Drive supporting system, drive supporting method, and program

Abstract

To provide a drive supporting system that can appropriately determine a risk degree for a drive of a vehicle by a driver.SOLUTION: A drive supporting system 100 comprises: a driving scene detection unit 101 that detects a driving scene of a vehicle V according to outputs from one or more sensors mounted on the vehicle V; a cognitive behavior detection unit 102 that detects a cognitive behavior of a driver who drives the vehicle V according to the outputs from the one or more sensors mounted on the vehicle V; and a risk determination unit 105 that determines a risk degree for the driver's driving of the vehicle V based on at least the detected driving scene and cognitive behavior.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to a system for assisting vehicle driving. [Background technology]

[0002] A management support system has been proposed that supports management of the driving condition of a driver who drives a vehicle (see, for example, Patent Document 1). The management support system of Patent Document 1 includes an in-vehicle system installed in each vehicle owned by a transportation company or the like, a server, and an administrator terminal. The in-vehicle system installed in the vehicle has an imaging device, a first monitoring device, and a second monitoring device. The imaging device captures images of the area in front of the vehicle. The first monitoring device and the second monitoring device monitor the driving condition of the vehicle and the physical condition of the driver of the vehicle while driving. When the monitoring results satisfy a predetermined condition, the in-vehicle system transmits video information obtained by imaging with the imaging device to a server. The predetermined condition is, for example, a condition that the driver's driving condition is at a caution level. The video information transmitted to the server is information obtained by imaging when the driver's driving condition is at a caution level. Such video information is transmitted from the server to the administrator terminal.

[0003] Therefore, by checking the video information, the user of the administrator terminal, for example, the transportation company's operations manager, can understand how the vehicle was being driven by a driver who is in a driving condition requiring caution. In other words, the operations manager can identify near-miss incidents (medical incidents) and encourage the driver to improve their driving awareness. Note that the in-vehicle system included in the management support system may also be called a driving support system, as it is used to support safe driving. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 6714036 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the in-vehicle system included in the management support system of Patent Document 1, i.e., the driving support system, has a problem in that it may not be able to appropriately determine the degree of risk involved in the driver's driving of the vehicle. In other words, the driving support system of Patent Document 1 monitors the driver's physical condition as a driving condition, but does not monitor the driver's cognitive behavior. Therefore, even if the cognitive behavior required for driving is not performed, if the driver's physical condition is good, the driver's driving condition will not be determined to be at a level requiring caution, and there is a possibility that near-miss events will be overlooked.

[0006] Therefore, the present disclosure provides a driving assistance system and the like that can appropriately determine the degree of risk involved in driving a vehicle by a driver. [Means for solving the problem]

[0007] A driving assistance system according to one embodiment of the present disclosure includes a driving scene detection unit that detects a driving scene of a vehicle in accordance with output from one or more sensors mounted on the vehicle, a cognitive behavior detection unit that detects cognitive behavior of a driver driving the vehicle in accordance with output from the one or more sensors mounted on the vehicle, and a risk assessment unit that determines the degree of risk associated with the driver's driving of the vehicle based on at least the detected driving scene and the detected cognitive behavior.

[0008] These comprehensive or specific aspects may be realized as an apparatus, a method, an integrated circuit, a computer program, or a computer-readable recording medium such as a CD-ROM, or may be realized as any combination of an apparatus, a method, an integrated circuit, a computer program, and a recording medium. The recording medium may also be a non-transitory recording medium. [Effects of the Invention]

[0009] The driving assistance system of the present disclosure can appropriately determine the degree of risk involved in the driver's driving of the vehicle.

[0010] Further advantages and effects of one aspect of the present disclosure will become apparent from the specification and drawings. Such advantages and / or effects are provided by some of the embodiments and configurations described in the specification and drawings, but not all of the configurations are necessarily required. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of a management system including a driving assistance system according to the first embodiment. [Figure 2] FIG. 2 is a block diagram showing an example of a functional configuration of the driving assistance system according to the first embodiment. [Figure 3] FIG. 3 is a diagram showing an example of a driving scene detected by the driving scene detection unit according to the first embodiment. [Figure 4] FIG. 4 is a diagram showing another example of a driving scene detected by the driving scene detection unit according to the first embodiment. [Figure 5] FIG. 5 is a diagram showing still another example of a driving scene detected by the driving scene detection unit according to the first embodiment. [Figure 6] FIG. 6 is a diagram showing still another example of a driving scene detected by the driving scene detection unit according to the first embodiment. [Figure 7] FIG. 7 is a diagram illustrating the detection of cognitive behavior by the cognitive behavior detection unit according to the first embodiment. [Figure 8] FIG. 8 is a flowchart showing an example of the overall processing operation of the driving assistance system according to the first embodiment. [Figure 9] FIG. 9 is a flowchart showing a detailed example of the scoring criterion selection process according to the first embodiment. [Figure 10]FIG. 10 is a flowchart showing a detailed example of the cognitive behavior scoring process according to the first embodiment. [Figure 11] FIG. 11 is a flowchart showing a detailed example of the vehicle behavior scoring process according to the first embodiment. [Figure 12] FIG. 12 is a block diagram showing an example of a functional configuration of the driving assistance system according to the second embodiment. [Figure 13] FIG. 13 is a diagram for explaining detection of a dangerous operation by the dangerous operation detection unit in the second embodiment. [Figure 14] FIG. 14 is a diagram for explaining visual detection of a surrounding object by a surrounding visual detection unit according to the second embodiment. [Figure 15] FIG. 15 is a flowchart showing an example of the overall processing operation of the driving assistance system according to the second embodiment. [Figure 16] FIG. 16 is a flowchart showing a detailed example of the cognitive behavior scoring process and the surrounding visual scoring process in Example 1 of the second embodiment. [Figure 17] FIG. 17 is a flowchart showing a detailed example of the vehicle behavior scoring process in Example 1 of the second embodiment. [Figure 18] FIG. 18 is a flowchart showing a detailed example of the cognitive behavior scoring process and the surrounding visual scoring process in Example 2 of the second embodiment. [Figure 19] FIG. 19 is a flowchart showing a detailed example of the vehicle behavior scoring process in Example 2 of the second embodiment. [Figure 20] FIG. 20 is a flowchart showing a detailed example of the risky operation scoring process in Example 2 of the second embodiment. [Figure 21] FIG. 21 is a flowchart showing a detailed example of the cognitive behavior scoring process in Example 3 of the second embodiment. [Figure 22] FIG. 22 is a flowchart showing a detailed example of the surrounding visual scoring process in Example 3 of Embodiment 2. [Figure 23] FIG. 23 is a flowchart showing a detailed example of the vehicle behavior scoring process in Example 3 of the second embodiment. [Figure 24] FIG. 24 is a flowchart showing a detailed example of the cognitive behavior scoring process in Example 4 of the second embodiment. [Figure 25] FIG. 25 is a flowchart showing a detailed example of the surrounding visual scoring process in Example 4 of Embodiment 2. [Figure 26] FIG. 26 is a flowchart showing a detailed example of the vehicle behavior scoring process in Example 4 of the second embodiment. [Figure 27] FIG. 27 is a flowchart showing a detailed example of the cognitive behavior scoring process in Example 5 of the second embodiment. [Figure 28] FIG. 28 is a flowchart showing a detailed example of the vehicle behavior scoring process in Example 5 of the second embodiment. [Figure 29] FIG. 29 is a flowchart showing a detailed example of the cognitive behavior scoring process in Example 6 of the second embodiment. [Figure 30] FIG. 30 is a flowchart showing a detailed example of the vehicle behavior scoring process in Example 6 of the second embodiment. [Figure 31] FIG. 31 is a block diagram showing an example of a functional configuration of a driving assistance system according to the third embodiment. [Figure 32] FIG. 32 is a diagram illustrating an example of driving history data stored in the driving history storage unit according to the third embodiment. [Figure 33] FIG. 33 is a diagram showing a more specific example of driving history data stored in the driving history storage unit in the third embodiment. [Figure 34] FIG. 34 is a diagram illustrating an example of the processing operation of the time condition update unit according to the third embodiment. [Figure 35] FIG. 35 is a flowchart showing an example of the overall processing operation of the driving assistance system according to the third embodiment. [Figure 36] FIG. 36 is a flowchart showing an example of processing operations related to driving history performed by the risk determining unit and the individual threshold value determining unit according to the third embodiment. [Figure 37]FIG. 37 is a diagram showing an example of accident history included in driving history data in a modification of the third embodiment. [Figure 38] FIG. 38 is a diagram illustrating the conditional individual risk threshold for a location in the modification of the third embodiment. [Figure 39] FIG. 39 is a diagram showing another example of accident history included in driving history data in the modification of the third embodiment. [Figure 40] FIG. 40 is a flowchart showing an example of processing operations related to driving history by the risk determining unit and the individual threshold value determining unit in the modification of the third embodiment. [Figure 41] FIG. 41 is a flowchart showing another example of the processing operation related to the driving history by the risk determining unit and the personal threshold value determining unit in the modification of the third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] A driving assistance system according to a first aspect of the present disclosure includes a driving scene detection unit that detects a driving scene of a vehicle in accordance with output from one or more sensors mounted on the vehicle, a cognitive behavior detection unit that detects cognitive behavior of a driver driving the vehicle in accordance with output from one or more sensors mounted on the vehicle, and a risk assessment unit that assesses the degree of risk associated with the driver's driving of the vehicle based on at least the detected driving scene and cognitive behavior.

[0013] As a result, the driving scene and cognitive behavior are detected, and the degree of risk for the driver's driving of the vehicle is determined based on the detection results, so that the degree of risk can be appropriately determined. In other words, even if the driver is in good physical condition, if the cognitive behavior required for the driving scene, such as visually checking to the left when turning left at an intersection, is not performed, it can be determined that there is a high risk for driving in that driving scene. In other words, it can be appropriately determined whether unsafe driving has been performed.

[0014] In the driving assistance system according to a second aspect, the risk determination unit may determine the degree of risk by scoring the driving of the vehicle by the driver, and a driving score obtained by the scoring may be smaller as the degree of risk increases. Note that the second aspect may be dependent on the first aspect.

[0015] This allows the degree of risk to be determined by the driving score, making it possible to evaluate the degree of risk in an easy-to-understand manner.

[0016] In addition, in a driving assistance system according to a third aspect, the risk determination unit may further determine whether the driving score is less than a risk threshold, and the driving assistance system may further include an upload unit that uploads driving video data obtained by capturing images of the driving scene to a server when it is determined that the driving score is less than the risk threshold. Note that the third aspect may be dependent on the second aspect.

[0017] As a result, driving video data with low driving scores, i.e., driving video data when unsafe driving occurs, is uploaded to the server, so that the driver of the vehicle or the manager who manages the vehicle can access the server and check the driving video data. As a result, the driver or manager can recognize unsafe driving and improve their awareness of safe driving. In other words, after the driver has completed work that involves driving a vehicle, the driver or manager can review the work by checking the driving video data, i.e., the video when unsafe driving occurred.

[0018] The driving assistance system according to a fourth aspect may further include a video editing unit that generates the driving video data by selecting and editing portions corresponding to the driving scenes detected by the driving scene detection unit from one or more video data sets obtained by capturing images using a camera mounted on the vehicle. Note that the fourth aspect may be dependent on the third aspect.

[0019] As a result, even if video data for a predetermined recording time is output from the camera, for example, only a portion corresponding to the driving scene is extracted from one or more of the output video data and uploaded as driving video data. In a specific example, the recording time is one minute. Therefore, video data showing a scene different from the detected driving scene, or even a portion of the video data, can be prevented from being uploaded to the server. As a result, the transmission load of unnecessary data can be reduced. Furthermore, if a driving scene is shown across multiple video data, the multiple video data can be edited into a single driving video data. As a result, when the driving video data is played back, the driving scene video is displayed without interruption, allowing the video of the driving scene to be properly confirmed.

[0020] Furthermore, the driving assistance system according to a fifth aspect may further include a surroundings detection unit that detects moving objects around the vehicle in accordance with outputs from one or more sensors mounted on the vehicle, and the risk determination unit may determine the degree of risk based on the detection result of the moving objects by the surroundings detection unit. Note that the fifth aspect may be subordinate to any one of the first to fourth aspects. Note that the moving objects may be, for example, people, bicycles, motorcycles, other vehicles, etc. Also, the surroundings of the vehicle may be within a radius R [m] from the vehicle, where R is an arbitrary numerical value.

[0021] This allows the degree of risk to be determined as high if there are moving objects around the vehicle, and low if there are no moving objects around the vehicle. In other words, if there are people or other objects around the vehicle, the danger increases, so the degree of risk can be determined as high. As a result, the degree of risk can be determined more appropriately.

[0022] In addition, a driving assistance system according to a sixth aspect may further include a surroundings visual detection unit that detects the driver's visual inspection of one or more objects around the vehicle in accordance with outputs from one or more sensors mounted on the vehicle, and the risk determination unit may determine the degree of risk based on the visual inspection detection result by the surroundings visual detection unit. Note that the sixth aspect may be dependent on any one of the first to fifth aspects. Note that the one or more objects may be oncoming traffic or moving objects such as an oncoming vehicle.

[0023] This allows the degree of risk to be determined depending on whether or not the driver has visually checked the oncoming lane or vehicle, for example, in a driving situation where a vehicle is turning right at an intersection. In other words, if the driver has not visually checked, the degree of risk can be determined as high, and conversely, if the driver has visually checked, the degree of risk can be determined as low. As a result, the degree of risk can be determined more appropriately.

[0024] In addition, the driving assistance system according to a seventh aspect may further include a vehicle behavior detection unit that detects a behavior of the vehicle, and the risk determination unit may determine the degree of the risk based on a detection result of the behavior by the vehicle behavior detection unit. Note that the seventh aspect may be dependent on any one of the first to sixth aspects.

[0025] This allows the degree of risk to be determined depending on whether the vehicle is slowing down, for example, in a driving scene where the vehicle is turning left at an intersection. In other words, if the vehicle is not slowing down, the degree of risk can be determined as high, and conversely, if the vehicle is slowing down, the degree of risk can be determined as low. As a result, the degree of risk can be determined more appropriately.

[0026] In addition, the driving assistance system according to an eighth aspect may further include a dangerous operation detection unit that detects a predetermined device operation by the driver as a dangerous operation in accordance with outputs from one or more sensors mounted on the vehicle, and the risk determination unit may determine the degree of the risk based on a detection result of the dangerous operation by the dangerous operation detection unit. Note that the eighth aspect may be dependent on any one of the first to seventh aspects.

[0027] This makes it possible to determine the degree of risk in a driving situation, such as when a vehicle is turning right at an intersection, depending on whether the driver has performed a dangerous operation on a smartphone or the like. In other words, if no dangerous operation has been performed, the degree of risk can be determined as low, and conversely, if a dangerous operation has been performed, the degree of risk can be determined as high. As a result, the degree of risk can be determined more appropriately.

[0028] In addition, in a driving assistance system according to a ninth aspect, when the risk determination unit determines that the driving score is less than a risk threshold, the risk determination unit may further select an individual risk threshold associated with the driver from a plurality of individual risk thresholds and determine whether the driving score is less than the selected individual risk threshold, and the upload unit may upload the driving video data to the server when the driving score is less than the risk threshold and also less than the individual risk threshold. Note that the ninth aspect may be dependent on any one of the fourth to eighth aspects, which are dependent on the third aspect.

[0029] As a result, even if the driving score is below the risk threshold, unless it is below the personal risk threshold associated with the driver, uploading of driving video data will not be performed. Therefore, even if a driver engages in unsafe driving, whether or not to upload driving video data can be switched depending on the driver. For example, a low personal risk threshold is associated with an experienced driver who normally drives safely. As a result, even if the driver happens to engage in unsafe driving, uploading of driving video data can be suppressed. Therefore, uploading can be optimized according to the driver. By associating a low personal risk threshold with an experienced driver who has been driving safely for many years (i.e., a safe experienced driver), uploading of driving video data can be made more difficult. On the other hand, by associating a high personal risk threshold with an experienced driver who has been driving dangerously for many years (i.e., a dangerous experienced driver), uploading of driving video data can be made more easy. As a result, uploading of driving video data can be optimized and made more efficient.

[0030] The driving assistance system according to a tenth aspect may further include an individual threshold value determining unit that determines the individual risk threshold values corresponding to the plurality of people based on the driving histories of the plurality of people. Note that the tenth aspect may be dependent on the ninth aspect.

[0031] This makes it possible to determine an individual risk threshold for each of a plurality of people according to that person's type as a driver (i.e., driver type).

[0032] In addition, in a driving assistance system according to an eleventh aspect, the personal threshold determination unit may, in determining the plurality of personal risk thresholds, determine a personal risk threshold corresponding to each of a plurality of combinations, each of the plurality of combinations including one of the plurality of people including the driver and one of a plurality of driving scenes, and the risk assessment unit may, in selecting the personal risk threshold, select from the plurality of personal risk thresholds a personal risk threshold associated with the combination of the driving scene and the driver detected by the driving scene detection unit. Note that the eleventh aspect may be dependent on the tenth aspect.

[0033] This allows for the selection and use of a personal risk threshold corresponding to the driver who engaged in unsafe driving and the driving scene in which the unsafe driving occurred, thereby further optimizing and streamlining the uploading of driving video data. For example, if a driver who normally drives safely tends to engage in unsafe driving in driving scenes involving backing into a parking space, a high personal risk threshold can be selected for that driver only when backing into a parking space. As a result, it becomes easier to upload driving video data only when backing into a parking space.

[0034] In a driving assistance system according to a twelfth aspect, the personal threshold determination unit may, in determining the plurality of personal risk thresholds, determine a personal risk threshold corresponding to each of a plurality of combinations, each of the plurality of combinations including the driver, and one of a plurality of locations, and the risk assessment unit may, in selecting the personal risk threshold, select from the plurality of personal risk thresholds a personal risk threshold associated with a combination of the driver and a location where the vehicle is traveling in the driving scene detected by the driving scene detection unit. Note that the twelfth aspect may be dependent on the tenth aspect or an eleventh aspect which is dependent on the tenth aspect.

[0035] This allows for the selection and use of a personal risk threshold corresponding to the driver who engaged in unsafe driving and the location of the driving scene in which the unsafe driving occurred, thereby further optimizing and streamlining the uploading of driving video data. For example, if a driver who normally drives safely tends to engage in unsafe driving on a highway, a high personal risk threshold can be selected for that driver only when the vehicle is traveling on the highway. As a result, it becomes easier to upload driving video data only when the vehicle is traveling on the highway.

[0036] In addition, a driving assistance system according to a thirteenth aspect may further include a time condition determination unit that determines a predetermined time corresponding to each of a plurality of people based on driving histories of the plurality of people, and the risk determination unit, in determining the degree of risk, may select a predetermined time associated with the driver from the plurality of predetermined times determined by the time condition determination unit, and determine the degree of risk by comparing a time during which the cognitive behavior detected by the cognitive behavior detection unit was performed with the predetermined time. Note that the thirteenth aspect may be dependent on any one of the first to twelfth aspects.

[0037] As a result, a specified time according to the driver is selected, and the degree of risk is determined by comparing the time during which the driver performs cognitive behavior with the specified time, so that the risk can be determined more appropriately. For example, there is a tendency that the time required for cognitive behavior (e.g., the time spent looking to the right) required for safe driving differs between experienced drivers and novice drivers. Therefore, by using a specified time according to the driver, the degree of risk associated with driving by that driver can be determined appropriately.

[0038] Hereinafter, the embodiments will be specifically described with reference to the drawings.

[0039] The embodiments described below are all comprehensive or specific examples. The numerical values, shapes, materials, components, component placement and connection configurations, steps, and step sequences shown in the following embodiments are merely examples and are not intended to limit the present invention. Furthermore, among the components in the following embodiments, components that are not described in the independent claims that represent the highest concepts are described as optional components. Furthermore, each drawing is a schematic diagram and is not necessarily an exact illustration. Furthermore, the same components are designated by the same reference numerals in each drawing.

[0040] (Embodiment 1) FIG. 1 is a diagram showing an example of the configuration of a management system including a driving assistance system according to this embodiment.

[0041] The management system 1000 is a system for managing a plurality of vehicles V, and includes a plurality of driving assistance systems 100, a management server 200, and a management terminal 300, which are connected to each other via a communication network Nt.

[0042] Each of the multiple driving assistance systems 100 is mounted on a vehicle V. The vehicle V may be a passenger car, or may be a commercial vehicle such as a taxi or truck.

[0043] The management server 200 acquires and stores driving video data uploaded from each of the multiple driving assistance systems 100. The driving video data shows unsafe driving of the vehicle V that uploaded the driving video data. Note that unsafe driving may also be called dangerous driving.

[0044] The management terminal 300 is operated by an administrator who manages multiple vehicles V, accesses the management server 200, and downloads driving video data from the management server 200. The administrator, who is the user of the management terminal 300, then views the driving video data played back by the management terminal 300 to understand what kind of unsafe driving is occurring and takes measures to prevent the occurrence of such unsafe driving. Note that each driver of multiple vehicles V may operate the management terminal 300 to play back the driving video data and check the unsafe driving depicted in the driving video data.

[0045] FIG. 2 is a block diagram showing an example of the functional configuration of the driving assistance system 100 according to this embodiment.

[0046] The driving assistance system 100 assists the driving of the vehicle V in accordance with outputs from a plurality of sensors mounted on the vehicle V equipped with the driving assistance system 100. The driving assistance is, for example, uploading the driving video data described above. The plurality of sensors include, for example, a GPS (Global Positioning System) unit 11, a driver monitor 12, an acceleration sensor 13, a drive recorder 14, a CAN (Controller Area Network) 15, and the like.

[0047] The GPS unit 11 receives signals from GPS satellites, identifies the position of the vehicle V on which the GPS unit 11 is mounted, and outputs a position signal indicating the position of the vehicle V. Any sensor may be used instead of the GPS unit 11 as long as it is a sensor that identifies the position of the vehicle V using the GNSS (Global Navigation Satellite System) and outputs a position signal indicating the position. The driver monitor 12 is equipped with a camera that captures images of the driver of the vehicle V and outputs video image data obtained by capturing the image with the camera. The acceleration sensor 13 measures the acceleration of the vehicle V and outputs an acceleration signal indicating the acceleration. The drive recorder 14 is equipped with a camera that captures images of the surroundings of the vehicle V and outputs video image data obtained by capturing the image with the camera. The CAN 15 is a communication network including, for example, multiple ECUs (Electronic Control Units) mounted on the vehicle V, and outputs driving signals that indicate the driving conditions of the vehicle V, such as the vehicle speed and steering angle.

[0048] The driving assistance system 100 in this embodiment includes a driving scene detection unit 101, a cognitive behavior detection unit 102, a vehicle behavior detection unit 103, a surroundings detection unit 104, a risk determination unit 105, a video storage unit 106, and an upload unit 107.

[0049] The driving scene detection unit 101 detects a driving scene of the vehicle V based on outputs from one or more sensors mounted on the vehicle V. The one or more sensors are, for example, a GPS unit 11 and a drive recorder 14. Note that the output from the CAN 15 may be used to detect the driving scene. The driving scene may be, for example, a left or right turn at an intersection, backing into a parking lot, changing lanes, etc.

[0050] The cognitive behavior detection unit 102 detects the cognitive behavior of the driver who drives the vehicle V according to the output from one or more sensors mounted on the vehicle V. The one or more sensors are, for example, the driver monitor 12. The cognitive behavior is the driver's visual observation, and more specifically, the position or direction in the vehicle V to which the driver's gaze is directed.

[0051] The vehicle behavior detection unit 103 detects the behavior of the vehicle V according to outputs from one or more sensors mounted on the vehicle V. The one or more sensors are, for example, an acceleration sensor 13, a drive recorder 14, and a CAN 15. The behavior of the vehicle V includes the vehicle speed, steering angle, traveling position, and changes therein of the vehicle V. More specifically, the behavior of the vehicle V may include slowing down, sudden deceleration, sudden acceleration, an action of issuing a direction indication, moving left or right, and the like.

[0052] The surroundings detection unit 104 detects objects around the vehicle V as surrounding objects according to outputs from one or more sensors mounted on the vehicle V. The detected surrounding objects may be moving objects such as people, bicycles, other vehicles, etc. The surroundings detection unit 104 may also determine the type of the detected surrounding object, and may determine whether the surrounding object is a stationary object or a moving object. The type of surrounding object may be a vehicle, an oncoming vehicle, a person, a bicycle, a truck, a parking space, etc.

[0053] The risk determination unit 105 determines the degree of risk associated with the driver's driving of the vehicle V based on at least the detected driving scene and cognitive behavior. In this embodiment, the risk determination unit 105 determines the degree of risk not only based on the driving scene and cognitive behavior, but also based on the detection results of the behavior of the vehicle V by the vehicle behavior detection unit 103. Specifically, the risk determination unit 105 determines the degree of risk by scoring the driver's driving of the vehicle V. The driving score, which is the score obtained by the scoring, is smaller as the degree of risk increases. Furthermore, the risk determination unit 105 determines whether the driving score is less than a risk threshold. If the risk determination unit 105 determines that the driving score is less than the risk threshold, it determines that the driving of the vehicle V is unsafe.

[0054] The moving image storage unit 106 is a recording medium for storing moving image data obtained by capturing images using cameras provided in the driver monitor 12 and the drive recorder 14. For example, the moving image storage unit 106 is a hard disk drive, a RAM (Random Access Memory), a ROM (Read Only Memory), or a semiconductor memory. Note that such moving image storage unit 106 may be volatile or non-volatile.

[0055] The upload unit 107 uploads, as driving video data, one or more video data sets obtained by capturing images of the detected driving scenes, among the one or more video data sets stored in the video storage unit 106, to the management server 200. The driving scenes are driving scenes in which the above-mentioned unsafe driving was performed. In other words, when the risk determination unit 105 determines that the driving score is below the risk threshold, the upload unit 107 uploads the driving video data obtained by capturing images of the driving scenes to the management server 200.

[0056] FIG. 3 is a diagram showing an example of a driving scene detected by the driving scene detection unit 101. As shown in FIG.

[0057] 3, the driving scene detection unit 101 detects a left turn at an intersection as a driving scene. Specifically, the driving scene detection unit 101 identifies the position and orientation of the vehicle V on a map based on at least one of the position signal output from the GPS unit 11 and the video data output from the drive recorder 14. Here, when the driving scene detection unit 101 identifies that the vehicle V has entered an intersection and the traveling direction of the vehicle V has changed, for example, by approximately 90 degrees to the left, it detects a left turn at the intersection as a driving scene of the vehicle V.

[0058] FIG. 4 is a diagram showing another example of a driving scene detected by the driving scene detection unit 101. In FIG.

[0059] 4, the driving scene detection unit 101 detects a right turn at an intersection as a driving scene. Specifically, the driving scene detection unit 101 determines that the vehicle V has entered the intersection and the traveling direction of the vehicle V has changed, for example, by approximately 90 degrees to the right, based on at least one of the position signal output from the GPS unit 11 and the video data output from the drive recorder 14. In this case, the driving scene detection unit 101 detects a right turn at the intersection as a driving scene of the vehicle V.

[0060] FIG. 5 is a diagram showing yet another example of a driving scene detected by the driving scene detection unit 101. In FIG.

[0061] As shown in FIG. 5, the driving scene detection unit 101 detects backing into a parking lot as a driving scene. Backing into a parking lot is an action in which the vehicle V backs into a parking space and stops. Specifically, the driving scene detection unit 101 determines that the vehicle V has entered the parking lot based on at least one of the position signal output from the GPS unit 11 and the video image data output from the drive recorder 14. Furthermore, the driving scene detection unit 101 determines whether the vehicle V has changed its direction of travel by, for example, 90 degrees or more, whether the vehicle has changed direction multiple times, whether the vehicle has stopped after switching from forward to reverse, etc. In this case, the driving scene detection unit 101 detects backing into a parking lot as a driving scene for the vehicle V.

[0062] FIG. 6 is a diagram showing yet another example of a driving scene detected by the driving scene detection unit 101. In FIG.

[0063] As shown in FIG. 6 , the driving scene detection unit 101 detects a lane change or overtaking as a driving scene. Specifically, the driving scene detection unit 101 determines that the vehicle V is traveling on a road with two or more lanes, based on at least one of the position signal output from the GPS unit 11 and the video data output from the drive recorder 14. Furthermore, the driving scene detection unit 101 determines that the vehicle V has crossed a white line. In this case, the driving scene detection unit 101 detects a lane change as a driving scene of the vehicle V. Thereafter, the driving scene detection unit 101 further determines that the vehicle V has crossed the above-mentioned white line again and overtaken another vehicle, based on at least one of the position signal output from the GPS unit 11 and the video data output from the drive recorder 14. In this case, the driving scene detection unit 101 detects overtaking as a driving scene of the vehicle V.

[0064] When detecting a right or left turn using only video data, the driving scene detection unit 101 may detect white lines and detect the right or left turn based on the pattern of the white line detection. In the case of a left turn, a white line is detected in the direction of travel, but no white lines are detected within the intersection. When the white line of the lane begins to be detected after a left turn, it is detected diagonally relative to the direction of travel, but gradually becomes aligned with the direction of travel. Similarly, in the case of a right turn, the angle at which the white line is detected diagonally is reversed in the left-right direction compared to the angle for a left turn. To improve accuracy, if sound data is output along with the video data, the sound data, specifically, sound data of the turn signal, may be utilized. When detecting backing up of the vehicle V, the driving scene detection unit 101 may utilize sound data indicating the sound generated when backing up. Many vehicles repeatedly emit a specific sound pattern while the turn signal is on and while backing up. To detect this specific pattern with high accuracy, it is recommended that the video data and sound data when the turn signal is on or when the vehicle is in reverse gear be explicitly registered in advance in the driving assistance system 100.

[0065] FIG. 7 is a diagram for explaining the detection of cognitive behavior by the cognitive behavior detection unit 102. As shown in FIG.

[0066] The cognitive behavior detection unit 102 identifies a gaze point ahead of the driver's line of sight based on the video data output from the driver monitor 12. Then, the cognitive behavior detection unit 102 identifies a part of the vehicle V that overlaps with the gaze point or a part of the vehicle V that is close to the gaze point. As a result, the cognitive behavior detection unit 102 detects that the driver is looking at the identified part as the driver's cognitive behavior.

[0067] For example, as shown in FIG. 7, if the gaze point is in area a4, the cognitive behavior detection unit 102 detects that the driver is looking at the left mirror (also called a left door mirror) as the cognitive behavior of the driver. Note that area a4 is an area that includes the left mirror of the vehicle V. Similarly, if the gaze point is in area a3, the cognitive behavior detection unit 102 detects that the driver is looking at the right mirror (also called a right door mirror) as the cognitive behavior of the driver. Note that area a3 is an area that includes the right mirror of the vehicle V. Furthermore, if the gaze point is in area a1, the cognitive behavior detection unit 102 detects that the driver is looking at the rearview mirror as the cognitive behavior of the driver. Note that area a1 is an area that includes the rearview mirror of the vehicle V. Furthermore, if the gaze point is in area a5, the cognitive behavior detection unit 102 detects that the driver is looking at a car navigation system (also called a car navigation device) as the cognitive behavior of the driver. Note that area a5 is an area that includes the car navigation system of the vehicle V. Furthermore, if the gaze point is in area a2, the cognitive behavior detection unit 102 detects that the driver is looking at the front or windshield of the vehicle V as the driver's cognitive behavior. Area a2 is an area that includes part or the entire windshield of the vehicle V. Similarly, if the gaze point is in an area that includes the left or right glass window of the vehicle V, the cognitive behavior detection unit 102 detects that the driver is looking to the left or to the right of the vehicle V as the driver's cognitive behavior.

[0068] FIG. 8 is a flowchart showing an example of the overall processing operation of the driving assistance system 100 according to this embodiment.

[0069] First, the driving scene detection unit 101 detects a driving scene of the vehicle V (step S1). Next, the risk determination unit 105 performs a scoring criterion selection process (step S100) based on the driving scene detected by the driving scene detection unit 101 and the detection result of surrounding objects by the surroundings detection unit 104. This scoring criterion selection process is a process for selecting a scoring criterion for scoring the driving score described above.

[0070] Next, the risk determination unit 105 performs a cognitive behavior scoring process (step S200) based on the driving scene detected by the driving scene detection unit 101 and the detection result of the cognitive behavior by the cognitive behavior detection unit 102. This cognitive behavior scoring process is a process for scoring driving behavior points for the cognitive behavior of the detected scene based on the scoring criteria selected in step S100.

[0071] Furthermore, the risk determination unit 105 performs a vehicle behavior scoring process based on the driving scene detected by the driving scene detection unit 101 and the detection result of the behavior of the vehicle V by the vehicle behavior detection unit 103 (step S400). This vehicle behavior scoring process is a process for scoring a driving action score for the vehicle behavior of the detected scene based on the scoring criteria selected in step S100. Note that the driving assistance system 100 may perform the process of step S4 depending on the processes of steps S200 and S400. Furthermore, the processes of steps S200 and S400 may be performed in any order.

[0072] Then, the risk determination unit 105 determines the degree of risk associated with the driver's driving of the vehicle V based on the processing results of steps S200 and S400. That is, the risk determination unit 105 determines the degree of risk by scoring the driver's driving of the vehicle V (step S2). Specifically, the risk determination unit 105 scores the driving score by performing weighted addition on the driving behavior scores scored in steps S200 and S400, respectively.

[0073] Next, the risk determination unit 105 determines whether the driving score assigned in step S2 is less than the risk threshold (step S3). If the risk determination unit 105 determines that the driving score is less than the risk threshold (Yes in step S3), it instructs the upload unit 107 to upload the driving video data. As a result, the upload unit 107 uploads the driving video data to the management server 200 (step S4). That is, the upload unit 107 extracts the driving video data obtained by capturing the driving scene detected in step S1 from the video storage unit 106 and uploads it to the management server 200. Note that the driving video data includes one or more video data sets output from the camera of the driver monitor 12 and one or more video data sets output from the camera of the drive recorder 14.

[0074] On the other hand, if the risk assessment unit 105 determines that the driving score is not less than the risk threshold (No in step S3), that is, if it determines that the driving score is greater than or equal to the risk threshold, it terminates processing for the driving scene detected in step S1 without issuing an upload instruction.

[0075] 9 is a flowchart showing a detailed example of the scoring criterion selection process, which illustrates the detailed process of step S100 in FIG.

[0076] The risk determination unit 105 determines whether or not there are any moving objects (hereinafter also referred to as surrounding moving objects) around the vehicle V based on the detection result of the surrounding objects by the surrounding detection unit 104 (step S101). As described above, surrounding moving objects are, for example, people, bicycles, other vehicles, etc. Such surrounding moving objects may be moving objects that are predetermined for each driving scene detected in step S100. For example, if the driving scene is turning right at an intersection, the surrounding moving object may be an oncoming vehicle, and if the driving scene is turning right, the surrounding moving object may be a person, bicycle, etc.

[0077] Then, when the risk determination unit 105 determines in step S101 that there are no surrounding moving objects (No in step S101), it adopts the first scoring criterion (step S103). On the other hand, when the risk determination unit 105 determines in step S101 that there are surrounding moving objects (Yes in step S101), it adopts the second scoring criterion (S102). Note that the first scoring criterion and the second scoring criterion are each criteria for scoring the driving score, that is, criteria for determining the degree of risk associated with driving. Note that the second scoring criterion may adopt a lower score for the same driving than the first scoring criterion. In other words, the second scoring criterion is stricter than the first scoring criterion.

[0078] As described above, in this embodiment, the risk determination unit 105 determines the degree of risk based on the detection result of the moving object by the surroundings detection unit 104. As a result, if there is a moving object around the vehicle V, it is possible to determine the degree of risk as high, i.e., to assign a low driving score. Conversely, if there is no moving object around the vehicle V, it is possible to determine the degree of risk as low, i.e., to assign a high driving score. In other words, if there is a person or the like around the vehicle V, the danger increases, and therefore it is possible to determine the degree of risk as high. As a result, it is possible to more appropriately determine the degree of risk.

[0079] Fig. 10 is a flowchart showing a detailed example of the cognitive behavior scoring process. Fig. 10 shows detailed processing of step S200 in Fig. 8 when a left turn at an intersection is detected as a driving scene in step S1 in Fig. 8 and the first scoring criterion is adopted in step S100 in Fig. 8 (specifically, step S103 in Fig. 9). That is, Fig. 10 shows a detailed example of the cognitive behavior scoring process when a left turn is made when there are no surrounding moving objects.

[0080] <Check left mirror> First, the risk determination unit 105 determines whether the driver has visually checked the left mirror of the vehicle V for a specified period of time at a predetermined timing based on the result of the cognitive behavior detection by the cognitive behavior detection unit 102 (step S201). Note that the predetermined timing may be the timing from the start to the end of a roughly 90-degree change in the traveling direction of the vehicle V, or the timing from n seconds (n is any integer equal to or greater than 0) before the vehicle V enters the intersection until it passes through the intersection. Also, the specified period of time may be a period within a predetermined range of times ta1 to ta2 (ta2 > ta1). Here, when the risk determination unit 105 determines that the driver has visually checked the left mirror of the vehicle V for a specified period of time at a predetermined timing (Yes in step S201), it determines the driving behavior score for the cognitive behavior of the left mirror to be 100 points (step S204).

[0081] On the other hand, when the risk determination unit 105 determines that the driver has not visually checked the left mirror of the vehicle V for a specified time at a predetermined timing (No in step S201), it determines whether the driver has visually checked the left mirror for a time different from the specified time at the predetermined timing (step S202). Here, when the risk determination unit 105 determines that the driver has visually checked the left mirror for a time different from the specified time at the predetermined timing (Yes in step S202), it determines the driving action score for the cognitive action of the left mirror to be 90 points (step S205). For example, if the driver glances at the left mirror at a predetermined timing, 90 points is determined.

[0082] On the other hand, when the risk determination unit 105 determines that the driver has not visually checked the left mirror for a period of time different from the specified time at the predetermined timing (No in step S202), it determines whether or not the left mirror was visually checked at a timing shifted from the predetermined timing (step S203). Here, when the risk determination unit 105 determines that the left mirror was visually checked at a timing shifted from the predetermined timing (Yes in step S203), it determines the driving action score for the cognitive behavior of the left mirror to be 70 points (step S206). On the other hand, when the risk determination unit 105 determines that the left mirror was not visually checked at a timing shifted from the predetermined timing (No in step S203), it instructs the upload unit 107 to upload driving video data (step S207). For example, when the driver is not looking at the left mirror, the upload is instructed. As a result, the process of step S4 in FIG. 8 is executed.

[0083] <Check left> Furthermore, the risk determination unit 105 determines whether the driver has visually checked the left side of the vehicle V for a specified period of time at a specified timing based on the detection result of the cognitive behavior by the cognitive behavior detection unit 102 (step S211). The specified timing and the specified period of time may be the same as or different from the specified timing and the specified period of time in step S201. Here, when the risk determination unit 105 determines that the driver has visually checked the left side of the vehicle V for a specified period of time at a specified timing (Yes in step S211), it determines the driving behavior score for the cognitive behavior to the left as 100 points (step S214).

[0084] On the other hand, when the risk determination unit 105 determines that the driver has not looked to the left of the vehicle V for a specified time at a predetermined timing (No in step S211), it determines whether the driver has looked to the left for a time different from the specified time at the predetermined timing (step S212). Here, when the risk determination unit 105 determines that the driver has looked to the left for a time different from the specified time at the predetermined timing (Yes in step S212), it determines the driving action score for the cognitive action to the left to be 90 points (step S215). For example, if the driver glances to the left at a predetermined timing, 90 points is determined.

[0085] On the other hand, when the risk determination unit 105 determines that the driver has not looked to the left at the predetermined timing for a time period different from the specified time (No in step S212), it determines whether or not a left look was made at a timing shifted from the predetermined timing (step S213). Here, when the risk determination unit 105 determines that a left look was made at a timing shifted from the predetermined timing (Yes in step S213), it determines the driving action score for the cognitive action to the left to be 70 points (step S216). On the other hand, when the risk determination unit 105 determines that a left look was not made at a timing shifted from the predetermined timing (No in step S213), it instructs the upload unit 107 to upload driving video data (step S217). For example, uploading is instructed when the driver is not looking to the left. As a result, the process of step S4 in FIG. 8 is executed.

[0086] Fig. 11 is a flowchart showing a detailed example of the vehicle behavior scoring process. Fig. 11 shows detailed processing of step S400 in Fig. 8 when a left turn at an intersection is detected as the driving scene in step S1 in Fig. 8 and the first scoring criterion is adopted in step S100 in Fig. 8 (specifically, step S103 in Fig. 9). In other words, Fig. 11 shows a detailed example of the vehicle behavior scoring process when a left turn is made when there are no surrounding moving objects.

[0087] <Slow down> First, the risk determination unit 105 determines whether the vehicle V has slowed down based on the detection result of the behavior of the vehicle V by the vehicle behavior detection unit 103 (step S401). For example, the risk determination unit 105 determines that the vehicle V has slowed down when the vehicle V has traveled at a speed of 10 km / h or 20 km / h or less. Here, when the risk determination unit 105 determines that the vehicle V has slowed down (Yes in step S401), it determines the driving action score for the vehicle speed to be 100 points (step S404).

[0088] On the other hand, if the risk determination unit 105 determines that the vehicle V is not moving slowly (No in step S401), it determines whether the vehicle V is traveling faster than 20 km / h but not more than 30 km / h, or whether the vehicle V has suddenly decelerated (step S402). Here, if the risk determination unit 105 determines that the vehicle V is traveling faster than 20 km / h but not more than 30 km / h, or if it determines that the vehicle V has suddenly decelerated (Yes in step S402), it sets the driving action score for the vehicle speed to 90 points (step S405). Note that if the negative acceleration of the vehicle V is equal to or less than a predetermined lower limit, it is determined that the vehicle V has suddenly decelerated.

[0089] On the other hand, if the risk determination unit 105 determines that the vehicle V is not traveling faster than 20 km / h and not at or below 30 km / h, and that the vehicle V is not suddenly decelerating (No in step S402), it determines whether the vehicle V is traveling faster than 30 km / h and not at or below 40 km / h (step S403). Here, if the risk determination unit 105 determines that the vehicle V is traveling faster than 30 km / h and not at or below 40 km / h (Yes in step S403), it determines the driving action score for the vehicle speed to be 70 points (step S406). On the other hand, if the risk determination unit 105 determines that the vehicle V is not traveling faster than 30 km / h and not at or below 40 km / h (No in step S403), it instructs the upload unit 107 to upload driving video data (step S407). For example, if the vehicle V is traveling at a speed faster than 40 km / h, uploading is instructed. As a result, the process of step S4 in FIG. 8 is executed.

[0090] <Direction instructions> Furthermore, the risk determination unit 105 determines whether or not the vehicle V has issued a turn signal (i.e., a blinker) within 30 m of the intersection based on the detection result of the behavior of the vehicle V by the vehicle behavior detection unit 103 (step S411). Here, when the risk determination unit 105 determines that the vehicle V has issued a turn signal (Yes in step S411), it determines the driving action score for the turn signal to be 100 points (step S414).

[0091] On the other hand, if the risk determination unit 105 determines that the vehicle V has not issued a direction indication within 30 meters of the intersection (No in step S411), it determines whether the vehicle V has issued a direction indication within 20 meters of the intersection (step S412).Here, if the risk determination unit 105 determines that the vehicle V has issued a direction indication (Yes in step S412), it determines the driving action score for the direction indication to be 90 points (step S415).

[0092] On the other hand, if the risk determination unit 105 determines that the vehicle V has not issued a turn signal within 20 meters of the intersection (No in step S412), it determines whether the vehicle V has issued a turn signal within 10 meters of the intersection (step S413). Here, if the risk determination unit 105 determines that the vehicle V has issued a turn signal (Yes in step S413), it determines the driving action score for the turn signal to be 70 points (step S416). On the other hand, if the risk determination unit 105 determines that the vehicle V has not issued a turn signal within 10 meters of the intersection (No in step S413), it instructs the upload unit 107 to upload driving video data (step S417). As a result, the processing of step S4 in FIG. 8 is executed.

[0093] <Left justified> Furthermore, the risk determination unit 105 determines whether the vehicle V has drifted to the left before turning left, based on the detection result of the behavior of the vehicle V by the vehicle behavior detection unit 103 (step S421). If the risk determination unit 105 determines that the vehicle V has drifted to the left (Yes in step S421), it determines the driving action score for pulling left to be 100 points (step S422). On the other hand, if the risk determination unit 105 determines that the vehicle V has not drifted to the left (No in step S421), it determines the driving action score for pulling left to be 90 points (step S423).

[0094] As a result, in step S2 of Fig. 8, the risk determination unit 105 calculates a driving score for the driver's driving of the vehicle V by weighting and adding the driving action scores for the left mirror cognitive action, the leftward cognitive action, the vehicle speed, the direction indication, and the leftward pull-up. For example, each of the above driving action scores has a maximum score of 100, and the risk determination unit 105 weights and adds these driving action scores so that the maximum score is 100. Alternatively, the same weight may be assigned to each driving action score.

[0095] 10 and 11 show examples of the cognitive behavior scoring process and the vehicle behavior scoring process when the first scoring criterion is adopted in a situation where there are no surrounding moving objects. On the other hand, in a situation where there are surrounding moving objects, the second scoring criterion is adopted as shown in Fig. 9. Therefore, in a situation where there are surrounding moving objects, the cognitive behavior scoring process and the vehicle behavior scoring process will assign a lower driving behavior score than the examples in Fig. 10 and 11.

[0096] In this way, in this embodiment, a driving scene and cognitive behavior are detected, and the degree of risk associated with the driver's driving of the vehicle V is determined based on the detection results, so that the degree of risk can be appropriately determined. In other words, even if the driver is in good physical condition, if the cognitive behavior required for the driving scene, such as visually checking to the left when turning left at an intersection, is not performed, it is possible to determine that there is a high risk associated with driving in that driving scene. In other words, it is possible to appropriately determine whether unsafe driving has occurred.

[0097] Furthermore, in this embodiment, the degree of risk is determined by the driving score, so that the degree of risk can be evaluated in an easy-to-understand manner.

[0098] Furthermore, in this embodiment, driving video data with a low driving score, i.e., driving video data when unsafe driving is performed, is uploaded to the management server 200. Therefore, the manager who manages the vehicle V (i.e., the user of the management terminal 300) can access the management server 200 from the management terminal 300 and check the driving video data. As a result, the manager can recognize unsafe driving and improve awareness of safe driving.

[0099] Furthermore, in this embodiment, the degree of risk is determined based on the detection result of the behavior of the vehicle V by the vehicle behavior detection unit 103. Therefore, for example, in a driving scene in which the vehicle V turns left at an intersection, the degree of risk can be determined depending on whether the vehicle V is driving slowly. In other words, if the vehicle V is not driving slowly, the degree of risk can be determined to be high, and conversely, if the vehicle V is driving slowly, the degree of risk can be determined to be low. As a result, the degree of risk can be determined more appropriately.

[0100] (Embodiment 2) FIG. 12 is a block diagram showing an example of the functional configuration of the driving assistance system according to this embodiment.

[0101] The driving assistance system 100a in this embodiment includes the components of the driving assistance system 100 in the first embodiment, and further includes a dangerous operation detection unit 111, a surroundings visual detection unit 112, and a video editing unit 113.

[0102] The dangerous operation detection unit 111 detects a predetermined device operation by the driver as a dangerous operation in accordance with outputs from one or more sensors mounted on the vehicle V. The one or more sensors are, for example, a camera of the driver monitor 12.

[0103] The surrounding visual detection unit 112 detects the driver's visual inspection of one or more objects around the vehicle V in accordance with outputs from one or more sensors mounted on the vehicle V. The one or more sensors are, for example, the cameras of the driver monitor 12 and the drive recorder 14. Specifically, the surrounding visual detection unit 112 acquires a result of the cognitive behavior detection unit 102 detecting the cognitive behavior from the video image data output from the camera of the driver monitor 12. Furthermore, the surrounding visual detection unit 112 acquires a result of the peripheral object detection unit 104 detecting the peripheral object from the video image data output from the camera of the drive recorder 14. Then, the surrounding visual detection unit 112 detects the driver's visual inspection of one or more objects around the vehicle V based on the result of the cognitive behavior detection and the result of the peripheral object detection. The one or more objects around the vehicle V are the one or more peripheral objects detected by the surrounding detection unit 104. That is, when the driver's gaze point overlaps with a detected surrounding object or a bounding box surrounding the surrounding object, the surrounding visual detection unit 112 detects that the driver is visually observing the surrounding object.

[0104] In this embodiment, the risk determination unit 105 determines the degree of risk based not only on the detection results of cognitive behavior and the behavior of the vehicle V, but also on the detection results of dangerous operations by the dangerous operation detection unit 111 and the visual detection results by the surrounding visual detection unit 112.

[0105] The video editing unit 113 generates driving video data by selecting and editing portions of one or more video data sets obtained by capturing images using a camera mounted on the vehicle V, the portions corresponding to the driving scenes detected by the driving scene detection unit 101. The camera mounted on the vehicle V is at least one of the driver monitor 12 and the drive recorder 14. Such a camera continuously outputs video data sets obtained by capturing images for each predetermined recording time and stores the video data sets in the video storage unit 106. The recording time is, for example, one minute. When the video data sets of the driving scenes detected by the driving scene detection unit 101 are displayed across multiple video data sets, the video editing unit 113 generates one driving video data set by combining the multiple video data sets. Furthermore, when the video data set of the first of the multiple video data sets does not display the driving scene at the beginning, the video editing unit 113 may delete the beginning of the video data set. Similarly, if the last of the multiple video data sets does not show the driving scene video at its rear end, video editing unit 113 may delete that rear end from the video data set. Alternatively, if the driving scene video detected by driving scene detection unit 101 is shown in only a portion of one video data set, video editing unit 113 may generate driving video data by extracting that portion from the one video data set.

[0106] For example, in a driving scene of a left or right turn at an intersection, the video editing unit 113 generates driving video data showing moving images during a period in which the vehicle V travels from 60 m before a change point in the vehicle V's traveling direction (i.e., vector) to 20 m beyond that change point. Note that the change point is a point where the traveling direction changes by approximately 90 degrees. Alternatively, in the driving scene, the video editing unit 113 generates driving video data showing moving images during a period from 5 seconds before the vehicle V arrives at the change point in the vehicle V's traveling direction to 3 seconds after passing that change point. Furthermore, in a driving scene of backing into a parking space, the video editing unit 113 generates driving video data showing moving images during a period in which the vehicle V travels from 30 m before the change point in the vehicle V's traveling direction to 10 m beyond that change point. Alternatively, in the driving scene, the video editing unit 113 generates driving video data showing moving images during a period from 3 seconds before the vehicle V arrives at the change point in the vehicle V's traveling direction to 1 second after passing that change point. In addition, in a driving scene of a lane change, the video editing unit 113 generates driving video data showing a moving image from 5 seconds before the vehicle V crosses a white line to 5 seconds after the vehicle V crosses the white line. Note that the above numerical values such as the number of meters and the number of seconds are merely examples, and other numerical values may be used.

[0107] FIG. 13 is a diagram for explaining the detection of a dangerous operation by the dangerous operation detection unit 111. As shown in FIG.

[0108] The unsafe operation detection unit 111 identifies a gaze point in the driver's line of sight based on the video data output from the driver monitor 12. Then, as shown in FIG. 13 , if the gaze point is in area a5 and the driver's hand or finger is displayed in area a5 while driving, the unsafe operation detection unit 111 detects the driver's operation of the car navigation system as a unsafe operation. Alternatively, if the gaze point is in an area where the driver's smartphone is located and the driver's hand or finger is displayed in that area while driving, the unsafe operation detection unit 111 detects the driver's operation of the smartphone as a unsafe operation. Note that the unsafe operation detection unit 111 may detect the driver's operation of the car navigation system or smartphone in a driving scene detected by the driving scene detection unit 101 as a unsafe operation, depending on the driving scene.

[0109] FIG. 14 is a diagram for explaining the visual detection of surrounding objects by the surrounding visual detection unit 112. In FIG.

[0110] The surroundings visual detection unit 112 detects the driver's visual inspection of a surrounding object based on the detection result of the cognitive behavior and the detection result of the surrounding object, for example, as shown in Fig. 14. Specifically, when the driver's gaze point overlaps with a truck that is a surrounding object, the surroundings visual detection unit 112 detects the driver's visual inspection of the truck.

[0111] FIG. 15 is a flowchart showing an example of the overall processing operation of the driving assistance system 100a in this embodiment.

[0112] The driving assistance system 100a in this embodiment not only performs the processes of the steps included in the flowchart of FIG. 8 in the first embodiment, but also performs the processes of steps S300 and S500.

[0113] That is, the risk determination unit 105 performs a surroundings visual scoring process (step S300) based on the driving scene detected by the driving scene detection unit 101 and the detection result of the visual scoring of surrounding objects by the surroundings visual scoring detection unit 112. This surroundings visual scoring process is a process for scoring a driving behavior score for the visual scoring of surrounding objects in the detected scene based on the scoring criteria selected in step S100.

[0114] Next, the risk determination unit 105 performs a dangerous operation scoring process (step S500) based on the driving scene detected by the driving scene detection unit 101 and the detection result of the dangerous operation by the dangerous operation detection unit 111. This dangerous operation scoring process is a process for scoring driving action points for the dangerous operation in the detected scene based on the scoring criteria selected in step S100.

[0115] Note that the driving assistance system 100a may perform the processing of step S4 depending on the processing of steps S200, S300, S400, and S500. The processing of steps S200, S300, S400, and S500 may be performed in any order. The surroundings visual scoring processing of step S300 and the risky operation scoring processing of step S500 may be omitted depending on the driving scene detected by the driving scene detection unit 101 and the detection result of surrounding objects by the surroundings detection unit 104.

[0116] In this embodiment, the risk determination unit 105 determines the degree of risk associated with the driver's driving of the vehicle V based on the processing results of steps S200, S300, S400, and S500. That is, the risk determination unit 105 determines the degree of risk by scoring the driver's driving of the vehicle V (step S2). A driving score is thereby assigned. Specifically, the risk determination unit 105 assigns a driving score by performing weighted addition on the driving behavior scores assigned in steps S200, S300, S400, and S500.

[0117] After the process of step S2, the driving assistance system 100a executes the processes of steps S3 and S4, similarly to the first embodiment.

[0118] Specific processes of the driving assistance system 100a according to this embodiment will be described below with reference to examples 1 to 6.

[0119] [Example 1 (Driving scene: Turn right, Surrounding moving objects: None)] Example 1 is an example of specific processing by the driving assistance system 100a in a driving scene in which the vehicle is turning right at an intersection and there is no oncoming vehicle as a nearby moving object.

[0120] Fig. 16 is a flowchart showing a detailed example of the cognitive behavior scoring process and the surroundings visual scoring process. Fig. 16 shows detailed processing of steps S200 and S300 in Fig. 15 when a right turn is detected as a driving scene in step S1 in Fig. 15 and the first scoring criterion is adopted in step S100 in Fig. 15 (specifically, step S103 in Fig. 9). That is, Fig. 16 shows a detailed example of the cognitive behavior scoring process and the surroundings visual scoring process when a right turn is made when there is no oncoming vehicle as a surrounding moving object.

[0121] <Check the right side> First, the risk determination unit 105 determines whether the driver has visually checked the right side of the vehicle V for a specified period of time at a specified timing based on the result of the cognitive behavior detection by the cognitive behavior detection unit 102 (step S221). The specified timing may be the same as or different from the specified timing for turning left in the first embodiment. The specified period of time may be the same as or different from the specified period of time in the first embodiment. For example, the specified period of time may be within a predetermined range of times tb1 to tb2 (tb2>tb1). Here, when the risk determination unit 105 determines that the driver has visually checked the right side of the vehicle V for a specified period of time at a specified timing (Yes in step S221), it determines the driving behavior score for the cognitive behavior to the right to be 100 points (step S224).

[0122] On the other hand, when the risk determination unit 105 determines that the driver has not looked to the right of the vehicle V for a specified time at a predetermined timing (No in step S221), it determines whether the driver has looked to the right for a time different from the specified time at the predetermined timing (step S222). Here, when the risk determination unit 105 determines that the driver has looked to the right for a time different from the specified time at the predetermined timing (Yes in step S222), it determines the driving action score for the cognitive action to the right to be 90 points (step S225). For example, if the driver glances to the right at a predetermined timing, 90 points is determined.

[0123] On the other hand, when the risk determination unit 105 determines that the driver has not looked to the right at the predetermined timing for a time period different from the specified time (No in step S222), it determines whether or not a right-side look was made at a timing shifted from the predetermined timing (step S223). Here, when the risk determination unit 105 determines that a right-side look was made at a timing shifted from the predetermined timing (Yes in step S223), it determines the driving action score for the cognitive action to the right to be 70 points (step S226). On the other hand, when the risk determination unit 105 determines that a right-side look was not made at a timing shifted from the predetermined timing (No in step S223), it instructs the upload unit 107 to upload driving video data (step S227). For example, uploading is instructed when the driver is not looking to the right. As a result, the process of step S4 in FIG. 15 is executed.

[0124] <Check for oncoming traffic> Furthermore, the risk assessment unit 105 determines whether the driver visually assessed the oncoming lane as a surrounding object for a specified period of time at a specified timing based on the visual assessment detection result by the surrounding visual assessment detection unit 112 (step S301). The specified timing is, for example, the timing when the vehicle V is positioned just before entering the oncoming lane of the intersection. The specified period of time may be the same as or different from the specified period of time in the first embodiment. For example, the specified period of time may be a period within a predetermined range of times tc1 to tc2 (tc2 > tc1). If there are multiple oncoming lanes, the risk assessment unit 105 may determine whether the driver visually assessed the multiple oncoming lanes. Here, when the risk assessment unit 105 determines that the driver visually assessed the oncoming lane for the specified period of time at a specified timing (Yes in step S301), the risk assessment unit 105 assigns a driving action score of 100 points for visually assessing the oncoming lane (step S304).

[0125] On the other hand, when the risk determination unit 105 determines that the driver has not visually inspected the oncoming lane for a specified time at a predetermined timing (No in step S301), it determines whether the driver has visually inspected the oncoming lane for a time different from the specified time at the predetermined timing (step S302). Here, when the risk determination unit 105 determines that the driver has visually inspected the oncoming lane for a time different from the specified time at the predetermined timing (Yes in step S302), it determines the driving action score for visually inspecting the oncoming lane to be 90 points (step S305). For example, if the driver glances at the oncoming lane at a predetermined timing, 90 points will be determined.

[0126] On the other hand, if the risk determination unit 105 determines that the driver has not visually checked the oncoming lane for a time period different from the specified time at the predetermined timing (No in step S302), it determines whether or not the oncoming lane was visually checked at a time shifted from the predetermined timing (step S303). Here, if the risk determination unit 105 determines that the oncoming lane was visually checked at a time shifted from the predetermined timing (Yes in step S303), it determines the driving action score for visually checking the oncoming lane to be 70 points (step S306). On the other hand, if the risk determination unit 105 determines that the oncoming lane was not visually checked at a time shifted from the predetermined timing (No in step S303), it instructs the upload unit 107 to upload driving video data (step S307). For example, uploading is instructed when the driver is not looking at the oncoming lane. As a result, the process of step S4 in FIG. 15 is executed.

[0127] Fig. 17 is a flowchart showing a detailed example of the vehicle behavior scoring process. Fig. 17 shows detailed processing of step S400 in Fig. 15 when a right turn is detected as the driving scene in step S1 in Fig. 15 and the first scoring criterion is adopted in step S100 in Fig. 15 (specifically, step S103 in Fig. 9). That is, Fig. 17 shows a detailed example of the vehicle behavior scoring process when a right turn is made when there is no oncoming vehicle as a surrounding moving object.

[0128] <Slow down and give directions> First, the risk determination unit 105 executes the processes of steps S401 to S407 and steps S411 to S417, similarly to the example shown in Fig. 11. As a result, a driving action score for the vehicle speed and a driving action score for the direction instruction are determined.

[0129] <Right justified> Furthermore, the risk determination unit 105 determines whether the vehicle V has drifted to the right before turning right, based on the detection result of the behavior of the vehicle V by the vehicle behavior detection unit 103 (step S431). If the risk determination unit 105 determines that the vehicle V has drifted to the right (Yes in step S431), it determines the driving action score for the right pull-up to be 100 points (step S432). On the other hand, if the risk determination unit 105 determines that the vehicle V has not drifted to the right (No in step S431), it determines the driving action score for the right pull-up to be 90 points (step S433).

[0130] As a result, in step S2 of Fig. 15, the risk determination unit 105 calculates a driving score for the driver's driving of vehicle V by weighting and adding the driving action scores for each of the cognitive action to the right, visual inspection of the oncoming lane, vehicle speed, direction indication, and pulling to the right. Note that in Example 1, since there is no oncoming vehicle as a surrounding moving object, the processing of step S500 of Fig. 15 is omitted.

[0131] [Example 2 (Driving scene: Turn right, Surrounding moving objects (oncoming vehicles): Yes)] Example 2 is an example of specific processing by the driving assistance system 100a when the driving scene involves turning right at an intersection and there is an oncoming vehicle as a nearby moving object.

[0132] Fig. 18 is a flowchart showing a detailed example of the cognitive behavior scoring process and the surroundings visual scoring process. Fig. 18 shows detailed processing of steps S200 and S300 in Fig. 15 when a right turn is detected as a driving scene in step S1 in Fig. 15 and the second scoring criterion is adopted in step S100 in Fig. 15 (specifically, step S102 in Fig. 9). That is, Fig. 18 shows a detailed example of the cognitive behavior scoring process and the surroundings visual scoring process when a right turn is made when there is an oncoming vehicle as a surrounding moving object.

[0133] <Check the right side> First, as shown in FIG. 18, the risk determination unit 105 performs the same processes as steps S221 to S227 in FIG. 16, but performs the processes of steps S225a and S226a instead of steps S225 and S226.

[0134] That is, when the risk determination unit 105 determines in step S222 that the driver has looked to the right at a predetermined timing for a time period different from the specified time (Yes in step S222), it determines the driving action score for the cognitive action to the right as 60 points (step S225a). For example, if the driver glances to the right at a predetermined timing, 60 points is determined. In example 1 where there is no oncoming vehicle (i.e., the example in FIG. 16), 90 points is determined, whereas in example 2 where there is an oncoming vehicle (i.e., the example in FIG. 18), 60 points, which is lower than 90 points, is determined. In this way, the second scoring criteria provide stricter scoring than the first scoring criteria.

[0135] Similarly, when the risk determination unit 105 determines in step S223 that a visual check to the right was performed at a timing that was shifted from the predetermined timing (Yes in step S223), it determines the driving action score for the cognitive action to the right to be 40 points (step S226a). In example 1 (i.e., the example in FIG. 16) where there is no oncoming vehicle, 70 points is determined, whereas in example 2 (i.e., the example in FIG. 18) where there is an oncoming vehicle, 40 points, which is lower than 70 points, is determined. In this way, the second scoring criteria provide stricter scoring than the first scoring criteria.

[0136] <Check for oncoming vehicles> Furthermore, the risk assessment unit 105 determines whether the driver has visually inspected the oncoming vehicle for a specified time based on the visual inspection detection result by the surroundings visual inspection detection unit 112 (step S311). The specified time may be the same as or different from the specified time in the first embodiment. For example, the specified time may be a time within a predetermined range from td1 to td2 (td2>td1). Furthermore, when there are oncoming vehicles traveling in each of multiple oncoming lanes, it may be determined whether the driver has visually inspected these multiple oncoming vehicles. Here, when the risk assessment unit 105 determines that the driver has visually inspected the oncoming vehicle for the specified time (Yes in step S311), it sets the driving action score for visually inspecting the oncoming vehicle to 100 points (step S314).

[0137] On the other hand, when the risk determination unit 105 determines that the driver has not visually observed the oncoming vehicle for a specified time period at a predetermined timing (No in step S311), it determines whether the driver has visually observed the oncoming vehicle for a time period different from the specified time period (step S312). Here, when the risk determination unit 105 determines that the driver has visually observed the oncoming vehicle for a time period different from the specified time period (Yes in step S312), it determines the driving action score for visually observing the oncoming vehicle to be 60 points (step S315). For example, if the driver glances at the oncoming vehicle, 60 points will be determined.

[0138] On the other hand, if the risk determination unit 105 determines that the driver has not visually observed the oncoming vehicle for a period of time different from the specified time (No in step S312), it determines whether or not the point of gaze is around the oncoming vehicle (step S313). Here, if the risk determination unit 105 determines that the point of gaze is around the oncoming vehicle (Yes in step S313), it determines the driving action score for visually observing the oncoming vehicle to be 40 points (step S316). On the other hand, if the risk determination unit 105 determines that the point of gaze is not around the oncoming vehicle (No in step S313), it instructs the upload unit 107 to upload driving video data (step S317). For example, uploading is instructed when the driver does not look at the oncoming vehicle at all. As a result, the process of step S4 in FIG. 15 is executed.

[0139] Fig. 19 is a flowchart showing a detailed example of the vehicle behavior scoring process. Fig. 19 shows detailed processing of step S400 in Fig. 15 when a right turn is detected as the driving scene in step S1 in Fig. 15 and the second scoring criterion is adopted in step S100 in Fig. 15 (specifically, step S102 in Fig. 9). That is, Fig. 19 shows a detailed example of the vehicle behavior scoring process when a right turn is made when there is an oncoming vehicle as a nearby moving object.

[0140] <Pause / Accelerate> First, the risk determination unit 105 determines whether the vehicle V has safely stopped and accelerated in a predetermined manner based on the detection result of the behavior of the vehicle V by the vehicle behavior detection unit 103 (step S441). For example, the risk determination unit 105 determines that the vehicle V has safely stopped when the vehicle V has stopped at an intersection without suddenly decelerating. The predetermined acceleration is acceleration performed at a predetermined timing with an acceleration within an allowable range. For example, the risk determination unit 105 determines that the vehicle V has safely accelerated in a predetermined timing with an acceleration within an allowable range sufficiently before an oncoming vehicle enters the intersection by traveling straight or after the oncoming vehicle has passed the intersection by traveling straight. The sufficiently before timing is a predetermined number sc1 seconds after the oncoming vehicle enters the intersection by traveling straight, where sc1 may be, for example, a numerical value greater than or equal to 1. Here, when the risk determination unit 105 determines that the vehicle V has safely stopped temporarily and accelerated in a predetermined manner (Yes in step S441), it determines the driving action score for the vehicle speed to be 100 points (step S444).

[0141] On the other hand, when the risk determination unit 105 determines that the vehicle V has not safely stopped and accelerated as specified (No in step S441), it determines whether the vehicle V has suddenly accelerated (step S442). For example, the risk determination unit 105 determines that the vehicle V has suddenly accelerated when the acceleration of the vehicle V is greater than the above-mentioned allowable range. Here, when the risk determination unit 105 determines that the vehicle V has suddenly accelerated (Yes in step S442), it sets the driving action score for the vehicle speed to 80 points (step S445).

[0142] On the other hand, if the risk determination unit 105 determines that the vehicle V did not suddenly accelerate (No in step S442), it determines whether the vehicle V suddenly accelerated without stopping (step S443). Here, if the risk determination unit 105 determines that the vehicle V suddenly accelerated without stopping (Yes in step S443), it determines the driving action score for the vehicle speed to be 40 points (step S446). On the other hand, if the risk determination unit 105 determines that the vehicle V did not suddenly accelerate without stopping (No in step S443), it determines that the vehicle V suddenly stopped, and instructs the upload unit 107 to upload driving video data (step S447). As a result, the processing of step S4 in FIG. 15 is executed.

[0143] <Direction instructions> Furthermore, the risk determination unit 105 executes the processes of steps S411 to S417 in the same manner as in the example shown in Fig. 11. As a result, a driving action score for the direction instruction is determined.

[0144] <Right justified> Furthermore, the risk determination unit 105 executes the processes of steps S431 to S433 in the same manner as in the example shown in Fig. 17. As a result, the driving action score for pulling to the right is determined.

[0145] Fig. 20 is a flowchart showing a detailed example of the dangerous operation scoring process. Fig. 20 shows the detailed process of step S500 in Fig. 15 when a right turn is detected as a driving scene in step S1 in Fig. 15 and the second scoring criterion is adopted in step S100 in Fig. 15 (specifically, step S102 in Fig. 9). That is, Fig. 20 shows a detailed example of the dangerous operation scoring process when a right turn is made when there is an oncoming vehicle as a nearby moving object.

[0146] <Smartphone> First, the risk determination unit 105 determines whether the driver has operated a smartphone based on the detection result of the dangerous operation by the dangerous operation detection unit 111 (step S501). If the risk determination unit 105 determines that the driver has operated a smartphone (Yes in step S501), it determines the driving action score for the smartphone operation to be 60 points (step S502). On the other hand, if the risk determination unit 105 determines that the driver has not operated a smartphone (No in step S501), it determines the driving action score for the smartphone operation to be 100 points (step S503).

[0147] <Car navigation system> Furthermore, the risk determination unit 105 determines whether or not the driver has operated the car navigation system based on the detection result of the dangerous operation by the dangerous operation detection unit 111 (step S504). If the risk determination unit 105 determines that the driver has operated the car navigation system (Yes in step S504), it determines the driving action score for the car navigation system operation to be 60 points (step S505). On the other hand, if the risk determination unit 105 determines that the driver has not operated the car navigation system (No in step S504), it determines the driving action score for the car navigation system operation to be 100 points (step S506).

[0148] As a result, in step S2 of Figure 15, the risk assessment unit 105 calculates a driving score for the driver's driving of vehicle V by weighting and adding the driving action scores for each of the cognitive actions to the right, visual inspection of oncoming vehicles, vehicle speed, direction indication, pulling to the right, smartphone operation, and car navigation operation.

[0149] [Example 3 (Driving scene: Backing into a parking space, No moving objects around)] Example 3 is a specific example of processing by the driving assistance system 100a when the driving scene is backing into a parking lot and there are no surrounding moving objects. The surrounding moving objects are, for example, people or other vehicles in the parking lot.

[0150] Fig. 21 is a flowchart showing a detailed example of the cognitive behavior scoring process. Fig. 21 shows detailed processing of step S200 in Fig. 15 in the case where backing into parking is detected as the driving scene in step S1 in Fig. 15 and the first scoring criterion is adopted in step S100 in Fig. 15 (specifically, step S103 in Fig. 9). That is, Fig. 21 shows a detailed example of the cognitive behavior scoring process in the case where backing into parking is performed when there are no surrounding moving objects.

[0151] <Check left and right mirrors> First, the risk determination unit 105 determines whether the driver visually checked the left and right mirrors of the vehicle V for a specified period of time at a predetermined timing based on the result of the cognitive behavior detection by the cognitive behavior detection unit 102 (step S231). The specified timing may be when the vehicle V is backing up, or may be a timing sc2 seconds or later before the backing starts. sc2 is an arbitrary value greater than 0. The specified period of time may be the same as or different from the specified period of time in the first embodiment. For example, the specified period of time may be a period of time within a predetermined range of te1 to te2 (te2 > te1). Here, when the risk determination unit 105 determines that the driver visually checked the left and right mirrors of the vehicle V for a specified period of time at a predetermined timing (Yes in step S231), it determines the driving behavior score for the cognitive behavior of the left and right mirrors to be 100 points (step S234).

[0152] On the other hand, when the risk determination unit 105 determines that the driver has not visually checked the left and right mirrors of the vehicle V for a specified time at a predetermined timing (No in step S231), it determines whether the driver has visually checked the left and right mirrors for a time different from the specified time at the predetermined timing (step S232). Here, when the risk determination unit 105 determines that the driver has visually checked the left and right mirrors for a time different from the specified time at the predetermined timing (Yes in step S232), it determines the driving action score for the cognitive action of the left and right mirrors to be 90 points (step S235). For example, if the driver glances at the left and right mirrors at a predetermined timing, 90 points is determined.

[0153] On the other hand, if the risk determination unit 105 determines that the driver has not visually checked the left and right mirrors for a period of time different from the specified time at the predetermined timing (No in step S232), it determines whether the left and right mirrors were visually checked at a time shifted from the predetermined timing (step S233). Here, if the risk determination unit 105 determines that the left and right mirrors were visually checked at a time shifted from the predetermined timing (Yes in step S233), it determines the driving action score for the cognitive behavior of the left and right mirrors to be 70 points (step S236). On the other hand, if the risk determination unit 105 determines that the left and right mirrors were not visually checked at a time shifted from the predetermined timing (No in step S233), it instructs the upload unit 107 to upload driving video data (step S237). For example, if the driver has not looked at the left and right mirrors, the upload is instructed. As a result, the process of step S4 in FIG. 15 is executed.

[0154] <Backward confirmation> Furthermore, the risk determination unit 105 determines whether the driver has visually checked the rear of the vehicle V for a specified period of time at a specified timing based on the detection result of the cognitive behavior by the cognitive behavior detection unit 102 (step S241). The specified timing and specified period of time may be the same as or different from the specified timing and specified period of time used in step S231. Here, when the risk determination unit 105 determines that the driver has visually checked the rear of the vehicle V for a specified period of time at a specified timing (Yes in step S241), it determines the driving behavior score for the rearward cognitive behavior to be 100 points (step S244).

[0155] On the other hand, when the risk determination unit 105 determines that the driver has not visually checked the rear of the vehicle V for a specified time at a predetermined timing (No in step S241), it determines whether the driver has visually checked the rear for a time different from the specified time at the predetermined timing (step S242). Here, when the risk determination unit 105 determines that the driver has visually checked the rear for a time different from the specified time at the predetermined timing (Yes in step S242), it determines the driving action score for the cognitive behavior of looking behind to be 90 points (step S245). For example, if the driver glances behind at a predetermined timing, 90 points is determined.

[0156] On the other hand, when the risk determination unit 105 determines that the driver has not visually checked the rear at the predetermined timing for a time period different from the specified time (No in step S242), it determines whether or not a rearward check was performed at a timing shifted from the predetermined timing (step S243). Here, when the risk determination unit 105 determines that a rearward check was performed at a timing shifted from the predetermined timing (Yes in step S243), it determines the driving action score for the rearward cognitive behavior to be 70 points (step S246). On the other hand, when the risk determination unit 105 determines that a rearward check was not performed at a timing shifted from the predetermined timing (No in step S243), it instructs the upload unit 107 to upload driving video data (step S247). For example, uploading is instructed when the driver is not looking rearward. As a result, the process of step S4 in FIG. 15 is executed.

[0157] Fig. 22 is a flowchart showing a detailed example of the surroundings visual scoring process. Fig. 22 shows the detailed process of step S300 in Fig. 15 when backing into parking is detected as the driving scene in step S1 in Fig. 15 and the first scoring criterion is adopted in step S100 in Fig. 15 (specifically, step S103 in Fig. 9). That is, Fig. 22 shows a detailed example of the surroundings visual scoring process when backing into parking is performed when there are no surrounding moving objects.

[0158] <Check parking space> First, the risk assessment unit 105 determines whether the driver visually assessed the parking space for a specified period of time at a predetermined timing based on the visual assessment detection result by the surroundings visual assessment detection unit 112 (step S321). The predetermined timing may be the same as or different from the example in FIG. 21. The specified period of time may be the same as or different from the specified period of time in the first embodiment. For example, the specified period of time may be within a predetermined range of times tf1 to tf2 (tf2>tf1). The parking space is, for example, a space surrounded by a white line in a parking lot where no vehicle is parked. Here, when the risk assessment unit 105 determines that the driver visually assessed the parking space for the specified period of time at a predetermined timing (Yes in step S321), it assigns a driving action score of 100 points to the visual assessment of the parking space (step S324).

[0159] On the other hand, if the risk assessment unit 105 determines that the driver has not visually inspected the parking space for the specified time at the predetermined timing (No in step S321), it determines whether the driver has visually inspected the parking space for a time different from the specified time at the predetermined timing (step S322). Here, if the risk assessment unit 105 determines that the driver has visually inspected the parking space for a time different from the specified time at the predetermined timing (Yes in step S322), it determines the driving action score for visually inspecting the parking space to be 90 points (step S325). For example, if the driver glances at the parking space at the predetermined timing, 90 points will be determined.

[0160] On the other hand, if the risk determination unit 105 determines that the driver has not visually checked the parking space for a period of time different from the specified time at the predetermined timing (No in step S322), it determines whether the parking space has been visually checked at a time shifted from the predetermined timing (step S323). Here, if the risk determination unit 105 determines that the parking space has been visually checked at a time shifted from the predetermined timing (Yes in step S323), it determines the driving action score for the visual checking of the parking space to be 70 points (step S326). On the other hand, if the risk determination unit 105 determines that the parking space has not been visually checked at a time shifted from the predetermined timing (No in step S323), it instructs the upload unit 107 to upload driving video data (step S327). For example, uploading is instructed when the driver has not looked at the parking space. As a result, the process of step S4 in FIG. 15 is executed.

[0161] <Check the space ahead> Furthermore, the risk determination unit 105 determines whether the driver has visually checked the space ahead for a specified period of time at a specified timing based on the visual check detection result by the surroundings visual check detection unit 112 (step S331). The specified timing and specified period of time may be the same as or different from the specified timing and specified period of time used in step S321. The space ahead is a space around the front end of the vehicle V or around the front left and right ends of the vehicle V. Here, when the risk determination unit 105 determines that the driver has visually checked the space ahead for a specified period of time at a specified timing (Yes in step S331), it determines the driving action score for visually checking the space ahead to be 100 points (step S334).

[0162] On the other hand, when the risk determination unit 105 determines that the driver has not visually checked the space ahead for a specified time period at a predetermined timing (No in step S331), it determines whether the driver has visually checked the space ahead for a time period different from the specified time period at a predetermined timing (step S332). Here, when the risk determination unit 105 determines that the driver has visually checked the space ahead for a time period different from the specified time period at a predetermined timing (Yes in step S332), it determines the driving action score for visually checking the space ahead to be 90 points (step S335). For example, if the driver glances at the space ahead at a predetermined timing, 90 points is determined.

[0163] On the other hand, when the risk determination unit 105 determines that the driver has not visually checked the space ahead for a period of time different from the specified time at the predetermined timing (No in step S332), it determines whether or not the space ahead has been visually checked at a timing shifted from the predetermined timing (step S333). Here, when the risk determination unit 105 determines that the space ahead has been visually checked at a timing shifted from the predetermined timing (Yes in step S333), it determines the driving action score for the visual checking of the space ahead to be 70 points (step S336). On the other hand, when the risk determination unit 105 determines that the space ahead has not been visually checked at a timing shifted from the predetermined timing (No in step S333), it instructs the upload unit 107 to upload driving video data (step S337). For example, uploading is instructed when the driver has not looked at the space ahead. As a result, the process of step S4 in FIG. 15 is executed.

[0164] Fig. 23 is a flowchart showing a detailed example of the vehicle behavior scoring process. Fig. 23 shows detailed processing of step S400 in Fig. 15 when backing into parking is detected as the driving scene in step S1 in Fig. 15 and the first scoring criterion is adopted in step S100 in Fig. 15 (specifically, step S103 in Fig. 9). In other words, Fig. 22 shows a detailed example of the vehicle behavior scoring process when backing into parking is performed when there are no surrounding moving objects.

[0165] <vehicle speed> First, the risk determination unit 105 determines whether the vehicle V has slowed down and stopped at an acceleration within a first acceleration range based on the detection result of the behavior of the vehicle V by the vehicle behavior detection unit 103 (step S451). The first acceleration range is a predetermined range of negative acceleration. Here, when the risk determination unit 105 determines that the vehicle V has slowed down and stopped (Yes in step S451), it determines the driving action score for the vehicle speed to be 100 points (step S453).

[0166] On the other hand, when the risk determination unit 105 determines that the vehicle V is moving slowly and not stopped (No in step S451), it determines whether the vehicle V has suddenly decelerated (step S452). For example, the risk determination unit 105 determines that the vehicle V has suddenly decelerated when the vehicle V backs up and stops at an acceleration in a second acceleration range that is smaller than the first acceleration range. Here, when the risk determination unit 105 determines that the vehicle V has suddenly decelerated (Yes in step S452), it sets the driving action score for the vehicle speed to 90 points (step S454).

[0167] On the other hand, if the risk determination unit 105 determines that the vehicle V has not suddenly decelerated (No in step S452), it determines that the vehicle V has traveled at an acceleration smaller than the second acceleration range. In other words, the risk determination unit 105 determines that the vehicle V has suddenly stopped at a buffer stop. As a result, the risk determination unit 105 instructs the upload unit 107 to upload the driving video data (step S455). As a result, the processing of step S4 in FIG. 15 is executed.

[0168] Then, in step S2 of Figure 15, the risk assessment unit 105 calculates a driving score for the driver's driving of vehicle V by weighting and adding the driving action scores for each of the cognitive actions of the left and right mirrors, the cognitive actions behind, the visual inspection of the parking space, the visual inspection of the space ahead, and the vehicle speed.

[0169] [Example 4 (Driving scene: Backing into a parking space, surrounding moving objects (oncoming vehicles): Yes)] Example 4 is an example of specific processing by the driving assistance system 100a when the driving scene is backing into a parking lot and there is a nearby moving object.

[0170] Fig. 24 is a flowchart showing a detailed example of the cognitive behavior scoring process. Fig. 24 shows detailed processing of step S200 in Fig. 15 in the case where backing into parking is detected as the driving scene in step S1 in Fig. 15 and the second scoring criterion is adopted in step S100 in Fig. 15 (specifically, step S102 in Fig. 9). That is, Fig. 24 shows a detailed example of the cognitive behavior scoring process in the case where backing into parking is performed when there is a nearby moving object.

[0171] <Check left and right mirrors> First, as shown in FIG. 24, the risk assessment unit 105 performs the same processes as steps S231 to S237 in FIG. 21, but performs the processes of steps S235a and S236a instead of steps S235 and S236.

[0172] That is, when the risk determination unit 105 determines in step S232 that the driver has visually checked the left and right mirrors for a period of time different from the specified time at a predetermined timing (Yes in step S232), it determines the driving action score for the cognitive action of checking the left and right mirrors as 60 points (step S235a). For example, if the driver glances at the left and right mirrors at a predetermined timing, 60 points is determined. In example 3 (i.e., the example in FIG. 21) where there is no surrounding moving object, 90 points is determined, whereas in example 4 (i.e., the example in FIG. 24) where there is a surrounding moving object, 60 points, which is lower than 90 points, is determined. In this way, the second scoring criteria provide stricter scoring than the first scoring criteria.

[0173] Similarly, when the risk determination unit 105 determines in step S233 that the left and right mirrors were visually checked at a timing that differs from the predetermined timing (Yes in step S233), it determines the driving action score for the cognitive behavior of the left and right mirrors to be 40 points (step S236a). In example 3 (i.e., the example in FIG. 21) where there is no surrounding moving object, 70 points is determined, whereas in example 4 (i.e., the example in FIG. 24) where there is a surrounding moving object, 40 points, which is lower than 70 points, is determined. In this way, the second scoring criteria provide stricter scoring than the first scoring criteria.

[0174] <Check left> Next, the risk determination section 105 performs the same processes as steps S241 to S247 in FIG. 21, but performs the processes of steps S245a and S246a instead of steps S245 and S246.

[0175] That is, when the risk determination unit 105 determines in step S242 that the driver has visually looked behind at a predetermined timing for a time period different from the specified time (Yes in step S242), it determines the driving action score for the cognitive behavior behind as 60 points (step S245a). For example, if the driver glances to the left at a predetermined timing, 60 points is determined. In example 3 (i.e., the example in FIG. 21) where there is no surrounding moving object, 90 points is determined, whereas in example 4 (i.e., the example in FIG. 24) where there is a surrounding moving object, 60 points, which is lower than 90 points, is determined. In this way, the second scoring criteria provide stricter scoring than the first scoring criteria.

[0176] Similarly, in step S243, when the risk determination unit 105 determines that a rearward visual check was performed at a timing shifted from the predetermined timing (Yes in step S243), the risk determination unit 105 determines the driving action score for the rearward cognitive action to be 40 points (step S246a). In example 3 where there is no surrounding moving object (i.e., the example in FIG. 21), 70 points is determined, whereas in example 4 where there is a surrounding moving object (i.e., the example in FIG. 24), 40 points, which is lower than 70 points, is determined. In this way, the second scoring criteria provide stricter scoring than the first scoring criteria.

[0177] Fig. 25 is a flowchart showing a detailed example of the surroundings visual scoring process. Fig. 25 shows the detailed process of step S300 in Fig. 15 when backing into parking is detected as the driving scene in step S1 in Fig. 15 and the second scoring criterion is adopted in step S100 in Fig. 15 (specifically, step S102 in Fig. 9). That is, Fig. 25 shows a detailed example of the surroundings visual scoring process when backing into parking is performed when there are surrounding moving objects.

[0178] <Check parking space> First, as shown in FIG. 25, the risk assessment unit 105 performs the same processes as steps S321 to S327 in FIG. 22, but performs the processes of steps S325a and S326a instead of steps S325 and S326.

[0179] That is, when the risk assessment unit 105 determines in step S322 that the driver visually inspected the parking space for a period of time different from the specified time at a predetermined timing (Yes in step S322), it determines the driving action score for visually inspecting the parking space as 60 points (step S325a). For example, if the driver glances at the parking space at a predetermined timing, 60 points is determined. In example 3 (i.e., the example in FIG. 22) where there is no surrounding moving object, 90 points is determined, whereas in example 4 (i.e., the example in FIG. 25) where there is a surrounding moving object, 60 points, which is lower than 90 points, is determined. In this way, the second scoring criteria result in stricter scoring than the first scoring criteria.

[0180] Similarly, when the risk determination unit 105 determines in step S323 that the parking space was visually inspected at a timing different from the predetermined timing (Yes in step S323), it determines the driving action score for the visual inspection of the parking space to be 40 points (step S326a). In example 3 where there is no surrounding moving object (i.e., the example in FIG. 22), 70 points is determined, whereas in example 4 where there is a surrounding moving object (i.e., the example in FIG. 25), 40 points, which is lower than 70 points, is determined. In this way, the second scoring criteria provide stricter scoring than the first scoring criteria.

[0181] <Check the space ahead> Next, the risk assessment section 105 performs the same processes as steps S331 to S337 in FIG. 22, but performs the processes of steps S335a and S336a instead of steps S335 and S336.

[0182] That is, when the risk determination unit 105 determines in step S332 that the driver has visually inspected the space ahead at a predetermined timing for a time period different from the specified time (Yes in step S332), it determines the driving action score for visually inspecting the space ahead to be 60 points (step S335a). For example, if the driver glances at the space ahead at a predetermined timing, 60 points is determined. In example 3 (i.e., the example in FIG. 22) where there is no surrounding moving object, 90 points is determined, whereas in example 4 (i.e., the example in FIG. 25) where there is a surrounding moving object, 60 points, which is lower than 90 points, is determined. In this way, the second scoring criteria are stricter in scoring than the first scoring criteria.

[0183] Similarly, when the risk determination unit 105 determines in step S333 that the visual inspection of the space ahead was performed at a timing shifted from the predetermined timing (Yes in step S333), it determines the driving action score for the visual inspection of the space ahead to be 40 points (step S336a). In example 3 where there is no surrounding moving object (i.e., the example in FIG. 22), 70 points is determined, whereas in example 4 where there is a surrounding moving object (i.e., the example in FIG. 25), 40 points, which is lower than 70 points, is determined. In this way, the second scoring criteria provide stricter scoring than the first scoring criteria.

[0184] Fig. 26 is a flowchart showing a detailed example of the vehicle behavior scoring process. Fig. 26 shows detailed processing of step S400 in Fig. 15 when backing into parking is detected as the driving scene in step S1 in Fig. 15 and the second scoring criterion is adopted in step S100 in Fig. 15 (specifically, step S102 in Fig. 9). In other words, Fig. 24 shows a detailed example of the vehicle behavior scoring process when backing into parking is performed when there is a nearby moving object.

[0185] <vehicle speed> As shown in FIG. 26, the risk assessment unit 105 performs the same processes as steps S451 to S455 in FIG. 23, but performs the process of step S454a instead of step S454.

[0186] That is, when the risk determination unit 105 determines in step S452 that the vehicle V has suddenly decelerated (Yes in step S452), it determines the driving action score for the vehicle speed to be 60 points (step S454a). In example 3 where there is no surrounding moving object (i.e., the example in FIG. 23), a score of 90 is determined, whereas in example 4 where there is a surrounding moving object (i.e., the example in FIG. 26), a score of 60 points, which is lower than 90 points, is determined. In this way, the second scoring criteria provide stricter scoring than the first scoring criteria.

[0187] Then, in step S2 of Figure 15, the risk assessment unit 105 calculates a driving score for the driver's driving of vehicle V by weighting and adding the driving action scores for each of the cognitive actions of the left and right mirrors, the cognitive actions behind, the visual inspection of the parking space, the visual inspection of the space ahead, and the vehicle speed.

[0188] [Example 5 (Driving scene: Lane change to the right, surrounding moving objects: none)] Fig. 27 is a flowchart showing a detailed example of the cognitive behavior scoring process. Fig. 27 shows detailed processing of step S200 in Fig. 15 in the case where a lane change to the right is detected as a driving scene in step S1 in Fig. 15 and the first scoring criterion is adopted in step S100 in Fig. 15 (specifically, step S103 in Fig. 9). That is, Fig. 27 shows a detailed example of the cognitive behavior scoring process in the case where a lane change to the right is performed when there is no surrounding moving object. The surrounding moving object is, for example, another vehicle traveling in the same direction as vehicle V.

[0189] <Check the right mirror> First, the risk determination unit 105 determines whether the driver has visually checked the right mirror of the vehicle V for a specified period of time at a predetermined timing based on the result of the cognitive behavior detection by the cognitive behavior detection unit 102 (step S251). The predetermined timing may be, for example, sc3 seconds or later before the lane change is initiated. sc3 is an arbitrary value greater than 0. The specified period of time may be the same as or different from the specified period of time in the first embodiment. For example, the specified period of time may be within a predetermined range of times tg1 to tg2 (tg2 > tg1). Here, when the risk determination unit 105 determines that the driver has visually checked the right mirror of the vehicle V for a specified period of time at a predetermined timing (Yes in step S251), it determines the driving behavior score for the cognitive behavior of the right mirror to be 100 points (step S254).

[0190] On the other hand, when the risk determination unit 105 determines that the driver has not visually checked the right mirror of the vehicle V for a specified time at a predetermined timing (No in step S251), it determines whether the driver has visually checked the right mirror for a time different from the specified time at a predetermined timing (step S252). Here, when the risk determination unit 105 determines that the driver has visually checked the right mirror for a time different from the specified time at a predetermined timing (Yes in step S252), it determines the driving action score for the cognitive action of the right mirror to be 90 points (step S255). For example, if the driver glances at the right mirror at a predetermined timing, 90 points is determined.

[0191] On the other hand, when the risk determination unit 105 determines that the driver has not visually checked the right mirror for a period of time different from the specified time at the predetermined timing (No in step S252), it determines whether or not the right mirror was visually checked at a timing different from the predetermined timing (step S253). Here, when the risk determination unit 105 determines that the right mirror was visually checked at a timing different from the predetermined timing (Yes in step S253), it determines the driving action score for the cognitive behavior of the right mirror to be 70 points (step S256). On the other hand, when the risk determination unit 105 determines that the right mirror was not visually checked at a timing different from the predetermined timing (No in step S253), it instructs the upload unit 107 to upload driving video data (step S257). For example, uploading is instructed when the driver is not looking at the right mirror. As a result, the process of step S4 in FIG. 15 is executed.

[0192] <Check the right side> Furthermore, the risk determination unit 105 executes the processes of steps S221 to S227, similarly to the example of FIG.

[0193] Fig. 28 is a flowchart showing a detailed example of the vehicle behavior scoring process. Fig. 28 shows detailed processing of step S400 in Fig. 15 in the case where a lane change to the right is detected as the driving scene in step S1 in Fig. 15 and the first scoring criterion is adopted in step S100 in Fig. 15 (specifically, step S103 in Fig. 9). In other words, Fig. 28 shows a detailed example of the vehicle behavior scoring process in the case where a lane change to the right is made when there is no surrounding moving object.

[0194] <vehicle speed> First, the risk determination unit 105 determines whether the vehicle V changed lanes at a constant speed based on the detection result of the behavior of the vehicle V by the vehicle behavior detection unit 103 (step S461). For example, if the difference between the vehicle speed va of the vehicle V at the time of the lane change and the vehicle speed vb of the vehicle V immediately before the lane change is within, for example, 10% of the vehicle speed vb, the risk determination unit 105 determines that the vehicle V changed lanes at a constant speed. Note that 10% is just an example, and the difference is not limited to this numerical value. Here, when the risk determination unit 105 determines that the vehicle V changed lanes at a constant speed (Yes in step S461), it sets the driving action score for the vehicle speed to 100 points (step S464).

[0195] On the other hand, if the risk determination unit 105 determines that the vehicle V has not changed lanes at a constant speed (No in step S461), it determines whether the vehicle V has suddenly decelerated (step S462). Note that the sudden deceleration does not include a sudden stop. For example, here, if the risk determination unit 105 determines that the vehicle V has suddenly decelerated (Yes in step S462), it determines the driving action score for the vehicle speed to be 90 points (step S465).

[0196] On the other hand, if the risk determination unit 105 determines that the vehicle V has not suddenly decelerated (No in step S462), it determines whether the vehicle V has suddenly accelerated (step S463). Here, if the risk determination unit 105 determines that the vehicle V has suddenly accelerated (Yes in step S463), it determines the driving action score for the vehicle speed to be 70 points (step S466). On the other hand, if the risk determination unit 105 determines that the vehicle V has not suddenly accelerated (No in step S463), it instructs the upload unit 107 to upload driving video data (step S467). For example, uploading is instructed when the vehicle V suddenly stops. As a result, the processing of step S4 in FIG. 15 is executed.

[0197] <Direction instructions> Furthermore, the risk determination unit 105 determines whether the vehicle V issued a direction instruction within 3 seconds before the lane change based on the detection result of the behavior of the vehicle V by the vehicle behavior detection unit 103 (step S471). Here, when the risk determination unit 105 determines that the vehicle V issued a direction instruction (Yes in step S471), it determines the driving action score for the direction instruction to be 100 points (step S474).

[0198] On the other hand, if the risk determination unit 105 determines that the vehicle V has not issued a direction indication within 3 seconds before the lane change (No in step S471), it determines whether the vehicle V issued a direction indication within 2 seconds before the lane change (step S472).Here, if the risk determination unit 105 determines that the vehicle V has issued a direction indication (Yes in step S472), it determines the driving action score for the direction indication to be 90 points (step S475).

[0199] On the other hand, if the risk determination unit 105 determines that the vehicle V has not issued a turn signal within two seconds before the lane change (No in step S472), it determines whether the vehicle V has issued a turn signal within one second before the lane change (step S473). Here, if the risk determination unit 105 determines that the vehicle V has issued a turn signal (Yes in step S473), it determines the driving action score for the turn signal to be 70 points (step S476). On the other hand, if the risk determination unit 105 determines that the vehicle V has not issued a turn signal within one second before the lane change (No in step S473), it instructs the upload unit 107 to upload driving video data (step S477). As a result, the processing of step S4 in FIG. 15 is executed.

[0200] Then, in step S2 of Figure 15, the risk assessment unit 105 calculates a driving score for the driver's driving of vehicle V by weighting and adding the driving action scores for the right mirror cognitive action, right-side cognitive action, vehicle speed, and direction indication.

[0201] [Example 6 (Driving scene: Lane change to the right, surrounding moving objects: Yes)] Fig. 29 is a flowchart showing a detailed example of the cognitive behavior scoring process. Fig. 29 shows detailed processing of step S200 in Fig. 15 in the case where a lane change to the right is detected as a driving scene in step S1 in Fig. 15 and the second scoring criterion is adopted in step S100 in Fig. 15 (specifically, step S102 in Fig. 9). That is, Fig. 29 shows a detailed example of the cognitive behavior scoring process in the case where a lane change to the right is made when there is a nearby moving object.

[0202] <Check the right mirror> First, as shown in FIG. 29, the risk assessment unit 105 performs the same processes as steps S251 to S257 in FIG. 27, but performs the processes of steps S255a and S256a instead of steps S255 and S256.

[0203] That is, when the risk determination unit 105 determines in step S252 that the driver has visually checked the right mirror for a period of time different from the specified time at a predetermined timing (Yes in step S252), it determines the driving action score for the cognitive action of checking the right mirror as 60 points (step S255a). For example, if the driver glances at the right mirror at a predetermined timing, 60 points is determined. In example 5 where there is no surrounding moving object (i.e., the example in FIG. 27), 90 points is determined, whereas in example 6 where there is a surrounding moving object (i.e., the example in FIG. 29), 60 points, which is lower than 90 points, is determined. In this way, the second scoring criteria provide stricter scoring than the first scoring criteria.

[0204] Similarly, when the risk determination unit 105 determines in step S253 that the right mirror was visually checked at a timing that is shifted from the predetermined timing (Yes in step S253), it determines the driving action score for the cognitive action of checking the right mirror to be 40 points (step S256a). In example 5 where there is no surrounding moving object (i.e., the example in FIG. 27), 70 points is determined, whereas in example 6 where there is a surrounding moving object (i.e., the example in FIG. 29), 40 points, which is lower than 70 points, is determined. In this way, the second scoring criteria provide stricter scoring than the first scoring criteria.

[0205] <Check the right side> Furthermore, the risk determination unit 105 executes the processes of steps S221 to S227, similarly to the example of FIG.

[0206] Fig. 30 is a flowchart showing a detailed example of the vehicle behavior scoring process. Fig. 30 shows detailed processing of step S400 in Fig. 15 when a lane change to the right is detected as the driving scene in step S1 in Fig. 15 and the second scoring criterion is adopted in step S100 in Fig. 15 (specifically, step S102 in Fig. 9). That is, Fig. 30 shows a detailed example of the vehicle behavior scoring process when a lane change to the right is made when there is a nearby moving object.

[0207] <vehicle speed> First, as shown in FIG. 30, the risk assessment section 105 performs the same processes as steps S461 to S467 in FIG. 28, but performs the processes of steps S465a and S466a instead of steps S465 and S466.

[0208] That is, when the risk determination unit 105 determines in step S462 that the vehicle V has suddenly decelerated (Yes in step S462), it determines the driving action score for the vehicle speed to be 80 points (step S465a). In example 5 where there is no surrounding moving object (i.e., the example in FIG. 28), a score of 90 is determined, whereas in example 6 where there is a surrounding moving object (i.e., the example in FIG. 30), a score of 80 points, which is lower than 90 points, is determined. In this way, the second scoring criteria provide stricter scoring than the first scoring criteria.

[0209] Similarly, when the risk determination unit 105 determines in step S463 that the vehicle V has suddenly accelerated (Yes in step S463), it determines the driving action score for the vehicle speed to be 40 points (step S466a). In example 5 (i.e., the example in FIG. 28) where there is no surrounding moving object, 70 points is determined, whereas in example 6 (i.e., the example in FIG. 30) where there is a surrounding moving object, 40 points, which is lower than 70 points, is determined. In this way, the second scoring criteria provide stricter scoring than the first scoring criteria.

[0210] <Direction instructions> Furthermore, the risk assessment section 105 performs the same processes as steps S471 to S477 in FIG. 28, but performs the processes of steps S475a and S476a instead of steps S475 and S476.

[0211] That is, when the risk determination unit 105 determines in step S472 that the vehicle V has issued a direction instruction (Yes in step S472), it determines the driving action score for the direction instruction to be 60 points (step S475a). In example 5 where there is no surrounding moving object (i.e., the example in FIG. 28), a score of 90 is determined, whereas in example 6 where there is a surrounding moving object (i.e., the example in FIG. 30), a score of 60 points, which is lower than 90 points, is determined. In this way, the second scoring criteria provide stricter scoring than the first scoring criteria.

[0212] Similarly, when the risk determination unit 105 determines in step S473 that the vehicle V has issued a direction instruction (Yes in step S473), it determines the driving action score for the direction instruction to be 40 points (step S476a). In example 5 (i.e., the example in FIG. 28) where there is no surrounding moving object, 70 points is determined, whereas in example 6 (i.e., the example in FIG. 30) where there is a surrounding moving object, 40 points, which is lower than 70 points, is determined. In this way, the second scoring criteria provide stricter scoring than the first scoring criteria.

[0213] <Following distance> Furthermore, the risk determination unit 105 determines whether the inter-vehicle distance in the lane direction between vehicle V and another vehicle traveling in the adjacent lane to the right is equal to or greater than a specified distance, based on the detection result of the behavior of vehicle V by the vehicle behavior detection unit 103 (step S481). The specified distance is a predetermined distance, and may be longer as the vehicle speed of vehicle V is faster. Here, if the risk determination unit 105 determines that the inter-vehicle distance is equal to or greater than the specified distance (Yes in step S481), it determines the driving action score for the inter-vehicle distance to be 100 points (step S482). On the other hand, if the risk determination unit 105 determines that the inter-vehicle distance is not equal to or greater than the specified distance (No in step S481), it determines the driving action score for the inter-vehicle distance to be 90 points (step S483).

[0214] Then, in step S2 of Figure 15, the risk assessment unit 105 calculates a driving score for the driver's driving of vehicle V by weighting and adding the driving action scores for each of the cognitive action of the right mirror, cognitive action to the right, vehicle speed, direction indication, and inter-vehicle distance.

[0215] As described above, in this embodiment, driving video data is generated by editing one or more video data segments by the video editing unit 113, and the driving video data is uploaded to the management server 200. As a result, even if video data for a predetermined recording time is output from the camera of the driver monitor 12 or the drive recorder 14, for example, at predetermined recording times, only a portion corresponding to a driving scene is extracted from the one or more output video data segments and uploaded as driving video data. In a specific example, the recording time is one minute. Therefore, video data segments depicting scenes different from the detected driving scene, or even a portion of the video data segments, can be prevented from being uploaded to the server. As a result, the transmission load of unnecessary data can be reduced. Furthermore, if a driving scene is captured across multiple video data segments, the multiple video data segments can be edited into a single driving video data segment. As a result, the driving scene is displayed without interruption when the driving video data is played back, allowing the video of the driving scene to be properly confirmed.

[0216] Furthermore, in this embodiment, the degree of risk is determined based on the visual inspection detection result by the surrounding visual inspection detection unit 112. As a result, for example, in a driving scene in which the vehicle V turns right at an intersection, the degree of risk can be determined depending on whether or not the driver has visually inspected the oncoming lane or oncoming vehicle. In other words, if no visual inspection has been performed, the degree of risk can be determined to be high, and conversely, if visual inspection has been performed, the degree of risk can be determined to be low. As a result, the degree of risk can be determined more appropriately.

[0217] Furthermore, in this embodiment, the degree of risk is determined based on the detection result of the dangerous operation by the dangerous operation detection unit 111. As a result, for example, in a driving scene in which the vehicle V turns right at an intersection, the degree of risk can be determined depending on whether or not the driver has performed an operation on a smartphone or the like as a dangerous operation. In other words, if no dangerous operation is being performed, the degree of risk can be determined to be low, and conversely, if a dangerous operation is being performed, the degree of risk can be determined to be high. As a result, the degree of risk can be determined more appropriately.

[0218] (Embodiment 3) In the first and second embodiments, if the driving score is less than the risk threshold, the driving video data is uploaded. On the other hand, in the present embodiment, even if the driving score is less than the risk threshold, if the driving score is less than the personal risk threshold associated with the driver, the driving video data is not uploaded. As a result, even if the driver's driving of the vehicle V is unsafe, the uploading of driving video data of the unsafe driving can be suppressed depending on the characteristics of the driver, etc., thereby reducing the load associated with uploading. Furthermore, in the present embodiment, the above-mentioned specified time is determined according to the driver's driving history. As a result, the degree of risk associated with driving the vehicle V can be more appropriately determined.

[0219] FIG. 31 is a block diagram showing an example of the functional configuration of the driving assistance system according to this embodiment.

[0220] The driving assistance system 100b of this embodiment includes the components of the driving assistance system 100a of the second embodiment, and further includes a time condition updating unit 121, a driving history storage unit 122, and a personal threshold value determining unit 123.

[0221] The driving history storage unit 122 is a recording medium that stores driving history data indicating the driving histories of each of a plurality of people. For example, the driving history storage unit 122 is a hard disk drive, a RAM (Random Access Memory), a ROM (Read Only Memory), or a semiconductor memory. Note that such a driving history storage unit 122 may be volatile or non-volatile.

[0222] The time condition update unit 121 determines, as a time condition, the above-mentioned specified time corresponding to each of the plurality of people based on the driving histories of the plurality of people indicated in the driving history data stored in the driving history storage unit 122. The time condition update unit 121 is also called a time condition determination unit.

[0223] The personal threshold determination unit 123 determines a plurality of personal risk thresholds corresponding to the plurality of people based on the driving histories of the people indicated in the driving history data stored in the driving history storage unit 122.

[0224] When the risk determination unit 105 in this embodiment determines that the driving score is less than the risk threshold, it further selects an individual risk threshold associated with the driver of vehicle V from the multiple individual risk thresholds determined by the individual threshold determination unit 123. Then, the risk determination unit 105 determines whether the driving score is less than the selected individual risk threshold. Here, when the risk determination unit 105 determines that the driving score is less than the individual risk threshold, it instructs the upload unit 107 to upload driving video data. In other words, the upload unit 107 in this embodiment uploads the driving video data to the management server 200 when the driving score is less than the risk threshold and also less than the individual risk threshold.

[0225] FIG. 32 is a diagram illustrating an example of driving history data stored in the driving history storage unit 122. For example, the driving history data indicates the number of unsafe driving incidents per month for each of a plurality of people. The plurality of people are people who could potentially be drivers of the vehicle V. The number of unsafe driving incidents is the number of times unsafe driving occurred. Each of (a) to (c) in FIG. 32 is a graph showing the number of unsafe driving incidents per month for each month. (a) in FIG. 32 is a graph showing the number of unsafe driving incidents by a person who drives dangerously, (b) in FIG. 32 is a graph showing the number of unsafe driving incidents by a person who drives safely, and (c) in FIG. 32 is a graph showing the number of unsafe driving incidents by a person with little driving experience.

[0226] Specifically, the driving history data indicates that dangerous drivers have a high number of unsafe driving incidents each month, as shown in Figure 32(a). The driving history data also indicates that safe drivers have a low number of unsafe driving incidents each month, as shown in Figure 32(b). The driving history data also indicates that inexperienced drivers have no unsafe driving incidents each month, as shown in Figure 32(c).

[0227] FIG. 33 is a diagram showing a more specific example of driving history data stored in the driving history storage unit 122.

[0228] For example, as shown in FIG. 31 , the driving history data indicates the driving history of each of a plurality of people by associating their user ID, name, age, years of driving, driver type, unsafe history, frequency of unsafe incidents, and accident history with each other. The user ID is identification information of the person corresponding to the user ID. The years of driving indicate the number of years that the person corresponding to the years of driving has experience driving a vehicle. The number of years may also be the number of years that the person has driven a vehicle professionally. Driver types include, for example, a safe veteran, a risky veteran, and a beginner. A safe veteran is a type of person whose years of driving are longer than a first career threshold and whose frequency of unsafe incidents is lower than the first frequency threshold. A risky veteran is a type of person whose years of driving are longer than the first career threshold and whose frequency of unsafe incidents is higher than a second frequency threshold. A beginner is a type of person whose years of driving are shorter than the second career threshold. The second career threshold may be smaller than or equal to the first career threshold. The first frequency threshold may be smaller than or equal to the second frequency threshold.

[0229] The unsafety history indicates the number of unsafe driving incidents per month by the person corresponding to the unsafety history. For example, the unsafety history C1 is represented as the graph shown in FIG. 32(b), the unsafety history C2 is represented as the graph shown in FIG. 32(a), and the unsafety history C3 is represented as the graph shown in FIG. 32(c). The unsafety frequency is the moving average of the number of unsafe driving incidents per month by the person corresponding to the unsafety frequency. Note that the moving average may be a weighted moving average. The moving average may also be an average value for the most recent three months, and is not limited to three months, but may also be an average value for a period of two months, four months, or more. The accident history indicates the number of accidents that the person corresponding to the accident history has caused in the past. For example, in the driving history data, the person with user ID "001" is associated with accident history D1, the person with user ID "002" is associated with accident history D2, and the person with user ID "003" is associated with accident history D3.

[0230] The individual threshold determination unit 123 determines, for each user ID indicated in the driving history data, the individual risk threshold of the person identified by the user ID based on the person's driving history. For example, the individual threshold determination unit 123 identifies the driver of the vehicle V based on video data output from the driver monitor 12 and specifies the driver's user ID. Then, the individual threshold determination unit 123 references the driving history data stored in the driving history storage unit 122 and specifies the driver type associated with the user ID. For example, if the identified driver type is a safe veteran, the individual threshold determination unit 123 determines x1 [%] of the risk threshold as the individual risk threshold of the driver. Note that x1 is a numerical value greater than 0 and less than 100. For example, if the identified driver type is a risky veteran, the individual threshold determination unit 123 determines x2 [%] of the risk threshold as the individual risk threshold of the driver. Note that x2 is a numerical value greater than 0 and less than 100, and is greater than x1. Furthermore, for example, if the identified driver type is a beginner, the personal threshold determination unit 123 determines the risk threshold to be the above-mentioned personal risk threshold of the driver.

[0231] As a result, if the driver is a beginner, all of the driving video data of the driver's unsafe driving will be uploaded. Also, if the driver is a dangerous veteran, most of the driving video data of all of the driver's unsafe driving will be uploaded. On the other hand, if the driver is a safe veteran, the number of driving video data of all of the driver's unsafe driving will be reduced.

[0232] The personal threshold determination unit 123 may determine a personal risk threshold for a person according to the frequency of unsafety of the person, regardless of the driver type. For example, the more the frequency of unsafety of the person, the higher the personal threshold determination unit 123 may determine a personal risk threshold for the person, and conversely, the less the frequency of unsafety of the person, the lower the personal risk threshold may determine a personal risk threshold for the person.

[0233] Fig. 34 is a diagram for explaining an example of the processing operation of the time condition update unit 121. Note that the horizontal axis of the graph in Fig. 34 represents the visual inspection time, and the vertical axis represents the accident occurrence probability.

[0234] The time condition update unit 121 determines the specified time for veterans, the specified time for beginners, etc. based on the driving history data stored in the driving history storage unit 122. The specified time for veterans may be the same as the specified time for safe veterans and the specified time for dangerous veterans. Furthermore, the time condition update unit 121 may update the specified time based on the latest driving history data every time the driving history data is updated.

[0235] Specifically, as shown in FIG. 34, the time condition update unit 121 compiles statistics for each driver type based on the driving history data, relating the visual observation time to the probability of an accident occurring due to driving based on that visual observation time (i.e., the accident probability). Here, the shorter the visual observation time, the higher the accident probability. Similarly, the longer the visual observation time, the higher the accident probability. Therefore, the time condition update unit 121 determines, for each driver type, a visual observation time equal to or less than a predetermined probability threshold as the specified time. That is, the time condition update unit 121 determines each visual observation time among the visual observation times t11 to t12 as the specified time for an experienced driver. Similarly, the time condition update unit 121 determines each visual observation time among the visual observation times t21 to t22 as the specified time for a beginner. In this way, the specified time corresponding to each driver type of a plurality of people is determined based on the driving histories of the plurality of people. That is, the specified time is determined according to the driving proficiency of the driver.

[0236] The risk determination unit 105 determines a driving action score using the specified time determined by the time condition update unit 121. That is, the risk determination unit 105 identifies the driver of the vehicle V based on the video data output from the driver monitor 12, and specifies the user ID of the driver. Furthermore, the risk determination unit 105 refers to the driving history data stored in the driving history storage unit 122, and specifies the driver type associated with the user ID. Then, the risk determination unit 105 determines the specified time for the specified driver type by inquiring of the time condition update unit 121. In other words, the risk determination unit 105 selects the specified time for the specified driver type from the multiple specified times determined by the time condition update unit 121. The risk determination unit 105 determines a driving action score for a cognitive behavior using the specified time determined or selected in this way.

[0237] That is, the risk determination unit 105 in this embodiment selects a specified time associated with the driver of the vehicle V from a plurality of specified times determined by the time condition update unit 121. Then, the risk determination unit 105 determines the degree of risk by comparing the time during which cognitive behavior detected by the cognitive behavior detection unit 102 was performed with the selected specified time. As a result, a specified time appropriate for the driver is selected, and the degree of risk is determined by comparing the time during which cognitive behavior by the driver was performed with the specified time, so that the risk can be more appropriately determined. For example, there is a tendency for the time required for cognitive behavior (e.g., the time spent looking to the right) required for safe driving to differ between experienced drivers and novice drivers. Therefore, by using a specified time appropriate for the driver, the degree of risk associated with driving by that driver can be appropriately determined.

[0238] FIG. 35 is a flowchart showing an example of the overall processing operation of the driving assistance system 100b in this embodiment.

[0239] The driving assistance system 100b in this embodiment performs the processing operations of the driving assistance system 100a shown in FIG. 15, and further executes the processing of step S3a.

[0240] That is, when the risk determination unit 105 determines that the driving score assigned in step S2 is less than the risk threshold (Yes in step S3), it further determines whether the driving score is less than the personal risk threshold (step S3a). The personal risk threshold is a threshold determined by the personal threshold determination unit 123 for the driver of vehicle V. Then, when the risk determination unit 105 determines that the driving score assigned in step S2 is less than the personal risk threshold (Yes in step S3a), it instructs the upload unit 107 to upload the driving video data. As a result, the upload unit 107 uploads the driving video data to the management server 200 (step S4).

[0241] On the other hand, even if the driving score is less than the risk threshold (Yes in step S3), if the risk determination unit 105 determines that the driving score is not less than the personal risk threshold (No in step S3a), it terminates processing for the driving scene detected in step S1 without instructing uploading.

[0242] FIG. 36 is a flowchart showing an example of processing operations related to driving history by the risk determination unit 105 and the individual threshold determination unit 123 in this embodiment.

[0243] As described above, the risk determination unit 105 determines whether the driving score scored in step S2 of Fig. 35 is less than the risk threshold (step S3). Here, if the risk determination unit 105 determines that the driving score is less than the risk threshold (Yes in step S3), it updates the driving history data stored in the driving history storage unit 122 (step S11). That is, the risk determination unit 105 updates the unsafe history and unsafe frequency shown in Fig. 33 out of the driving history data.

[0244] After the process of step S11, the personal threshold determination unit 123 updates the personal risk threshold of the driver corresponding to the driving score described above, based on the unsafe frequency of the updated driving history data (step S12).

[0245] As described above, in this embodiment, even if the driving score is below the risk threshold, unless it is below the personal risk threshold associated with the driver, uploading of driving video data is not performed. Therefore, even if a driver engages in unsafe driving, whether or not to upload driving video data can be switched depending on the driver. For example, a low personal risk threshold is associated with an experienced driver who normally drives safely. As a result, even if the driver happens to engage in unsafe driving, uploading of driving video data can be suppressed. Therefore, uploading can be optimized according to the driver. By associating a low personal risk threshold with an experienced driver who has been driving safely for many years (i.e., a safe experienced driver), uploading of driving video data can be made more difficult. On the other hand, by associating a high personal risk threshold with an experienced driver who has been driving dangerously for many years (i.e., a dangerous experienced driver), uploading of driving video data can be made more easy. As a result, uploading of driving video data can be optimized and made more efficient.

[0246] In this embodiment, a plurality of individual risk thresholds corresponding to a plurality of individuals are determined based on the driving histories of the individuals, thereby enabling a personal risk threshold to be determined for each of the individuals according to their type as a driver (i.e., driver type). (Variation) The above-mentioned personal risk threshold is a threshold that is commonly used for all driving by the person driving the vehicle V, regardless of the type of driving performed. On the other hand, in this modified example, a personal risk threshold that corresponds to each driving scene (i.e., each driving scene) or each location (or region) where driving is performed is used. Such personal risk thresholds are also called conditional personal risk thresholds. Furthermore, a personal risk threshold that corresponds to a driving scene is also called a conditional personal risk threshold for a driving scene, and a personal risk threshold that corresponds to a location (or region) is also called a conditional personal risk threshold for a location.

[0247] FIG. 37 is a diagram showing an example of an accident history included in the driving history data.

[0248] For example, the accident history includes the date and time of the accident, the driving scene of the accident, the unsafe type that is considered to be the cause of the accident, etc. The date and time of the accident indicates the year, month, day, and time. The unsafe type is the type of driving that is distinguished by the driving behavior with the lowest driving behavior score in the driving scene.

[0249] In a specific example, the accident history D1 indicates that a person with user ID "001" caused an accident of unsafe type "failure to check left" in a driving scene "turn left at intersection" on the date and time of occurrence "a." The driving scene "turn left at intersection" is a driving scene of turning left at the above-mentioned intersection. The unsafe type "failure to check left" is the type of driving that received the lowest driving behavior score for cognitive behavior on the left side out of all the driving behavior scores scored for the driving in the accident.

[0250] The personal threshold determination unit 123 determines the personal risk threshold of the person with user ID "001" based on the driver type of the driving history data shown in FIG. 33. Furthermore, the personal threshold determination unit 123 determines the personal risk threshold for each driving scene of the person with user ID "001" as the conditional personal risk threshold for the driving scene based on the above-mentioned accident history D1. For example, for each driving scene shown in the accident history D1, the personal threshold determination unit 123 adds a difference value corresponding to the number of accidents that have occurred in that driving scene to the personal risk threshold. The difference value may be a negative value or a positive value. More specifically, the more the number of accidents that have occurred, the larger the positive difference value that the personal threshold determination unit 123 adds to the personal risk threshold. In this way, the personal risk threshold for each driving scene, i.e., the conditional personal risk threshold for the driving scene, is determined.

[0251] 35, the risk determination unit 105 compares the conditional individual risk threshold for the driving scene detected in step S1 and the driving scene determined by the individual threshold determination unit 123 for the driver of vehicle V with the driving score scored in step S2. As a result, the more accidents that occur in a driving scene, the more likely it is that driving video data of unsafe driving in that driving scene will be uploaded.

[0252] Fig. 38 is a diagram for explaining the conditional individual risk threshold for a location, showing a map of the location where the vehicle V travels.

[0253] For example, a person with user ID "001" drives vehicle V as the driver. This driver has a high number of accidents on expressways and a low number of accidents off expressways. In this case, it is desirable for this person's personal risk threshold to be high when driving on expressways and low when driving off expressways. Also, a person with user ID "002" drives vehicle V as the driver. This driver has a high number of accidents in urban areas and a low number of accidents off expressways. In this case, it is desirable for this person's personal risk threshold to be high when driving in urban areas and low when driving off expressways.

[0254] Therefore, the accident history may indicate the driving location, as in the example shown in FIG.

[0255] FIG. 39 is a diagram showing another example of the accident history included in the driving history data.

[0256] For example, the accident history includes the date and time of an accident that occurred while driving, the location of the driving, etc. In a specific example, the accident history D1 indicates that a person with user ID "001" had an accident at the location "expressway" on the date and time of the accident "e." Note that the location may be a location identified by a location signal output from the GPS unit 11.

[0257] The individual threshold determination unit 123 determines the individual risk threshold of the person with user ID "001" based on the driver type of the driving history data shown in FIG. 33. Furthermore, the individual threshold determination unit 123 determines the individual risk threshold for each location of driving by the person with user ID "001" as the conditional individual risk threshold for the location based on the above-mentioned accident history D1. For example, for each location shown in the accident history D1, the individual threshold determination unit 123 adds a difference value corresponding to the number of accidents that have occurred at that location to the individual risk threshold. The difference value may be a negative value or a positive value. More specifically, the individual threshold determination unit 123 adds a larger positive difference value to the individual risk threshold as the number of accidents that have occurred at that location increases. This determines the individual risk threshold for each location, i.e., the conditional individual risk threshold for the location.

[0258] 35, the risk determination unit 105 compares the driving score scored in step S2 with the conditional individual risk threshold for the location where the vehicle V is being driven and for the driver of the vehicle V, determined by the individual threshold determination unit 123. As a result, the more accidents that occur in a location, the more likely it is that driving video data of unsafe driving at that location will be uploaded.

[0259] The accident history may include the content shown in FIG. 37 and the content shown in FIG. 39. That is, the accident history may include the date and time of an accident caused by driving, the driving scene of the driving, and the location of the driving. In this case, the personal threshold determination unit 123 adds a difference value corresponding to the number of accidents occurring in the combination of driving scenes and locations shown in the accident history to the personal risk threshold. More specifically, the more accidents occur in the driving scenes and locations included in the combination, the larger the positive difference value added by the personal threshold determination unit 123 to the personal risk threshold. This determines the personal risk threshold for each combination, i.e., the personal risk threshold for each combination of driving scene and location. This personal risk threshold for each combination of driving scene and location is also referred to as the conditional personal risk threshold for the driving scene and location.

[0260] FIG. 40 is a flowchart showing an example of processing operations related to driving history by the risk determination unit 105 and the individual threshold determination unit 123 in this modification.

[0261] As described above, the risk determination unit 105 determines whether the driving score scored in step S2 is less than the risk threshold (step S3). Here, if the risk determination unit 105 determines that the driving score is less than the risk threshold (Yes in step S3), it updates the driving history data stored in the driving history storage unit 122 (step S11). That is, the risk determination unit 105 updates the unsafe frequency shown in Fig. 33 among the driving history data.

[0262] After processing step S11, the personal threshold determination unit 123 updates the personal risk threshold of the driver of vehicle V based on the unsafe frequency of the updated driving history data (step S12). Furthermore, the personal threshold determination unit 123 updates the conditional personal risk threshold of the driver of vehicle V based on the accident history of the updated driving history data (step S13). The conditional personal risk threshold may be a conditional personal risk threshold of a driving scene, or a conditional personal risk threshold of a location. Furthermore, the conditional personal risk threshold may be a conditional personal risk threshold of a driving scene and a location.

[0263] FIG. 41 is a flowchart showing another example of the processing operation related to the driving history by the risk determining unit 105 and the individual threshold determining unit 123 in this modified example.

[0264] The processing operation relating to this driving history includes the processing of each step in the flowchart shown in FIG. 40, and further includes the processing of step S6.

[0265] For example, when the driving history data stored in the driving history storage unit 122 is updated (step S11), the personal threshold determination unit 123 determines whether updating the personal risk threshold is prohibited (step S6). For example, based on the accident history, the personal threshold determination unit 123 determines that updating the personal risk threshold is prohibited if a predetermined period has not elapsed since the date and time of the most recent accident. The predetermined period may be, for example, one year, or a period other than one year, and may depend on the interval between accidents. Conversely, when a predetermined period has elapsed since the date and time of the accident, the personal threshold determination unit 123 determines that updating the personal risk threshold is not prohibited. Then, when the personal threshold determination unit 123 determines that updating the personal risk threshold is not prohibited (No in step S6), it executes the processes of steps S12 and S13. On the other hand, when the personal threshold determination unit 123 determines that updating the personal risk threshold is prohibited (Yes in step S6), it terminates the processing operation related to the personal risk threshold without executing the processes of steps S12 and S13.

[0266] In the example of FIG. 41, the processing of step S6 is performed, so the personal risk threshold and the conditional personal risk threshold can be maintained at appropriate values. In other words, immediately after an accident occurs, that is, during a predetermined period from the date and time of the most recent accident, even a driver who normally drives unsafely tends to avoid unsafe driving. If the personal risk threshold and the conditional personal risk threshold are updated to large values due to a small number of unsafe driving incidents during such a predetermined period, the uploading of appropriate driving video data may be hindered. Therefore, the processing of step S6 prevents the uploading of appropriate driving video data from being hindered, and the personal risk threshold and the conditional personal risk threshold can be maintained at appropriate values.

[0267] In the example of FIG. 41 , the personal threshold determination unit 123 prohibits updating of the personal risk threshold if a predetermined period has not elapsed since the date and time of the most recent accident. However, updating of the personal risk threshold may be suppressed without prohibiting it. That is, the personal threshold determination unit 123 may limit the amount of change in the personal risk threshold. For example, if a predetermined period has elapsed since the date and time of the most recent accident caused by a person, the personal threshold determination unit 123 determines a value corresponding to the frequency of unsafety for that person as the personal risk threshold. On the other hand, if a predetermined period has not elapsed since the date and time of the most recent accident caused by a person, the personal threshold determination unit 123 determines a value corresponding to the frequency of unsafety for that person as the provisional personal risk threshold, as described above. Then, the personal threshold determination unit 123 calculates a difference value by multiplying the difference between the personal risk threshold immediately before updating and the provisional personal risk threshold by a weight smaller than 1. The personal threshold determination unit 123 determines the personal risk threshold for that person by adding the difference value to the personal risk threshold immediately before updating.

[0268] As described above, in this modification, the personal threshold determination unit 123 determines a conditional personal risk threshold for a driving scene. That is, the personal threshold determination unit 123 determines a personal risk threshold corresponding to each of a plurality of combinations. Each of the plurality of combinations includes one of a plurality of people, including the driver of the vehicle V, and one of a plurality of driving scenes. The risk determination unit 105 then selects, from the plurality of personal risk thresholds, a personal risk threshold associated with the combination of the driving scene detected by the driving scene detection unit 101 and the driver of the vehicle V as the conditional personal risk threshold for the driving scene. This allows for the selection and use of a personal risk threshold corresponding to the driver who engaged in unsafe driving and the driving scene in which the unsafe driving occurred, thereby further optimizing and streamlining the uploading of driving video data. For example, if a driver who normally drives safely tends to engage in unsafe driving in driving scenes involving backing into a parking space, a high personal risk threshold can be selected for that driver only when backing into a parking space. As a result, it becomes easier to upload driving video data only when backing into a parking space is performed.

[0269] Furthermore, in this modification, the personal threshold determination unit 123 determines a location-conditional personal risk threshold. That is, the personal threshold determination unit 123 determines a personal risk threshold corresponding to each of a plurality of combinations. Each of the plurality of combinations includes one of a plurality of people, including the driver of the vehicle V, and one of a plurality of locations. The risk determination unit 105 then selects, from the plurality of personal risk thresholds, a personal risk threshold associated with the combination of the location where the vehicle V is traveling in the driving scene detected by the driving scene detection unit 101 and the driver of the vehicle V, as the location-conditional personal risk threshold. This selects and uses a personal risk threshold corresponding to the driver who engaged in unsafe driving and the location in the driving scene where the unsafe driving occurred, thereby further optimizing and streamlining the uploading of driving video data. For example, if a driver who normally drives safely tends to engage in unsafe driving on highways, a high personal risk threshold can be selected for that driver only when the vehicle is traveling on the highway. As a result, it becomes easier to upload driving video data only when the vehicle is traveling on the highway.

[0270] Although the driving assistance system according to one or more aspects of the present disclosure has been described above based on several embodiments and modifications, the present disclosure is not limited to these embodiments and modifications. Various modifications conceivable by those skilled in the art may also be included in the present disclosure, as long as they do not deviate from the spirit of the present disclosure. Furthermore, combinations of multiple embodiments and modifications may also be included in the present disclosure.

[0271] For example, in each of the above embodiments and modifications, driving video data is transmitted from the vehicle V to the management server 200, but the timing of the transmission may be real time or not. Real time is the timing when unsafe driving is determined. On the other hand, non-real time may be, for example, the timing when the vehicle V returns to the parking lot of the person or company that owns the vehicle V.

[0272] Furthermore, at least one component included in each of the driving assistance systems 100, 100a, and 100b may be included not in the driving assistance system itself but in a cloud server such as the management server 200. There are several possible patterns in which the at least one component may be included in a cloud server.

[0273] In the first pattern, at least one component included in the cloud server is a plurality of components included in each of the driving assistance systems 100, 100a, and 100b, excluding the video storage unit 106 and the upload unit 107. In this case, the cloud server acquires outputs (excluding video image data) from sensors such as the GPS unit 11 mounted on the vehicle V at any time via the communication network Nt. Then, the risk determination unit 105 included in the cloud server requests the upload unit 107 mounted on the vehicle V to transmit driving video data corresponding to driving scenes when unsafe driving occurred. Then, the cloud server acquires the driving video data transmitted from the upload unit 107 of the vehicle V in response to the request via the communication network Nt.

[0274] In the second pattern, at least one component included in the cloud server is a risk assessment unit 105. In this case, multiple detection units, such as the driving scene detection unit 101 and the cognitive behavior detection unit 102, installed in the vehicle V transmit detection results to the risk assessment unit 105 via the communication network Nt. The risk assessment unit 105 included in the cloud server then acquires the transmitted detection results and determines the degree of risk associated with driving. Based on the determination result, the risk assessment unit 105 may request the upload unit 107 installed in the vehicle V to transmit driving video data corresponding to driving scenes where unsafe driving occurred. Furthermore, when all video data obtained by the vehicle V has been transmitted from the vehicle V to the cloud server, the risk assessment unit 105 may generate driving video data corresponding to driving scenes where unsafe driving occurred from all of the video data.

[0275] In the third pattern, all components included in each of the driving assistance systems 100, 100a, and 100b are included in a cloud server. In this case, outputs from all sensors, such as the GPS unit 11, mounted on the vehicle V are transmitted to the cloud server, and the cloud server performs all processing similar to that of the above-described embodiments and modifications. The outputs from all sensors may be transmitted to the cloud server via Wi-Fi when the vehicle V enters a Wi-Fi (registered trademark) communication area. That is, outputs from all sensors are accumulated in the vehicle V, and when the vehicle enters the above-described communication area, all accumulated outputs are transmitted to the cloud server. The Wi-Fi (registered trademark) communication area may be, for example, a parking lot for the owner or company of the vehicle V. In this third pattern, the system configuration included in the vehicle V can be significantly simplified.

[0276] 15 and 35, the processes of steps S300 and S500 may or may not be performed depending on the driving scene detected in step S1. In addition, the items of the driving action score scored in steps S200, S300, S400, and S500 in Fig. 8, 15, and 35 may be determined depending on the driving scene detected in step S1. The items of the driving action score include, for example, cognitive behavior to the right, visual inspection of the oncoming lane, vehicle speed, turn signals, smartphone operation, etc.

[0277] In each of the above embodiments, each component may be configured with dedicated hardware or circuitry, or may be realized by executing a software program appropriate for each component. Each component may be realized by a program execution unit such as a CPU (Central Processing Unit) or processor reading and executing a software program recorded on a recording medium such as a hard disk or semiconductor memory. Here, the program, which is software that realizes a system such as the driving assistance system of each of the above embodiments, causes a computer to execute each step included in the flowcharts of Figures 8 to 11, 15 to 30, 35, 36, 40, and 41.

[0278] The following cases are also included in this disclosure:

[0279] (1) Specifically, the above system may be a computer system consisting of a microprocessor, ROM (Read Only Memory), RAM (Random Access Memory), a hard disk unit, a display unit, a keyboard, a mouse, etc. A computer program is stored in the RAM or hard disk unit. The above system achieves its functions when the microprocessor operates in accordance with the computer program. Here, the computer program is composed of a combination of multiple instruction codes that indicate commands for the computer to achieve a predetermined function.

[0280] (2) Some or all of the components constituting the above system may be configured as a single system LSI (Large Scale Integration). A system LSI is an ultra-multifunctional LSI manufactured by integrating multiple components on a single chip, and specifically, is a computer system configured including a microprocessor, ROM, RAM, etc. A computer program is stored in the RAM. The system LSI achieves its functions when the microprocessor operates in accordance with the computer program.

[0281] (3) Some or all of the components constituting the above system may be configured as an IC card or a standalone module that can be attached to the system. The IC card or module is a computer system consisting of a microprocessor, ROM, RAM, etc. The IC card or module may include the above-mentioned ultra-multifunctional LSI. The IC card or module achieves its functions when the microprocessor operates according to a computer program. This IC card or module may be tamper-resistant.

[0282] (4) The present disclosure may be embodied as the methods described above, a computer program for implementing these methods on a computer, or a digital signal comprising the computer program.

[0283] The present disclosure may also be a computer program or a digital signal recorded on a computer-readable recording medium, such as a flexible disk, a hard disk, a CD (Compact Disc)-ROM, a DVD, a DVD-ROM, a DVD-RAM, a BD (Blu-ray (registered trademark) Disc), a semiconductor memory, etc. Alternatively, the present disclosure may be a digital signal recorded on such a recording medium.

[0284] The present disclosure may also be applied to transmitting a computer program or digital signal via a telecommunications line, a wireless or wired communication line, a network such as the Internet, data broadcasting, or the like.

[0285] Furthermore, the program or digital signal may be recorded on a recording medium and transferred, or the program or digital signal may be transferred via a network or the like, so that the program or digital signal may be implemented by another independent computer system. [Industrial Applicability]

[0286] The driving assistance system of the present disclosure has the effect of being able to appropriately determine the risk associated with driving a vehicle, and can be applied, for example, to a system installed in the vehicle. [Explanation of symbols]

[0287] 11 GPS unit 12 Driver monitor 13 Acceleration sensor 14 Drive Recorder 15 CAN 100, 100a, 100b Driver Assistance Systems 101 Driving scene detection unit 102 Cognitive Behavior Detection Unit 103 Vehicle behavior detection unit 104 Surrounding detection unit 105 Risk Assessment Department 106 Video storage section 107 Upload Section 111 Dangerous operation detection unit 112 Surrounding visual detection unit 113 Video Editing Department 121 Time condition update unit (time condition determination unit) 122 Operation history storage unit 123 Individual Threshold Determination Unit 200 Management Server 300 Management terminal 1000 Management System V vehicle

Claims

1. a driving scene detection unit that detects a driving scene of the vehicle in accordance with outputs from one or more sensors mounted on the vehicle; a cognitive behavior detection unit that detects cognitive behavior of a driver who drives the vehicle in accordance with outputs from one or more sensors mounted on the vehicle; a risk determination unit that determines a degree of risk to the driver's driving of the vehicle based on at least the detected driving scene and the detected cognitive behavior; A driving assistance system equipped with:

2. The risk determination unit determining the degree of risk by scoring the driver's driving of the vehicle; The driving score obtained by the scoring is smaller as the degree of risk increases. The driving assistance system according to claim 1 .

3. The risk determination unit further determining whether the driving score is less than a risk threshold; The driving assistance system further comprises: an upload unit that uploads driving video data obtained by capturing an image of the driving scene to a server when the driving score is determined to be less than the risk threshold; The driving assistance system according to claim 2 .

4. The driving assistance system further comprises: a video editing unit that generates the driving video data by selecting and editing a portion of one or more video data obtained by capturing images using a camera mounted on the vehicle, the portion corresponding to the driving scene detected by the driving scene detection unit; The driving assistance system according to claim 3 .

5. The driving assistance system further comprises: a surroundings detection unit that detects moving objects around the vehicle in response to outputs from one or more sensors mounted on the vehicle; The risk determination unit determining the degree of the risk based on a result of the detection of the moving object by the surroundings detection unit; The driving assistance system according to claim 1 .

6. The driving assistance system further comprises: a surroundings visual detection unit that detects visual inspection by the driver of one or more objects around the vehicle in response to outputs from one or more sensors mounted on the vehicle; The risk determination unit determining the degree of the risk based on the visual detection result by the surrounding visual detection unit; The driving assistance system according to claim 1 .

7. The driving assistance system further comprises: a vehicle behavior detection unit that detects the behavior of the vehicle; The risk determination unit determining the degree of risk based on the detection result of the behavior by the vehicle behavior detection unit; The driving assistance system according to claim 1 .

8. The driving assistance system further comprises: a dangerous operation detection unit that detects a predetermined device operation by the driver as a dangerous operation in accordance with outputs from one or more sensors mounted on the vehicle; The risk determination unit determining the degree of the risk based on a result of the detection of the dangerous operation by the dangerous operation detection unit; The driving assistance system according to claim 1 .

9. The risk determination unit When it is determined that the driving score is less than the risk threshold, selecting a personal risk threshold associated with the driver from a plurality of personal risk thresholds; determining whether the driving score is less than the selected personal risk threshold; The upload unit If the driving score is less than the risk threshold and less than the personal risk threshold, upload the driving video data to the server. The driving assistance system according to claim 3 .

10. The driving assistance system further comprises: and an individual threshold determination unit that determines the plurality of individual risk thresholds corresponding to the plurality of people based on the driving histories of the plurality of people. The driving assistance system according to claim 9.

11. The personal threshold determination unit determining the plurality of individual risk thresholds; For each of the plurality of combinations, determining an individual risk threshold corresponding to the combination; each of the plurality of combinations includes one of the plurality of people including the driver and one of a plurality of driving scenes; The risk determination unit In selecting the individual risk threshold, selecting, from the plurality of personal risk thresholds, a personal risk threshold associated with a combination of the driving scene detected by the driving scene detection unit and the driver; The driving assistance system according to claim 10.

12. The personal threshold determination unit determining the plurality of individual risk thresholds; For each of the plurality of combinations, determining an individual risk threshold corresponding to the combination; each of the plurality of combinations includes one of the plurality of people including the driver and one of a plurality of locations; The risk determination unit In selecting the individual risk threshold, selecting, from the plurality of personal risk thresholds, a personal risk threshold associated with a combination of a location where the vehicle is traveling and the driver in the driving scene detected by the driving scene detection unit; The driving assistance system according to claim 10.

13. The driving assistance system further comprises: a time condition determination unit that determines a specified time corresponding to each of the plurality of people based on the driving history of the plurality of people; The risk determination unit In determining the degree of risk, selecting a specified time associated with the driver from a plurality of specified times determined by the time condition determination unit; determining the degree of risk by comparing a time during which the cognitive behavior detected by the cognitive behavior detection unit was performed with the specified time; The driving assistance system according to claim 1 .

14. A driving assistance method implemented by a computer, comprising: Detecting a driving scene of the vehicle according to outputs from one or more sensors mounted on the vehicle; Detecting a cognitive behavior of a driver who drives the vehicle according to an output from one or more sensors mounted on the vehicle; determining a degree of risk to the driver's driving of the vehicle based on at least the detected driving scene and the detected cognitive behavior; Driving assistance methods.

15. Detecting a driving scene of the vehicle according to outputs from one or more sensors mounted on the vehicle; Detecting a cognitive behavior of a driver who drives the vehicle according to an output from one or more sensors mounted on the vehicle; determining a degree of risk to the driver's driving of the vehicle based on at least the detected driving scene and the detected cognitive behavior; A program that makes a computer do something.

Images