JP2020016950A - Collision determination device and collision determination method - Google Patents

Abstract

To provide a technique of reducing a situation where an own vehicle collides with another vehicle because of a lane change of the other vehicle.SOLUTION: A collision determination device for determining a possibility of a collision of a vehicle includes: a detection section for detecting another vehicle existing at a front side of an own vehicle on the basis of a captured image from an imaging section mounted on the own vehicle; and a determination section for determining whether or not a collision between the own vehicle and the other vehicle is in a predicted dangerous state on the basis of position relation between the detected other vehicle and the own vehicle and time change information of the position relation.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a collision determination device and a collision determination method for a vehicle.

  2. Description of the Related Art Conventionally, there has been known an apparatus that detects a blind spot area of another vehicle and appropriately supports traveling of the own vehicle. For example, the driving support device disclosed in Patent Document 1 discloses a mirror camera that captures an image of another vehicle existing outside the host vehicle and a driver of the other vehicle from an image of the other vehicle obtained by the imaging of the mirror camera. A head detection unit that detects, a blind spot area detection unit that detects a blind spot area of another vehicle based on the detection result by the head detection unit, and a case where the own vehicle exists in the blind spot area detected by the blind spot area detection unit. A vehicle control unit and an alarm device that notify an occupant of the host vehicle that the host vehicle is in the blind spot area.

JP 2009-211309 A

  When the own vehicle exists in the blind spot area of the other vehicle, there is a high possibility that the other vehicle changes lanes without noticing the own vehicle. For this reason, when it is determined that the own vehicle exists in the blind spot area of the other vehicle as in Patent Literature 1, notifying the occupant of the own vehicle to that effect is effective for preventing collision. However, even when the own vehicle does not exist in the blind spot area, the other vehicle may suddenly change lanes and cause a collision between the other vehicle and the own vehicle.

  The present invention has been made in view of the above circumstances, and has as its object to provide a technique capable of reducing a situation in which a host vehicle collides with another vehicle due to a lane change of another vehicle.

  In order to achieve the above object, a collision judging device of the present invention is a collision judging device for judging a possibility of a collision of a vehicle, the collision judging device being mounted on a front side of the own vehicle based on a photographed image from a photographing unit mounted on the own vehicle. A detection unit that detects another vehicle existing in the vehicle, and a collision between the own vehicle and the other vehicle is predicted based on the detected positional relationship between the other vehicle and the own vehicle and time change information of the positional relationship. (A first configuration) that includes a determination unit that determines whether the vehicle is in a dangerous state.

  In the collision determination device according to the first configuration, the determination unit determines whether or not the vehicle is in the danger state and, when the other vehicle changes lanes to a lane in which the own vehicle runs, the own vehicle and the other vehicle. It is preferable to adopt a configuration (second configuration) for determining that the vehicle is in the dangerous state when it is expected that the vehicle will collide with the vehicle.

  In the collision determination device according to the first or second configuration, the determination unit may determine whether the vehicle is in the danger state by determining that the relationship between the host vehicle and the other vehicle in a left-right direction is a predetermined period. A configuration (third configuration) for checking whether or not the process has been continued may be employed.

  In the collision judging device according to any one of the first to third configurations, the judging unit is configured to determine whether or not the vehicle is in the danger state by a predetermined rearward position in which the own vehicle is set behind the other vehicle. A configuration (fourth configuration) for confirming whether or not the relationship of approaching the other vehicle beyond the threshold distance has continued for a certain period of time may be employed.

  In the collision determination device according to any one of the first to fourth configurations, the determination unit may be configured to determine whether or not the vehicle is in the danger state, wherein the determination unit is closer than a predetermined forward threshold distance set in front of the host vehicle. (Fifth configuration) for confirming whether or not the relationship that the other vehicle located at a distance from the own vehicle has continued for a certain period of time.

  In the collision determination device according to any one of the first to fifth configurations, the determination unit may determine whether or not the vehicle is in the dangerous state based on a speed relationship between the other vehicle and the host vehicle. 6).

  In the collision determination device according to any one of the first to sixth configurations, the determination unit may determine whether or not the vehicle is in the danger state based on an acceleration relationship between the other vehicle and the host vehicle. 7 configuration).

  In the collision determination device according to any one of the first to seventh configurations, the determination unit may determine whether the vehicle is in the dangerous state based on the presence or absence of an object in front of the other vehicle (eighth configuration). Configuration).

  In the collision determination device according to any one of the first to eighth configurations, the determination unit may determine whether the vehicle is in the danger state based on a state of a turn signal of the other vehicle (a ninth configuration). Configuration).

  In the collision determination device according to any one of the first to ninth configurations, the determination unit may recognize the positional relationship based on at least one of a rear end of the other vehicle, a side mirror, and a tire. (Tenth configuration) is preferable.

  In the collision determination device according to the tenth configuration, the determination unit recognizes the positional relationship based on a side mirror of the other vehicle and determines whether the vehicle is in the danger state based on an image reflected on the side mirror. (Eleventh configuration).

  The collision determination device according to any one of the first to eleventh configurations further includes a notification processing unit that operates a notification unit that notifies a occupant of the host vehicle of danger when it is determined that the vehicle is in the danger state. (Twelfth configuration) is preferable.

  In the collision determination device having the twelfth configuration, the notification processing unit may have a configuration (a thirteenth configuration) in which an operation of the notification unit is changed according to the level of the danger.

  In the collision determination device according to the twelfth or thirteenth configuration, the notification processing unit controls the notification unit when it is determined that the dangerous state occurs simultaneously on the left front side and the right front side. A configuration (a fourteenth configuration) may be used in which the left and right danger states are simultaneously distinguished and notified.

  In the collision determination device according to the twelfth or thirteenth configuration, the notification processing unit controls the notification unit when it is determined that the dangerous state occurs simultaneously on the left front side and the right front side. A configuration (15th configuration) for giving priority to the higher danger level and notifying may be used.

  In order to achieve the above object, a collision determination method according to the present invention is a collision determination method performed by a collision determination device that determines a possibility of a vehicle collision. A detection step of detecting another vehicle existing on the front side of the own vehicle based on the own vehicle and the other vehicle based on the detected positional relationship between the other vehicle and the own vehicle and time change information of the positional relationship. (A sixteenth configuration) comprising: a determination step of determining whether or not the vehicle is in a danger state in which a collision is expected.

  ADVANTAGE OF THE INVENTION According to this invention, the situation where the own vehicle and another vehicle collide due to the lane change of another vehicle can be reduced.

FIG. 1 is a diagram illustrating a configuration example of a driving support system to which a collision determination device according to the present invention is applied. 9 is a flowchart illustrating an example of a process performed by the collision determination device. Diagram schematically showing the relationship between own vehicle and other vehicles 9 is a flowchart illustrating an example of a dangerous state determination process according to the first embodiment performed by the determination unit. Schematic diagram for explaining a dangerous state determination process according to the first embodiment. 9 is a flowchart illustrating a modified example of the dangerous state determination process according to the first embodiment. 9 is a flowchart illustrating an example of a dangerous state determination process according to the second embodiment performed by the determination unit. Schematic diagram for explaining a dangerous state determination process according to the second embodiment. 9 is a flowchart illustrating a modified example of the dangerous state determination process according to the second embodiment. 9 is a flowchart illustrating an example of a dangerous state determination process according to the third embodiment performed by the determination unit. Schematic diagram for explaining a dangerous state determination process according to the third embodiment. Flowchart showing a modified example of the dangerous state determination process according to the third embodiment.

  Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the drawings. In the following description, the direction in which the vehicle travels straight, and the direction from the driver's seat to the steering wheel is referred to as “forward”. The direction in which the vehicle travels straight, that is, the direction from the steering wheel to the driver's seat is referred to as “rearward direction”. Further, a direction which is perpendicular to the straight traveling direction of the vehicle and the vertical line and which is directed from the right side to the left side of the driver who is facing forward is referred to as “left direction”. In addition, the direction that is perpendicular to the straight traveling direction and the vertical line of the vehicle and that is directed from the left side to the right side of the driver who is facing forward is referred to as “right direction”. The left and right directions with respect to the front and the rear are referred to as "sideways". In addition, a vehicle equipped with the driving support system including the collision determination device of the present invention is referred to as “own vehicle”.

<1. Driving support system>
FIG. 1 is a diagram illustrating a configuration example of a driving support system SYS1 to which a collision determination device 1 according to the present invention is applied. The driving support system SYS1 of the present embodiment is a system for notifying the occupant of the own vehicle when there is a risk of collision with another vehicle. As shown in FIG. 1, the driving support system SYS1 includes a collision determination device 1, an imaging unit 2, a vehicle sensor 3, an in-vehicle network 4, and a notification unit 5. These are all mounted on the own vehicle.

  The collision determination device 1 is a device that determines the possibility of a vehicle collision. Specifically, the collision determination device 1 determines whether or not the own vehicle and the other vehicle are in a dangerous state with a high possibility of colliding when another vehicle changes lanes. Details of the collision determination device 1 will be described later.

  The photographing unit 2 is provided so as to be able to photograph another vehicle existing in front of the own vehicle. The other vehicles existing on the front side of the own vehicle include other vehicles that are at least partially located on the front side of the own vehicle and that are located on the left or right side of the own vehicle. The photographing unit 2 is configured by a camera mounted on the own vehicle. The photographing unit 2 may be composed of only a single camera, or may be composed of a plurality of cameras. The imaging unit 2 includes, for example, at least one of a front camera in which the front of the lens is arranged toward the front of the own vehicle and a side camera in which the front of the lens is arranged toward the side of the own vehicle. Is preferably performed. When the imaging unit 2 is configured to include a side camera, the imaging unit 2 includes a left side camera in which the front of the lens is arranged to the left and a right side camera in which the front of the lens is arranged to the right. It is preferable to be comprised including.

  The lens constituting the camera may be, for example, a fisheye lens having a horizontal angle of view exceeding 180 °. Further, the photographing unit 2 may be a camera or the like built in a vehicle-mounted device such as a drive recorder.

  The imaging unit 2 is connected to the collision determination device 1 by wire or wirelessly. That is, the photographing unit 2 is provided so that a photographed image can be input to the collision determination device 1. The collision determination device 1 determines the possibility of a collision between the own vehicle and another vehicle based on a captured image input from the imaging unit 2.

  The vehicle sensor 3 is a sensor mounted on the own vehicle. In the present embodiment, the vehicle sensor 3 includes a vehicle speed sensor and an acceleration sensor. The vehicle speed sensor is a sensor that detects the speed of the host vehicle. The acceleration sensor detects acceleration acting on the vehicle body of the host vehicle. The vehicle sensor 3 is connected to the collision determination device 1 via a vehicle-mounted network 4 such as a CAN (Controller Area Network). That is, the collision determination device 1 can acquire the speed information and the acceleration information of the own vehicle via the in-vehicle network 4.

  The notification unit 5 is connected to the collision determination device 1 by wire or wirelessly. The notification unit 5 notifies the occupant of the own vehicle of danger under the control of the collision determination device 1. The means by which the notification unit 5 notifies the danger is not particularly limited, but preferably includes, for example, at least one of sound, light, vibration, and screen display. The sound may include an alarm sound or a sound. The notification unit 5 may be, for example, a sound generating device, a light emitting device, a vibration generating device, a display device, or the like.

<2. Collision determination device>
<2-1. Basic configuration of collision determination device>
As shown in FIG. 1, the collision determination device 1 includes a control unit 11 and a storage unit 12. The control unit 11 controls the entire collision determination device 1. The control unit 11 is configured by, for example, a computer including a CPU (Central Processing Unit), a RAM (Random Access Memory), and a ROM (Read Only Memory). The storage unit 12 stores computer programs and data necessary for the operation of the control unit 11 in a nonvolatile manner. As the storage unit 12, for example, an electrically erasable programmable read-only memory (EEPROM), a flash memory, or the like can be used.

  The image acquisition unit 111, the detection unit 112, the determination unit 113, and the notification processing unit 114 illustrated in FIG. 1 are functions of the control unit 11 realized by executing a computer program stored in the storage unit 12. In other words, the collision determination device 1 includes an image acquisition unit 111, a detection unit 112, a determination unit 113, and a notification processing unit 114.

  At least one of the functional units 111 to 114 included in the control unit 11 may be configured by hardware such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array). The function units 111 to 114 included in the control unit 11 are conceptual components. The functions performed by one component may be distributed to a plurality of components, or the functions of the plurality of components may be integrated into one component.

  The image acquiring unit 111 acquires an analog or digital photographed image from the photographing unit 2 temporally continuously at a predetermined cycle (for example, 1/30 second cycle). Then, when the acquired temporally continuous captured images are analog, the image acquiring unit 111 converts the analog captured images into digital captured images (A / D conversion).

  The detection unit 112 detects another vehicle existing on the front side of the own vehicle based on the captured image from the imaging unit 2 mounted on the own vehicle. Specifically, the detection unit 112 detects another vehicle based on the digital photographed image converted by the image acquisition unit 111. The detecting unit 112 detects other vehicles existing on the left front side and the right front side. The detection unit 112 may detect another vehicle using an image captured by one camera, but may detect the other vehicle using a composite image obtained by combining captured images captured by a plurality of cameras. May be.

  The detection unit 112 performs image processing such as edge extraction processing. Note that an edge refers to a light-dark boundary of an image, for example, a contour of a subject corresponds to the edge. The detection unit 112 obtains a change in the lightness and darkness of the image by obtaining a difference in density gradation between each pixel of the captured image and an adjacent pixel, and extracts a change point of the lightness and darkness that is equal to or more than a certain value as an edge. The detection unit 112 detects another vehicle by comparing the extracted edge shape with the pattern data of the edge shape stored in the storage unit 12. The edge shape used for detecting another vehicle may be the entire shape of the vehicle, or may be a shape of a part of the vehicle. The shape of a part of the vehicle may be, for example, a shape of a rear portion of the vehicle, a shape of a side mirror, a shape of a tire, and the like.

  The determining unit 113 determines whether or not the vehicle is in a dangerous state based on the detected positional relationship between the other vehicle and the own vehicle and the time change information of the positional relationship. The dangerous state is a state in which a collision between the own vehicle and another vehicle is expected. More specifically, the danger state is a state in which the host vehicle cannot be expected to avoid a collision with another vehicle due to a sudden lane change of the other vehicle. In the present embodiment, when determining whether or not the vehicle is in a danger state, the determination unit 113 may determine that a collision between the own vehicle and the other vehicle is expected if the other vehicle changes lanes to the lane in which the own vehicle runs. It is determined that it is in the state. By providing the determination unit 113, it is possible to notify the occupant of the own vehicle that the own vehicle is in a dangerous state in relation to the other vehicle on the front side. Further, when the determination unit 113 is provided, when the own vehicle is provided with an automatic driving function, it is possible to cause the self-vehicle to perform a departure action from the dangerous state using the automatic driving function.

  The notification processing unit 114 operates the notification unit 5 that notifies the occupant of the own vehicle of danger when the determination unit 113 determines that the vehicle is in a dangerous state. That is, the notification processing unit 114 controls the notification unit 5 to notify an alarm when there is a danger. Note that the notification processing unit 114 does not operate the notification unit 5 when the determination unit 113 does not determine that the vehicle is in a dangerous state. By providing the notification processing unit 114, the driver driving the own vehicle can easily know that the own vehicle is exposed to a dangerous state by another vehicle existing on the front side, and the other vehicle On the other hand, the own vehicle can be moved to a safe position.

  FIG. 2 is a flowchart illustrating an example of a process performed by the collision determination device 1. When the operation start event occurs, the collision determination device 1 starts the operation of the flowchart shown in FIG. Examples of the operation start event include a case where the ignition switch is turned on, a case where the own vehicle starts moving, a case where the speed of the own vehicle becomes equal to or higher than a predetermined speed, and the like.

  First, the detection unit 112 sequentially performs a process for detecting another vehicle existing on the front side of the own vehicle with respect to the photographed image input from the photographing unit 2 at a predetermined cycle. Then, the detection unit 112 monitors whether or not another vehicle existing on the front side of the own vehicle is detected (step S1). In the present embodiment, the front side of the host vehicle includes the left front side and the right front side. That is, when another vehicle is detected on at least one of the left front side and the right front side of the own vehicle, it means that another vehicle existing on the front side of the own vehicle is detected.

  When the detecting unit 112 detects another vehicle existing on the front side of the own vehicle (Yes in step S1), the determining unit 113 determines whether or not the vehicle is in a dangerous state (step S2). The details of the process of determining whether or not the vehicle is in a dangerous state will be described later. When other vehicles existing on the left front side and the right front side of the own vehicle are detected, it is determined whether each of the two detected other vehicles is in a dangerous state.

  When the determination unit 113 determines that the vehicle is in a dangerous state (Yes in step S2), the notification processing unit 114 performs a notification process (step S3). Specifically, the notification processing unit 114 performs control processing of the notification unit 5 and causes the notification unit 5 to operate. By the notification process, the notification unit 5 notifies the occupant of the host vehicle of an alarm indicating that the vehicle is in a dangerous state. The notification operation of the notification unit 5 is continued, for example, until the dangerous state is resolved.

  In addition, with respect to the other vehicles existing on the left front side and the right front side of the own vehicle, both may be determined to be in a dangerous state. An example of the notification process in this case will be described with reference to FIG. FIG. 3 is a diagram schematically illustrating the relationship between the host vehicle V and the other vehicle OV. In FIG. 3, the host vehicle V is traveling in the middle lane LN2 of the three lanes. The first other vehicle OV1 travels on the leftmost lane LN1 of the three lanes, and is located on the left front side of the host vehicle V. The second other vehicle OV2 travels on the rightmost lane LN3 among the three lanes, and is located on the right front side of the host vehicle V. Lanes LN1 and LN2 are traveling lanes, and lane LN3 is an overtaking lane. The host vehicle V is determined to be in a dangerous state in relation to the first other vehicle OV1 and the second other vehicle OV2.

  As a first notification processing example, the notification processing unit 114 controls the notification unit 5 to simultaneously determine the left and right danger states when it is determined that the danger state occurs simultaneously on the left front side and the right front side. Let them know separately. For example, the notification unit 5 is a light emitting device, and is disposed at the front left and front right inside the host vehicle V, respectively. When it is determined that a dangerous state occurs simultaneously on the left front side and the right front side, the notification processing unit 114 turns on or blinks the left and right light emitting devices. The light emitting device is merely an example, and the notification processing unit 114 may notify that a dangerous state has occurred on the left and right front sides by, for example, sound or screen display. According to the example of the notification process, the occupant of the host vehicle can easily grasp the danger state occurring simultaneously on the left and right sides, and can take appropriate measures for the danger state.

  As a second notification processing example, the notification processing unit 114 controls the notification unit 5 to determine a higher danger level when it is determined that a dangerous state occurs simultaneously on the left front side and the right front side. Give priority to inform. In the example in which the notification unit 5 is the above-described light emitting device, the notification processing unit 114 turns on or blinks only the light emitting device corresponding to the one determined to have the higher danger level among the left and right danger states. Also in this case, the notification unit 5 may be other than the light emitting device. When it is determined that a dangerous state occurs at the left front side and the right front side at the same time, the driver of the own vehicle V is panicked with the dangerous state by giving priority to the side with the higher risk level. It is possible to deal calmly without.

  The danger level may be determined based on, for example, the type of lane in which the other vehicle OV travels. In the example illustrated in FIG. 3, for example, it may be determined that the other vehicle OV traveling in one of the overtaking lane LN3 and the traveling lane LN1 has a higher danger level. For example, since the first other vehicle OV1 traveling on the traveling lane LN1 may suddenly change lanes at a merging point or the like, the traveling lane LN1 side may be determined to have a high danger level.

  The danger level may be determined according to the positional relationship between the host vehicle V and the other vehicle OV. For example, it may be determined that the dangerous level is higher on the other vehicle OV side where the host vehicle V has more portions arranged in the left-right direction with the other vehicle OV. In the example shown in FIG. 3, in the host vehicle V, the first other vehicle OV1 has more portions arranged in the left-right direction than the second other vehicle OV2. For this reason, it is determined that the danger level is higher on the first other vehicle OV1 side.

  The danger level may be determined according to the movement of the other vehicle OV. For example, the side on which another vehicle OV that is performing a movement different from the expected movement according to the traveling lane is determined to have a high danger level. In the example shown in FIG. 3, a case where the first other vehicle OV1 running on the traveling lane LN1 is moving away, or a case where the second other vehicle OV2 running on the passing lane LN3 is moving approaching In addition, the side on which the other vehicle OV runs may be determined to have a high danger level.

  Note that, in the case of the first notification processing example, the notification processing unit 114 may be configured to make the notification such that the higher the danger level is more noticeable than the lower danger level. For example, in the example in which the notification unit 5 is the light emitting device described above, a configuration may be adopted in which the side with a higher danger level blinks and the one with a lower danger level is simply turned on.

  Returning to FIG. 2, after the notification processing by the notification processing unit 114, it is confirmed whether or not an operation end event has occurred (step S4). Examples of the operation end event include a case where the ignition switch is turned off, a case where the host vehicle stops, and a case where the speed of the host vehicle becomes slower than a predetermined speed.

  If the operation end event has not occurred (No in step S4), the process returns to step S1, and the processes after step S1 are repeated. On the other hand, when the operation end event has occurred (Yes in step S4), the collision determination device 1 ends the operation of the flowchart illustrated in FIG.

<2-2. Details of dangerous state judgment processing>
<2-2-1. First Embodiment>
FIG. 4 is a flowchart illustrating an example of a dangerous state determination process according to the first embodiment performed by the determination unit 113. FIG. 4 is a detailed example of step S2 in FIG. FIG. 5 is a schematic diagram for explaining the dangerous state determination process according to the first embodiment.

  As shown in FIG. 4, first, the determination unit 113 checks whether or not the host vehicle V has a portion aligned with the other vehicle OV in the left-right direction (step S11). The determination unit 113 performs the confirmation based on the captured image acquired from the imaging unit 2. Note that the determination unit 113 may perform the confirmation using a captured image captured by one camera, but may perform the confirmation using a composite image obtained by combining captured images captured by a plurality of cameras. Is also good.

  In the present embodiment, the determination unit 113 recognizes the positional relationship between the host vehicle V and the other vehicle OV based on the rear end of the other vehicle OV. The position of the rear end of the other vehicle OV can be obtained by, for example, edge extraction processing, pattern matching processing, or the like. When it is determined that the front end of the own vehicle V is ahead of the rear end of the other vehicle OV based on the captured image, the determination unit 113 determines that the own vehicle V has a portion aligned with the other vehicle OV in the left-right direction. to decide. In FIG. 5, the host vehicle V has a portion arranged in the left-right direction with the other vehicle OV. Note that the camera constituting the photographing unit 2 is fixed at a predetermined position of the own vehicle, and the coordinates of each pixel in the image taken by the camera correspond to the coordinates of the own vehicle V in an actual space on a one-to-one basis. ing. For this reason, the positional relationship between the host vehicle V and the other vehicle OV can be obtained based on the captured image.

  In the present embodiment, the determination unit 113 is configured to recognize the positional relationship between the host vehicle V and the other vehicle OV based on the rear end of the other vehicle OV, but this is merely an example. For example, the determination unit 113 may be configured to recognize the positional relationship between the own vehicle V and the other vehicle OV based on a side mirror or tires of the other vehicle OV instead of the rear end of the other vehicle OV. Further, for example, the determination unit 113 may be configured to recognize the positional relationship between the host vehicle V and the other vehicle OV based on at least one of the rear end of the other vehicle OV, the side mirror, and the tire. The rear end of the other vehicle OV, the side mirrors, and the tires are easily detected by image processing, and the positional relationship between the host vehicle V and the other vehicle OV can be accurately detected.

  When the determination unit 113 determines that the host vehicle V has a portion aligned with the other vehicle OV in the left-right direction (Yes in step S11), the determination unit 113 checks whether the positional relationship has continued for a certain period of time (step S12). Based on the captured image used for the determination in step S11 and at least one captured image obtained thereafter, the determination unit 113 determines whether the relationship in which the host vehicle V has a portion arranged in the left-right direction with the other vehicle OV has continued for a certain period of time. Determine whether or not.

  If the determination unit 113 determines that the relationship in which the host vehicle V has a portion arranged in the left-right direction with the other vehicle OV has continued for a certain period of time (Yes in step S12), the determination unit 113 determines that the vehicle is in a dangerous state (step S13). On the other hand, when it is not determined in step S11 that the vehicle V has a portion aligned with the other vehicle OV in the left-right direction (No in step S11), the determination unit 113 determines that the vehicle V has continued for a certain period of time in step S12. If not performed (No in step S12), it is determined that the vehicle is not in a dangerous state (step S14).

  As described above, in the present embodiment, when determining whether or not the vehicle is in a danger state, the determination unit 113 determines whether or not the relationship in which the host vehicle V has a portion arranged in the left-right direction with the other vehicle OV has continued for a certain period of time. Confirm. According to this, not only the case where the own vehicle V exists in the blind spot area but also a case where a collision may occur due to a sudden lane change of the other vehicle OV can be widely regarded as a dangerous state. The safety of the vehicle V can be improved. Further, according to the configuration of the present embodiment, not only that the own vehicle V has a dangerous positional relationship with respect to the other vehicle OV, but also that the own vehicle V is in a dangerous state on condition that the dangerous positional relationship has continued for a certain period of time. to decide. For this reason, it is possible to prevent the notification unit 5 from operating even when the speed difference between the own vehicle V and the other vehicle OV is large and the vehicle is immediately avoided from the dangerous state. That is, it is possible to notify the occupant of the vehicle V of the danger state at an appropriate timing.

  In the present embodiment, the determination unit 113 is configured to determine only whether or not the vehicle is in a dangerous state. However, this is only an example. The determining unit 113 may be configured to determine the level of danger when it is determined to be in a danger state. For example, the risk level may be divided into a plurality of stages according to the length X (see FIG. 5, the length in the front-rear direction) of a portion of the host vehicle V arranged in the left-right direction with the other vehicle OV. For example, the longer the length X, the higher the risk level may be. Further, a configuration may be employed in which the danger level is divided into a plurality of stages according to information on the speed relationship between the own vehicle V and the other vehicle OV. When the determination unit 113 determines the level of danger, the notification processing unit 114 may change the operation of the notification unit 5 according to the level of danger. For example, as compared with the case where the level of danger is low, the notification sound may be louder, the light for notifying the danger blinks faster, or the color of the screen display for notifying the danger may be changed. According to this, the driver who drives the vehicle V can know how dangerous the vehicle is, and can perform an appropriate evacuation action according to the level of the risk.

  FIG. 6 is a flowchart illustrating a modified example of the dangerous state determination process according to the first embodiment. In the modification, the conditions for determining whether or not the vehicle is in a dangerous state are increased as compared with the case of the first embodiment. Specifically, after step S12 in FIG. 4, the number of processes (step S15) for checking whether or not a specific condition is satisfied is increased. That is, it is determined that the vehicle V is in the danger state only when a specific condition is satisfied in addition to the fact that the relationship that the host vehicle V has a portion arranged in the left-right direction with the other vehicle OV continues for a certain period. Note that the timing for executing step S15 is not limited to the timing shown in FIG. For example, step S15 may be performed before step S11.

  The specific condition may be one condition, or may be composed of a plurality of conditions. Specific conditions are exemplified below, but these may be used alone or in combination with others.

  The specific condition may be a condition relating to a speed relationship between the own vehicle V and the other vehicle OV. In other words, the determination unit 113 may determine whether or not the vehicle is in a dangerous state based on the speed relationship between the other vehicle OV and the own vehicle V. Note that the speed of the host vehicle V can be acquired from a vehicle speed sensor included in the vehicle sensor 3 via the in-vehicle network 4, for example. However, this is only an example, and may be calculated from, for example, a change over time of the position information obtained from the GPS sensor. Further, the speed of the other vehicle OV may be calculated from the speed of the own vehicle V and the relative speed of the other vehicle OV with respect to the own vehicle V obtained based on a plurality of temporally continuous captured images. However, when the own vehicle V includes a measuring device such as a radar device that measures the speed information of the other vehicle OV, the speed of the other vehicle OV may be obtained from information from the measuring device.

  For example, the specific condition may be a condition that the speed between the host vehicle V and the other vehicle OV is equal to or higher than a predetermined speed. According to this, it is possible to notify the alarm only when the own vehicle V and the other vehicle OV are emitting a certain speed. The specific condition may be a condition that the relative speed of the host vehicle V with respect to the other vehicle OV is lower than a predetermined speed. According to this, it is possible to issue an alarm only when it is expected that the time during which the own vehicle V exists at a dangerous position with respect to the other vehicle OV will be long.

  The specific condition may be a condition relating to an acceleration relationship between the own vehicle V and the other vehicle OV. In other words, the determination unit 113 may determine whether the vehicle is in a dangerous state based on the acceleration relationship between the other vehicle OV and the own vehicle V. Note that the acceleration of the host vehicle V can be acquired from, for example, an acceleration sensor included in the vehicle sensor 3 via the in-vehicle network 4. The acceleration of the host vehicle V may be obtained based on the speed obtained from the speed sensor. The acceleration of the other vehicle OV is obtained, for example, by obtaining a time change of the speed of the other vehicle OV obtained based on the vehicle speed sensor of the own vehicle V and the captured image. For example, the specific condition may be a condition that the acceleration between the host vehicle V and the other vehicle OV is equal to or higher than a predetermined acceleration. According to this, for example, the warning can be issued only when the accelerator is simultaneously depressed by the own vehicle V and the other vehicle OV and the danger of a collision accompanying the lane change is increased.

  The specific condition may be a condition regarding the presence or absence of an object in front of another vehicle OV. In other words, the determination unit 113 may determine whether the vehicle is in a dangerous state based on the presence or absence of an object in front of the other vehicle OV. The presence or absence of an object in front of the other vehicle OV may be determined based on a captured image, but may be determined based on information from an object detection device such as a radar device. The object may be, for example, a vehicle. The object may be a stationary object or a moving object. For example, the specific condition may be a condition that a vehicle exists in front of another vehicle OV. According to this, it is possible to issue an alarm only when there is a high possibility that another vehicle OV will change lanes due to the effect of a vehicle existing ahead.

<2-2-2. Second Embodiment>
FIG. 7 is a flowchart illustrating an example of a dangerous state determination process according to the second embodiment performed by the determination unit 113. FIG. 7 is a detailed example of step S2 in FIG. FIG. 8 is a schematic diagram illustrating a dangerous state determination process according to the second embodiment. The description of the contents overlapping with the first embodiment will be omitted unless it is particularly necessary.

  As shown in FIG. 7, first, the determination unit 113 checks whether or not the host vehicle V has exceeded a predetermined rear threshold distance d1 (see FIG. 8) set behind the other vehicle OV (step S21). . The determination unit 113 performs the confirmation based on the captured image acquired from the imaging unit 2. In the present embodiment, the rear threshold distance d1 is set assuming a distance from the rear end of the other vehicle OV to the front end of the host vehicle V. For this purpose, the determination unit 113 extracts the rear end of the other vehicle OV from the captured image, and obtains the distance between the front end of the host vehicle V and the rear end of the other vehicle OV. Then, when the distance becomes smaller than the rear threshold distance d1, the determination unit 113 determines that the host vehicle V has exceeded the rear threshold distance d1. In the example shown in FIG. 8, the host vehicle V exceeds the rear threshold distance d1. In the present embodiment, the determination unit 113 also confirms that the own vehicle V is away from the other vehicle OV based on the captured image.

  The method of setting the rear threshold distance d1 may be changed as appropriate. For example, the rear threshold distance d1 may be set assuming, for example, a side mirror or a tire of the other vehicle OV, instead of the rear end of the other vehicle OV. In this case, the determination unit 113 extracts a side mirror or a tire from the captured image instead of the rear end of the other vehicle OV, and determines whether the host vehicle V has exceeded the rear threshold distance d1.

  When the determination unit 113 determines that the host vehicle V has exceeded the rear threshold distance d1 (Yes in step S21), the determination unit 113 checks whether the host vehicle V is approaching another vehicle OV (step S22). The determination unit 113 confirms whether or not the own vehicle V is in a state of approaching the other vehicle OV based on the captured image used for the determination in step S21 and at least one captured image obtained after the captured image. In the present embodiment, a configuration is used to check whether or not the host vehicle V is approaching the other vehicle OV based on the rear end of the other vehicle OV, but this is an example. For example, a configuration may be used in which it is determined whether or not the own vehicle V is approaching the other vehicle OV based on a side mirror, tires, and the like of the other vehicle OV.

  When determining that the vehicle V is approaching the other vehicle OV (Yes in step S22), the determination unit 113 checks whether the relationship has continued for a certain period of time (step S23). Based on the captured image used for the determination in step S22 and at least one captured image obtained after the captured image, the determination unit 113 causes the own vehicle V to approach the other vehicle OV beyond the rear threshold distance d1. Check whether the status has continued for a certain period of time.

  When determining that the vehicle V has approached the other vehicle OV beyond the rear threshold distance d1 for a certain period of time (Yes in Step S23), the determining unit 113 determines that the vehicle is in a dangerous state (Step S24). . On the other hand, when it is not determined that the vehicle V has exceeded the rear threshold distance d1 in step S21 (No in step S21), the determination unit 113 is in a state where the vehicle V approaches the other vehicle OV in step S22. If it is not determined (No in step S22), and if it is not determined that it has continued for a certain period in step S23 (No in step S23), it is determined that the state is not dangerous (step S25).

  As described above, in the present embodiment, when determining whether or not the vehicle is in a danger state, the determination unit 113 determines that the own vehicle V exceeds the predetermined rear threshold distance d1 set behind the other vehicle OV, and the other vehicle It is checked whether the relationship approaching OV has continued for a certain period. According to this, not only the case where the own vehicle V exists in the blind spot area but also a case where a collision may occur due to a sudden lane change of the other vehicle OV can be widely regarded as a dangerous state. The safety of the vehicle V can be improved. Further, according to the configuration of the present embodiment, not only that the own vehicle V has a dangerous positional relationship with respect to the other vehicle OV, but also that the own vehicle V is in a dangerous state on condition that the dangerous positional relationship has continued for a certain period of time. to decide. For this reason, for example, the notification unit 5 can be prevented from operating even when the own vehicle V only momentarily approaches the other vehicle OV and the dangerous state is immediately avoided. That is, it is possible to notify the occupant of the vehicle V of the danger state at an appropriate timing.

  Note that, also in the present embodiment, the determination unit 113 may be configured to determine the level of danger when it is determined to be in a danger state. The notification processing unit 114 may change the operation of the notification unit 5 according to the level of danger. For example, a plurality of danger levels may be set based on the magnitude of the relative speed of the own vehicle V approaching the other vehicle OV with respect to the other vehicle OV. In this case, the risk level may be set higher when the relative speed is low.

  FIG. 9 is a flowchart illustrating a modified example of the dangerous state determination process according to the second embodiment. In the modification, the conditions for determining whether or not the vehicle is in a dangerous state are increased as compared with the case of the second embodiment. Specifically, after step S23 in FIG. 7, the number of processes (step S26) for checking whether or not the specific condition is satisfied is increased. That is, it is determined that the host vehicle V is in a dangerous state only when a specific condition is satisfied, in addition to the state where the host vehicle V approaches the other vehicle OV beyond the rear threshold distance d1 for a certain period of time. Note that the timing for executing step S26 is not limited to the timing shown in FIG. For example, step S26 may be performed before step S21 or step S22.

  The specific conditions include, as in the first embodiment, conditions relating to the speed relationship between the host vehicle V and the other vehicle OV, conditions relating to the acceleration relationship between the host vehicle V and the other vehicle OV, and the condition of the object in front of the other vehicle OV. At least one of the conditions related to existence may be included. The specific condition may be one or a combination of a plurality of conditions.

  In addition, the specific condition may be a condition relating to a state of a turn signal indicator (a blinker) of another vehicle OV. In other words, the determination unit 113 may determine whether or not the vehicle is in a dangerous state based on the state of the turn signal of the other vehicle OV. In the present embodiment, the state of the turn signal is whether or not the turn signal is in a lighting (flashing) state. The state of the direction indicator may be determined based on a plurality of captured images acquired successively in time. The determining unit 113 determines that the direction indicator is lit, for example, when detecting a local region in which the luminance changes in a predetermined cycle. For example, the specific condition may be a condition of detecting lighting of the turn signal of the other vehicle OV. According to this, the warning can be issued only when there is a high possibility that the other vehicle OV will change lanes. Note that the specific condition may be applied to the first embodiment in some cases.

  The specific condition may be a condition relating to an image reflected on a side mirror of another vehicle. In other words, the determination unit 113 may determine whether or not the vehicle is in a dangerous state based on an image reflected on a side mirror of another vehicle. In this case, the determination unit 113 recognizes the positional relationship between the own vehicle V and the other vehicle OV based on the side mirror of the other vehicle OV. The image reflected on the side mirror of another vehicle may be determined based on the captured image. For example, the specific condition may be a condition that a face is not reflected in a side mirror of another vehicle OV. Further, the specific condition may be a condition that only a part of the face is reflected on the side mirror of the other vehicle OV. Further, the specific condition may be a condition that an eye is not reflected on a side mirror of another vehicle OV. The face and eyes can be extracted by, for example, pattern matching. When the face or eyes are not reflected on the side mirror of the other vehicle OV, there is a possibility that the driver of the other vehicle OV does not recognize the own vehicle V. For this reason, it is possible to specify a case where there is a high possibility that a collision will occur due to a change in the lane of another vehicle OV, and to issue an alarm. Note that the specific condition may be applied to the first embodiment in some cases.

<2-2-3. Third Embodiment>
FIG. 10 is a flowchart illustrating an example of a dangerous state determination process according to the third embodiment performed by the determination unit 113. FIG. 10 is a detailed example of step S2 in FIG. FIG. 11 is a schematic diagram illustrating a dangerous state determination process according to the third embodiment. In the third embodiment, a dangerous state is determined in relation to another vehicle OV which may be performing an overtaking operation for changing lanes. The description of the contents overlapping with those of the first and second embodiments will be omitted unless particularly necessary.

  As shown in FIG. 10, first, the determination unit 113 checks whether or not the other vehicle OV is located closer than a predetermined forward threshold distance d2 (see FIG. 11) set ahead of the host vehicle V (see FIG. 10). Step S31). The determination unit 113 performs the confirmation based on the captured image acquired from the imaging unit 2. In the present embodiment, the predetermined front threshold distance d2 is set assuming a distance from the front end of the own vehicle V to the rear end of the other vehicle OV. For this purpose, the determination unit 113 extracts the rear end of the other vehicle OV from the captured image, and obtains the distance between the front end of the host vehicle V and the rear end of the other vehicle OV. When the distance is smaller than the front threshold distance d2, the determination unit 113 determines that the other vehicle OV is located closer to the front threshold distance d2. In the example shown in FIG. 11, the other vehicle OV is located closer than the forward threshold distance d2. In the present embodiment, the determination unit 113 also confirms that the other vehicle OV is separated from the own vehicle V forward based on the captured image.

  Note that the method of setting the forward threshold distance d2 may be changed as appropriate. For example, the front threshold distance d2 may be set assuming, for example, a side mirror or a tire of the other vehicle OV instead of the rear end of the other vehicle OV. In this case, the determination unit 113 extracts a side mirror or a tire from the captured image, not the rear end of the other vehicle OV, and determines whether the other vehicle OV is located closer to the front threshold distance d2. I do.

  When determining that the other vehicle OV is located closer than the forward threshold distance d2 (Yes in step S31), the determination unit 113 checks whether or not the other vehicle OV is away from the host vehicle V (step S32). . The determination unit 113 checks whether or not the other vehicle OV is in a state of being separated from the host vehicle V based on the captured image used for the determination in step S31 and at least one captured image obtained after the captured image. In the present embodiment, a configuration is used to check whether or not the host vehicle V is separated from the other vehicle OV based on the rear end of the other vehicle OV. However, this is an example. For example, a configuration may be used in which it is determined whether or not the other vehicle OV is moving away from the host vehicle V based on a side mirror or tires of the other vehicle OV.

  When determining that the other vehicle OV is moving away from the host vehicle V (Yes in step S32), the determination unit 113 checks whether the relationship has continued for a certain period of time (step S33). Based on the captured image used for the determination in step S32 and at least one captured image obtained after the captured image, the determination unit 113 separates the other vehicle OV located closer than the forward threshold distance d2 from the own vehicle V. It is checked whether or not the moving state has continued for a certain period.

  When the determining unit 113 determines that the state in which the other vehicle OV located closer to the front threshold distance d2 has left the own vehicle V has continued for a certain period of time (Yes in step S33), it determines that the vehicle is in a dangerous state (step S33). S34). On the other hand, when it is not determined in step S31 that the other vehicle OV is located closer than the forward threshold distance d2 (No in step S31), the determination unit 113 moves the other vehicle OV away from the host vehicle V in step S32. If it is not determined that the condition is (No in step S32), and if it is not determined that the process has continued for a certain period in step S33 (No in step S33), it is determined that the vehicle is not in a dangerous state (step S35).

  As described above, in the present embodiment, when determining whether or not the vehicle is in a danger state, the determination unit 113 determines whether the other vehicle OV located closer to the front threshold distance d2 set in front of the host vehicle V is closer to the vehicle. It is confirmed whether or not the state of leaving the host vehicle V has continued for a certain period. According to this, not only the case where the own vehicle V exists in the blind spot area but also a case where a collision may occur due to a sudden lane change of the other vehicle OV can be widely regarded as a dangerous state. The safety of the vehicle V can be improved. Further, according to the configuration of the present embodiment, not only that the own vehicle V has a dangerous positional relationship with respect to the other vehicle OV but also that the own vehicle V is in a dangerous state on condition that the dangerous positional relationship has continued for a certain period of time. to decide. For this reason, for example, it is possible to prevent the notification unit 5 from operating only by instantaneously detecting a state in which the speed of the other vehicle OV is increased for the lane change. That is, the notification unit 5 can be prevented from operating even when the speed of the other vehicle OV is sufficiently higher than the own vehicle V and the dangerous state is immediately avoided.

  Note that, also in the present embodiment, the determination unit 113 may be configured to determine the level of danger when it is determined to be in a danger state. The notification processing unit 114 may change the operation of the notification unit 5 according to the level of danger. For example, the danger level may be set in a plurality of levels based on the magnitude of the relative speed of the other vehicle OV moving away from the own vehicle V with respect to the own vehicle V. In this case, the risk level may be set higher when the relative speed is low.

  FIG. 12 is a flowchart illustrating a modified example of the dangerous state determination process according to the third embodiment. In the modification, the conditions for determining whether or not the vehicle is in a dangerous state are increased as compared with the case of the third embodiment. Specifically, after step S32 in FIG. 12, the number of processes (step S36) for checking whether or not a specific condition is satisfied has increased. That is, it is determined that the vehicle is in a dangerous state only when the specific condition is satisfied in addition to the state where the other vehicle OV located closer to the front threshold distance d2 is separated from the host vehicle V for a certain period of time. The timing at which step S36 is executed is not limited to the timing shown in FIG. For example, step S36 may be performed before steps S31 and S32.

  As in the first embodiment and the second embodiment, the specific conditions include a condition relating to a speed relationship between the own vehicle V and another vehicle OV, a condition relating to an acceleration relationship between the own vehicle V and another vehicle OV, and a condition ahead of the other vehicle OV. At least one of a condition relating to the presence or absence of an object, a condition relating to the state of the direction indicator of the other vehicle OV, and a condition relating to an image reflected on the side mirror.

<3. Notes>
Various technical features disclosed in the present specification can be variously modified without departing from the spirit of the technical creation in addition to the above-described embodiment. That is, the above-described embodiment is to be considered in all respects as illustrative and not restrictive. The technical scope of the present invention is defined not by the description of the embodiment but by the claims. It is to be understood that all changes that fall within the meaning and scope equivalent to the appended claims are included. In addition, a plurality of embodiments and modifications shown in this specification may be implemented in combination within a possible range.

  For example, in the above description, the collision determination device is provided with the notification processing unit and configured to control the notification unit. However, this is merely an example, and the collision determination device may simply determine that it is in a dangerous state. , The determination information may be transmitted to the notification unit. In this case, the collision determination device does not need to include the notification processing unit.

  In the above description, the driving support system including the collision determination device includes the notification unit. However, this is merely an example. The driving support system may be configured not to include the notification unit. In this case, the collision determination device does not need to include the notification processing unit. For example, the driving support system may be configured to include an automatic driving control device that controls automatic driving of the own vehicle, instead of the notification unit. In this case, the driving support system may be a system that automatically leaves the own vehicle from the dangerous state when the collision determination device determines that the own vehicle is in a dangerous state in which a collision with another vehicle is expected. Can be. The collision determination device may be configured to transmit a determination result as to whether or not the vehicle is in a dangerous state to the automatic operation control device. The automatic driving control device that has been input that the vehicle is in a dangerous state may perform a process of controlling the operation of the own vehicle and leaving the dangerous state.

  Note that the automatic driving here refers to a state in which at least one of the driving force, the braking force, and the steering is automatically performed without operation of the driver. That is, in the automatic driving, all of the steering, acceleration / deceleration, and braking may be automatically performed, but, for example, only the brake assist may be automatically performed.

DESCRIPTION OF SYMBOLS 1 ... Collision determination apparatus 2 ... Imaging part 112 ... Detection part 113 ... Judgment part 114 ... Notification processing part d1: Rear threshold distance d2 ... Front threshold distance OV ... Other vehicle V ... own vehicle

Claims (16)

A collision determination device that determines the possibility of a vehicle collision,
A detection unit that detects another vehicle existing on the front side of the host vehicle based on a captured image from a shooting unit mounted on the host vehicle,
A determining unit configured to determine, based on the detected positional relationship between the other vehicle and the own vehicle and time change information of the positional relationship, whether or not a collision between the own vehicle and the other vehicle is expected to be in a dangerous state; When,
A collision determination device comprising:   The determination unit, when determining whether or not the danger state, when the other vehicle is expected to collide with the other vehicle when the other vehicle changes lane to the lane in which the own vehicle runs, The collision judging device according to claim 1 which judges that it is in danger.   3. The method according to claim 1, wherein the determination unit determines whether or not the relationship in which the host vehicle has a portion arranged in the left-right direction with the other vehicle has continued for a certain period of time when determining whether or not the vehicle is in the danger state. 4. The collision judging device according to the above.   The determination unit determines whether or not the relationship that the own vehicle approaches the other vehicle beyond a predetermined rear threshold distance set for the other vehicle has continued for a predetermined period when determining whether the vehicle is in the danger state. The collision determination device according to any one of claims 1 to 3, wherein the collision determination device determines whether the collision has occurred.   The determination unit is configured to determine whether or not the vehicle is in the danger state, and the relationship that the other vehicle located closer to a predetermined front threshold distance set for the vehicle is separated from the vehicle continues for a certain period of time. The collision determination device according to claim 1, wherein the collision determination device determines whether the collision has occurred.   The collision determination device according to claim 1, wherein the determination unit determines whether the vehicle is in the danger state based on a speed relationship between the other vehicle and the host vehicle.   The collision determination device according to any one of claims 1 to 6, wherein the determination unit determines whether the vehicle is in the dangerous state based on an acceleration relationship between the other vehicle and the host vehicle.   The collision determination device according to any one of claims 1 to 7, wherein the determination unit determines whether the vehicle is in the dangerous state based on the presence or absence of an object in front of the other vehicle.   The collision determination device according to any one of claims 1 to 8, wherein the determination unit determines whether or not the vehicle is in the danger state based on a state of a turn indicator of the other vehicle.   The collision determination device according to claim 1, wherein the determination unit recognizes the positional relationship based on at least one of a rear end of the other vehicle, a side mirror, and a tire.   The collision determination device according to claim 10, wherein the determination unit recognizes the positional relationship based on a side mirror of the other vehicle, and determines whether or not the vehicle is in the dangerous state based on an image reflected on the side mirror. .   The collision determination device according to any one of claims 1 to 11, further comprising: a notification processing unit that operates a notification unit that notifies a driver of the host vehicle of a danger when the danger state is determined.   The collision judging device according to claim 12, wherein the notification processing unit changes an operation of the notification unit according to the level of the danger.   The notification processing unit, when it is determined that the dangerous state is occurring at the left front side and the right front side at the same time, controls the notification unit to simultaneously notify the left and right danger states and notify. 14. The collision determination device according to 12 or 13.   The notification processing unit, when it is determined that the danger state occurs simultaneously on the left front side and the right front side, the notification unit is controlled to give priority to the higher danger level to notify. Item 14. The collision determination device according to item 12 or 13. A collision determination method performed by a collision determination device that determines a possibility of a vehicle collision,
A detection step of detecting another vehicle existing on the front side of the own vehicle based on a photographed image from a photographing unit mounted on the own vehicle,
A determining step of determining, based on the detected positional relationship between the other vehicle and the own vehicle and time change information of the positional relationship, whether or not a collision between the own vehicle and the other vehicle is expected; When,
A collision determination method comprising:

Images