caused the abnormal reaction of the detected biological information. It is determined whether or not there is a near-miss event (step 15).
The near-miss event is determined by comparing the detected vehicle information and environmental information with a predetermined determination condition to determine whether or not the corresponding condition is satisfied.

FIG. 7 is a diagram illustrating an example of a determination item for determining whether the event is a near-miss event and its conditions.
As shown in FIG. 7, for example, the vehicle speed is 5 km / h or more, the vehicle acceleration is −0.3 G or less, and the brake pedaling force is an average value ± 3Σ or more (or a brake operation situation average value ± 3Σ or more). In this case, the ECU 10 determines that a near-miss event related to sudden braking (condition example 1) has been encountered (a near-miss event has occurred).
Similarly, when the conditions of sudden approach (condition example 2) and sudden handle (condition example 3) are satisfied, it is determined that a near-miss event has been encountered.
In the sudden braking (condition example 1), the sudden braking determination is made based on the OR condition of the brake pedal force and the brake operation status. You may make it perform based on the result of the cluster analysis plotted.

In the rapid approach (condition example 2), the determination of the rapid approach is made based on the AND condition when the distance between the vehicles ahead of the vehicle is equal to or smaller than the average value. You may make it perform based on the result of the cluster analysis which plotted the inter-vehicle distance two-dimensionally.
In the sudden handle (condition example 3), the determination of the sudden handle is performed based on the AND condition of the handle operation torque and the handle operation status. The determination of the sudden handle is based on the OR condition of the handle operation torque and the handle operation status. You may make it perform based on.
The determination of the abrupt handle may be made based on the result of cluster analysis in which the handle operation torque and the handle operation situation are two-dimensionally plotted.

Returning to the description of FIG. 5, when it is determined that a near-miss event has not occurred (step 15; N), the ECU 10 returns to step 11 and continues detecting biometric information.
When it is determined that a near-miss event has occurred (step 15; Y), the ECU 10 determines the point where the change in the biological state (abnormal reaction of the driver's biological information) is detected as the encounter point of the near-miss event. The position information is acquired from the detection result of the current position detection device 11 (step 16). The ECU 10 temporarily stores the position information acquired here in the RAM.

Subsequently, the ECU 10 reads out the vehicle identification information such as the vehicle number, vehicle type / format, and color data stored in advance in the storage device 15 and temporarily stores it in the RAM (step 17).
The ECU 10 stores the vehicle information, environmental information, position information, own vehicle identification information, etc. stored in the RAM as the near hat information in the near hat information storage memory 155 (step 18).

Further, the ECU 10 detects 30 seconds (15 seconds before detection + 15 after detection) before and after the detection of the near-miss reaction based on the time information when the change in the biological state (abnormal reaction of the driver's biological information) is detected. Second) image data is read out from the image storage memory 158 and stored in the near-miss information storage memory 155 (step 19).
Then, the ECU 10 transmits the near-miss information stored (stored) in the near-miss information storage memory 155 to the data center 2 using the communication device 20 (step 20), and returns the process.
The information stored (stored) in the near-miss information storage memory 155 is deleted after the transmission to the data center 2 is completed.

  It should be noted that all the information other than the position information and the time information at the time of detecting the near-miss reaction does not necessarily have to be transmitted to the data center 2 as the near-miss information. For example, only necessary information may be arbitrarily selected, or a plurality of necessary information may be combined and transmitted to the data center 2.

In the above-described embodiment, each time information is stored in the near-miss information storage memory 155, that is, every time one near-hat information is generated, the information is transmitted to the data center 2. However, the transmission timing of the near-miss information is not limited to this.
For example, information in the near-miss information storage memory 155 may be read periodically and transmitted to the data center 2 every predetermined time (for example, 10 minutes).
In this case, until the next transmission process is executed, the ECU 10 repeatedly executes a series of information processing related to the near-miss area from step 11 to step 19 described above, and the acquired information is The information is stored in the hat information storage memory 155.

Next, a description will be given of processing in the data center 2 that creates a reproduced moving image of a near-miss event based on the received near-hat information and updates the database 28 storing these reproduced moving images.
FIG. 8 is a flowchart showing an operation procedure for creating the database 28 in the data center 2.
First, the CPU 21 in the data center 2 receives the near-miss information transmitted from the probe car 1 (step 31) and stores it in the near-hat information storage unit 272 in the storage unit 15. When the near-miss information is received, it is sequentially stored in the near-miss information storage unit 272.

Then, the CPU 21 stores the stored near-hat information based on the position information and time information in which the change in the biological state (abnormal reaction of the driver's biological information) is detected, which is included in each near-miss information. Then, grouping (grouping) into related event units (step 32).
That is, a group of near-miss information related to an abnormal reaction of biological information presumed to be caused by the same event is created. The plurality of near-miss information collected in the same group is used as related event information.
For example, with reference to the time and position specified by the first detected near miss information, within a predetermined determination time (for example, 3 minutes) from the reference time and within a predetermined determination distance (for example, a radius of 1 km) from the reference position Combine near-miss information into the same group. Note that the determination time and the determination distance can be changed to arbitrary values.

The CPU 21 counts the number of near-miss information for each related event information, that is, the number of cases, as information for determining a near-miss level described later (step 33).
Next, the CPU 21 determines whether or not the necessary near-miss information for each related event information has been prepared (step 34). Here, for example, it is determined whether or not a predetermined determination time (for example, 3 minutes) has elapsed since the time when the first near-miss information was received. Further, it may be determined whether or not a predetermined time (for example, 1 minute) has elapsed since the last near-miss information was received.
When it is determined that the necessary near-miss information for each related event information is not available (step 34; N), the processing returns to step 31.

On the other hand, when it is determined that the necessary near-miss information for each related event information has been prepared (step 34; Y), the CPU 21 determines the near-hat level for the related event information (step 35).
FIG. 9 is a diagram showing an example of a near / hat level determination condition.
As shown in FIG. 9, the number of near-miss information (number of case counts) for each related event information counted in the processing of step 33, and “sudden brake” and “sudden handle” are included in the near-miss information. Based on whether or not there is a near-miss level determination condition. However, the set values of these determination conditions are appropriately changed depending on conditions such as the driving environment.

Based on such a determination condition, the CPU 21 determines the near-miss level of each related event information, that is, the risk of a near-miss event.
For example, when the case count number is “1”, the sudden brake is “present”, and the sudden handle is “present”, it is determined that the near-miss level is the “medium” level.
In this embodiment, when the case count number is “3 or more”, it is determined that the near-miss level is the “high” level without depending on the detection results of “sudden brake” and “sudden steering wheel”. To do.
Note that the near / hat level determination condition is not limited to this, and is appropriately changed according to changes in the driving environment.

Returning to the description of FIG. 8, the CPU 21 performs a process of identifying the cause vehicle that has started the near-miss event based on various analysis results of the related event information (near-miss information) (step 36).
FIG. 10 is an explanatory diagram of the occurrence of a near-miss event (interrupt).
Here, for example, as shown in FIG. 10, a case where the cause vehicle interrupts the front direction of the vehicle A and the vehicle B across two lanes will be described.
The exclamation mark “!” Shown in the figure indicates a vehicle that has detected a near-miss reaction.

In this case, a near-miss reaction is detected in the rear vehicle A and vehicle B due to the interruption of the cause vehicle, and the near-hat information is transmitted from the vehicles A and B to the data center 2. Then, the CPU 21 of the data center 2 groups (groups) the near-miss information received from the vehicle A and the vehicle B as information related to the same event.
Then, the CPU 21 specifies image data (30 seconds before and after the detection of the near-miss reaction) in the near-miss information transmitted from the vehicle A and the vehicle B in order to identify the cause vehicle of the near-miss event (interrupt). The image analysis process is executed.
Based on the result of the image analysis process, identification information of the cause vehicle, for example, vehicle number, vehicle type / format, and color information is specified.
The CPU 21 stores the result of the image analysis process in the cause vehicle information storage unit 273.

Returning to the description of FIG. 8, based on the identified identification information of the cause vehicle, the CPU 21 addresses the information processing apparatus mounted on the cause vehicle from the data table stored in the data table storage unit 271 of the storage unit 27. Read information.
The CPU 21 transmits predetermined information to the cause vehicle using the read address information of the cause vehicle (step 37).
That is, the CPU 21 requests the cause vehicle to provide, for example, position information / vehicle information of the own vehicle (cause vehicle) at the time of occurrence of the near-miss event, image information before and after the occurrence of the near-miss event, and the like. Send a message to the vehicle that caused the near-miss event.

When the ECU 10 of the cause vehicle (probe car 1) receives the information transmitted from the data center 2 (CPU 21), based on the received information, the position information / vehicle information of the vehicle at the time of occurrence of a near-miss event Is read from the vehicle information storage memory 156, and image information before and after the occurrence of the near-miss event is read from the image storage memory 158 and stored in the RAM.
Further, the ECU 10 of the cause vehicle is determined to be the cause vehicle of the near-miss event in the history information storage memory 160 provided in the storage device 15 based on the received message indicating that the cause of the near-hat event has occurred. Remember. For example, the history information storage memory stores the occurrence time, occurrence location, event content, etc. of a near-miss event in which the host vehicle is determined to be the cause vehicle.
Then, the cause vehicle ECU 10 transmits the information read from the storage device 15 as the cause vehicle information, that is, the information requested to be provided to the data center 2 to the data center 2 using the communication device 20. .

In the above-described example, the near-miss reaction of the cause vehicle is not detected in the data center 2 because no near-miss reaction is detected in the cause vehicle. Therefore, in order to acquire the near / miss information of the cause vehicle, the data center 2 is configured to perform the processing of step 37 and step 38.
However, in the vehicle identified as the cause vehicle, when a near-miss reaction is detected when the near-miss event occurs, the near-hat information is received from the cause vehicle. The processing of step 38 is not performed, and the near-miss information of the cause vehicle is processed as the cause vehicle information.

When receiving the cause vehicle information from the cause vehicle (step 38), the CPU 21 of the data center 2 stores the cause vehicle information together with the related event information of the near-miss event in a predetermined storage area.
Then, the CPU 21 creates a reproduced moving image of the near-miss event based on the acquired related event information and cause vehicle information (step 39).
Here, an example of a method for creating a reproduction video of a near-miss event will be described.
For example, when creating a reproduction video of a near-miss event (interrupt) as shown in FIG. 10 described above, the CPU 21 performs vehicle A and vehicle B based on image information and vehicle information (operation information) of related event information. The travel locus a, b is calculated.
Then, the CPU 21 performs image processing based on the calculated travel trajectories a and b, the cause vehicle information, and the image information of the vehicles A and B to create a reproduction moving image.

The method of creating a near-miss event reproduction video is not limited to the above-described method. For example, in the case of a near-miss event (interrupt) as shown in FIG. 10, the cause vehicle, vehicle A and vehicle B, That is, a time-series travel locus for all the vehicles related to the event may be calculated, and a reproduction moving image may be created based on the calculated travel locus (exercise status) of each vehicle and vehicle identification information.
A CG (computer graphics) image of the cause vehicle may be created based on the cause vehicle information acquired from the cause vehicle, and a reproduction movie may be created by combining the CG images of the cause vehicle.
Even if you create a reproduction video of a near-miss event by combining the image information of all the vehicles related to the event, not only the viewpoint is fixed, but also the near-hat from multiple viewpoints You may make it produce the reproduction video of an event.

In addition, when a reproduced moving image is created by combining images transmitted from a plurality of vehicles, it is preferable to preferentially combine the images in which the cause vehicle is most reflected.
In addition, when synthesizing images transmitted from a plurality of vehicles in time series, for example, it is preferable to extract a section in which the cause vehicle is projected from each image and combine them so as to be joined in time series. .

The CPU 21 adds the reproduced moving image of the near-miss event created in this way, updates the database 28 (step 40), and returns to the main routine.
In the embodiment described above, a series of processing from step 35 to step 40 has been described for one near-miss event in order to avoid complicating the explanation. These processes are performed for each near-miss event (grouped). These processes may be performed in parallel for a plurality of near-miss events, or may be performed for each near-miss event.

In the present embodiment, access information to the data center 2, more specifically, the URL of the near-event event reproduction video that the cause vehicle wants to see (the storage location of the near-event event reproduction movie caused by the cause vehicle) Is transmitted to the cause vehicle to urge the driver of the cause vehicle to access the URL. That is, the driver of the cause vehicle is prompted to view a reproduction video of a near-miss event resulting from his / her driving operation.
The driver of the cause vehicle can reproduce (view) the reproduced moving image by accessing the received URL via a terminal device such as the information processing device, the PC 3, or the mobile phone 4.
FIG. 11 is a diagram illustrating an example of browsing a reproduced moving image of a near-miss event.
As shown in FIG. 11, the PC 3 and the mobile phone 4 can browse the reproduced moving image of the near-miss event by accessing the database 28 of the data center 2 and starting a predetermined moving image reproducing program.
When a reproduced video is selected (designated), for example, as shown in FIG. 11, a map image or a near-hat that displays the position where the near-miss event has occurred is displayed as report information, including the reproduced moving image. Related information such as event occurrence time, near-miss level (risk level), and the like is provided from the data center 2.

Note that the method for browsing the reproduced moving image is not limited to the method for directly accessing the specific reproduced moving image as described above.
For example, the database 28 is accessed from the terminal device, and the reproduction video is narrowed down and selected by designating the near-hat event occurrence area (position information), time zone, and related vehicle identification information. The reproduced video may be browsed.
As described above, according to the present embodiment, by accessing the database 28 of the data center 2, it is possible to confirm information on the near-miss event that has occurred.
In addition, according to the present embodiment, by creating a reproduction video of a near-miss event that has occurred based on information provided from each probe car 1, it is possible to easily grasp the situation at the time of the near-miss event occurring. can do.
In this way, the driver of the cause vehicle, that is, the driver who has performed a dangerous driving operation is encouraged to view a reproduction video of the near-miss event, thereby causing a dangerous driving operation situation to the driver himself. Can be objectively recognized easily and driving manners can be expected to improve.

Moreover, you may make it transmit the information of URL for accessing the data center 2 to the vehicle relevant to a near-miss event. In this case, the URL information may be transmitted to all vehicles related to the near-miss event, or may be transmitted only to some related vehicles. When transmitting to some related vehicles, for example, the vehicle that transmitted the near-miss information due to the near-miss event first (or the last), and the vehicle closest to the cause vehicle Identify and send.
As described above, according to the present embodiment, not only the driver of the cause vehicle but also the driver of the vehicle in which the dangerous reaction (near-miss) is detected, the reproduction video of the near-miss event is viewed. By prompting, it is possible to objectively recognize the situation of the event that caused the dangerous reaction.

Next, a modified example of the method for identifying the cause vehicle in the process of step 36 of the database creation process described above will be described.
First, an example of a near-miss event pattern used in the description of the modification will be described.
12 and 13 are diagrams showing examples of near-miss event patterns.
As a pattern of the near-miss event, in addition to the interrupt shown in FIG. 10 described above, “rapid braking” (pattern A) shown in FIG. 12A and “interrupt-1” shown in FIG. 12B (pattern) B), “Interrupt-2” (pattern C) shown in FIG. 12C, “Fluctuation” (pattern D) shown in FIG. 13A, “Jump out of intersection” (pattern E) shown in FIG. , There are patterns such as “protruding to the opposite lane” (pattern F) shown in FIG.

“Sudden braking” shown in FIG. 12A is a pattern of a near-miss event due to sudden braking due to carelessness of the vehicle ahead.
“Interrupt-1” shown in FIG. 12 (b) is a cause vehicle at a point where a cause vehicle is forcedly interrupted from one lane to the other lane on a two-lane road or merged from two lanes to one lane. This is a pattern of near-miss events caused by unreasonable interruptions.
“Interrupt-2” shown in FIG. 12C is a pattern of a near-miss event caused by a lane change across the two lanes of the cause vehicle on a three-lane road.

“Fluctuation” shown in FIG. 13A is a pattern of a near-miss event caused by the cause vehicle trying to change lanes but stopping halfway.
The “intersection jumping out” shown in FIG. 13B is a pattern of a near-miss event caused by the cause vehicle jumping out ahead of the vehicle traveling on the priority road at the intersection.
“Extrusion to oncoming lane” shown in FIG. 13C is a pattern of a near-miss event that is caused by an onset of the cause vehicle to the oncoming lane due to a curve or the like.

[Example 1]
For example, when a near-miss event (fluctuation, interrupt) shown in patterns B to F occurs, the CPU 21 of the data center 2 performs image processing on the image information of the related event information, so that it is reflected in each image information. Identify the vehicle number, type and color of the vehicle. The CPU 21 searches for a vehicle that is common to all the image information of the related event information, collates with the vehicle number of the own vehicle specifying information of each vehicle, deletes the vehicle that is not the cause vehicle, and specifies the cause vehicle.
In this way, the cause vehicle may be specified by the elimination method.

[Example 2]
For example, when a near-miss event shown in the patterns B to D occurs, the information processing device of each vehicle that has detected a near-hat reaction is performed by image processing instead of the data center 2 to actually identify the cause vehicle. It may be.
FIG. 14 is a flowchart showing an operation procedure for identifying the cause vehicle of a near-miss event in the vehicle.
The ECU 10 of the probe car 1 recognizes a forward vehicle by processing an image in front of the vehicle captured by the image input device 16 (stereo camera 161) by the image processing device 132 (step 51).
Then, the ECU 10 performs an analysis process on the captured image in front of the vehicle in the image processing device 132, and based on the analysis result, whether or not the vehicle in front of the vehicle fluctuates or an interruption of another vehicle has occurred in front of the vehicle. Judgment is made (step 52).
When the wobbling of the vehicle ahead or the occurrence of an interruption of another vehicle is detected (step 52; Y), the ECU 10 performs the biological information acquisition device 12 in the same manner as step 15 in the acquisition processing of the near-miss information described above. Based on information such as the presence / absence of a change in the biometric information detected by the above, it is determined whether a near-miss reaction has been detected in response to a wobbling of the vehicle ahead or an interruption of another vehicle (step 53).
When it is determined that the fluctuation of the vehicle ahead or the interruption of the other vehicle corresponds to a near-miss event (step 53; Y), the ECU 10 identifies (recognizes) that the front vehicle is the cause vehicle of the near-miss event. (Step 54).

Then, the ECU 10 acquires the information of the vehicle number, the vehicle type / format, and the color data as the specific information of the specified cause vehicle by performing the analysis processing of the captured image in front of the vehicle by the image processing device 132 (step 55).
The ECU 10 transmits the acquired specific information of the cause vehicle together with the near / miss information to the data center 2 using the communication device 20 (step 56), and returns to the main routine.
On the other hand, when the occurrence of a wobbling of the vehicle ahead or the occurrence of an interruption of another vehicle is not detected (step 52; N), a near-miss reaction is detected depending on the wobbling of the vehicle ahead or the interruption of another vehicle. If not (step 53; N), the ECU 10 returns to the process of step 51.

The CPU 21 of the data center 2 specifies the cause vehicle of the near-miss event by acquiring the cause vehicle information transmitted from the probe car 1.
In this way, by causing the probe car 1 to share the processing for identifying the cause vehicle of the near-miss event, not only can the load on the data center 2 be reduced, but also the accuracy of identifying the cause vehicle can be improved.

[Example 3]
When the data center 2 is in a communicable state with all the vehicles in the area (cover area) under its jurisdiction, in detail, the traveling position information of all the vehicles in the area is continuously transmitted to the data center 2. A description will be given of a method for identifying the cause vehicle of a near-miss event when the traveling locus of each vehicle can be identified.
For example, when a near-miss event shown in the patterns B to D occurs, the CPU 21 of the data center 2 calculates the travel locus of each vehicle based on the travel position information of the vehicle transmitted from each vehicle.
Then, the CPU 21 may identify the cause vehicle based on the relative positional relationship between the calculated travel locus of each vehicle and the vehicle that has transmitted the near-miss information. CPU21 specifies the vehicle from which the direction of a locus vector differs from other vehicles, for example as a cause vehicle.
The trajectory vector is a vector indicating the travel direction and speed of the vehicle at a specific time point calculated based on the travel trajectory of each vehicle.

[Example 4]
Similarly to [Example 3] described above, when the data center 2 is in a communicable state with all the vehicles in the area (cover area) under its jurisdiction, the data center 2 informs the data center 2 of the travel position information of all the vehicles in the area. A description will be given of a method of identifying the cause vehicle of a near-miss event in the case where the traveling locus of each vehicle can be identified by continuously transmitting and the like.
For example, when a near-miss event shown in the patterns B to D occurs, the CPU 21 of the data center 2 exists in the area of the same near-miss event (in the related area of the near-miss event) and is near the near-miss event. A vehicle that has not transmitted information (a vehicle in which no dangerous reaction has been detected) is searched.
Here, CPU21 lists the vehicle which exists in the area of the same near-miss event, for example. Then, the vehicle list and the transmission source list of the near-miss information are compared, and the vehicle that has transmitted the near-miss information is deleted from the vehicle list to search for a vehicle in which no dangerous reaction is detected. .

The near-miss event area is, for example, from the reference position within a predetermined determination time (for example, 3 minutes) from the reference time based on the time and position specified by the first detected near-miss information. An area having a predetermined determination distance (for example, a radius of 50 m) is shown.
Then, the CPU 21 may identify a vehicle that has transmitted near-miss information among a plurality of searched vehicles, that is, a vehicle that is closest to a vehicle in which a dangerous reaction has been detected, as a cause vehicle.
However, the identification method of the cause vehicle shown in the above [Example 1] to [Example 4] assumes a case where no dangerous reaction is detected in the cause vehicle.

[Example 5]
Causes of a near-miss event when vehicles related to a near-miss event are in a state where vehicle-to-vehicle communication is possible, that is, when the probe cars 1 can directly communicate with each other without going through the data center 2. A vehicle identification method will be described.
In this case, as in [Example 1], not the data center 2 but the information processing apparatus of the vehicle (probe car 1) related to the near-miss event is actually identified. .

For example, when a near-miss event occurs due to a sudden braking due to carelessness of the vehicle ahead as shown in pattern A, the ECU 10 of the information processing device in each probe car 1 uses the image input device 16 (stereo camera 161). Based on the captured image information ahead of the vehicle, it is recognized that the brake lamp of the vehicle ahead is lit.
Thereafter, when a driver's dangerous reaction (near-miss reaction) is detected, the ECU 10 recognizes the lighting of the brake lamp of the preceding vehicle until the driver's dangerous reaction is detected. Then, it is determined whether the own vehicle is the cause vehicle or the preceding vehicle is the cause vehicle.
Here, for example, if one or more seconds elapse from the lighting of the brake lamp of the preceding vehicle until the driver's dangerous reaction is detected, the vehicle is determined to be the cause vehicle, that is, the driver's own failure. Judged as a dangerous reaction due to attention. On the other hand, when the elapsed time is less than 1 second, it is determined that the preceding vehicle is the cause vehicle, that is, it is determined that the reaction is a dangerous reaction caused by the sudden braking of the preceding vehicle.

When it is determined that the preceding vehicle is the cause vehicle, the ECU 10 transmits a determination result indicating that the preceding vehicle is the cause vehicle to the preceding vehicle.
On the other hand, when determining that the host vehicle is the cause vehicle, the ECU 10 transmits a determination result indicating that the host vehicle is the cause vehicle to the data center 2 together with the near-miss information.
When it is determined that the own vehicle is the cause vehicle, when the “determination result that the preceding vehicle (= own vehicle) is the cause vehicle” is received from the following vehicle, the own vehicle The judgment result indicating that the vehicle is the cause is transmitted.
Then, the CPU 21 of the data center 2 may identify the cause vehicle of the near-miss event by acquiring the cause vehicle information transmitted from the probe car 1.

Next, a specific procedure for identifying the cause vehicle using inter-vehicle communication will be described.
FIG. 15 is a flowchart showing an operation procedure for identifying a cause vehicle of a near-miss event using inter-vehicle communication.
Here, an explanation will be given by taking as an example the occurrence of a near-miss event caused by sudden braking due to carelessness in front of the cause vehicle as shown in FIG. 12A. Specifically, vehicle A (cause vehicle), vehicle A method for identifying a cause vehicle using inter-vehicle communication in B and vehicle C will be described.

The ECU 10 of the vehicle A recognizes that the brake lamp of the front vehicle is lit based on the image information ahead of the vehicle imaged by the image input device 16 (stereo camera 161) (step 71).
Then, after one second or more has passed since the lighting of the brake lamp of the preceding vehicle is recognized, the ECU 10 of the vehicle A operates based on information such as the presence or absence of a change in biological information detected by the biological information acquisition device 12. A near-miss reaction (dangerous reaction) is detected (step 72).
The ECU 10 of the vehicle A detects the driver's sudden braking operation based on the value detected by the brake pedal force sensor 142 (step 73).

The ECU 10 of the vehicle B recognizes the brake lamp that the vehicle A is turned on based on the image information in front of the vehicle (step 81).
Then, within one second after the lighting of the brake lamp of the preceding vehicle is recognized, the ECU 10 of the vehicle B is based on information such as the presence or absence of a change in the biological information detected by the biological information acquisition device 12. A near-miss reaction (dangerous reaction) is detected (step 82).
The ECU 10 of the vehicle B detects the driver's sudden braking operation based on the value detected by the brake pedal force sensor 142 (step 83).
The ECU 10 of the vehicle B transmits the near-miss information detected in the own vehicle (vehicle B) to the preceding vehicle (vehicle A) (step 84).

The ECU 10 of the vehicle C recognizes the brake lamp that is turned on by the sudden brake operation in the vehicle B based on the image information in front of the vehicle (step 91).
Then, within one second after the lighting of the brake lamp of the preceding vehicle is recognized, the ECU 10 of the vehicle C, based on information such as the presence / absence of a change in the biological information detected by the biological information acquisition device 12, A near-miss reaction (dangerous reaction) is detected (step 92).
The ECU 10 of the vehicle C detects the driver's sudden braking operation based on the value detected by the brake pedal force sensor 142 (step 93).
The ECU 10 of the vehicle C transmits the near-miss information detected in the own vehicle (vehicle C) to the preceding vehicle (vehicle B) (step 94).
The process of transmitting the detected near-miss information to the preceding vehicle in the above-described step 84 and step 94 is performed in such a manner that the driver's near-miss reaction occurs within one second after the lighting of the brake lamp of the preceding vehicle is recognized. It is executed only when it is detected.

The ECU 10 of the vehicle A receives the near-miss information transmitted from the vehicle B (step 74).
Then, the ECU 10 of the vehicle A determines whether or not the own vehicle is the cause vehicle (step 75). Here, for example, the ECU 10 determines that the host vehicle is the cause vehicle when 1 second or more has elapsed after the driver's dangerous reaction is detected after the brake lamp of the vehicle ahead is recognized.
When it is determined that the host vehicle is the cause vehicle (step 75; Y), the ECU 10 of the vehicle A uses the data center 2 as information (cause vehicle information) indicating that the host vehicle (vehicle A) is the cause vehicle. And it transmits to the vehicle B (step 76), and returns to the main routine.
On the other hand, when it is determined that the host vehicle is not the cause vehicle (step 75; N), the process directly returns to the main routine.

When the cause vehicle information is transmitted from the vehicle A, the ECU 10 of the vehicle B receives the cause vehicle information transmitted from the vehicle A (step 85).
Further, the ECU 10 of the vehicle B receives the near-miss information detected by the vehicle C transmitted from the vehicle C (step 86).
The ECU 10 of the vehicle B transmits information (cause vehicle information) indicating that the vehicle A is the cause vehicle to the vehicle C (step 87), and returns to the main routine.
The ECU 10 of the vehicle C receives information (cause vehicle information) transmitted from the vehicle B and indicating that the vehicle A is the cause vehicle (step 95), and returns to the main routine.
The CPU 21 of the data center 2 can identify the cause vehicle of the near-miss event by acquiring the cause vehicle information transmitted from the vehicle A, that is, the cause vehicle.

[Example 6]
Next, when communication with a traffic monitoring camera installed in an area (cover area) managed by the data center 2 is possible, that is, to acquire a camera image in which the data center 2 is monitoring the area. A method for identifying the cause vehicle of a near-miss event that can be used in the case where it is possible will be described.
For example, when a near-miss event shown in patterns A to F occurs, the CPU 21 of the data center 2 identifies the occurrence location of the near-miss event based on the related event information, and monitors the specific area (corresponding area). Get the camera image.

The CPU 21 analyzes the acquired camera video, and each vehicle obtained from the analysis result of the camera video and related event information, specifically, own vehicle specifying information and image information transmitted from the vehicle related to the near-miss event. Check the information and identify the cause vehicle.
In this way, the cause vehicle may be specified using the captured image information of the traffic monitoring camera. By using the captured image information of the traffic monitoring camera, the cause vehicle can be identified based on more evidence information, so that the cause vehicle identification accuracy can be further improved.

[Example 7]
For example, when a near-miss event shown in patterns B to F occurs, in the information processing apparatus for a vehicle (probe car 1) related to the near-miss event, the own vehicle or another vehicle located in the vicinity is the cause vehicle. It is determined that there is a possibility, and the determination result is transmitted to the data center 2. Then, the data center 2 may identify the cause vehicle based on the information on the determination result transmitted from the related vehicle.

FIG. 16 is a flowchart illustrating an operation procedure of a process for identifying a vehicle having a possibility of being a cause vehicle of a near-miss event in the vehicle.
The ECU 10 of the probe car 1 reads an image in front of the vehicle imaged by the image input device 16 (stereo camera 161) into the image processing device 132 and analyzes it, and recognizes the traveling lane (step 101).
Then, the ECU 10 performs an analysis process on the captured image in front of the vehicle in the image processing device 132, and determines whether or not the own vehicle has deviated from the traveling lane based on the analysis result (step 102).
If the host vehicle has not deviated from the travel lane (step 102; N), the ECU 10 returns to step 101 and repeats the process.

When the host vehicle deviates from the driving lane (step 102; Y), the ECU 10 determines whether there is a change in the biological information detected by the biological information acquisition device 12 as in step 15 in the acquisition processing of the near-miss information described above. Based on such information, it is determined whether the lane departure of the host vehicle corresponds to a near-miss event (step 103).
When it is determined that the lane departure of the host vehicle does not correspond to a near-miss event (step 103; N), the ECU 10 uses the determination result indicating that the host vehicle may be the cause vehicle as identification information of the host vehicle. At the same time, the data is transmitted to the data center 2 (step 104), and the process returns to the main routine.
On the other hand, when it is determined that the lane departure of the host vehicle corresponds to a near-miss event (step 103; Y), the ECU 10 determines the surroundings of the host vehicle based on the measurement result of the inter-vehicle distance / relative speed measuring device 131, for example. It is determined whether or not there is another vehicle (around) (step 105).

If there is no other vehicle around (around) the host vehicle (step 105; N), the process returns to the main routine.
When there is another vehicle around (around 105) the host vehicle (step 105; Y), the ECU 10 gives a determination result that the other vehicle around the host vehicle (around) may be the cause vehicle. Then, it is transmitted to the data center 2 together with the specific information of the own vehicle (step 106), and the process returns to the main routine.
Then, the CPU 21 of the data center 2 identifies the cause vehicle of the near-miss event based on the information on the vehicle having the possibility of being the cause vehicle transmitted from the probe car 1.
By providing the data center 2 with information indicating that there is a possibility that the vehicle is the cause vehicle, the identification accuracy of the cause vehicle can be further improved.

It should be noted that the method for identifying the cause vehicle of the near-miss event described in the above-described embodiment and [Example 1] to [Example 7] may be used independently according to the facility environment of the system. A plurality of methods may be used in combination.
For example, the method of actually identifying the cause vehicle on the vehicle (probe car 1) side described in [Example 2] and the method of identifying the cause vehicle on the data center 2 side described in [Example 6] are used in combination. You may do it.

It is a figure showing the structure of the database creation system in this embodiment. It is the figure which showed schematic structure of the data center. It is the figure which showed the example of the detection result of a brake pedaling force sensor. It is the figure which showed the example of the information stored in a near-miss information storage memory. It is the flowchart which showed the operation | movement procedure of the acquisition process of near-miss information in a probe car. It is the figure which showed the example of determination of the change of biometric information. It is the figure which showed an example of the judgment item of whether it is a near-miss event, and its conditions. It is the flowchart which showed the operation | movement procedure of the production process of a database in a data center. It is the figure which showed an example of the judgment conditions of a near-miss level. It is explanatory drawing of the generation | occurrence | production state of a near-miss event (interruption). It is the figure which showed the browsing example of the reproduction animation of a near-miss event. It is the figure which showed the example pattern of a near-miss event. It is the figure which showed the example pattern of a near-miss event. It is the flowchart which showed the operation | movement procedure of the specific process of the cause vehicle of the near-miss event in a vehicle. It is the flowchart which showed the operation | movement procedure of the specific process of the cause vehicle of the near-miss event using inter-vehicle communication. It is the flowchart which showed the operation | movement procedure of the specific process of the vehicle which may be a cause vehicle of the near-miss event in a vehicle.

Explanation of symbols

1 Probe car 2 Data center 3 PC
4 Cellular Phone 5 Vehicle 6 Network 11 Current Position Detection Device 12 Biological Information Acquisition Device 13 Environmental Information Acquisition Device 14 Vehicle Information Acquisition Device 15 Storage Device 16 Image Input Device 17 Display Device 18 Audio Output Device 19 Input Device 20 Communication Device 21 CPU
22 ROM
23 RAM
24 input device 25 output device 26 communication device 27 storage unit 28 database 121 heart rate sensor 122 skin impedance sensor 131 inter-vehicle distance / relative speed measuring device 132 image processing device 141 steering wheel angle sensor 142 brake pedal force sensor 143 accelerator pedal force sensor 144 vehicle speed sensor 145 Gyro sensor 151 Biological information processing program 152 Vehicle information processing program 153 Environmental information processing program 154 Near-miss determination program 155 Near-miss information storage memory 156 Vehicle information storage memory 157 Environment information storage memory 158 Image storage memory 159 Map database 161 Stereo camera

Claims (6)

Detection information acquisition means for acquiring detection information including a detection time and a vehicle position from the vehicle that has detected a change in the biological information of the driver due to an event based on a specific cause vehicle ,
Based on the acquired detection information, related vehicle specifying means for specifying related vehicles related to each event;
Image acquisition means for acquiring a surrounding image of each identified related vehicle;
Reproduction image creation means for creating a reproduction image of each event based on the acquired image;
A cause vehicle identifying means for identifying the cause vehicle;
Access information transmitting means for transmitting access information for viewing the created reproduced image to the identified cause vehicle;
A database creation device characterized by comprising: 2. The database creation device according to claim 1, wherein the cause vehicle specifying means specifies the cause vehicle from the acquired surrounding image of each related vehicle . The image acquisition means acquires, as a surrounding image of the related vehicle, a moving image around the vehicle before and after the detection of a change in the biological information of the driver from the vehicle from which the detection information has been acquired.
The database creation device according to claim 1, wherein the reproduction image creation unit creates a reproduction image of each event based on the acquired moving image. The access information transmitting means further transmits access information for viewing the created reproduction image to at least one of the identified related vehicles. Or the database creation apparatus of Claim 3.   5. The database creation device according to claim 1, wherein the reproduction image creation unit creates a reproduction image of a cause vehicle behavior as a reproduction image of the event. . In a database creation device comprising detection information acquisition means, related vehicle identification means, image acquisition means, reproduction image creation means, cause vehicle identification means, and access information transmission means ,
The detection information acquisition means includes detection information including a detection time at which a change in the biological information is detected and a vehicle position from a vehicle in which the change in the biological information of the driver due to an event based on the specific cause vehicle is detected. Detection information acquisition function to acquire,
The related vehicle specifying means specifies a related vehicle related to each event based on the acquired detection information;
An image acquisition function in which the image acquisition means acquires a surrounding image of each identified related vehicle;
The reproduction image creation means creates a reproduction image of each event based on the acquired image, a reproduction image creation function,
The cause vehicle specifying means specifies a cause vehicle specifying function for specifying the cause vehicle;
The access information transmitting means transmits an access information for browsing the created reproduced image to the identified cause vehicle, and an access information transmitting function.
A computer-readable database creation program for realizing the above.

Images