Driving assist device

Abstract

To provide a drive assist device which can determine deterioration of concentration of a driver on driving with high accuracy.SOLUTION: An object detection unit 2 detects risk objects 13, 15 at the risk of causing an accident on a route 11 of a self vehicle 10 and in the surroundings of the route 11. An avoidance driving detection unit 3 detects characteristics of avoidance driving for performing a driving operation to avoid the risk of the accident by a driver when the self vehicle 10 approaches and passes through the risk objects 13, 15. A concentration determination unit 4 determines concentration of the driver on the driving on the basis of the characteristics of the avoidance driving of the driver detected by the avoidance driving detection unit 3. A notification unit 5 performs notification for prompting safe driving when the concentration determination unit 4 determines that the concentration of the driver on the driving is deteriorated.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present disclosure relates to a driving assistance device that assists a driver in driving a vehicle. [Background technology]

[0002] As an example of a driving support device, Patent Document 1 describes a warning system that detects the driver's level of consciousness and issues a warning to the driver depending on the situation. This driving support device processes images of the driver's face captured by an in-vehicle camera or the like, detects changes in the opening and closing time of the eyelids when blinking or the direction of the driver's gaze, and detects the driver's level of consciousness based on the state of inattentiveness. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2003-317197 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, as a result of detailed investigation by the inventors, it was found that the driving assistance device described in Patent Document 1 has the following problems. (i) Image processing using an in-vehicle camera may not be able to reliably detect the object the driver is looking at. In particular, the further away the driver looks, the more difficult it becomes to determine what the driver is looking at. (ii) Even if there is an object (hereinafter referred to as a "hazardous object") that may cause an accident in the direction the driver is looking, the driver may simply gaze blankly in that direction and not recognize it as a hazardous object. (iii) Even if a driver is aware of a dangerous object in the direction of his or her line of sight, the driver may become anxious about a problem and thus lose his or her sense of safety in order to avoid the dangerous object.

[0005] Due to the problems (i), (ii), and (iii) described above, the driving assistance device described in Patent Document 1 may not be able to accurately detect the driver's level of awareness, i.e., a decrease in the driver's concentration on driving, and may not be able to effectively provide notifications to encourage safe driving.

[0006] In view of the above, an object of the present disclosure is to provide a driving assistance device that can determine a decrease in a driver's concentration on driving with high accuracy. [Means for solving the problem]

[0007] According to one aspect of the present disclosure, a driving assistance device includes: an object detection unit (2) for detecting dangerous objects (13, 15) that may cause an accident on a path (11) of a host vehicle (10) and around the path (12, 16); an avoidance driving detection unit (3) for detecting characteristics of avoidance driving in which a driver performs a driving operation to avoid the risk of an accident when the host vehicle approaches and passes a dangerous object detected by the object detection unit; a concentration level determination unit (4) that determines a concentration level of the driver on driving based on the characteristics of the driver's avoidance driving detected by the avoidance driving detection unit; The vehicle is provided with a notification unit (5) that issues a notification to encourage safe driving when the concentration level determination unit determines that the driver's concentration level on driving has decreased.

[0008] According to this, the driving support device of the present disclosure judges the driver's concentration level on driving based on the characteristics of avoidance driving in which the driver avoids an object (i.e., a dangerous object) that may cause an accident to the vehicle by driving operations. Therefore, compared with a method of judging the driver's concentration level on driving based only on the driver's facial expression or line of sight as in the above-mentioned Patent Document 1, the driving support device of the present disclosure can judge the decrease in the driver's concentration level on driving with high accuracy. For example, the driving support device of the present disclosure can judge the decrease in the driver's concentration level on driving even if the driver sees a dangerous object but just stares blankly and does not recognize it as a dangerous object. Also, for example, the driving support device of the present disclosure can judge the decrease in the driver's concentration level on driving even if the driver recognizes it as a dangerous object but has a decreased safety consciousness due to impatience or the like. Therefore, the driving support device of the present disclosure can effectively encourage the driver to drive safely by reducing the annoyance felt by the driver in response to the notification as being acceptable to the driver.

[0009] The reference symbols in parentheses attached to each component indicate an example of the correspondence between the component and specific components described in the embodiments described below. [Brief description of the drawings]

[0010] [Figure 1] 1 is a block diagram showing a schematic configuration of a driving assistance device according to a first embodiment. [Diagram 2] 4 is an explanatory diagram illustrating a state in which a concentration level decrease determination is executed when the host vehicle approaches and passes a vehicle parked on the roadside in the driving assistance device according to the first embodiment. FIG. [Diagram 3] FIG. 3 is an explanatory diagram illustrating how a notification is executed when the host vehicle passes a vehicle parked on the roadside, following FIG. 2; [Figure 4] 4 is a flowchart showing an algorithm executed by the driving assistance device according to the first embodiment. [Diagram 5] 5 is a flowchart showing the process of S200 in FIG. 4. [Figure 6] 5 is a flowchart showing the process of S300 in FIG. 4. [Figure 7] FIG. 11 is a block diagram showing a schematic configuration of a driving assistance device according to a second embodiment. [Figure 8] FIG. 11 is an explanatory diagram illustrating how the characteristics of the driver's avoidance driving are learned when the host vehicle approaches and passes a vehicle parked on the shoulder in the driving assistance device according to the second embodiment. [Figure 9] FIG. 9 is an explanatory diagram illustrating how a determination of a decrease in concentration level and a notification are performed when the vehicle approaches and passes another vehicle parked on the roadside in a scene different from that in FIG. 8. [Figure 10] 10 is a flowchart showing an algorithm executed by a driving assistance device according to a second embodiment. [Figure 11] 11 is a flowchart showing the process of S200 in FIG. 10. [Figure 12] 11 is a flowchart showing the process of S300 in FIG. 10. [Figure 13] 11 is a flowchart showing the process of S500 in FIG. 10. [Figure 14] FIG. 13 is an explanatory diagram illustrating how the characteristics of the driver's avoidance driving are learned when the host vehicle approaches a vehicle parked on the shoulder in the driving assistance device according to the third embodiment. [Figure 15] FIG. 15 is an explanatory diagram illustrating how a determination of a decrease in concentration level and a notification are made when the vehicle approaches another vehicle parked on the roadside in a scene different from that in FIG. 14. [Figure 16] 13 is a flowchart showing an algorithm executed by a driving assistance device according to a third embodiment. [Figure 17] 17 is a flowchart showing the process of S200 in FIG. 16. [Figure 18] 17 is a flowchart showing the process of S300 in FIG. 16. [Figure 19] 17 is a flowchart showing the process of S500 in FIG. 16. [Figure 20]FIG. 13 is an explanatory diagram illustrating how the driving support device according to the fourth embodiment learns characteristics of the driver's avoidance driving when a crossing pedestrian is present around the path of the host vehicle. [Figure 21] FIG. 21 is an explanatory diagram illustrating how a determination of a decrease in concentration level and a notification are executed when there is another crossing pedestrian in the vicinity of the path of the host vehicle in a scene different from that in FIG. 20. [Figure 22] 13 is a flowchart showing an algorithm executed by a driving assistance device according to a fourth embodiment. [Figure 23] 23 is a flowchart showing the process of S200 in FIG. 22. [Figure 24] 23 is a flowchart showing the process of S300 in FIG. 22. [Diagram 25] 23 is a flowchart showing the process of S500 in FIG. 22. [Figure 26] 13 is an explanatory diagram illustrating a state in which a concentration level decrease determination is executed when the host vehicle approaches and passes a vehicle parked on the shoulder of the road in a driving assistance device according to a fifth embodiment. FIG. [Figure 27] FIG. 27 is an explanatory diagram illustrating how a notification is given when the host vehicle passes a vehicle parked on the shoulder of the road, following FIG. 26 . [Figure 28] 13 is a flowchart showing an algorithm executed by a driving assistance device according to a fifth embodiment. [Figure 29] 30 is a flowchart showing the process of S200 in FIG. 28. [Diagram 30] 30 is a flowchart showing the process of S300 in FIG. 28. [Diagram 31] FIG. 13 is an explanatory diagram illustrating how the driving assistance device according to the sixth embodiment learns characteristics of the driver's avoidance driving when a crossing pedestrian is present around the path of the host vehicle. [Diagram 32] FIG. 32 is an explanatory diagram illustrating how a determination of a decrease in concentration level and a notification are performed when there is another crossing pedestrian in the vicinity of the path of the host vehicle in a scene different from that in FIG. 31 . [Diagram 33]33 is an explanatory diagram illustrating how a notification is executed when there is another crossing pedestrian in the vicinity of the path of the host vehicle in a scene different from that in FIG. 31 and FIG. 32. FIG. [Diagram 34] 33 is an explanatory diagram illustrating how a notification is given when the host vehicle approaches a vehicle parked on the shoulder of the road in a scene different from that in FIG. 31 and FIG. 32. [Diagram 35] 13 is a flowchart showing an algorithm executed by a driving assistance device according to a sixth embodiment. [Diagram 36] FIG. 36 is a state transition diagram for explaining the process of S300 in FIG. 35. [Figure 37] FIG. 23 is an explanatory diagram illustrating how the characteristics of the driver's avoidance driving are learned when the host vehicle approaches and passes a vehicle parked on the shoulder in the driving assistance device according to the seventh embodiment. [Figure 38] FIG. 38 is an explanatory diagram illustrating how a determination of a decrease in concentration level and a notification are made when the vehicle approaches and passes another vehicle parked on the roadside in a scene different from that in FIG. 37. [Figure 39] FIG. 39 is an explanatory diagram illustrating how a notification is executed when a crossing pedestrian is present around the path of the host vehicle in a scene different from that in FIG. 37 and FIG. 38. [Diagram 40] FIG. 39 is an explanatory diagram illustrating how a notification is given when the host vehicle approaches another vehicle parked on the roadside in a scene different from that in FIGS. 37 and 38. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In the following embodiments, the same or equivalent parts are denoted by the same reference numerals, and the description thereof will be omitted.

[0012] (First embodiment) A first embodiment will be described. The driving support device of the first embodiment issues a notification to encourage safe driving when the driver's concentration on driving is decreasing. In addition, this driving support device can reduce the annoyance felt by the driver in response to the notification by determining the decrease in the driver's concentration on driving with high accuracy, and can effectively encourage the driver to drive safely.

[0013] The driving assistance device is composed of a microcomputer including a processor and memories such as ROM and RAM, and an electronic control device having its peripheral circuits. The driving assistance device is connected to an on-board camera, various sensors, and other electronic control devices (not shown) mounted on the vehicle through an in-vehicle LAN or a wire harness using CAN communication, for example. CAN is an abbreviation for Controller Area Network, and LAN is an abbreviation for Local Area Network.

[0014] As shown in FIG. 1, the driving assistance device 1 functions as an object detection unit 2, an evasive driving detection unit 3, a concentration determination unit 4, an alarm unit 5, etc., by a processor executing a program stored in a memory, and performs various control processes.

[0015] The object detection unit 2 detects objects (hereinafter referred to as "hazardous objects") that may cause accidents on the path of the vehicle (hereinafter simply referred to as "vehicle") to which the driving assistance device 1 is applied and around the vehicle, for example, by analyzing image data transmitted from an on-board camera. The surroundings of the vehicle's path include, for example, the shoulder, sidewalk, sidewalk, bicycle lane, etc. The dangerous objects include, for example, the shoulder of the road around the vehicle's path or a vehicle parked on the shoulder that protrudes from the shoulder into the vehicle's path, a pedestrian crossing the vehicle's path or a pedestrian crossing the vehicle's path. The object detection unit 2 transmits information about the detected dangerous objects to the avoidance driving detection unit 3, the concentration determination unit 4, the notification unit 5, etc.

[0016] The avoidance driving detection unit 3 detects the characteristics of avoidance driving (hereinafter referred to as "characteristics of the driver's avoidance driving"), which is a driving operation performed by the driver to avoid the risk of an accident when the vehicle approaches and passes a dangerous object.

[0017] One of the characteristics of the driver's avoidance driving is, for example, the distance between the host vehicle and a dangerous object when the host vehicle passes beside the object (hereinafter referred to as "lateral distance Lm"). When the dangerous object is a vehicle parked on the roadside, the lateral distance Lm may be the distance between the center of the host vehicle and the outer wall surface of the vehicle parked on the roadside that is closer to the host vehicle, or may be the distance between the outer wall surface of the vehicle parked on the roadside that is closer to the host vehicle and the outer wall surface of the vehicle parked on the roadside that is closer to the host vehicle.

[0018] One of the characteristics of the driver's avoidance driving is, for example, the time to collision when the driver starts braking in front of a dangerous object. Hereinafter, the time to collision is referred to as TTC. TTC is an abbreviation for Time To Collision. TTC is a value obtained by dividing the distance between the position of the host vehicle and the dangerous object by the current vehicle speed of the host vehicle. The distance between the host vehicle and the dangerous object is detected by, for example, a millimeter wave radar or LiDAR. LiDAR is an abbreviation for Light Detection and Ranging or Laser Imaging Detection and Ranging. The vehicle speed of the host vehicle is detected by a vehicle speed sensor.

[0019] The avoidance driving detection unit 3 conveys the characteristics of the driver's avoidance driving to the concentration level determination unit 4.

[0020] The concentration level determination unit 4 determines the driver's concentration level on driving based on the characteristics of the driver's avoidance driving. In the first embodiment, the concentration level determination unit 4 determines that the driver's concentration level on driving has decreased when the lateral distance Lm is smaller than a preset distance threshold. The distance threshold is set in advance through experiments or the like and stored in the memory of the driving support device 1. The distance threshold may be set to a greater distance as the vehicle speed of the vehicle increases, or may be a function related to the vehicle speed of the vehicle, or may have a different value depending on the presence or absence of a parallel vehicle or an oncoming vehicle in an adjacent lane.

[0021] Furthermore, when the TTC is smaller than a preset time threshold, the concentration level determination unit 4 determines that the driver's concentration level on driving is decreasing. The time threshold is preset through experiments or the like, and is stored in the memory of the driving support device 1. The time threshold may be set to a longer time as the vehicle speed of the vehicle increases, or may be a function related to the vehicle speed of the vehicle, or may have a different value depending on the presence or absence of a parallel vehicle or an oncoming vehicle in an adjacent lane.

[0022] The concentration level determination unit 4 notifies the notification unit 5 of the determination result of the driver's concentration level on driving.

[0023] When the concentration level determination unit 4 determines that the driver's concentration level on driving is decreasing, the notification unit 5 issues a notification to encourage safe driving. The notification unit 5 controls, for example, a speaker in the vehicle (not shown) to output a warning sound of "pop" followed by a voice saying "Watch out for vehicles rushing out from the shoulder parking area." Alternatively, the notification unit 5 may control, for example, a head-up display (not shown) to surround the vehicle parked on the shoulder with a color frame and display the text "Watch out for vehicles rushing out from the shoulder parking area."

[0024] Next, when a vehicle to which the driving assistance device 1 of the first embodiment is applied approaches and passes a vehicle parked on the side of the road, the driving assistance device 1 determines a decrease in the driver's concentration on driving from the characteristics of the driver's evasive driving and issues an alert, as shown in Figures 2 and 3.

[0025] 2, a shoulder-parked vehicle 13 is parked on a shoulder 12 ahead of a course 11 (i.e., a driving lane) on which the host vehicle 10 is traveling. In this case, there is a possibility that a pedestrian (not shown) may jump out from a blind spot 14 of the shoulder-parked vehicle 13. For this reason, it is preferable for the host vehicle 10 to pass the shoulder-parked vehicle 13 by slowing down its speed and keeping the lateral distance Lm between the host vehicle 10 and the shoulder-parked vehicle 13 as large as possible, in terms of safe driving.

[0026] However, the driver may become impatient and lose concentration on driving, and may pass the shoulder-parked vehicle 13 without decelerating, shortening the lateral distance Lm between the shoulder-parked vehicle 13 and the vehicle 10, as shown by the arrow R in Fig. 2. In the first embodiment, the concentration determination unit 4 determines that the driver's concentration on driving has decreased when the lateral distance Lm, which is a characteristic of the driver's avoidance driving, is smaller than the distance threshold value.

[0027] In this case, as shown in Fig. 3, after the vehicle 10 passes the roadside parked vehicle 13, the notification unit 5 controls the in-vehicle speaker or the like to issue a notification encouraging safe driving. This enables the driving support device 1 to encourage the driver to regain concentration on driving in the subsequent driving.

[0028] Meanwhile, a conventional general automatic braking system issues a warning or executes brake control to avoid a collision with a dangerous object when the driver's concentration on driving decreases and the collision is imminent. In contrast, the driving support device 1 of the present embodiment detects a tendency for the driver's concentration to decrease in advance and provides a warning to help the driver recover his / her concentration on driving so that a dangerous situation in which a collision with a dangerous object is imminent does not occur in the first place.

[0029] Next, an example of an algorithm executed by the driving support device 1 according to the first embodiment will be described with reference to Fig. 4 to Fig. 6. In this algorithm, a dangerous object is a vehicle 13 parked on the roadside, and a characteristic of the driver's avoidance driving is a lateral distance Lm. This algorithm is repeatedly executed at a predetermined control cycle interval (for example, 100 ms interval).

[0030] In the following explanation and flowcharts, steps are simply represented as "S".

[0031] 4, the object detection unit 2 detects dangerous objects that exist on the path of and around the vehicle 10. The dangerous objects are, for example, vehicles 13 parked on the shoulder of the road.

[0032] Next, in S200, the avoidance driving detection unit 3 detects the characteristics of the driver's avoidance driving against a dangerous object. The process of S200 will be described in detail with reference to FIG.

[0033] 5, the evasive driving detection unit 3 determines whether the object detection unit 2 is detecting the shoulder-parked vehicle 13 as a dangerous object. If the evasive driving detection unit 3 determines in S211 that the object detection unit 2 is detecting the shoulder-parked vehicle 13, the process proceeds to S212. In S212, the evasive driving detection unit 3 calculates the lateral distance Lm when the host vehicle 10 passes beside the shoulder-parked vehicle 13 as a feature of the driver's evasive driving, and then ends the process of S200 and proceeds to S300. The evasive driving detection unit 3 holds in a memory the minimum value of the lateral distance Lm between the host vehicle 10 and the shoulder-parked vehicle 13 detected before and after the host vehicle 10 passes the shoulder-parked vehicle 13.

[0034] On the other hand, if the avoidance driving detection unit 3 determines in S211 that the object detection unit 2 is not detecting the vehicle 13 parked on the roadside, the avoidance driving detection unit 3 ends the process of S200 and advances the process to S300.

[0035] 4, in S300, the concentration determination unit 4 determines whether the driver's concentration on driving has decreased. The process of S300 will be described in detail with reference to FIG.

[0036] 6, the concentration degree determination unit 4 determines whether or not the host vehicle 10 has passed the shoulder-parked vehicle 13. When the concentration degree determination unit 4 determines that the host vehicle 10 has passed the shoulder-parked vehicle 13, the concentration degree determination unit 4 advances the process to S312.

[0037] In S312, the concentration degree determination unit 4 determines whether or not the lateral distance Lm when the vehicle 10 passes the shoulder-parked vehicle 13 is equal to or less than the distance threshold. If the concentration degree determination unit 4 determines that the lateral distance Lm is equal to or less than the distance threshold, the process proceeds to S313. In S313, the concentration level determination unit 4 determines that the driver's concentration level on driving is decreasing, ends the process of S300, and advances the process to S400.

[0038] On the other hand, if the concentration degree determination unit 4 determines in S311 that the host vehicle 10 has not yet passed the shoulder-parked vehicle 13, that is, has not yet passed (i.e. is approaching) or is passing, the process proceeds to S314. Also, if the concentration degree determination unit 4 determines in S312 that the lateral distance Lm when the host vehicle 10 passed the shoulder-parked vehicle 13 is greater than the distance threshold, the process proceeds to S314. In S314, the concentration level determination unit 4 determines that the driver's concentration level on driving is not decreasing, ends the process of S300, and advances the process to S400.

[0039] 4, in S400, when the concentration level determination unit 4 determines that the driver's concentration level on driving is decreasing, the notification unit 5 issues a notification to encourage safe driving. On the other hand, when the concentration level determination unit 4 determines that the driver's concentration level on driving is not decreasing, the notification unit 5 does not issue a notification. After S400, the process ends once, and is executed again repeatedly from S100 at a predetermined control period.

[0040] The driving support device 1 of the first embodiment described above provides the following advantageous effects. (1) In the first embodiment, the driving assistance device 1 includes an object detection unit 2, an evasive driving detection unit 3, a concentration determination unit 4, and an alert unit 5 as functional units that perform various control processes. The object detection unit 2 detects a path 11 of the vehicle 10 and dangerous objects present around the path 11. The evasive driving detection unit 3 detects characteristics of the driver's evasive driving for dangerous objects. The concentration determination unit 4 determines the driver's concentration on driving based on the characteristics of the driver's evasive driving. The alert unit 5 issues an alert to encourage safe driving when it is determined that the driver's concentration on driving has decreased. According to this, the driving support device 1 judges the driver's concentration level on driving based on the characteristics of the driver's avoidance driving. Therefore, compared with the method of judging the driver's concentration level on driving based only on the driver's facial expression or line of sight as in the above-mentioned Patent Document 1, the driving support device 1 of the first embodiment can judge the decrease in the driver's concentration level on driving with high accuracy. Therefore, this driving support device 1 can effectively encourage the driver to drive safely by treating the notification encouraging safe driving as acceptable to the driver and reducing the annoyance felt by the driver in response to the notification.

[0041] (2) In the first embodiment, one of the characteristics of the driver's avoidance driving is the lateral distance Lm between the vehicle 10 and a dangerous object when the vehicle 10 passes beside the object. This makes it possible to determine with high accuracy the decrease in the driver's concentration on driving.

[0042] (3) In addition, in the first embodiment, as one of the characteristics of the driver's avoidance driving, the TTC when the driver starts braking in front of a dangerous object is exemplified. This also makes it possible to determine the decrease in the driver's concentration on driving with high accuracy.

[0043] Second embodiment The second embodiment will be described. In the second embodiment, the configuration and algorithm of the driving support device 1 are partially changed from those in the first embodiment, and the rest is the same as in the first embodiment, so only the parts that are different from the first embodiment will be described.

[0044] 7, the driving support device 1 of the second embodiment includes a learning unit 6 as functional units that perform various control processes, in addition to an object detection unit 2, an evasive driving detection unit 3, a concentration level determination unit 4, and a notification unit 5. That is, the driving support device 1 of the second embodiment functions as the object detection unit 2, the evasive driving detection unit 3, the concentration level determination unit 4, the notification unit 5, and the learning unit 6 and performs various control processes by the processor executing a program stored in the memory.

[0045] The learning unit 6 receives information about dangerous objects from the object detection unit 2, and information about the characteristics of the driver's evasive driving from the evasive driving detection unit 3. Based on that information, the learning unit 6 learns the characteristics of the driver's evasive driving under normal circumstances. The learning value obtained by the learning unit 6 is transmitted to the concentration level determination unit 4. If the learning value is valid, the concentration level determination unit 4 uses the learning value in place of the distance threshold or the time threshold described in the first embodiment.

[0046] 8 and 9 show how the driving support device 1 learns the characteristics of the driver's avoidance driving when the vehicle 10 to which the driving support device 1 of the second embodiment is applied approaches and passes a roadside parked vehicle 13, and then determines that the driver's concentration on driving has decreased and issues a warning. In each of the figures referred to in the following description, a jagged arrow B typically indicates that the vehicle 10 has decelerated, and a curved arrow R indicates the driving route of the vehicle 10.

[0047] As shown in FIG. 8, a shoulder-parked vehicle 13a is parked as a hazardous object on the shoulder 12 ahead of the path 11 (i.e., the driving lane) on which the vehicle 10 is traveling. According to the driver's avoidance driving under normal circumstances, the vehicle 10 decelerates before the shoulder-parked vehicle 13a and passes the shoulder-parked vehicle 13a while keeping the lateral distance Lm between the vehicle 10 and the shoulder-parked vehicle 13a as large as possible. If the lateral distance Lm when the vehicle 10 passes the shoulder-parked vehicle 13a is equal to or greater than the distance threshold value described in the first embodiment, the driver's avoidance driving under normal circumstances can be said to be performing the avoidance driving. In addition, if the TTC when the vehicle 10 starts braking before the shoulder-parked vehicle 13a is equal to or greater than the time threshold value described in the first embodiment, the driver's avoidance driving under normal circumstances can be said to be performing the avoidance driving.

[0048] In this case, the learning unit 6 stores the lateral distance Lm when the vehicle 10 passes the shoulder-parked vehicle 13a as a characteristic of the driver's evasive driving under normal circumstances, and calculates the lateral distance learning value Lm_l based on the lateral distance Lm one or more times. As will be described later in the third embodiment, the learning unit 6 may store the TTC when the driver starts braking in front of the shoulder-parked vehicle 13a as a feature of the driver's avoidance driving under normal circumstances, and calculate the TTC learning value TTCf_l based on one or more TTCs. In this case, the learning unit 6 saves the TTC when the driver starts braking in front of the shoulder-parked vehicle 13a in a predetermined folder TTCf set in the memory, and calculates the TTC learning value TTCf_l based on the TTCf.

[0049] Next, Fig. 9 shows a scene after the host vehicle 10 has passed the scene of Fig. 8 and continued to travel, that is, a different scene from Fig. 8. As shown in Fig. 9, a shoulder-parked vehicle 13b different from the one shown in Fig. 8 is parked on the shoulder 12 ahead of the path 11 (i.e., the travel lane) on which the host vehicle 10 travels. In this case, the driver may become impatient and lose focus on driving, and may pass the shoulder-parked vehicle 13b with the shoulder-parked vehicle 13b close to the shoulder-parked vehicle 13b as shown by the arrow R in Fig. 9 due to overlooking the shoulder-parked vehicle 13b. If the lateral distance Lm, which is a characteristic of the driver's avoidance driving, is smaller than the lateral distance learning value Lm_l obtained by learning from the scene shown in Fig. 8, the concentration level determination unit 4 determines that the driver's concentration level on driving has decreased.

[0050] In this case, after the vehicle 10 passes the shoulder-parked vehicle 13b, the notification unit 5 controls the in-vehicle speaker or the like to issue a notification encouraging safe driving. This allows the driving support device 1 to encourage the driver to regain concentration on driving in the subsequent driving.

[0051] Next, an example of an algorithm executed by the driving support device 1 according to the second embodiment will be described with reference to Fig. 10 to Fig. 13. In this algorithm, a dangerous object is a vehicle 13 parked on the roadside, the characteristics of the driver's avoidance driving are a lateral distance Lm, and a learning value acquired by learning the learning unit 6 is a lateral distance learning value Lm_l. This algorithm is repeatedly executed at a predetermined control cycle interval (for example, 100 ms interval).

[0052] 10, the object detection unit 2 detects dangerous objects present on the path of and around the vehicle 10. An example of a dangerous object is a vehicle 13 parked on the shoulder of the road.

[0053] Next, in S200, the avoidance driving detection unit 3 detects the characteristics of the driver's avoidance driving against a dangerous object. The process of S200 will be described in detail with reference to FIG.

[0054] 11, the evasive driving detection unit 3 determines whether the object detection unit 2 is detecting the shoulder-parked vehicle 13 as a dangerous object. If the evasive driving detection unit 3 determines in S221 that the object detection unit 2 is detecting the shoulder-parked vehicle 13, the process proceeds to S222. In S222, the evasive driving detection unit 3 calculates the lateral distance Lm when the host vehicle 10 passes the shoulder-parked vehicle 13 as a feature of the driver's evasive driving, and then ends the process of S200 and proceeds to S300. The evasive driving detection unit 3 holds in a memory the minimum value of the lateral distance Lm between the host vehicle 10 and the shoulder-parked vehicle 13 detected before and after the host vehicle 10 passes the shoulder-parked vehicle 13.

[0055] On the other hand, if the avoidance driving detection unit 3 determines in S221 that the object detection unit 2 is not detecting the vehicle 13 parked on the roadside, the avoidance driving detection unit 3 ends the process of S200 and advances the process to S300.

[0056] 10, in S300, the concentration determination unit 4 determines whether the driver's concentration on driving has decreased. The process of S300 will be described in detail with reference to FIG.

[0057] 12, the concentration degree determination unit 4 determines whether or not the host vehicle 10 has passed the shoulder-parked vehicle 13. When the concentration degree determination unit 4 determines that the host vehicle 10 has passed the shoulder-parked vehicle 13, the concentration degree determination unit 4 advances the process to S322.

[0058] In S322, the concentration level determination unit 4 determines whether the lateral distance learning value Lm_l obtained by the learning unit 6 through learning is valid or not. This determination is made, for example, based on whether the lateral distance learning value Lm_l has been updated a predetermined number of times or more after the traveling switch of the host vehicle 10 is turned on. The number of updates of the lateral distance learning value Lm_l is reset when the traveling switch of the host vehicle 10 is turned off. The method of updating the lateral distance learning value Lm_l will be described later. If the lateral distance learning value Lm_l has been updated a predetermined number of times or more after the traveling switch of the host vehicle 10 is turned on, the concentration level determination unit 4 determines that the lateral distance learning value Lm_l is valid, and proceeds to S323.

[0059] In S323, the concentration degree determination unit 4 determines whether or not the lateral distance Lm when the vehicle 10 passed the shoulder-parked vehicle 13 is equal to or less than the lateral distance learning value Lm_l. If the concentration degree determination unit 4 determines that the lateral distance Lm is equal to or less than the lateral distance learning value Lm_l, the process proceeds to S324. In S324, the concentration level determination unit 4 determines that the driver's concentration level on driving is decreasing, ends the process of S300, and advances the process to S400.

[0060] On the other hand, if the concentration degree determination unit 4 determines in S321 that the host vehicle 10 has not yet passed the shoulder-parked vehicle 13, i.e., has not yet passed or is passing, the process proceeds to S325. Also, if the concentration degree determination unit 4 determines in S322 that the lateral distance learning value Lm_l obtained by the learning unit 6 through learning is invalid, the process proceeds to S325. Also, if the concentration degree determination unit 4 determines in S323 that the lateral distance Lm when the host vehicle 10 passed the shoulder-parked vehicle 13 is greater than the lateral distance learning value Lm_l, the process proceeds to S325. In S325, the concentration level determination unit 4 determines that the driver's concentration level on driving is not decreasing, ends the process of S300, and advances the process to S400.

[0061] 10, in S400, if the concentration level determination unit 4 determines that the driver's concentration level on driving is decreasing, the notification unit 5 issues a notification to encourage safe driving and proceeds to S500. On the other hand, if the concentration level determination unit 4 determines that the driver's concentration level on driving is not decreasing, the notification unit 5 does not issue a notification and proceeds to S500.

[0062] At S500, the learning unit 6 learns the characteristics of the driver's avoidance driving under normal circumstances. The process at S500 will be described in detail with reference to FIG.

[0063] 13, the learning unit 6 determines whether or not the process of S300 has determined that the driver's concentration on driving is decreasing at the time when the host vehicle 10 has passed the shoulder-parked vehicle 13. If the process of S300 has not determined that the driver's concentration on driving is decreasing at S521, the learning unit 6 advances the process to S522.

[0064] In S522, the learning unit 6 updates the lateral distance learning value Lm_l using the lateral distance Lm when the vehicle 10 passes the vehicle 13 parked on the roadside. The case where it is not determined that the driver's concentration on driving is decreasing means a normal time. Therefore, when it is not determined that the driver's concentration on driving is decreasing, the lateral distance learning value Lm_l is updated using the lateral distance Lm, which is a characteristic of the driver's avoidance driving in a normal time. For example, a moving average method is used to update the lateral distance learning value Lm_l. In the moving average method, the average of the lateral distance Lm of the past N scenes and the lateral distance Lm detected this time is calculated.

[0065] On the other hand, in the above S521, if it is determined in the process of S300 that the driver's concentration on driving is decreasing, the learning unit 6 ends the process without updating the lateral distance learning value Lm_l. In this case, the lateral distance Lm when the vehicle 10 passes the shoulder-parked vehicle 13 is not characteristic of the driver's avoidance driving under normal circumstances, so the lateral distance learning value Lm_l is not updated.

[0066] After S500, the process is temporarily terminated, and then the process is repeated from S100 again at a predetermined control period.

[0067] The driving support device 1 of the second embodiment described above provides the following advantageous effects. (1) In the second embodiment, the driving assistance device 1 further includes a learning unit 6 as functional units for performing various control processes, in addition to the object detection unit 2, the evasive driving detection unit 3, the concentration determination unit 4, and the notification unit 5. The learning unit 6 learns the characteristics of the driver's evasive driving under normal circumstances. The concentration determination unit 4 compares the characteristics of the driver's evasive driving learned by the learning unit 6 with the characteristics of the driver's evasive driving detected by the evasive driving detection unit 3, and determines that the driver's concentration on driving has decreased. According to this, the driver receives a notification encouraging safe driving when the characteristics of the avoidance driving in the driver's normal driving differ from the characteristics of the avoidance driving in the current driving, so that the driver feels convinced of the notification and the driver feels less bothered. Therefore, the driving support device 1 can effectively encourage the driver to drive safely.

[0068] (2) In the second embodiment, if the characteristics of the driver's avoidance driving do not satisfy a predetermined criterion, the learning unit 6 does not reflect the characteristics of the avoidance driving at that time in the learning. On the other hand, if the characteristics of the driver's avoidance driving satisfy the predetermined criterion, the learning unit 6 reflects the characteristics of the driver's avoidance driving at that time in the learning. According to this, the learning unit 6 learns the characteristics of the driver's avoidance driving only when the characteristics of the driver's avoidance driving satisfy a predetermined standard. Therefore, the characteristics of the driver's avoidance driving learned by the learning unit 6 are changed to be on the safer side with respect to a predetermined threshold setting. Therefore, the driving support device 1 can provide a highly safe notification corresponding to each driver's normal driving. The predetermined criteria may be, for example, the time threshold and distance threshold described in the first embodiment. In addition, a case where the characteristics of the driver's avoidance driving do not satisfy the predetermined criteria means, for example, that the driver's concentration level is determined to be decreasing. On the other hand, a case where the characteristics of the driver's avoidance driving satisfy the predetermined criteria means, for example, that the driver's concentration level is not determined to be decreasing, that is, that the driver is in a normal state.

[0069] Third embodiment The third embodiment will be described below. The third embodiment has the same configuration as the second embodiment, but the algorithm is partly changed from the second embodiment.

[0070] Figures 14 and 15 show how, when a vehicle 10 to which the driving assistance device 1 of the third embodiment is applied approaches and passes a vehicle 13 parked on the side of the road, the driving assistance device 1 learns the characteristics of the driver's evasive driving, and then determines that the driver's concentration on driving has decreased and issues an alert.

[0071] As shown in FIG. 14, a shoulder-parked vehicle 13a is parked as a hazardous object on the shoulder 12 ahead of the path 11 (i.e., the driving lane) on which the vehicle 10 is traveling. According to the driver's avoidance driving under normal circumstances, the vehicle 10 decelerates before the shoulder-parked vehicle 13a and passes the shoulder-parked vehicle 13a while keeping the lateral distance Lm between the vehicle 10 and the shoulder-parked vehicle 13 as large as possible. If the TTC when the vehicle 10 starts braking before the shoulder-parked vehicle 13a is equal to or greater than the time threshold value described in the first embodiment, it can be said that the driver is avoiding the obstacle under normal circumstances. Note that if the lateral distance Lm when the vehicle 10 passes the shoulder-parked vehicle 13a is equal to or greater than the distance threshold value described in the first embodiment, it can be said that the driver is avoiding the obstacle under normal circumstances.

[0072] In this case, the learning unit 6 stores the TTC when the driver starts braking in front of the shoulder-parked vehicle 13a as a feature of the driver's avoidance driving in TTCf, and calculates the TTC learning value TTCf_l based on the TTCf. As described in the second embodiment, the learning unit 6 may store the lateral distance Lm when the host vehicle 10 passes the shoulder-parked vehicle 13a as a feature of the driver's avoidance driving in normal times, and calculate the lateral distance learning value Lm_l.

[0073] Next, Fig. 15 shows a scene after the host vehicle 10 has passed the scene of Fig. 14 and continued to travel, that is, a different scene from Fig. 14. As shown in Fig. 15, a shoulder-parked vehicle 13b different from the one shown in Fig. 14 is parked on the shoulder 12 ahead of the path 11 (i.e., the travel lane) on which the host vehicle 10 travels. In this case, the driver may become impatient and lose concentration on driving, and the TTC may become shorter due to overlooking the shoulder-parked vehicle 13b. If the TTC as a characteristic of the driver's avoidance driving is smaller than the TTC learning value TTCf_l obtained by learning from the scene shown in Fig. 14, the concentration determination unit 4 determines that the driver's concentration on driving has decreased.

[0074] In this case, the notification unit 5 controls the speaker in the vehicle interior to issue a notification to encourage safe driving. The notification may be issued before the vehicle 10 passes the shoulder-parked vehicle 13b (i.e., while approaching) or after the vehicle 10 passes the shoulder-parked vehicle 13b. This allows the driving support device 1 to encourage the driver to regain concentration on driving in the subsequent driving.

[0075] Next, an example of an algorithm executed by the driving support device 1 according to the third embodiment will be described with reference to Fig. 16 to Fig. 19. In this algorithm, a dangerous object is a roadside parked vehicle 13, a characteristic of the driver's avoidance driving is a TTC, and a learning value acquired by learning the learning unit 6 is a TTC learning value TTCf_l. This algorithm is repeatedly executed at a predetermined control cycle interval (for example, 100 ms interval).

[0076] 16, the object detection unit 2 detects dangerous objects present on the path of and around the vehicle 10. The dangerous objects are, for example, vehicles 13 parked on the shoulder of the road.

[0077] Next, in S200, the avoidance driving detection unit 3 detects the characteristics of the driver's avoidance driving against a dangerous object. The process of S200 will be described in detail with reference to FIG.

[0078] 17, the evasive driving detection unit 3 determines whether or not the object detection unit 2 is detecting a shoulder-parked vehicle 13 as a dangerous object. If the evasive driving detection unit 3 determines in S231 that the object detection unit 2 is detecting a shoulder-parked vehicle 13, the process proceeds to S232. In S232, the avoidance driving detection unit 3 determines whether or not the avoidance driving of the driver has been detected. If the avoidance driving detection unit 3 determines in S232 that the avoidance driving of the driver has not been detected, the process proceeds to S233.

[0079] In S233, the evasive driving detection unit 3 determines whether or not an evasive driving operation by the driver has been detected. If an evasive driving operation by the driver has been detected in S233, the evasive driving detection unit 3 advances the process to S234. In S234, the evasive driving detection unit 3 stores (latches) the TTCn when the driver starts braking just before the shoulder-parked vehicle 13 as a characteristic of the driver's evasive driving in TTCf as the TTC when evasive driving around the shoulder-parked vehicle, and then proceeds to processing at S300.

[0080] On the other hand, if the evasive driving detection unit 3 determines in the above S231 that the object detection unit 2 is not detecting the shoulder-parked vehicle 13, it ends the process of S200 and proceeds to S300. Also, if the evasive driving detection unit 3 determines in the above S232 that the driver's evasive driving has been detected, it ends the process of S200 and proceeds to S300. Also, if the evasive driving detection unit 3 does not detect the driver's evasive driving in the above S233, it ends the process of S200 and proceeds to S300.

[0081] 16, in S300, the concentration determination unit 4 determines whether the driver's concentration on driving has decreased. The process of S300 will be described in detail with reference to FIG.

[0082] 18, the concentration degree determination unit 4 determines whether or not the object detection unit 2 is detecting a shoulder-parked vehicle 13 as a hazardous object. If the concentration degree determination unit 4 determines in S331 that the object detection unit 2 is detecting a shoulder-parked vehicle 13, the process proceeds to S332. In S332, the concentration level determination unit 4 determines whether or not the evasive driving detection unit 3 has detected the evasive driving of the driver. If the concentration level determination unit 4 determines in S332 that the evasive driving detection unit 3 has not detected the evasive driving of the driver, the process proceeds to S333.

[0083] In S333, the concentration level determination unit 4 determines whether the TTC learning value TTCf_l obtained by the learning unit 6 through learning is valid or not. This determination is made, for example, based on whether the TTC learning value TTCf_l has been updated a predetermined number of times or more after the travel switch of the host vehicle 10 is turned on. The number of updates of the TTC learning value TTCf_l is reset when the travel switch of the host vehicle 10 is turned off. Note that a method of updating the TTC learning value TTCf_l will be described later. If the TTC learning value TTCf_l has been updated a predetermined number of times or more after the travel switch of the host vehicle 10 is turned on, the concentration level determination unit 4 determines that the TTC learning value TTCf_l is valid, and proceeds to S334.

[0084] In S334, the concentration level determination unit 4 determines whether or not the value obtained by subtracting the TTCn to the shoulder-parked vehicle 13 in real time from the TTC learning value TTCf_l (i.e., TTCf_l-TTCn) is equal to or greater than a predetermined threshold. Note that TTCn is a value obtained by dividing the distance from the vehicle to the hazardous object by the vehicle speed, regardless of whether the driver has started braking. Therefore, even before the driver has started braking, the subtraction process is performed in real time, and the concentration level determination is performed. In S334, if the concentration level determination unit 4 determines that TTCf_l-TTCn is equal to or greater than the predetermined threshold, the process proceeds to S335. The predetermined threshold can be 0 or a positive number. In S335, the concentration level determination unit 4 determines that the driver's concentration level on driving is decreasing, ends the process of S300, and advances the process to S400.

[0085] On the other hand, if the concentration level determination unit 4 determines in S331 that the object detection unit 2 is not detecting the shoulder-parked vehicle 13, the process proceeds to S336. Also, if the concentration level determination unit 4 determines in S332 that the evasive driving detection unit 3 has already detected the driver's evasive driving, the process proceeds to S336. Also, if the concentration level determination unit 4 determines in S333 that the TTC learning value TTCf_l obtained by the learning unit 6 through learning is invalid, the process proceeds to S336. Also, if the concentration level determination unit 4 determines in S333 that TTCf_l-TTCn is less than a predetermined threshold, the process proceeds to S336. In S336, the concentration level determination unit 4 determines that the driver's concentration level on driving is not decreasing, ends the process of S300, and advances the process to S400.

[0086] 16, in S400, if the concentration level determination unit 4 determines that the driver's concentration level on driving is decreasing, the notification unit 5 issues a notification to encourage safe driving and proceeds to S500. On the other hand, if the concentration level determination unit 4 determines that the driver's concentration level on driving is not decreasing, the notification unit 5 does not issue a notification and proceeds to S500.

[0087] In S500, the learning unit 6 learns the characteristics of the driver's avoidance driving under normal circumstances. The process of S500 will be described in detail with reference to FIG.

[0088] 19, the learning unit 6 determines whether or not the process of S300 has determined that the driver's concentration on driving is decreasing before the host vehicle 10 passes the shoulder-parked vehicle 13, at the timing when the host vehicle 10 has completely passed the shoulder-parked vehicle 13. If the learning unit 6 has not determined in S531 that the driver's concentration on driving is decreasing before the host vehicle 10 passes the shoulder-parked vehicle 13, the process proceeds to S532.

[0089] In S532, the learning unit 6 updates the TTC learning value TTCf_l for the shoulder-parked vehicle 13 using the TTCn at which the vehicle 10 starts braking before the shoulder-parked vehicle 13. The case where it is not determined that the driver's concentration on driving is decreasing means a normal time. Therefore, when it is not determined that the driver's concentration on driving is decreasing, the TTC learning value TTCf_l for the shoulder-parked vehicle 13 is updated using the TTCf, which is a characteristic of the driver's avoidance driving in normal time. For example, the moving average method is used to update the TTC learning value TTCf_l. In the moving average method, the average of the TTC of the past N scenes and the TTCn detected this time is calculated.

[0090] On the other hand, in the above S531, if it is determined in the processing of S300 that the driver's concentration on driving is decreasing at the timing when the host vehicle 10 has completely passed the shoulder-parked vehicle 13 before the host vehicle 10 passes the shoulder-parked vehicle 13, the learning unit 6 ends the processing without updating the TTC learning value TTCf_l. In that case, since this is not a characteristic of the driver's avoidance driving under normal circumstances, the TTC learning value TTCf_l is not updated.

[0091] After S500, the process is temporarily terminated, and then the process is repeated from S100 again at a predetermined control period.

[0092] The driving assistance device 1 of the third embodiment described above also achieves the same effects as those of the second embodiment.

[0093] (Fourth embodiment) A fourth embodiment will now be described. The fourth embodiment also has the same configuration as the second embodiment, but the algorithm is partly changed from the second embodiment.

[0094] When a vehicle 10 to which the driving assistance device 1 of the fourth embodiment is applied approaches and passes a crossing pedestrian, the driving assistance device 1 learns the characteristics of the driver's evasive driving, and then determines that the driver's concentration on driving has decreased and issues an alert, as shown in Figures 20 and 21.

[0095] As shown in Fig. 20, there is a crossing pedestrian 15a as a dangerous object that is about to cross the path 11 of the vehicle 10 on the sidewalk 16 ahead of the path 11 (i.e., the driving lane) on which the vehicle 10 is traveling. According to the driver's avoidance driving under normal circumstances, the vehicle 10 starts braking before the crossing pedestrian 15a, stops temporarily before the crosswalk if there is one, and stops temporarily or slows down if there is no crosswalk, so as not to obstruct the passage of the crossing pedestrian 15a. If the TTC when the vehicle 10 starts braking before the crossing pedestrian 15a is equal to or greater than the time threshold value described in the first embodiment, it can be said that the driver is performing the avoidance driving under normal circumstances.

[0096] In this case, the learning unit 6 stores the TTC when the driver starts braking in front of the crossing pedestrian 15a as a characteristic of the driver's evasive driving under normal circumstances in a predetermined folder TTCf set in the memory, and calculates the TTC learning value TTCf_l based on that TTCf.

[0097] Next, Fig. 21 shows a scene after the host vehicle 10 has passed the scene of Fig. 20 and continued traveling, that is, a different scene from Fig. 20. As shown in Fig. 21, a crossing pedestrian 15b different from the one shown in Fig. 20 is about to cross the path 11 of the host vehicle 10 on the sidewalk 16 ahead of the path 11 (i.e., the traveling lane) on which the host vehicle 10 is traveling. In this case, the driver may become impatient and decrease the concentration on driving, and the TTC may become shorter. If the TTC as a feature of the driver's avoidance driving is smaller than the TTC learning value TTCf_l obtained by learning from the scene shown in Fig. 20, the concentration determination unit 4 determines that the driver's concentration on driving has decreased.

[0098] In this case, the notification unit 5 controls the speaker in the vehicle interior to issue a notification to encourage safe driving. The notification may be issued before the vehicle 10 passes the crossing pedestrian 15b (i.e., while approaching), or after the vehicle 10 passes the crossing pedestrian 15b. This allows the driving support device 1 to encourage the driver to recover his / her concentration on driving.

[0099] Next, an example of an algorithm executed by the driving support device 1 according to the fourth embodiment will be described with reference to Fig. 22 to Fig. 25. In this algorithm, a dangerous object is a crossing pedestrian 15, the characteristics of the driver's avoidance driving are TTC, and the learning value acquired by the learning unit 6 through learning is the TTC learning value TTCf_l. This algorithm is repeatedly executed at a predetermined control cycle interval (for example, 100 ms interval).

[0100] 22, the object detection unit 2 detects dangerous objects existing on the path of the vehicle 10 and around the vehicle 10. An example of a dangerous object is a pedestrian 15 crossing the road.

[0101] Next, in S200, the avoidance driving detection unit 3 detects the characteristics of the driver's avoidance driving against a dangerous object. The process of S200 will be described in detail with reference to FIG.

[0102] 23, the evasive driving detection unit 3 determines whether or not the object detection unit 2 is detecting a crossing pedestrian 15 as a dangerous object. When the evasive driving detection unit 3 determines that the object detection unit 2 is detecting a crossing pedestrian 15, the evasive driving detection unit 3 advances the process to S242. In S242, the avoidance driving detection unit 3 determines whether or not the avoidance driving of the driver has been detected. If the avoidance driving detection unit 3 determines that the avoidance driving of the driver has not been detected, the process proceeds to S243.

[0103] In S243, the avoidance driving detection unit 3 determines whether or not the avoidance driving by the driver is detected. If the avoidance driving detection unit 3 detects the avoidance driving by the driver, the avoidance driving detection unit 3 advances the process to S244. In S244, the evasive driving detection unit 3 stores (latches) the TTCn when the driver starts braking just before the crossing pedestrian 15 as a characteristic of the driver's evasive driving in TTCf as the TTC during evasive driving around a crossing pedestrian, and then proceeds to S300.

[0104] On the other hand, if the evasive driving detection unit 3 determines in the above S241 that the object detection unit 2 is not detecting the crossing pedestrian 15, it ends the process of S200 and proceeds to S300. Also, if the evasive driving detection unit 3 determines in the above S242 that the driver's evasive driving has been detected, it ends the process of S200 and proceeds to S300. Also, if the evasive driving detection unit 3 does not detect the driver's evasive driving in the above S243, it ends the process of S200 and proceeds to S300.

[0105] 22, in S300, the concentration determination unit 4 determines whether the driver's concentration on driving has decreased. The process of S300 will be described in detail with reference to FIG.

[0106] 24, the concentration degree determination unit 4 determines whether or not the object detection unit 2 is detecting a crossing pedestrian 15. If the concentration degree determination unit 4 determines in S341 that the object detection unit 2 is detecting a crossing pedestrian 15, the process proceeds to S342. In S342, the concentration level determination unit 4 determines whether or not the evasive driving detection unit 3 has detected the evasive driving of the driver. If the concentration level determination unit 4 determines in S342 that the evasive driving detection unit 3 has not detected the evasive driving of the driver, the process proceeds to S343.

[0107] In S343, the concentration level determination unit 4 determines whether the TTC learning value TTCf_l obtained by the learning unit 6 through learning is valid. This determination is made, for example, based on whether the TTC learning value TTCf_l has been updated a predetermined number of times or more after the travel switch of the host vehicle 10 is turned on. The number of updates of the TTC learning value TTCf_l is reset when the travel switch of the host vehicle 10 is turned off. If the TTC learning value TTCf_l has been updated a predetermined number of times or more after the travel switch of the host vehicle 10 is turned on, the concentration level determination unit 4 determines that the TTC learning value TTCf_l is valid, and proceeds to S344.

[0108] In S344, the concentration level determination unit 4 determines whether the value obtained by subtracting the TTCn up to the crossing pedestrian 15 in real time from the TTC learning value TTCf_l (i.e., TTCf_l-TTCn) is equal to or greater than a predetermined threshold. If the concentration level determination unit 4 determines in S344 that TTCf_l-TTCn is equal to or greater than the predetermined threshold, the process proceeds to S345. The predetermined threshold can be 0 or a positive number. In S345, the concentration level determination unit 4 determines that the driver's concentration level on driving is decreasing, ends the process of S300, and advances the process to S400.

[0109] On the other hand, if the concentration level determination unit 4 determines in S341 that the object detection unit 2 is not detecting a crossing pedestrian 15, the process proceeds to S346. Also, if the concentration level determination unit 4 determines in S342 that the evasive driving detection unit 3 has already detected the driver's evasive driving, the process proceeds to S346. Also, if the concentration level determination unit 4 determines in S343 that the TTC learning value TTCf_l obtained by the learning unit 6 through learning is invalid, the process proceeds to S346. Also, if the concentration level determination unit 4 determines in S344 that TTCf_l-TTCn is less than a predetermined threshold, the process proceeds to S346. In S346, the concentration level determination unit 4 determines that the driver's concentration level on driving is not decreasing, ends the process of S300, and advances the process to S400.

[0110] 22, in S400, if the concentration level determination unit 4 determines that the driver's concentration level on driving is decreasing, the notification unit 5 issues a notification to encourage safe driving and proceeds to S500. On the other hand, if the concentration level determination unit 4 determines that the driver's concentration level on driving is not decreasing, the notification unit 5 does not issue a notification and proceeds to S500.

[0111] In S500, the learning unit 6 learns the characteristics of the driver's avoidance driving under normal circumstances. The process of S500 will be described in detail with reference to FIG.

[0112] 25, the learning unit 6 determines whether or not the process of S300 has determined that the driver's concentration on driving is decreasing before the host vehicle 10 passes the crossing pedestrian 15, at the timing when the host vehicle 10 has completely passed the crossing pedestrian 15. If the process of S541 does not determine that the driver's concentration on driving is decreasing before the host vehicle 10 passes the crossing pedestrian 15, the learning unit 6 advances the process to S542.

[0113] In S542, the learning unit 6 updates the TTC learning value TTCf_l for the crossing pedestrian 15 using the TTCn at which the vehicle 10 starts braking before the crossing pedestrian 15. The case where it is not determined that the driver's concentration on driving is decreasing means a normal time. Therefore, when it is not determined that the driver's concentration on driving is decreasing, the TTC learning value TTCf_l for the crossing pedestrian 15 is updated using the TTCf, which is a characteristic of the driver's avoidance driving in the normal time. Note that, for example, a moving average method is used to update the TTC learning value TTCf_l.

[0114] On the other hand, in the above S541, if it is determined in the processing of S300 that the driver's concentration on driving is decreasing at the timing when the host vehicle 10 has completely passed the crossing pedestrian 15 before the host vehicle 10 passes the crossing pedestrian 15, the learning unit 6 ends the processing without updating the TTC learning value TTCf_l. In that case, since this is not a characteristic of the driver's avoidance driving under normal circumstances, the TTC learning value TTCf_l is not updated.

[0115] After S500, the process is temporarily terminated, and then the process is repeated from S100 again at a predetermined control period.

[0116] The driving assistance device 1 of the fourth embodiment described above also achieves the same effects as those of the second and third embodiments.

[0117] Fifth embodiment The fifth embodiment will be described. The fifth embodiment is different from the first embodiment in that a part of the algorithm of the driving support device 1 is changed, and the rest is the same as the first embodiment, so only the parts that are different from the first embodiment will be described.

[0118] The driving support device 1 of the fifth embodiment processes an image of the driver's face captured by an imaging device (not shown) mounted on the vehicle, and judges the driver's concentration on driving, including the driver's line of sight toward a dangerous object when the vehicle 10 approaches the dangerous object. As the imaging device, for example, a Driver Status Monitor (registered trademark) manufactured by Denso Corporation can be used, but it is not limited thereto, and other imaging devices may also be used.

[0119] Figures 26 and 27 show how, when a vehicle 10 to which the driving assistance device 1 of the fifth embodiment is applied approaches and passes a vehicle 13 parked on the side of the road, the driving assistance device 1 determines a decrease in the driver's concentration on driving from the characteristics of the driver's evasive driving and issues an alert.

[0120] As shown in Fig. 26, a shoulder-parked vehicle 13 is parked on the shoulder 12 ahead of a course 11 (i.e., a driving lane) on which the vehicle 10 is traveling. In this case, there is a possibility that a pedestrian or the like may jump out from a blind spot 14 of the shoulder-parked vehicle 13. For this reason, it is preferable for the vehicle 10 to pass the shoulder-parked vehicle 13 while reducing the vehicle speed and making the lateral distance Lm between the vehicle 10 and the shoulder-parked vehicle 13 as large as possible, and further paying attention to a pedestrian or the like jumping out from the blind spot 14 of the shoulder-parked vehicle 13, in terms of safe driving.

[0121] However, the driver may become impatient and lose concentration on driving, and may pass the shoulder-parked vehicle 13 without slowing down, with the lateral distance Lm between the shoulder-parked vehicle 13 and the vehicle 10 being close, as shown by arrow R in Fig. 26. The driver may also pass the shoulder-parked vehicle 13 without looking at the shoulder-parked vehicle 13 and its blind spot 14.

[0122] In the fifth embodiment, the concentration level determination unit 4 determines that the driver's concentration level on driving has decreased when the lateral distance Lm, which is a characteristic of the driver's avoidance driving, is smaller than the distance threshold value. Furthermore, the concentration level determination unit 4 determines that the driver's concentration level on driving has decreased when the driver's visibility of the shoulder-parked vehicle 13 and its blind spot 14 is insufficient when the vehicle 10 approaches the shoulder-parked vehicle 13.

[0123] In this case, as shown in Fig. 27, after the vehicle 10 passes the shoulder-parked vehicle 13, the notification unit 5 controls the in-vehicle speaker or the like to issue a notification encouraging safe driving. This enables the driving support device 1 to encourage the driver to regain concentration on driving in the subsequent driving.

[0124] Next, an example of an algorithm executed by the driving support device 1 according to the fifth embodiment will be described with reference to Fig. 28 to Fig. 30. In this algorithm, a dangerous object is a roadside parked vehicle 13, and a characteristic of the driver's avoidance driving is a lateral distance Lm. This algorithm is repeatedly executed at a predetermined control cycle interval (for example, 100 ms interval).

[0125] 28, the object detection unit 2 detects dangerous objects present on the path of the host vehicle 10 and around the vehicle 10. The dangerous objects are, for example, vehicles 13 parked on the roadside.

[0126] Next, in S200, the avoidance driving detection unit 3 detects the characteristics of the driver's avoidance driving against a dangerous object. The process of S200 will be described in detail with reference to FIG.

[0127] 29, the evasive driving detection unit 3 determines whether the object detection unit 2 is detecting the shoulder-parked vehicle 13 as a dangerous object. If the evasive driving detection unit 3 determines in S251 that the object detection unit 2 is detecting the shoulder-parked vehicle 13, the process proceeds to S252. In S252, the evasive driving detection unit 3 calculates the lateral distance Lm when the host vehicle 10 passes beside the shoulder-parked vehicle 13 as a feature of the driver's evasive driving, and then ends the process of S200 and proceeds to S300. The evasive driving detection unit 3 holds in a memory the minimum value of the lateral distance Lm between the host vehicle 10 and the shoulder-parked vehicle 13 detected before and after the host vehicle 10 passes the shoulder-parked vehicle 13.

[0128] On the other hand, if the evasive driving detection unit 3 determines in S251 that the object detection unit 2 is not detecting the vehicle 13 parked on the roadside, the evasive driving detection unit 3 ends the process of S200 and advances the process to S300.

[0129] 28, in S300, the concentration determination unit 4 determines whether the driver's concentration on driving has decreased. The process of S300 will be described in detail with reference to FIG.

[0130] 30, the concentration degree determination unit 4 determines whether or not the host vehicle 10 has passed the shoulder-parked vehicle 13. When the concentration degree determination unit 4 determines that the host vehicle 10 has passed the shoulder-parked vehicle 13, the concentration degree determination unit 4 advances the process to S352.

[0131] In S352, the concentration degree determination unit 4 determines whether or not the lateral distance Lm when the vehicle 10 passes the shoulder-parked vehicle 13 is equal to or less than the distance threshold. If the concentration degree determination unit 4 determines that the lateral distance Lm is equal to or less than the distance threshold, the process proceeds to S353.

[0132] In S353, the concentration level determination unit 4 performs image processing on the image of the driver's face captured by the imaging device when the vehicle 10 approaches the shoulder-parked vehicle 13, and determines whether the driver has sufficient visibility of the shoulder-parked vehicle 13 and its blind spot 14. If the concentration level determination unit 4 determines that the driver does not have sufficient visibility of the shoulder-parked vehicle 13 and its blind spot 14, it advances the process to S354.

[0133] In S354, the concentration level determination unit 4 determines that the driver's concentration level on driving is decreasing, ends the process of S300, and advances the process to S400.

[0134] On the other hand, if the concentration degree determination unit 4 determines in S351 that the host vehicle 10 has not yet passed the shoulder-parked vehicle 13, that is, has not yet passed or is passing, the process proceeds to S355. Also, if the concentration degree determination unit 4 determines in S352 that the lateral distance Lm when the host vehicle 10 passed the shoulder-parked vehicle 13 is greater than the distance threshold, the process proceeds to S355. Furthermore, if the concentration degree determination unit 4 determines in S353 that the driver's visibility of the shoulder-parked vehicle 13 and its blind spot 14 is sufficient, the process proceeds to S355.

[0135] In S355, the concentration level determination unit 4 determines that the driver's concentration level on driving is not decreasing, ends the process of S300, and advances the process to S400.

[0136] 28, in S400, when the concentration level determination unit 4 determines that the driver's concentration level on driving is decreasing, the notification unit 5 issues a notification to encourage safe driving. On the other hand, when the concentration level determination unit 4 determines that the driver's concentration level on driving is not decreasing, the notification unit 5 does not issue a notification.

[0137] The driving assistance device 1 of the fifth embodiment described above determines a decrease in the driver's concentration level based on the driver's line of sight to a dangerous object when the host vehicle 10 approaches the dangerous object, in addition to the characteristics of the driver's avoidance driving. According to this, the concentration level determination unit 4 detects whether the driver is paying attention to a dangerous object based on the driver's line of sight when the vehicle 10 approaches the dangerous object. Therefore, the direction in which the driver's line of sight is directed can be accurately captured, thereby improving the accuracy of determining the driver's concentration on driving.

[0138] Sixth embodiment The sixth embodiment will be described. The sixth embodiment is different from the second to fourth embodiments in that a part of the algorithm of the driving support device 1 is changed, and other parts are the same as the second to fourth embodiments, so only the parts different from the second to fourth embodiments will be described.

[0139] When a vehicle 10 to which the driving assistance device 1 of the sixth embodiment is applied approaches and passes a crossing pedestrian 15, the driving assistance device 1 learns the characteristics of the driver's avoidance driving, then determines a decrease in the driver's concentration on driving and issues an alert, as shown in Figures 31 to 34.

[0140] As shown in Fig. 31, there is a crossing pedestrian 15a as a dangerous object that is about to cross the path 11 of the vehicle 10 on the sidewalk 16 ahead of the path 11 (i.e., the driving lane) on which the vehicle 10 is traveling. According to the driver's avoidance driving under normal circumstances, the vehicle 10 starts braking before the crossing pedestrian 15a, stops temporarily before the crosswalk if there is one, and stops temporarily or slows down if there is no crosswalk, so as not to obstruct the passage of the crossing pedestrian 15a. If the TTC when the vehicle 10 starts braking before the crossing pedestrian 15a is equal to or greater than the time threshold value described in the first embodiment, it can be said that the driver is performing the avoidance driving under normal circumstances.

[0141] In this case, the learning unit 6 stores the TTC when the driver starts braking in front of the crossing pedestrian 15a as a characteristic of the driver's evasive driving under normal circumstances in a predetermined folder TTCf set in the memory, and calculates the TTC learning value TTCf_l based on that TTCf.

[0142] Next, Fig. 32 shows a scene after the host vehicle 10 has passed the scene of Fig. 31 and continued to travel, that is, a different scene from Fig. 31. As shown in Fig. 32, a crossing pedestrian 15b different from the one shown in Fig. 31 is about to cross the path 11 of the host vehicle 10 on the sidewalk 16 ahead of the path 11 on which the host vehicle 10 is traveling. In this case, the driver may become impatient and decrease his / her concentration on driving, resulting in a shorter TTC. If the TTC as a characteristic of the driver's avoidance driving is smaller than the TTC learning value TTCf_l obtained by learning from the scene shown in Fig. 31, the concentration determination unit 4 determines that the driver's concentration on driving has decreased.

[0143] Next, Fig. 33 shows a scene different from Fig. 31 and Fig. 32 after passing the scene of Fig. 32 and before a predetermined time (e.g., 30 minutes) has elapsed since passing the scene of Fig. 32. Alternatively, Fig. 33 may show a scene different from Fig. 31 and Fig. 32 after a scene in Fig. 32, in which it is determined that the driver's concentration on driving is decreasing, is repeated a predetermined number of times or more.

[0144] In Fig. 33, a crossing pedestrian 15c different from those shown in Fig. 31 and Fig. 32 is attempting to cross the path 11 of the vehicle 10 on the sidewalk 16 ahead of the path 11 on which the vehicle 10 is traveling. In this case, when the object detection unit 2 detects the crossing pedestrian 15c as a dangerous object, the notification unit 5 controls the in-vehicle speaker or the like to issue a notification encouraging safe driving when the TTC of the vehicle 10 for the crossing pedestrian 15c falls below a predetermined value. In this way, the driving support device 1 can prevent accidents by issuing a notification encouraging safe driving in advance before the vehicle 10 passes the crossing pedestrian 15c (i.e., while approaching).

[0145] Similarly to Fig. 33, Fig. 34 shows a scene different from Fig. 31 and Fig. 32 after passing the scene of Fig. 32 and before a predetermined time (e.g., 30 minutes) has elapsed since passing the scene of Fig. 32. Alternatively, Fig. 34 may show a scene different from Fig. 31 and Fig. 32 after a scene in Fig. 32, in which it is determined that the driver's concentration on driving is decreasing, is repeated a predetermined number of times or more.

[0146] In Fig. 34, a shoulder-parked vehicle 13c is parked on the shoulder 12 ahead of the path 11 along which the vehicle 10 is traveling. In this case, when the object detection unit 2 detects the shoulder-parked vehicle 13c as a dangerous object, the notification unit 5 controls the speaker in the vehicle cabin and the like to issue a notification encouraging safe driving when the TTC of the vehicle 10 to the shoulder-parked vehicle 13c falls below a predetermined value. In this way, the driving support device 1 can prevent accidents by issuing a notification encouraging safe driving well in advance before the vehicle 10 passes the shoulder-parked vehicle 13c (i.e., while approaching the vehicle).

[0147] Next, an example of an algorithm executed by the driving support device 1 according to the sixth embodiment will be described with reference to Fig. 35, Fig. 36, etc. This algorithm is repeatedly executed at a predetermined control cycle interval (for example, at 100 ms intervals).

[0148] In S100 of FIG. 35, the object detection unit 2 detects dangerous objects present on the path of the host vehicle 10 and around it.

[0149] Next, in S200, the avoidance driving detection unit 3 detects the characteristics of the driver's avoidance driving against a dangerous object. The process of S200 is the same as that described in the second to fourth embodiments with reference to FIG. 11, FIG. 17, or FIG. 23.

[0150] Next, in S300, the concentration level determination unit 4 determines whether the driver's concentration level with respect to driving has decreased. As the process of S300, first, the process described in the second to fourth embodiments with reference to Fig. 12, Fig. 18 or Fig. 24 is executed.

[0151] 36, the concentration level determination unit 4 in S300 holds the determination result that "the driver's concentration level on driving is not decreasing" under a predetermined condition in S361. Also, the concentration level determination unit 4 in S300 holds the determination result that "the driver's concentration level on driving is decreasing" under a predetermined condition in S361.

[0152] The predetermined condition for holding the determination result that "the driver's concentration on driving is not decreasing" in S361 is that the determination that the characteristics of the driver's avoidance driving detected by the avoidance driving detection unit 3 satisfy the characteristics of the driver's avoidance driving learned by the learning unit 6 has been made a predetermined number of times Y or more in succession. Note that the predetermined number of times Y is set to multiple times.

[0153] Alternatively, the specified condition for retaining the determination result that "the driver's concentration on driving is not decreasing" in S361 may be that a specified time (e.g., 30 minutes) has elapsed since the start of retaining the determination result that "the driver's concentration on driving is decreasing."

[0154] On the other hand, the predetermined condition for holding the determination result that "the driver's concentration on driving is decreasing" in S362 is that the determination that the characteristics of the driver's avoidance driving detected by the avoidance driving detection unit 3 do not satisfy the characteristics of the driver's avoidance driving learned by the learning unit 6 has been made a predetermined number of times or more consecutively. The predetermined number of times X is also set to multiple times. Note that the predetermined number of times X and Y may be the same value or different values.

[0155] Next, in S400, if all of the following requirements (A), (B), and (C) are met, the notification unit 5 issues a notification to encourage safe driving and proceeds to S500. On the other hand, if any one of the following requirements (A), (B), and (C) is not met, the notification unit 5 does not issue a notification and proceeds to S500.

[0156] (A) The object detection unit 2 detects a dangerous object (for example, a vehicle 13 parked on the side of the road or a pedestrian 15 crossing the road). (B) The concentration level determination unit 4 holds a determination result that "the driver's concentration level on driving is decreasing." (C) The TTC for the hazardous object falls below a predetermined value. The predetermined value is set, for example, as a value that allows the driver to avoid contact with the hazardous object by performing evasive driving after the notification.

[0157] Next, in S500, the learning unit 6 learns the characteristics of the driver's avoidance driving under normal circumstances. The process of S500 is the same as that described with reference to FIG. 13, FIG. 19 or FIG. 25 in the second to fourth embodiments. After S500, the process is temporarily terminated, and then the process is repeated from S100 again at a predetermined control period.

[0158] The driving support device 1 of the sixth embodiment described above provides the following advantageous effects. (1) In the sixth embodiment, the concentration level determination unit 4 can hold the determination result that the driver's concentration level on driving is decreasing for a predetermined period of time. When the concentration level determination unit 4 holds the determination result that "the driver's concentration level on driving is decreasing", the notification unit 5 issues a notification to encourage safe driving before the vehicle 10 passes the dangerous object when the object detection unit 2 detects a dangerous object. Note that "before the vehicle 10 passes the object" refers to, for example, when the TTC for the dangerous object becomes equal to or less than a predetermined value. According to this, when the driver's concentration on driving is decreasing, the alarm unit 5 can prevent accidents by issuing an alarm encouraging safe driving well before the vehicle 10 passes through a hazardous object (i.e., while approaching the object). The predetermined time can be set arbitrarily, for example, to 10 minutes, 30 minutes, 1 hour, or until the travel switch is turned off.

[0159] (2) In the sixth embodiment, the concentration level determination unit 4 can hold a determination result that "the driver's concentration level on driving is decreasing" under a predetermined condition. The predetermined condition is that the determination that the characteristics of the driver's avoidance driving detected by the avoidance driving detection unit 3 do not satisfy the characteristics of the driver's avoidance driving learned by the learning unit 6 is made consecutively for a predetermined number of times X or more. In addition, the concentration level determination unit 4 can hold a determination result that "the driver's concentration level on driving is not decreasing" under a predetermined condition. The predetermined condition is that the determination that the characteristics of the driver's avoidance driving detected by the avoidance driving detection unit 3 satisfy the characteristics of the driver's avoidance driving learned by the learning unit 6 is made consecutively for a predetermined number of times Y or more. If the concentration determination unit 4 holds a determination result that "the driver's concentration on driving is decreasing," when the object detection unit 2 detects an object, the notification unit 5 issues a notification encouraging safe driving before the vehicle 10 passes over the object. According to this, when the driver's concentration on driving is decreasing, the alarm unit 5 can prevent accidents by issuing an alarm encouraging safe driving well before the vehicle 10 passes through a hazardous object (i.e., while approaching the object).

[0160] Seventh embodiment A seventh embodiment will be described. The seventh embodiment is a modification of the sixth embodiment.

[0161] When a host vehicle 10 to which the driving assistance device 1 of the seventh embodiment is applied approaches and passes a vehicle 13 parked on the side of the road, the driving assistance device 1 learns the characteristics of the driver's evasive driving, then determines a decrease in the driver's concentration on driving and issues an alert, as shown in Figures 37 to 40.

[0162] As shown in FIG. 37, a shoulder-parked vehicle 13a is parked as a hazardous object on the shoulder 12 ahead of the path 11 (i.e., the driving lane) on which the vehicle 10 is traveling. According to the driver's avoidance driving under normal circumstances, the vehicle 10 decelerates before the shoulder-parked vehicle 13a and passes the shoulder-parked vehicle 13a while keeping the lateral distance Lm between the vehicle 10 and the shoulder-parked vehicle 13a as large as possible. If the TTC when the vehicle 10 starts braking before the shoulder-parked vehicle 13a is equal to or greater than the time threshold value described in the first embodiment, the driver's avoidance driving under normal circumstances can be said to be. Note that if the lateral distance Lm when the vehicle 10 passes the shoulder-parked vehicle 13a is equal to or greater than the distance threshold value described in the first embodiment, the driver's avoidance driving under normal circumstances can be said to be.

[0163] In this case, the learning unit 6 stores the TTC when the driver starts braking in front of the shoulder-parked vehicle 13a as a feature of the driver's avoidance driving in TTCf, and calculates the TTC learning value TTCf_l based on the TTCf. The learning unit 6 may also store the lateral distance Lm when the host vehicle 10 passes the shoulder-parked vehicle 13a as a feature of the driver's avoidance driving in normal times, and calculate the lateral distance learning value Lm_l.

[0164] Next, FIG. 38 shows a scene after the host vehicle 10 has passed the scene of FIG. 37 and continued traveling, that is, a different scene from FIG. 37. As shown in FIG. 38, a shoulder-parked vehicle 13b different from the one shown in FIG. 37 is parked on the shoulder 12 ahead of the path 11 on which the host vehicle 10 is traveling. In this case, the driver may become impatient and decrease his / her concentration on driving, resulting in a shorter TTC. If the TTC as a characteristic of the driver's avoidance driving is smaller than the TTC learning value TTCf_l obtained by learning from the scene shown in FIG. 37, the concentration determination unit 4 determines that the driver's concentration on driving has decreased.

[0165] Next, Fig. 39 shows a scene different from Fig. 37 and Fig. 38 after passing the scene of Fig. 38 and before a predetermined time (e.g., 30 minutes) has elapsed since passing the scene of Fig. 38. Alternatively, Fig. 39 may show a scene different from Fig. 37 and Fig. 38 after a scene in which it is determined that the driver's concentration on driving is decreasing as in Fig. 38 is repeated a predetermined number of times or more.

[0166] In Fig. 39, a crossing pedestrian 15c is about to cross the path of the vehicle 10 on the sidewalk 16 ahead of the path 11 on which the vehicle 10 is traveling. In this case, when the object detection unit 2 detects the crossing pedestrian 15c as a dangerous object, the notification unit 5 controls the speaker in the vehicle and the like to issue a notification encouraging safe driving when the TTC for the dangerous object falls below a predetermined value. In this way, the driving support device 1 can prevent accidents by issuing a notification encouraging safe driving in advance before the vehicle 10 passes the crossing pedestrian 15c (i.e., while approaching).

[0167] Similarly to Fig. 39, Fig. 40 shows a scene different from Fig. 37 and Fig. 38 after passing the scene of Fig. 38 and before a predetermined time (e.g., 30 minutes) has elapsed since passing the scene of Fig. 38. Alternatively, Fig. 40 may show a scene different from Fig. 37 and Fig. 38 after a scene in Fig. 38, in which it is determined that the driver's concentration on driving is decreasing, is repeated a predetermined number of times or more.

[0168] In Fig. 40, a shoulder-parked vehicle 13c is parked on the shoulder 12 ahead of the path 11 along which the vehicle 10 is traveling. In this case, when the object detection unit 2 detects the shoulder-parked vehicle 13c as a dangerous object, the notification unit 5 controls the in-vehicle speaker or the like to issue a notification encouraging safe driving when the TTC of the vehicle 10 for the dangerous object falls below a predetermined value. In this way, the driving support device 1 can prevent accidents by issuing a notification encouraging safe driving well in advance before the vehicle 10 passes the shoulder-parked vehicle 13c (i.e., while approaching the vehicle 10).

[0169] The driving assistance device 1 of the seventh embodiment described above also achieves the same effects as those of the sixth embodiment.

[0170] (Other embodiments) (1) In each of the above embodiments, the number of times the learning value is updated is explained as being reset when the vehicle's driving switch is turned off. However, this is not limited to this, and the number of times the learning value is updated may be reset by a specified switch operation or at a specified timing, or may not be reset.

[0171] (2) In each of the above embodiments, the moving average method is used to update the learned value. However, the present invention is not limited to this. For example, various averaging methods such as the grand average method, a predetermined function, the median, the mode, etc. may be used.

[0172] (3) In each of the above embodiments, the driving assistance device has been described as being configured by an electronic control unit mounted on a vehicle, but is not limited thereto. For example, the driving assistance device may assist the driver in driving through communication between a server and the vehicle.

[0173] (4) In each of the above embodiments, the driving assistance device 1 has been described as being configured by one electronic control device mounted on the vehicle, but the present invention is not limited to this and may be configured by multiple electronic control devices.

[0174] The present disclosure is not limited to the above-mentioned embodiments, and can be appropriately modified within the scope of the claims. In addition, the above-mentioned embodiments and parts thereof are not unrelated to each other, and can be appropriately combined except when the combination is clearly impossible. In addition, in each of the above-mentioned embodiments, it goes without saying that the elements constituting the embodiment are not necessarily essential, except when it is specifically stated that they are essential or when it is clearly considered to be essential in principle. In addition, in each of the above-mentioned embodiments, when the numbers, values, amounts, ranges, etc. of the components of the embodiment are mentioned, they are not limited to the specific numbers, except when it is specifically stated that they are essential or when it is clearly limited to a specific number in principle. In addition, in each of the above-mentioned embodiments, when the shape, positional relationship, etc. of the components are mentioned, they are not limited to the shape, positional relationship, etc., except when it is specifically stated that they are essential or when it is clearly limited to a specific shape, positional relationship, etc. in principle.

[0175] The control unit and the method described in the present disclosure may be implemented by a special-purpose computer provided by configuring a processor and a memory programmed to execute one or more functions embodied in a computer program. Alternatively, the control unit and the method described in the present disclosure may be implemented by a special-purpose computer provided by configuring a processor with one or more dedicated hardware logic circuits. Alternatively, the control unit and the method described in the present disclosure may be implemented by one or more special-purpose computers configured by a combination of a processor and a memory programmed to execute one or more functions and a processor configured with one or more hardware logic circuits. The computer program may also be stored in a computer-readable non-transitory tangible recording medium as instructions executed by the computer.

[0176] (In view of the present disclosure) The present disclosure described above can be understood from the following viewpoints, for example. [First viewpoint] In the driving assistance device, an object detection unit (2) for detecting dangerous objects (13, 15) that may cause an accident on a path (11) of a host vehicle (10) and around the path (12, 16); an avoidance driving detection unit (3) that detects characteristics of avoidance driving in which a driver performs a driving operation to avoid the risk of an accident when the host vehicle approaches and passes the dangerous object detected by the object detection unit; a concentration level determination unit (4) that determines a concentration level of the driver on driving based on the characteristics of the avoidance driving of the driver detected by the avoidance driving detection unit; and a notification unit (5) that issues a notification to encourage safe driving when the concentration level determination unit determines that the driver's concentration level on driving has decreased. [Second viewpoint] A learning unit (6) that learns characteristics of the driver's avoidance driving under normal circumstances, The driving assistance device according to a first aspect, wherein the concentration determination unit compares the characteristics of the driver's avoidance driving during the normal operation learned by the learning unit with the characteristics of the driver's avoidance driving detected by the avoidance driving detection unit to determine that the driver's concentration on driving has decreased. [Third Perspective] The driving assistance device according to a second aspect, wherein the learning unit does not reflect the characteristics of the driver's avoidance driving at that time in the learning if the characteristics of the driver's avoidance driving do not satisfy a predetermined criterion, and reflects the characteristics of the driver's avoidance driving at that time in the learning if the characteristics of the driver's avoidance driving satisfy a predetermined criterion. [Fourth viewpoint] The driving assistance device according to any one of the first to third aspects, wherein the concentration determination unit determines a decrease in the driver's concentration level based on the driver's line of sight toward the hazardous object when the vehicle approaches the hazardous object, in addition to the characteristics of the driver's evasive driving toward the hazardous object. [Fifth viewpoint] The concentration level determination unit holds a determination result that "the driver's concentration level on driving is decreasing" for a predetermined period of time, A driving assistance device as described in any one of the first to fourth aspects, wherein when the concentration determination unit holds a determination result that "the driver's concentration on driving is decreasing," when the object detection unit detects the dangerous object, the notification unit issues a notification to encourage safe driving before the vehicle passes over the dangerous object. [Sixth viewpoint] The concentration determination unit is When the determination that the characteristics of the avoidance driving of the driver detected by the avoidance driving detection unit do not satisfy the characteristics of the avoidance driving of the driver learned by the learning unit is repeated for a predetermined number of times (X) or more, the determination result that "the driver's concentration on driving is decreasing" is retained, When the determination that the characteristics of the avoidance driving of the driver detected by the avoidance driving detection unit satisfy the characteristics of the avoidance driving of the driver learned by the learning unit is repeated for a predetermined number of times (Y) or more, the determination result that "the driver's concentration on driving is not decreasing" is retained, A driving assistance device as described in any one of the second to fourth aspects, wherein when the concentration determination unit holds a determination result that "the driver's concentration on driving is decreasing," when the object detection unit detects the dangerous object, the notification unit issues a notification to encourage safe driving before the vehicle passes over the dangerous object. [Seventh viewpoint] The driving assistance device according to any one of first to sixth aspects, wherein the characteristic of the driver's avoidance driving is a time to collision (TTC) when the driver starts braking in front of the hazardous object. [Eighth viewpoint] The driving assistance device according to any one of the first to sixth aspects, wherein the characteristic of the driver's avoidance driving is a lateral distance (Lm) between the host vehicle and the hazardous object when the host vehicle passes beside the hazardous object. [Explanation of symbols]

[0177] 2. Object detection section 3. Avoidance driving detection section 4 Concentration level determination section 5. Notification Department 10 Vehicle 11 Career Paths 12 Shoulder 13 Vehicles parked on the side of the road (hazardous objects) 15. Pedestrians crossing the road (hazardous objects) 16 Sidewalk

Claims

1. In the driving assistance device, an object detection unit (2) for detecting dangerous objects (13, 15) that may cause an accident on a path (11) of a vehicle (10) and around the path (12, 16); an avoidance driving detection unit (3) that detects characteristics of avoidance driving in which a driver performs a driving operation to avoid the risk of an accident when the vehicle approaches and passes the dangerous object detected by the object detection unit; a concentration level determination unit (4) that determines a concentration level of the driver on driving based on the characteristics of the avoidance driving of the driver detected by the avoidance driving detection unit; and a notification unit (5) that issues a notification to encourage safe driving when the concentration level determination unit determines that the driver's concentration level on driving has decreased.

2. A learning unit (6) that learns characteristics of the driver's avoidance driving under normal circumstances, 2. The driving assistance device according to claim 1, wherein the concentration determination unit compares the characteristics of the driver's avoidance driving during the normal operation learned by the learning unit with the characteristics of the driver's avoidance driving detected by the avoidance driving detection unit to determine that the driver's concentration on driving has decreased.

3. 3. The driving assistance device according to claim 2, wherein the learning unit does not reflect the characteristics of the driver's avoidance driving at that time in the learning if the characteristics of the driver's avoidance driving do not satisfy a predetermined criterion, and reflects the characteristics of the driver's avoidance driving at that time in the learning if the characteristics of the driver's avoidance driving satisfy a predetermined criterion.

4. The driving assistance device according to any one of claims 1 to 3, wherein the concentration determination unit determines a decrease in the driver's concentration level based on the driver's line of sight toward the hazardous object when the vehicle approaches the hazardous object, in addition to the characteristics of the driver's avoidance driving toward the hazardous object.

5. The concentration level determination unit holds a determination result that "the driver's concentration level on driving is decreasing" for a predetermined period of time, 4. The driving assistance device according to claim 1, wherein when the concentration level determination unit holds a determination result indicating that "the driver's concentration level on driving is decreasing," and when the object detection unit detects the dangerous object, the notification unit issues a notification to encourage safe driving before the vehicle passes over the dangerous object.

6. The concentration determination unit is When the determination that the characteristics of the avoidance driving of the driver detected by the avoidance driving detection unit do not satisfy the characteristics of the avoidance driving of the driver learned by the learning unit is repeated a predetermined number of times (X) or more, the determination result that "the driver's concentration on driving is decreasing" is retained, When the determination that the characteristics of the avoidance driving of the driver detected by the avoidance driving detection unit satisfy the characteristics of the avoidance driving of the driver learned by the learning unit is repeated a predetermined number of times (Y) or more, the determination result that "the driver's concentration on driving is not decreasing" is retained, 4. The driving assistance device according to claim 2, wherein when the concentration level determination unit holds a determination result indicating that "the driver's concentration level on driving is decreasing," when the object detection unit detects the dangerous object, the notification unit issues a notification encouraging safe driving before the vehicle passes over the dangerous object.

7. 4. The driving support device according to claim 1, wherein the characteristic of the driver's avoidance driving is a time to collision (TTC) when the driver starts braking in front of the hazardous object.

8. 4. The driving support device according to claim 1, wherein the characteristic of the driver's avoidance driving is a lateral distance (Lm) between the host vehicle and the dangerous object when the host vehicle passes beside the dangerous object.

Images