Driving assistance device, driving assistance method, drive recorder, driving assistance control program

Abstract

To analyze driving operation information and information derived by monitoring the surrounds of vehicle traveling in driving a vehicle and thereby provide advice that corresponds to the driving operation of each individual driver.SOLUTION: In a state recognition unit 36, each of information pieces about the traveling point of a vehicle 10 and the behavior of the vehicle 10 are sent to a risk event assessment unit 40 (transmission of essential information). In the state recognition unit 36, each of information pieces regarding the ambient environment of the vehicle 10 and the driving posture of a driver 24 are sent to the risk event assessment unit 40 (transmission of discretionary information) on the basis of a request from a risk event processing control unit 38. In the risk event assessment unit 40, the presence of risk during driving of the vehicle 10 is assessed. Advice is notified to the driver 24 on the basis of risk.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a driving assistance device, a driving assistance method, a drive recorder, and a driving assistance control program that provide advice regarding driving operations to a driver. [Background technology]

[0002] In recent years, when a vehicle is driven by a driver, acceleration and speed information is used to send advice to the driver to correct problematic driving behavior, such as sudden acceleration / deceleration caused by sudden acceleration and braking, and exceeding the legal speed limit, if the driver exhibits problematic driving behavior relative to predetermined safe driving operations.

[0003] The above advice may not be able to identify driving situations in which drivers are weak (for example, when turning left, a driver may be slow to notice pedestrians and may have to suddenly brake) or the characteristic driving operations of each individual driver, and may not be able to provide appropriate and specific corrective advice for such weak driving situations.

[0004] Here, Patent Document 1 describes a safe driving promotion system for a vehicle that can improve the driver's awareness of safe driving by giving mental encouragement to the driver.

[0005] More specifically, the in-vehicle device of Patent Document 1 judges the driver's driving condition from the safe driving threshold and economical driving threshold stored in the storage means, driving condition information acquired by the driving condition acquisition means, safe driving evaluation history information, and economical driving evaluation history information, and if the driver's driving condition is within the judgment thresholds stored in the storage means, a message praising the driver is notified via the notification means, and if the driver's driving condition is outside the judgment thresholds, advice is given to the driver via the notification means.

[0006] This praise (message) can provide mental encouragement to the driver, which will result in improved awareness of safe driving through voluntary safe driving by the driver, and is expected to reduce traffic accidents. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] JP 2013-191230 A Summary of the Invention [Problem to be solved by the invention]

[0008] However, only general advice is provided, rather than advice on how to drive depending on the scene, taking into account the driving conditions of each individual driver.

[0009] The present invention aims to provide a driving assistance device, a driving assistance method, a drive recorder, and a driving assistance control program that can provide advice tailored to the driving operation of each driver by analyzing driving operation information and information monitored around the vehicle when driving a vehicle. [Means for solving the problem]

[0010] The driving assistance device of the present invention has a situation recognition unit that acquires vehicle position information, vehicle behavior information, and image information of the vehicle's surroundings and interior as information sources, and recognizes the situation of the vehicle based on the information sources, a judgment unit that analyzes the recognition results by the situation recognition unit and determines whether or not a risk event has occurred while the vehicle is traveling, and a notification unit that, when the judgment unit determines that a risk event has occurred, notifies at least the driver of the vehicle of advice for resolving the risk event.

[0011] The driving assistance method of the present invention is characterized in that it acquires vehicle position information, vehicle behavior information, and image information of the vehicle's surroundings and interior as information sources, recognizes the situation of the vehicle based on the information sources, analyzes the recognition results, determines whether or not there is a risk event that may occur while the vehicle is traveling, and if it is determined that there is a risk event, notifies at least the driver of the vehicle of advice for resolving the risk event.

[0012] The drive recorder according to the present invention is mounted on a vehicle, captures images of the surrounding environment including at least the area in front of the vehicle, records the captured images within a predetermined capacity range while rewriting the oldest images first, and, when an emergency is detected, stores images for a predetermined period before and after the emergency. The drive recorder has the above-mentioned driving assistance device.

[0013] A driving assistance control program according to the present invention is characterized in that it causes a computer to operate as each part of the driving assistance device described above. Effect of the Invention

[0014] According to the present invention, when driving a vehicle, advice can be provided according to the driving operation of each driver by analyzing driving operation information and information obtained by monitoring the periphery of the vehicle. [Brief description of the drawings]

[0015] [Figure 1] 1 is a schematic diagram of a driving assistance system that assists in autonomous driving of a vehicle according to a first embodiment. [Diagram 2] 1 is a functional block diagram of a driving assistance device in a drive recorder according to a first embodiment, mainly showing driving assistance control classified by function. [Diagram 3] 3 is a functional block diagram showing a detailed configuration of a situation recognition unit according to the first embodiment. FIG. [Figure 4] 4 is a table showing classification of items processed by the behavior-specific scoring processor in the first embodiment. FIG. [Diagram 5] 4 is a table showing classification of items processed by the location-specific scoring processing unit in the first embodiment. FIG. [Figure 6] 3A is a flowchart showing a driving assistance control analysis routine in the driving assistance system according to the first embodiment, and FIG. 3B is a flowchart showing a driving assistance control notification routine in the driving assistance system according to the first embodiment. [Figure 7] FIG. 2 is a time-CPU resource characteristic diagram of the drive recorder according to the first embodiment. [Figure 8] FIG. 13 is a functional block diagram showing a detailed configuration of a situation recognition unit according to a modified example of the first embodiment (CAN information added). [Figure 9] FIG. 11 is a functional block diagram of a driving assistance device in a drive recorder according to a second embodiment, mainly showing driving assistance control classified by function. [Figure 10] 10 is a flowchart showing a driving assistance control analysis routine in a driving assistance system according to a second embodiment. [Figure 11] FIG. 11 is a functional block diagram of a driving assistance device in a drive recorder according to a third embodiment, mainly showing driving assistance control classified by function. [Figure 12] 13 is a flowchart showing a driving assistance control analysis routine in a driving assistance system according to a third embodiment. [Figure 13] FIG. 13 is a functional block diagram of a driving assistance device in a drive recorder according to a fourth embodiment, mainly showing driving assistance control classified by function. [Figure 14] 13 is a flowchart showing a driving assistance control analysis routine in a driving assistance system according to a fourth embodiment. [Figure 15] FIG. 13 is a functional block diagram of a driving assistance device in a drive recorder according to a fifth embodiment, mainly showing driving assistance control classified by function. [Figure 16] 13 is a flowchart showing a driving assistance control analysis routine in a driving assistance system according to a fifth embodiment. [Figure 17] FIG. 13 is a functional block diagram of a driving assistance device in a drive recorder according to a sixth embodiment, mainly showing driving assistance control classified by function. [Figure 18] 13 is a flowchart showing a driving assistance control analysis routine in a driving assistance system according to a sixth embodiment. [Figure 19] FIG. 13 is a schematic diagram of a driving assistance system that assists in autonomous driving of a vehicle according to a seventh embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] [First embodiment]

[0017] FIG. 1 is a plan view of a vehicle 10 to which a driving assistance system according to a first embodiment is applied.

[0018] The vehicle 10 is equipped with a vehicle control device 12 and a drive recorder 14. The drive recorder 14 has a function as a driving assistance device 14A (shown in a dotted frame in FIG. 1) according to the first embodiment in addition to the original functions of the drive recorder 14 (monitoring by photographing the outside and inside of the vehicle, and storing images in case of an emergency).

[0019] The vehicle control device 14 executes control including the drive system (engine control, etc.) and the electrical system when the vehicle 10 is running.

[0020] A gyro sensor (Gyro sensor) 16 (referred to as "Gyro" in FIG. 1), an acceleration sensor (G sensor) 18 (referred to as "G" in FIG. 1), and a GPS receiver 20 (referred to as "GPS" in FIG. 1) are connected to the drive recorder 14.

[0021] The gyro sensor 16 detects directional information regarding the traveling direction of the vehicle 10 , the acceleration sensor 18 detects the acceleration and deceleration of the vehicle 10 , and the GPS receiver 20 detects position information of the vehicle 10 .

[0022] In addition, the vehicle control device 12 is connected to a radar group 22 including a plurality of millimeter wave radars and a LIDAR.

[0023] The radar group 22 detects obstacles and the like ahead of the vehicle 10 .

[0024] The vehicle control device 12 uses the received information to execute control including the drive system and the electrical system, or notifies the driver 24 who is driving the vehicle of position information, a route to the destination, etc. Note that in FIG. 1, the driver 24 is shown getting off the vehicle.

[0025] The drive recorder 14 is equipped with a group of cameras (FIG. 1 shows a front monitoring camera 26A and a rear monitoring camera 26B as an example) that capture images of the surroundings of the vehicle 10. The drive recorder 14 of the first embodiment is also equipped with an interior monitoring camera 26C that captures images of the interior of the vehicle 10 (hereinafter, the front monitoring camera 26A, the rear monitoring camera 26B, and the interior monitoring camera 26C are collectively referred to as the "camera group 26"). The camera group 26 may also include a right monitoring camera 26D that captures the right side of the vehicle 10 and a left monitoring camera 26E that captures the left side of the vehicle 10.

[0026] A driving support device 14A according to the first embodiment is mounted on the driving recorder 14. In Fig. 1, the inside of the driving recorder 14 is illustrated by a dotted line, but since the original function of the driving recorder 14 and the common devices are used, the driving recorder 14 is referred to as the driving recorder 14 (14A) in Fig. 2 and thereafter, and is used to distinguish between the original function of the driving recorder 14 and the function of the driving support device 14A.

[0027] 2 is a functional block diagram of the driving support device 14A, which mainly shows driving support control classified by function in the drive recorder 14. Note that each block does not limit the hardware configuration of the driving support device 14A. If necessary, some or all of the blocks may be operated by a microcomputer as a driving support program.

[0028] 2, the drive recorder 14 includes a drive recorder main control unit 30. A group of cameras 26 (a front monitoring camera 26A, a rear monitoring camera 26B, and a vehicle interior monitoring camera 28) are connected to the drive recorder main control unit 30, and the group of cameras constantly capture images at least while the vehicle is driving.

[0029] (Original function of Drive Recorder 14)

[0030] In the drive recorder main control unit 30, as an original function of the drive recorder 14, the images captured by the cameras 26 are rewritten in order from the oldest image to the newest image within a predetermined capacity (for example, about one hour). In other words, the cameras 26 are constantly capturing images, and recording is performed at staggered intervals.

[0031] Here, an emergency time (for example, sudden deceleration due to sudden braking, etc.) is detected (or a signal is received from the vehicle control device 12), and images for a predetermined period (for example, 15 minutes before and after) before and after the emergency time are stored. Also, the driver 24 can check the through image (image being recorded) on the monitor unit 32.

[0032] The images taken before and after an emergency can be used to analyze factors that lead to an emergency, for example. The monitor unit 30 can also be used to play back recorded images.

[0033] (Driving assistance functions)

[0034] Here, the drive recorder 14 of the first embodiment serves as a driving support device 14A. The driving support device 14A analyzes the driving situation of the driver 24 based on information from the gyro sensor 16, the acceleration sensor 18, the GPS receiver 20, and the camera group 26, determines whether or not there is a dangerous behavior that is not suitable for driving (hereinafter, referred to as a dangerous event), and notifies the driver of advice for reducing the dangerous event for the next drive. Note that information may be acquired from a radar group 22 connected to the vehicle control device 12.

[0035] The driving support device 14A includes a sensor information acquisition unit 34, and acquires Gryo information (direction information), G information (acceleration / deceleration information), GPS information (position information), and radar group analysis information (obstacle information).

[0036] The sensor information acquisition unit 34 is connected to the situation recognition unit 36, and sends the acquired Gyro information, G information, GPS information, and obstacle detection information to the situation recognition unit 36.

[0037] The situation recognition unit 36 ​​is connected to the drive recorder main control unit 30. The situation recognition unit 36 ​​acquires shooting information by the group of cameras 26, which is used by the drive recorder main control unit 30.

[0038] This allows the situation recognition unit 36 ​​to collect the driving history, behavior, and situation outside and inside the vehicle 10 (image information) while the driver 24 is driving, and to analyze and recognize the situation while driving.

[0039] (Detailed configuration of the situation recognition unit 36)

[0040] FIG. 3 is a functional block diagram showing a detailed configuration of the situation recognition unit 36. As shown in FIG.

[0041] The situation recognition unit 36 ​​includes a travel point analysis unit 36A, a behavior analysis unit 36B, a surrounding environment recognition unit 36C, and a driver situation recognition unit 36D.

[0042] The travel location analysis unit 36A acquires GPS information from the sensor information acquisition unit 34, and analyzes the travel location of the vehicle 10, for example, whether the vehicle 10 is traveling at an intersection, a parking lot, indoors, or a narrow road.

[0043] The behavior analysis unit 36B acquires G information (acceleration information) and Gyro information (direction information) from the sensor information acquisition unit 34, and analyzes the behavior of the vehicle 10, such as right / left turns, starting, stopping, vehicle speed, steering angle, and the like.

[0044] The surrounding environment recognition unit 36C acquires outside-vehicle image information from the drive recorder main control unit 30 and, if necessary, obstacle information from the sensor information acquisition unit 34, and analyzes the surrounding environment of the vehicle 10, such as the positions and movements of people around the vehicle, preceding vehicles, oncoming vehicles, etc., white and yellow lines on the road, and the distance between vehicles, etc.

[0045] The driver situation recognition unit 36D acquires in-vehicle image information from the drive recorder main control unit 30, and analyzes the driving posture of the driver 24 driving the vehicle 10, such as line of sight, facial direction, degree of eye opening / closing, bone structure, objects carried, etc.

[0046] The situation recognition unit 36 ​​sends information on the travel location of the vehicle 10 analyzed by the travel location analysis unit 36A and the behavior of the vehicle 10 analyzed by the behavior analysis unit 36B to the risk event processing control unit 38 (sending required information).

[0047] In addition, based on a request from the risk event processing control unit 38, the situation recognition unit 36 ​​sends to the risk event processing control unit 38 information on the surrounding environment of the vehicle 10 analyzed by the surrounding environment recognition unit 36C and the driving posture of the driver 24 analyzed by the driver situation recognition unit 36D (sending optional information).

[0048] 2, the risk event processing control unit 38 includes a risk event determination unit 40. The risk event determination unit 40 collects the required information, which is the traveling location of the vehicle 10 and the behavior of the vehicle 10, and the optional information, which is the surrounding environment of the vehicle 10 and the driving posture of the driver 24, sent from the situation recognition unit 36.

[0049] The risk event determination unit 40 has a role of determining whether or not there is a risk while the vehicle 10 is being driven.

[0050] The risk while driving is, for example, the degree of danger of a situation that may hinder safe driving in a behavioral pattern determined by comparing with the Road Traffic Act, etc., and a location pattern determined by the driving location and driving state of the vehicle 10, and in the first embodiment, this is quantified (scored).

[0051] Here, the risk event determination section 40 executes the determination of the presence or absence of risk in two stages.

[0052] The risk event determination unit 40 determines whether there is a risk of a presumable item based on the required information (the traveling location of the vehicle 10 and the behavior of the vehicle 10) and the optional information (the surrounding environment of the vehicle 10 and the driving posture of the driver 24). The presumable items may be selected, for example, from pre-registered risks that have occurred in the past, and, for example, in relation to the processing capacity, it is sufficient to assume 100 to 200 risk items, but this number of items is not limited. Each determination item is determined by a scene search formula and a determination formula.

[0053] Scene search formulas include search conditions based on locations such as intersections without signals, intersections with signals, and parking lots, search conditions based on behavior such as right turns, left turns, going straight, and stopping, search conditions based on the surrounding environment such as preceding vehicles, pedestrians on the left shoulder, and bicycles stopped at intersections, and search conditions based on the driver's state such as the driver's line of sight, drowsiness, and smartphone conversations.Each can be used alone or in combination.For scenes that match the scene search formula, a judgment formula is used to judge whether there is a risk.

[0054] The formula for judgment is whether or not the vehicle is only braking suddenly (specifically, whether or not the acceleration exceeds a predetermined value (e.g., 0.35G)), or if the distance between the vehicle and the preceding vehicle is too close (specifically, the distance between the vehicles is less than 1.5 × (vehicle speed)). 2 / 9.8), etc.

[0055] The vehicle speed can be calculated using the integrated value of the G sensor information (corresponding to the first recognition situation), and the distance from the preceding vehicle can be calculated from the size of the preceding vehicle recognized in the image captured by the forward monitoring camera 26A and the hardware parameters of the camera group 26 (corresponding to the second recognition situation).

[0056] In the first stage of judgment in the risk event judgment unit 40, the presence or absence of a risk of a conceivable item is judged based on the essential information (the traveling location of the vehicle 10 and the behavior of the vehicle 10).

[0057] The first-stage scene search formulas include search conditions based on locations such as intersections without traffic lights and intersections with traffic lights, and search conditions based on behavior such as turning right, turning left, going straight, and stopping. Each of these can be used alone or in combination.

[0058] The first stage judgment formula includes cases where only sudden braking is judged (specifically, whether or not the acceleration exceeds a predetermined value (e.g., 0.35G)), failure to stop at a stop sign (specifically, when passing through an intersection with a stop sign and the vehicle speed is not 0km / h), etc.

[0059] The vehicle speed can be calculated using the integrated value of the G sensor information (corresponding to the first recognition situation), and the distance from the preceding vehicle can be calculated from the size of the preceding vehicle recognized in the image captured by the forward monitoring camera 26A and the hardware parameters of the camera group 26 (corresponding to the second recognition situation).

[0060] Here, if the presence or absence of risk can be determined using only the required information, there is no problem, but if there is insufficient information, it may be difficult to determine the risk. Therefore, if there is insufficient information, the image request unit 42 requests the drive recorder main control unit 30 to send information from the camera group 26 to the situation recognition unit 36.

[0061] The situation recognition unit 36 ​​acquires information from the camera group 26, and sends the surrounding environment of the vehicle 10 from the surrounding environment recognition unit 36C (see Figure 2) to the risk event determination unit 40 as arbitrary information, and also sends the driving posture of the driver 24 from the driver situation recognition unit 36D (see Figure 2) to the risk event determination unit 40.

[0062] In the second stage of judgment in the risk event judgment unit 40, the presence or absence of risk for possible items is judged based on required information (the driving location of the vehicle 10 and the behavior of the vehicle 10) and optional information (the surrounding environment of the vehicle 10 and the driving posture of the driver 24).

[0063] In the scene search formula, search for [Place search] Intersection without traffic lights, [Behavior] Going straight, [Surrounding environment] There is a vehicle ahead, [Driver] Looking ahead. Next, in the judgment formula, to check whether there was a short distance between vehicles in the corresponding scene, search for distance between vehicles < 1.5 × (own vehicle speed). 2 It is determined whether the condition in 9.8 is met. The distance between the vehicles is calculated based on the size of the preceding vehicle obtained from the optional information image and the hardware parameters. The vehicle speed is calculated based on the integral value of the G value obtained from the required information. These values ​​are input into the judgment formula; if the condition is met, it is determined that there is a risk, and if not, there is no risk.

[0064] In other words, by determining whether or not there is a risk based on sensor information, which contains a relatively small amount of information and does not take much time to analyze, rather than using image information from the beginning, which contains a relatively large amount of information and does not take much time to analyze, it is possible to realize the function of the driving assistance device 14A without compromising the original function of the drive recorder 14.

[0065] The risk event determination unit 40 is connected to the behavior-specific scoring processing unit 44 and the location-specific scoring processing unit 46, and the determination results of whether or not there is risk for each item are sent to the behavior-specific scoring processing unit 44 and the location-specific scoring processing unit 46.

[0066] It is not necessary to transmit the results of the judgment as to whether or not there is risk for all items to the behavior-specific scoring processing unit 44 and the location-specific scoring processing unit 46, and it is possible to select items based on the concepts of behavior and location. For example, out of a total of 160 items, 130 items may be transmitted to the behavior-specific scoring processing unit 44 and the location-specific scoring processing unit 46, respectively.

[0067] Fig. 4 is a table showing classification of items processed by the behavior-specific scoring processing unit 44. As shown in Fig. 4, behavior-specific items are classified into major items, medium items, and minor items.

[0068] Examples of major categories include classification according to the Road Traffic Act, classification based on safe driving obligations, and classification based on driver status.

[0069] Examples of medium-sized categories include classifications related to road signs, classifications related to driver behavior, and classifications related to the driver's physical condition.

[0070] Subcategories include classification by road signs, classification by driver behavior, classification by driver physical condition, etc.

[0071] Each major, medium, and minor item is assigned an ID to identify it, and the minor item ID aggregates the IDs. By looking at the minor item ID, it is possible to identify the type of major item, medium item, and minor item.

[0072] The behavior-specific scoring processing unit 44 scores each subitem based on a predetermined calculation formula, and evaluates the total value. For example, a calculation formula may be constructed in which a driver 24 who drives averagely is given 50 points out of a total of 100 points for behavior, and points are subtracted if there is risk and points are added if there is no risk for a long period of time.

[0073] Fig. 5 is a table showing classification of items processed by the location-specific scoring processing unit 46. As shown in Fig. 5, location-specific items are classified into vehicle location conditions and vehicle driving conditions, and a scene is set by combining each of them.

[0074] Examples of the vehicle location status include the presence or absence of facilities such as traffic lights, the shape of the road, and a specific location.

[0075] The vehicle driving conditions include, for example, the speed, acceleration / deceleration, location, weather, and the like.

[0076] There are a number of scenes, each of which is a combination of a vehicle location situation and a vehicle driving state, equal to the number of types of vehicle location situations x the number of types of vehicle driving states. Each scene is assigned an ID that identifies it, and the scene can be identified by looking at the ID.

[0077] The location-specific scoring processing unit 46 scores each scene based on a predetermined calculation formula, and evaluates the driver 24 by the total score. For example, a calculation formula may be constructed in which a driver 24 who drives averagely in a location is given 50 points out of a total of 100 points, and points are subtracted if there is risk and points are added if there is no risk for a long period of time.

[0078] As shown in FIG. 2, the behavior-specific scoring processing unit 44 and the location-specific scoring processing unit 46 are each connected to a priority determination unit 48.

[0079] The priority determination unit 48 extracts risks to be notified to the driver 24 based on the respective scoring results of the behavior-specific scoring processing unit 44 and the location-specific scoring processing unit 46, and sets priorities for the extracted risks. It is assumed that risks are notified to the driver 24, but if there is a notification limit, it is necessary to notify in order of highest priority, and sends notifications of a predetermined number of risks to the advice notification unit 50.

[0080] The advice notification unit 50 executes specific risk consulting based on the received risk, and notifies the mobile terminal device 24A or the like carried by the driver 24. The notification can be sent to the mobile terminal device 24A via the mobile terminal device 24A using a Web application or a native application. The notification may be sent to the driver 24 using a voice notification function of the drive recorder 14 or the vehicle 10, not limited to the mobile terminal device 24A, or may be sent to the driver 24 via the cockpit or display of the vehicle 10.

[0081] The operation of the first embodiment will be described below with reference to the flow chart of FIG.

[0082] FIG. 6(A) is a flowchart showing a driving assistance control analysis routine in the driving assistance system according to the first embodiment, and FIG. 6(B) is a flowchart showing a driving assistance control notification routine in the driving assistance system according to the first embodiment.

[0083] First, the driving support control analysis routine will be described with reference to FIG.

[0084] In step 100, it is determined whether it is time to analyze information. The information may be analyzed either periodically or irregularly.

[0085] When performing image information processing, it is preferable to set, for example, predetermined conditions. The predetermined conditions are set based on the availability of hardware (CPU resources) mounted on the drive recorder 14, as shown in FIG.

[0086] The horizontal axis of Fig. 7 is time, and the vertical axis is CPU resource (usage), with the usage rate at which the function locks set to 95%. In this state, the CPU resource fluctuates over time, and is kept below 5%.

[0087] Here, in the driving assistance device 14A according to the first embodiment, the time period for executing image processing is set to be equal to or less than a predetermined threshold value (90% in FIG. 7). In other words, image processing does not need to be executed in real time, and it is sufficient that image processing is executed and appropriate advice is generated by the time when the driver 14 is notified. The notification may be executed before the driver gets off the vehicle during the current driving or before the driver gets off the vehicle during the next driving. As a means for recognizing the communication time, the drive recorder 14 may detect the on / off state of the ignition switch and the locked / unlocked state of the doors, or the distance may be acquired from a short-range communication function between the mobile terminal device 24A carried by the driver 24 and the drive recorder 14.

[0088] The threshold value is not limited to 90% and may be determined so as not to impede the original function of the drive recorder 14.

[0089] 6A, if the determination is negative, the process proceeds to step 126 to determine whether or not to continue the analysis. If the determination is negative in step 126, the process returns to step 100.

[0090] If the determination in step 100 is affirmative, the process proceeds to step 1002 where various sensor information is acquired, then the process proceeds to step 104 where the traveling location is analyzed from the GPS information, and the process proceeds to step 106.

[0091] In step 106, the vehicle behavior is analyzed from the GPS information, G information, and Gyro information, and the process proceeds to step 108.

[0092] In step 108, the first stage, that is, the risk event is judged based on the required information, and the process proceeds to step 110 to judge whether or not sufficient information has been input to the judgment formula for the risk event.

[0093] If the determination in step 110 is negative, that is, if it is determined that there is insufficient information input to the risk event determination formula, the process proceeds from step 110 to step 112, where image information of the outside and inside of the vehicle is requested.

[0094] In the next step 114, image information is obtained from the drive recorder main control unit 30, and the process proceeds to step .

[0095] In step 116, the second stage, i.e., the risk event including optional information in addition to required information, is judged, and the process proceeds to step 118. If the judgment in step 110 is affirmative, it is determined that the second stage risk judgment taking into account optional information is unnecessary, and the process proceeds to step 118.

[0096] In step 118, a behavior-specific scoring process is executed for each determined risk event (see FIG. 4), and then the process proceeds to step 120, where a location-specific scoring process is executed for each determined risk event (see FIG. 5), and the process proceeds to step 122. Note that, although the processes are executed in the order of step 118 and step 120, the order may be step 120 → step 118, or step 118 and step 120 may be executed simultaneously (parallel processing).

[0097] In step 122, a priority for notifying the risk event is determined based on the result of the scoring process, and the process proceeds to step 124. In step 124, a result storage process is executed, and the process proceeds to step 126.

[0098] In step 126, it is determined whether or not the analysis should be continued. If the determination is affirmative, the process returns to step 100 as described above, whereas if the determination is negative, the routine ends.

[0099] do.

[0100] Next, the driving support control notification routine will be described with reference to FIG.

[0101] In step 150, it is determined whether or not it is time to notify a risk event. If the determination in step 150 is negative, this routine ends. If the determination in step 150 is positive, the routine proceeds to step 152.

[0102] In the first embodiment, the notification is made when the ignition key is turned off, and as shown in FIG. 1, the notification is made when the driver 24 leaves the vehicle 10.

[0103] The notification timing is not limited to the latter time of the vehicle 10, and may be, for example, the next time the ignition key is turned on. Also, other notification timings may be used, such as when the key or mobile terminal device 24A held by the driver 24 goes outside a predetermined range between the drive recorder 14 and the drive recorder 14 and then returns to the predetermined range, using a short-distance communication function. Also, even if there is no driver around the vehicle, notification may be made via a web app or native app at any timing.

[0104] In step 152, the risk event stored in step 124 in FIG. 6(A) is read out, and the process proceeds to step 154, in which notification processing is executed to the mobile terminal device 24A or the like carried by the driver 24, and the process proceeds to step 156.

[0105] In step 156, an archiving process of the notified risk event is executed, and this routine ends. It is not necessary to keep all risk events as history, and some may be selected.

[0106] (Modification of the first embodiment)

[0107] In the first embodiment, as shown in Fig. 3, the situation recognition unit 36 ​​recognizes the situation in the traveling point analysis unit 36A, the behavior analysis unit 36B, the surrounding environment recognition unit 36C, and the driver situation recognition unit 36D, but as shown in Fig. 8, a vehicle operation mode recognition unit 36E that recognizes the vehicle operation mode based on the CAN data of the vehicle 10 may be added. This allows the behavior of the vehicle (vehicle 10) to be recognized in more detail.

[0108] (Example 1 of the first embodiment)

[0109] Risk event processing is performed according to the steps (a) to (c) below.

[0110] (a) Based on GPS information and acceleration information, the vehicle (vehicle 10) behavior and location information are derived. For example, turning left at an intersection with a traffic light.

[0111] (b) Obtain the image processing results.

[0112] The image information from the camera group 26 (outside the vehicle) is analyzed to estimate the surrounding situation. For example, when the traffic light is green and it takes 5 seconds to turn left, a bicycle is on the left side of the vehicle (vehicle 10).

[0113] Image information from the camera group 26 (inside the vehicle) is analyzed to estimate the state of the driver 24. For example, the driver does not check for safety within 5 seconds before turning left.

[0114] (c) Risk events are determined using data from (a) and (b) for a certain period of time only before and after the left turn scene.

[0115] Case 1: (b) When turning left, the driver suddenly brakes.

[0116] This was a risky situation in which the driver was late in realizing the problem and slammed on the brakes, which constitutes unsafe operation.

[0117] Case 2: (b) When no braking is performed when turning left

[0118] This is a risky situation in which someone may have passed by without noticing, and constitutes obstruction to pedestrians.

[0119] Case 3: Based on the judgment results in (c), specific advice is provided.

[0120] In Case 1, the system notifies the driver, "Before entering an intersection, be sure to check for safety before turning left." In Case 2, the system notifies the driver, "Be sure to slow down to a speed where you can stop, and check for safety before turning left."

[0121] (Example 2 of the first embodiment)

[0122] After analyzing the location x vehicle (vehicle 10) behavior, additional images are analyzed (using optional information) only when it is necessary to analyze a risk scene that requires image analysis.

[0123] An example of an event that does not require image analysis (an example that can be analyzed using only the essential information) is an event of speeding while traveling straight on a main road.

[0124] In this case, the location is a main road, the vehicle (vehicle 10) is moving straight, and the vehicle (vehicle 10) speed is compared with the maximum speed limit of the main road to detect whether the vehicle is speeding; image information is not required.

[0125] According to the second example of the first embodiment, since the timestamp for performing the image analysis is specified, it is possible to reduce the CPU resources required for the image analysis.

[0126] (Example 3 of the first embodiment)

[0127] The detected risk events are classified by location and action and scored to select the scene with the lowest score and calculate the priority of the risk event for which advice should be given first.

[0128] According to Example 3 of the first embodiment, by identifying particularly problematic driving scenes among the many types of risky driving and notifying them as a priority, it becomes possible to quickly correct problematic driving scenes.

[0129] [Second embodiment]

[0130] A second embodiment of the present invention will be described below with reference to Figures 9 and 10. Note that the same components as those in the first embodiment are given the same reference numerals, and the description of the components will be omitted.

[0131] The feature of the second embodiment is that it adds a function to edit (add, delete, change, etc.) the judgment threshold and risk event judgment items when the judgment result in the risk event judgment unit 40 deviates from the previously expected judgment or when a new risk is discovered that does not fall under any of the previously set risk event items.

[0132] FIG. 9 is a functional block diagram of a driving assistance device 14A in a drive recorder 14 according to the second embodiment of the present invention, in which driving assistance control is mainly classified by function.

[0133] A user interface (UI) 52 is connected to the driving support device 14A. The UI 52 is, for example, installed in a management department that remotely supports the driving support device 14A, and is a device that can be operated by an operator. Note that the UI 52 may be the driver 24 other than the operator.

[0134] The risk event determination unit 40 sends information about the determination result to the UI 52. The UI 52 analyzes the determination result and determines whether or not the threshold value needs to be adjusted and whether or not the determination level needs to be adjusted.

[0135] When the operator determines that the threshold adjustment is necessary, he / she operates the UI 52 to send threshold adjustment information to the threshold adjustment unit 54. The threshold adjustment unit 54 is connected to the situation recognition unit 36, and adjusts the threshold value to be compared with the sensor information executed by the situation recognition unit 36.

[0136] Furthermore, when the operator determines that it is necessary to adjust the judgment level, he / she operates the UI 52 to send judgment level adjustment information to the judgment level adjustment unit 56. The judgment level adjustment unit 56 is connected to the risk event judgment unit 40, and adjusts the judgment criteria for risk event judgment, etc., which are executed by the risk event judgment unit 40.

[0137] The operation of the second embodiment will be described below.

[0138] Fig. 10 is a flowchart showing a driving support control analysis routine in the driving support system according to the second embodiment. Note that the same steps as those in the first embodiment (see Fig. 6(A)) are given the same step numbers, and the description of the processing is omitted.

[0139] After it is determined in step 110 that there is sufficient input information into the risk event judgment formula, or after the second stage risk event judgment is performed in step 116, the judgment result is sent to UI 52 in step 128.

[0140] In the next step 130, the UI 52 determines whether or not adjustment of the judgment result (adjustment of the threshold value, adjustment of the judgment level) is necessary. If it is determined in step 130 that adjustment is necessary, the process proceeds to step 132, where judgment level adjustment processing and threshold value adjustment processing are performed, and the process returns to step 108 to repeat the above steps.

[0141] If the determination in step 132 is negative, it is determined that no adjustment of the determination result is necessary, and the process proceeds to step 118.

[0142] According to the second embodiment, for example, if general traffic manners change due to changes in regulations or social conditions, it is possible to continue providing advanced advice by editing the threshold value for risk event judgment and adjusting the judgment level, without having to develop new ones.

[0143] (Example 3 of the second embodiment)

[0144] The output of the risk event determination unit 40 is presented to the operator via the UI 52.

[0145] If there is an error in the judgment, the threshold value and judgment level are adjusted according to the following cases.

[0146] Case 1: If it is determined that there is no risk

[0147] If the risk event judgment target matches (a combination of location, vehicle behavior, etc.), but does not match the actual degree of risk of the judgment threshold, the judgment threshold is adjusted (for example, the threshold of the vehicle distance or relative speed, etc.).

[0148] Case 2: When the risk event judgment target is incorrect and it is determined that the rule should be changed to another pre-designed rule

[0149] For example, if there is an error in the detection result of the location or the like that is to be the target of risk event determination, the following process 1 and process 2 are executed.

[0150] Process 1: Correct the detection results such as location so that the judgment result of the corresponding dangerous event is output.

[0151] Process 2: The judgment threshold values ​​are modified so that the output of the state recognition unit 36 ​​is consistent with the result of modification in process 1.

[0152] Case 3: The risk event judgment target is incorrect and does not fall under any of the pre-designed rules

[0153] Add a new risk event judgment rule.

[0154] The above response may be performed manually by the operator (driver 24), or may be processed according to a predetermined program, or may be determined by machine learning using AI based on a huge amount of past historical information (so-called big data).

[0155] [Third embodiment]

[0156] The third embodiment of the present invention will be described below with reference to Figures 11 and 12. Note that the same components as those in the first and second embodiments are given the same reference numerals, and the description of the components will be omitted.

[0157] The feature of the third embodiment is that it adds a function to edit (add, delete, change, etc.) the judgment threshold and risk event judgment items, etc. when the judgment result and judgment accuracy in the risk event judgment unit 40 deviates from the previously expected judgment, or when a new risk is discovered that does not fall under any of the previously set risk event items, etc.

[0158] FIG. 11 is a functional block diagram of a driving assistance device 14A in a drive recorder 14 according to the second embodiment of the present invention, in which driving assistance control is mainly classified by function.

[0159] A user interface (UI) 52 is connected to the driving support device 14A. The UI 52 is, for example, installed in a management department that remotely supports the driving support device 14A, and is a device that can be operated by an operator. Note that the UI 52 may be the driver 24 other than the operator.

[0160] The risk event determination unit 40 sends information on the determination result and the determination accuracy to the UI 52. The UI 52 analyzes the determination result and the determination accuracy, and determines whether or not the threshold value needs to be adjusted, and whether or not the determination level needs to be adjusted.

[0161] When the operator determines that the threshold adjustment is necessary, he / she operates the UI 52 to send threshold adjustment information to the threshold adjustment unit 54. The threshold adjustment unit 54 is connected to the situation recognition unit 36, and adjusts the threshold value to be compared with the sensor information executed by the situation recognition unit 36.

[0162] Furthermore, when the operator determines that it is necessary to adjust the judgment level, he / she operates the UI 52 to send judgment level adjustment information to the judgment level adjustment unit 56. The judgment level adjustment unit 56 is connected to the risk event judgment unit 40, and adjusts the judgment criteria for risk event judgment, etc., which are executed by the risk event judgment unit 40.

[0163] The operation of the third embodiment will be described below.

[0164] 12 is a flowchart showing a driving support control analysis routine in a driving support system according to a third embodiment. Note that the same steps as those in the first embodiment (see FIG. 6(A)) are given the same step numbers, and the description of the processes is omitted.

[0165] After it is determined in step 110 that there is sufficient input information into the risk event judgment formula, or after the second stage risk event judgment is performed in step 116, the judgment result and judgment accuracy are sent to UI 52 in step 128.

[0166] In the next step 130, the UI 52 determines whether or not adjustment of the judgment result and judgment accuracy (adjustment of the threshold value and judgment level) is necessary. If it is determined in step 130 that adjustment is necessary, the process proceeds to step 132, where judgment level adjustment processing and threshold value adjustment processing are performed, and the process returns to step 108 to repeat the above steps.

[0167] If the determination in step 132 is negative, it is determined that adjustment of the determination result and the determination accuracy is not necessary, and the process proceeds to step 118.

[0168] According to the third embodiment, for example, if general traffic manners change due to changes in regulations or social conditions, it is possible to continue providing advanced advice by editing the threshold value for risk event judgment and adjusting the judgment level, without having to develop new ones.

[0169] In addition, according to the third embodiment, the output is filtered and sorted according to the judgment accuracy of the risk event judgment results, so that when checking the event judgment results on UI52, the results can be checked efficiently by checking them in order of lowest judgment accuracy.

[0170] (Example 4 of the third embodiment)

[0171] In addition to the configuration of the third embodiment of the second embodiment, the determination accuracy is added to the output of the risk event determination unit 40 and is presented to the operator via the UI 52.

[0172] If there is an error in the judgment content, in addition to the third embodiment, the following measures will be taken.

[0173] Case 1: Providing advice

[0174] For risk events with a determination accuracy below a certain value, no advice is provided or the priority of the advice is lowered.

[0175] Case 2: Display of UI52

[0176] Risk events with a judgment accuracy equal to or greater than a given value are preferentially displayed on the UI 52.

[0177] [Fourth embodiment]

[0178] The fourth embodiment of the present invention will be described below with reference to Figures 13 and 14. Note that the same components as those in the first embodiment are given the same reference numerals, and the description of the components will be omitted.

[0179] Here, in the first embodiment, an image request unit 42 is provided in the risk event judgment unit 40, and when it is determined that image information is necessary in the judgment (first stage) based on the required information of the risk event judgment unit 40, image information (optional information) is requested from the drive recorder main control unit 30, and a judgment is made taking the image information into account (second stage).

[0180] In contrast to this, in the fourth embodiment, the risk event determination unit 40 does not distinguish between required information and optional information in its determination, but instead determines a risk event using sensor information and image information from the beginning.

[0181] FIG. 14 is a functional block diagram of a driving assistance device 14A in a drive recorder 14 according to the fourth embodiment of the present invention, in which driving assistance control is mainly classified by function.

[0182] The situation recognition unit 36 ​​sends to a risk event processing control unit 38 information on the travel location of the vehicle 10 analyzed by the travel location analysis unit 36A and the behavior of the vehicle 10 analyzed by the behavior analysis unit 36B.

[0183] In addition, the situation recognition unit 36 ​​acquires information from the camera group 26 and sends the surrounding environment of the vehicle 10 from the surrounding environment recognition unit 36C (see Figure 2) to the risk event determination unit 40, and also sends the driver behavior of the driver 24 (looking away, glancing away, absent-mindedness, smartphone operation, poor posture, failure to check for safety, etc.) from the driver situation recognition unit 36D (see Figure 2) to the risk event determination unit 40.

[0184] The risk event determination unit 40 determines whether or not there is a risk of a conceivable item based on the driving location of the vehicle 10 and the behavior of the vehicle 10, as well as the surrounding environment of the vehicle 10 and the driving posture of the driver 24.

[0185] The operation of the fourth embodiment will be described below.

[0186] 10 is a flowchart showing a driving support control analysis routine in a driving support system according to a fourth embodiment. Note that the same steps as those in the first embodiment (see FIG. 6(A)) are given the same step numbers, and the description of the processes is omitted.

[0187] In step 106, the vehicle behavior is analyzed from the GPS information, G information, and Gyro information, and the process proceeds to step 114.

[0188] In step 114, image information is obtained from the drive recorder main control unit 30, and the process proceeds to step .

[0189] In step 116, a risk event is determined based on the location where the vehicle 10 is traveling and the behavior of the vehicle 10, as well as the surrounding environment of the vehicle 10 and the driving posture of the driver 24, and the process proceeds to step 118.

[0190] According to the fourth embodiment, it is possible to determine risk events corresponding to various scenes acquired while the vehicle 10 is traveling, and to provide appropriate advice for resolving the risk events.

[0191] In addition, according to the fourth embodiment, by identifying particularly problematic driving scenes among the many types of risky driving and notifying them as a priority, it becomes possible to quickly correct problematic driving scenes.

[0192] [Fifth embodiment]

[0193] The fifth embodiment of the present invention will be described below with reference to Figures 15 and 16. Note that the same components as those in the first embodiment are given the same reference numerals, and the description of the components will be omitted.

[0194] Here, in the first embodiment, the risk event determination unit 40 is equipped with a behavior-specific scoring processing unit 44 and a location-specific scoring processing unit 46, and sends the determination results of the presence or absence of risk for each item to the behavior-specific scoring processing unit 44 and the location-specific scoring processing unit 46. The behavior-specific scoring processing unit 44 scores each sub-item based on a predetermined calculation formula and evaluates the total value, while the location-specific scoring processing unit 46 scores each scene based on a predetermined calculation formula and evaluates the total value.

[0195] In contrast to this, in the fifth embodiment, in the judgment by the risk event judgment unit 40, no distinction or priority is made to the results judged by the risk event judgment unit 40, and the risk event is judged based on so-called raw data.

[0196] FIG. 15 is a functional block diagram of a driving assistance device 14A in a drive recorder 14 according to a fifth embodiment of the present invention, in which driving assistance control is mainly classified by function.

[0197] The risk event determination section 40 performs a determination of the presence or absence of a risk in two stages.

[0198] In the first stage of judgment in the risk event judgment unit 40, the presence or absence of a risk of a conceivable item is judged based on the essential information (the traveling location of the vehicle 10 and the behavior of the vehicle 10).

[0199] Here, there is no problem if the presence or absence of risk can be determined using only the required information, but there are cases where a lack of information makes it difficult to determine the risk.

[0200] Therefore, if there is a lack of information, the image request section 42 requests the drive recorder main control section 30 to send information from the camera group 26 to the situation recognition section 36 .

[0201] The situation recognition unit 36 ​​acquires information from the camera group 26, and sends the surrounding environment of the vehicle 10 from the surrounding environment recognition unit 36C (see Figure 2) to the risk event determination unit 40 as optional information, and also sends the driver behavior of the driver 24 (looking away, glancing away, absent-mindedness, smartphone operation, poor posture, failure to check for safety, etc.) from the driver situation recognition unit 36D (see Figure 2) to the risk event determination unit 40.

[0202] In the second stage of judgment in the risk event judgment unit 40, the presence or absence of risk for possible items is judged based on required information (the driving location of the vehicle 10 and the behavior of the vehicle 10) and optional information (the surrounding environment of the vehicle 10 and the driving posture of the driver 24).

[0203] In other words, by determining whether or not there is a risk based on sensor information, which contains a relatively small amount of information and does not take much time to analyze, rather than using image information from the beginning, which contains a relatively large amount of information and does not take much time to analyze, it is possible to realize the function of the driving assistance device 14A without compromising the original function of the drive recorder 14.

[0204] The risk event determined by the risk event determination unit 40 is sent to the advice notification unit 50.

[0205] The advice notification unit 50 executes specific risk consulting based on the received risk, and notifies the mobile terminal device 24A carried by the driver 24 or the like.

[0206] The operation of the fifth embodiment will be described below.

[0207] 16 is a flowchart showing a driving support control analysis routine in a driving support system according to a fifth embodiment. Note that the same steps as those in the first embodiment (see FIG. 6(A)) are given the same step numbers, and the description of the processes is omitted.

[0208] If the determination in step 110 is negative, that is, if it is determined that there is insufficient information input to the risk event determination formula, the process proceeds from step 110 to step 112, where image information of the outside and inside of the vehicle is requested.

[0209] In the next step 114, image information is obtained from the drive recorder main control unit 30, and the process proceeds to step .

[0210] In step 116, the second stage, i.e., the risk event that includes optional information in addition to required information, is judged, and the process proceeds to step 124. If the judgment in step 110 is affirmative, it is determined that the second stage risk judgment that takes into account optional information is not necessary, and the process proceeds to step 124.

[0211] In step 124, a result storage process is executed, and the process proceeds to step 126.

[0212] In step 126, it is determined whether or not the analysis should be continued. If the determination is affirmative, the process returns to step 100 as described above, whereas if the determination is negative, the routine ends.

[0213] do.

[0214] According to the fifth embodiment, it is possible to determine risk events corresponding to various scenes acquired while the vehicle 10 is traveling, and to provide appropriate advice for resolving the risk events.

[0215] Furthermore, according to the fifth embodiment, since the timestamp for performing image analysis is specified, it is possible to reduce CPU resources required for image analysis.

[0216] [Sixth embodiment]

[0217] The sixth embodiment of the present invention will be described below with reference to Figures 17 and 18. Note that the same components as those in the first embodiment are given the same reference numerals, and the description of the components will be omitted.

[0218] (Difference 1 from the first embodiment)

[0219] Here, in the first embodiment, an image request unit 42 is provided in the risk event judgment unit 40, and when it is determined that image information is necessary in the judgment (first stage) based on the required information of the risk event judgment unit 40, image information (optional information) is requested from the drive recorder main control unit 30, and a judgment is made taking the image information into account (second stage).

[0220] In contrast to this, in the sixth embodiment, the risk event determination unit 40 does not distinguish between required information and optional information in its determination, but instead determines a risk event using sensor information and image information from the beginning.

[0221] (Difference 2 from the first embodiment)

[0222] In the first embodiment, the risk event determination unit 40 is equipped with a behavior-specific scoring processing unit 44 and a location-specific scoring processing unit 46, and sends the determination results of the presence or absence of risk for each item to the behavior-specific scoring processing unit 44 and the location-specific scoring processing unit 46. The behavior-specific scoring processing unit 44 scores each sub-item based on a predetermined calculation formula and evaluates the total value, while the location-specific scoring processing unit 46 scores each scene based on a predetermined calculation formula and evaluates the total value.

[0223] In contrast to this, in the sixth embodiment, in the judgment by the risk event judgment unit 40, no distinction or priority is made to the results judged by the risk event judgment unit 40, and the risk event is judged based on so-called raw data.

[0224] FIG. 17 is a functional block diagram of a driving assistance device 14A in a drive recorder 14 according to the fourth embodiment of the present invention, in which driving assistance control is mainly classified by function.

[0225] The situation recognition unit 36 ​​sends to a risk event processing control unit 38 information on the travel location of the vehicle 10 analyzed by the travel location analysis unit 36A and the behavior of the vehicle 10 analyzed by the behavior analysis unit 36B.

[0226] In addition, the situation recognition unit 36 ​​acquires information from the camera group 26 and sends the surrounding environment of the vehicle 10 from the surrounding environment recognition unit 36C (see Figure 2) to the risk event determination unit 40, and also sends the driver behavior of the driver 24 (looking away, glancing away, absent-mindedness, smartphone operation, poor posture, failure to check for safety, etc.) from the driver situation recognition unit 36D (see Figure 2) to the risk event determination unit 40.

[0227] The risk event determination unit 40 determines whether or not there is a risk of a conceivable item based on the driving location of the vehicle 10 and the behavior of the vehicle 10, as well as the surrounding environment of the vehicle 10 and the driving posture of the driver 24.

[0228] The risk event determined by the risk event determination unit 40 is sent to the advice notification unit 50.

[0229] The advice notification unit 50 executes specific risk consulting based on the received risk, and notifies the mobile terminal device 24A carried by the driver 24 or the like.

[0230] The operation of the sixth embodiment will be described below.

[0231] 18 is a flowchart showing a driving support control analysis routine in a driving support system according to a sixth embodiment. Note that the same steps as those in the first embodiment (see FIG. 6(A)) are given the same step numbers, and the description of the processes is omitted.

[0232] In step 106, the vehicle behavior is analyzed from the GPS information, G information, and Gyro information, and the process proceeds to step 114.

[0233] In step 114, image information is obtained from the drive recorder main control unit 30, and the process proceeds to step .

[0234] In step 116, a risk event is determined based on the location where the vehicle 10 is traveling and the behavior of the vehicle 10, as well as the surrounding environment of the vehicle 10 and the driving posture of the driver 24, and the process proceeds to step 124.

[0235] In step 124, a result storage process is executed, and the process proceeds to step 126.

[0236] In step 126, it is determined whether or not the analysis should be continued. If the determination is affirmative, the process returns to step 100 as described above, whereas if the determination is negative, the routine ends.

[0237] do.

[0238] According to the sixth embodiment, it is possible to determine risk events corresponding to various scenes acquired while the vehicle 10 is traveling, and to provide appropriate advice for resolving the risk events.

[0239] Furthermore, in the judgment by the risk event judgment unit 40, it is possible to judge a risk event using the sensor information and image information from the beginning, without distinguishing between required information and optional information.

[0240] Furthermore, in the judgment by the risk event judgment unit 40, no distinction or priority is made to the results judged by the risk event judgment unit 40, and the risk event is judged based on so-called raw data.

[0241] In other words, according to the sixth embodiment, risk events corresponding to the driving scene of the vehicle 10 can be accurately determined with necessary and sufficient information without performing complex control, and the configuration can be said to be useful as a low-cost version due to the simplified device configuration.

[0242] [Seventh embodiment]

[0243] The seventh embodiment of the present invention will be described below with reference to Fig. 19. Note that the same components as those in the first embodiment are given the same reference numerals, and the description of the components will be omitted.

[0244] As shown in FIG. 19, in the seventh embodiment, a gyro sensor 16 (referred to as "Gyro" in FIG. 19), an acceleration sensor (G sensor) 18 (referred to as "G" in FIG. 19), and a GPS receiver 20 (referred to as "GPS" in FIG. 19) are connected to the vehicle control device 12.

[0245] In the seventh embodiment, information required by the travel point analysis unit 36A and the behavior analysis unit 36B, i.e., information from the gyro sensor 16, the acceleration sensor 18, and the GPS receiver 20, is acquired from the vehicle control device 12. In this case, obstacle information ahead of the vehicle 10 may also be acquired from the vehicle control device 12 as information from the radar group 22. In particular, at night or in rainy weather, the amount of information from the camera group 26 is small, and obstacles that are difficult to recognize can be recognized reliably.

[0246] According to the seventh embodiment, if the vehicle 10 has any of the existing sensors (gyro sensor 16, acceleration sensor 18, and GPS receiver 20), there is no need for the drive recorder 14 to have that function, and the necessary sensor information can be obtained from the vehicle control device 12, thereby simplifying the configuration.

[0247] In each embodiment (including modified examples) of the present invention, all controls (information acquisition, situation recognition, risk judgment, etc.) are performed in the drive recorder 14, but the gyro sensor 16, acceleration sensor 18, and GPS receiver 20, as well as the camera group 26 and drive recorder main control unit 30 may be configured as hardware mounted on the vehicle (drive recorder 14), and an output signal from the hardware may be transmitted to the cloud, and processing in place of the risk event processing unit 40, priority determination unit 48, advice notification unit 50, etc. may be performed on the cloud. In other words, there is no restriction on the arrangement between the hardware mounted on the vehicle 10 and the functions that process on the cloud. [Explanation of symbols]

[0248] 10 vehicle, 12 vehicle control device, 14 drive recorder, 14A driving assistance device, 16 gyro sensor (Gyro sensor), 18 acceleration sensor (G sensor), 20 GPS receiver, 22 radar group, 24 driver, 26A forward monitoring camera, 26B rearward monitoring camera, 26C vehicle interior monitoring camera, 26 camera group 26D rightward monitoring camera, 26E leftward monitoring camera, 30 drive recorder main control unit, 32 monitor unit, 34 sensor information acquisition unit, 36 situation recognition unit, 36A driving point analysis unit, 36B behavior analysis unit, 36C surrounding environment recognition unit, 36D driver situation recognition unit, 38 risk event processing control unit, 40 risk event determination unit, 42 image request unit 44 behavior-specific scoring processing unit 44, 46 location-specific scoring processing unit, 48 priority determination unit 50 advice notification unit

Claims

1. a situation recognition unit that acquires information on the position of a vehicle, information on the behavior of the vehicle, and image information on the surroundings and interior of the vehicle as information sources, and recognizes a situation of the vehicle based on the information sources; a determination unit (38, 40) that analyzes the recognition result by the situation recognition unit and determines whether or not there is a risk event that may occur while the vehicle is traveling; A notification unit (50) that notifies at least a driver who drives the vehicle of advice for eliminating the risk event when the determination unit determines that the risk event exists, The determination unit determines the presence or absence of the risk event by using a scene search formula for searching whether or not a scene matches at least one of a search condition based on a location where the vehicle is located, a search condition based on a behavior of the vehicle, a search condition based on a surrounding environment of the vehicle, and a search condition based on a state of a driver of the vehicle, and a determination formula for determining the presence or absence of the risk event for a scene searched by the scene search formula. Driving assistance device.

2. The situation recognition unit, The information sources are classified into mandatory information sources based on location information of the vehicle and behavior information of the vehicle, and optional information sources based on image information of the surroundings and interior of the vehicle; According to the recognition items used in the determination unit to determine the risk event, a first recognition function for recognizing a driving situation based on the essential information source as a first stage of situation recognition and outputting the recognition result to the determination unit; A second recognition function that recognizes the driving situation based on the required information source and the optional information source as a second-stage situation recognition when the first-stage output is insufficient as a recognition item used for determining a risk event, and outputs the second-stage situation recognition to the determination unit; The driving assistance device according to claim 1 .

3. 3. A driving assistance device as described in claim 1 or claim 2, wherein the risk events determined by the determination unit are classified into behavioral risk events related to at least violations of the Road Traffic Act and location-type risk events based on scenes determined at least by the driving location and the driving behavior of the vehicle, with possible overlapping, and a scoring process is performed for each of the behavioral risk events and the location-type risk events according to the importance of countermeasures, and a priority for notification by the notification unit is determined based on the results of the scoring process.

4. The driving assistance device according to any one of claims 1 to 3, further comprising a monitoring unit that acquires a judgment result from the judgment unit, monitors whether the judgment result from the judgment unit is good or bad, and updates a judgment criterion in the judgment unit based on the monitoring result.

5. The driving support device according to claim 4 , wherein the monitoring unit acquires determination accuracy information from the determination unit, and monitors the quality of the determination results in ascending order of determination accuracy.

6. The driving support device according to any one of claims 1 to 5, wherein information about the vehicle's running is added as information source acquired by the situation recognition unit.

7. Acquire information on the position of the vehicle, information on the behavior of the vehicle, and image information on the surroundings and interior of the vehicle as information sources, and recognize the situation of the vehicle based on the information sources; Analyzing the recognition result to determine whether or not there is a risk event that may occur while the vehicle is traveling; When it is determined that the risk event occurs, advice for eliminating the risk event is notified to at least a driver who drives the vehicle. Driving assistance methods.

8. a main control unit mounted on a vehicle, which captures an image of the surrounding environment including at least an area in front of the vehicle, records the captured images within a predetermined capacity while rewriting the oldest images first, and when an emergency period is detected, stores images for a predetermined period before and after the emergency period; A driving assistance device according to any one of claims 1 to 7, A drive recorder having the above configuration.

9. Computer, Operate the driving support device as each part of the driving support device according to any one of claims 1 to 6. Driver assistance control program.

Images