JP2011118570A - Device for avoiding collision of vehicle - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To activate a collision avoiding action against the other vehicle at precise timing after grasping a viewable range of an intersection. <P>SOLUTION: The viewable range capable of viewing a front part from a vehicle is calculated based on a position of an obstacle around the vehicle. The passing speed VT of the vehicle for passing through the intersection without colliding with a moving body approaching the vehicle from the outside of the viewable range is calculated based on a viewable range. Control for avoiding a collision is activated when the present speed VN of the vehicle is lower than the passable speed VT. The maximum stoppable speed VS capable of stopping the vehicle before the vehicle enters the intersection is further calculated, then the activation of control for avoiding the collision is suppressed even if the present speed is lower than the passing speed VT when the present speed VN of the vehicle is lower than the stoppable speed VS. The control of avoiding the collision is activated at appropriate timing while considering speed which can be taken by a driver in response to the viewable range since the passing speed VT is calculated according to the magnitude of the viewable range. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an apparatus for avoiding a vehicle collision.

  Various devices for avoiding a vehicle from colliding with a moving object such as another vehicle have been proposed. When it is determined in Patent Document 1 below that there is a T-shaped road ahead of the vehicle, it is determined whether the T-shaped road has a good line of sight using image data captured by the front camera. If it is determined that the T-shaped road has poor visibility, a blind corner camera capable of imaging the side of the vehicle is activated, and the captured image is displayed on the display. Further, in Patent Document 2 below, when a terminal device mounted on a vehicle determines the possibility of collision with another vehicle and there is a possibility of collision, the other vehicle side has already been notified of the collision. In the case, it is described that the notification of the collision in the host vehicle is suppressed.

JP 2007-140992 A JP 2005-10937 A

  When the host vehicle enters the intersection, the driver selects the speed of the host vehicle entering the intersection according to the size of the range that can be seen from the host vehicle. That is, the smaller the range that can be seen, the lower the speed because it cannot be recognized unless the approaching moving object approaches. On the other hand, when the range that can be seen is large, the presence or absence of an approaching object can be confirmed on the near side of the intersection as compared with the case where the line of sight is small. In such a case, if the risk is judged based only on the conditions such as the speed of the vehicle and the distance to the intersection, and the collision avoidance operation is activated, the collision avoidance operation may be excessively activated. There is a risk of feeling it.

  In addition, the size of the range that can be seen from the host vehicle is often affected by the presence of an obstacle, but in Patent Documents 1 and 2 described above, the size of the range that can be seen is not precisely determined.

  Therefore, by more accurately grasping the extent of the line of sight and considering the speed range that the driver can take, the collision avoidance operation can be activated at a more accurate timing without disturbing the driver. A technique that can be used is desired.

  According to one aspect of the present invention, an apparatus mounted on a vehicle and for avoiding a collision is disposed at a front portion of the vehicle and detects an obstacle around the vehicle, and the detection Means for calculating a line-of-sight range from which the vehicle can be seen forward based on the position of the obstacle, and without colliding with a moving object approaching the vehicle from outside the line-of-sight range, A means for calculating a passable speed at which the vehicle can pass through an intersection between a traveling path of the vehicle and a road approaching the moving object based on the calculated line-of-sight range; And means for invoking control for collision avoidance when the current speed of the vehicle is low.

  According to the present invention, based on the position of an obstacle around the vehicle, the range that can be seen from the vehicle is obtained, and the speed at which collision can be avoided (passable speed) is calculated based on the range that can be seen. Judging the possibility of a collision. Therefore, it is possible to activate the collision avoidance control while taking into consideration the speed range that the driver can take according to the range that can be seen, and therefore, excessive activation of the collision avoidance control is suppressed, and the driver is disturbed. The collision avoidance control can be activated at a more appropriate timing. For example, when the line of sight is good, the range that can be seen is calculated large. Since the driver can recognize a moving object approaching from a distance, the passable speed is calculated to be small. On the other hand, when the line of sight is bad, the viewable range is calculated to be small. Since the driver cannot recognize that the moving object is not approaching, the passable speed is greatly calculated. Since collision avoidance control is activated according to the calculated passable speed, it is possible to prevent the driver from being disturbed.

  According to an embodiment of the present invention, the vehicle includes means for calculating a maximum speed at which the vehicle can stop before entering the intersection, and the current speed of the vehicle is smaller than the stoppable speed. Even if it is smaller than the passable speed, the activation of the control for avoiding the collision is suppressed.

  Even when the current speed of the host vehicle is smaller than the passable speed, if the vehicle can be stopped before entering the intersection, the collision can be avoided.In this case, the activation of the control for avoiding the collision is suppressed. Excessive control can be avoided.

  According to one embodiment of the present invention, the obstacle detection means is an imaging device arranged in the front part of the vehicle, calculates an optical flow for an image acquired by the imaging device, and determines the magnitude of the optical flow. The end point of the obstacle is determined on the basis of a point on the image where the value changes by a predetermined value or more. The line-of-sight range is calculated based on the position of the end point of the obstacle.

According to this invention, since the end point of the obstacle is determined based on the optical flow, the line-of-sight range can be specified with better accuracy. In addition, since the line-of-sight can be determined based on the image obtained by the imaging device, the cost can be suppressed.
Other features and advantages of the present invention will be apparent from the detailed description that follows.

1 is a block diagram of an apparatus for collision avoidance according to one embodiment of the present invention. The figure for demonstrating the imaging range, the end point of an obstruction, and the line-of-sight range according to one Example of this invention. The figure which shows an example of the optical flow computed based on the picked-up image and this picked-up image according to one Example of this invention. The figure which shows the relationship between the distance of an obstruction, and an optical flow according to one Example of this invention. The figure which shows the relationship between the inclination of an obstruction, and an optical flow according to one Example of this invention. The figure for demonstrating the method of calculating | requiring the distance and inclination of an obstruction based on the optical flow obtained from the captured image according to one Example of this invention. The figure for demonstrating the method of determining the end point of an obstruction based on the distance and inclination to an obstruction according to one Example of this invention. The figure for demonstrating the method of calculating the passable speed and stop possible speed according to one Example of this invention. The figure which shows the example of the speed which can be passed, and the speed which can be stopped in the situation where the line of sight is good according to one Example of this invention. The figure for demonstrating the method of determining the possibility of a collision based on the speed which can be passed according to one Example of this invention, the speed which can be stopped, and the speed of the own vehicle. The figure which shows the example of the speed which can be passed, and the speed which can be stopped in the condition where the line of sight is not good according to one Example of this invention. The flowchart of the process of collision avoidance control according to one Example of this invention. The flowchart of the process of collision avoidance control according to one Example of this invention.

  Next, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 shows an apparatus 10 mounted on a vehicle and for avoiding a collision of the vehicle according to an embodiment of the present invention. The apparatus 10 includes an external sensor 11, a vehicle state sensor 12, a processing device 13, and a collision avoidance support device 15.

  In this embodiment, the external sensor 11 is an imaging device that is provided at the center of the front end of the vehicle and images the front of the vehicle. Preferably, the imaging device is a camera including a wide-angle lens having a wide angle of view (for example, an angle of view of 180 degrees or more) so that a range in which the driver can see ahead of the vehicle can be calculated. Image data captured by the imaging device is transferred to the processing device 13.

  The vehicle state sensor 12 includes a sensor for detecting the speed of the host vehicle. The sensor 12 can be realized by any known appropriate means, for example, a wheel speed sensor that detects the rotational speed (wheel speed) of the driving wheel of the host vehicle, or a vehicle speed that detects the speed of the host vehicle. It can be realized by a sensor or an acceleration sensor that detects acceleration acting on the vehicle body.

  The processing device 13 can be realized by an electronic control unit (ECU) that is a computer including a central processing unit (CPU) and a memory. In the figure, functions realized by the processing device 13 are represented as blocks. In this embodiment, the processing device 13 includes an obstacle determination unit 31, a host vehicle information acquisition unit 33, a line-of-sight range calculation unit 35, a passable speed calculation unit 37, a stoppable speed calculation unit 39, a collision possibility determination unit 41, And a collision avoidance control unit 43.

  The obstacle determination unit 31 acquires output signals (image data) of the external sensor 11 at predetermined time intervals, determines obstacles existing around the vehicle based on the output signals, and determines the end points of the obstacles. Determine the position.

  Here, referring to FIG. 2, an intersection where the road 101 and the road 103 intersect is shown. A range 111 in which the host vehicle A is traveling on the road 101 toward the intersection and is imaged by the external sensor 11 (imaging device) provided at the center of the front end of the host vehicle is shown. Obstacles D1 and D2 exist on both sides of the road 101. When the driver of the host vehicle A looks forward, the end points E1 and E2 of the obstacles D1 and D2 obstruct the field of view, so the range that can be seen is the line L1 connecting the driver and the end point E1 of the obstacle, and the driver And a line L2 connecting the end point E2 of the obstacle. Therefore, the driver cannot visually recognize the other vehicle B that exists outside the range that can be seen.

  In order to calculate the range that can be seen more accurately, it is necessary to determine the obstacles D1 and D2 as shown in the figure and to detect the positions of the end points E1 and E2. Therefore, the obstacle determination unit 31 determines the position of such an obstacle and its end point by performing predetermined image processing on the captured image received from the external sensor 11.

  Returning to FIG. 1, the line-of-sight range calculation unit 35 calculates a range that the driver of the host vehicle can see through based on the determined end points of the obstacle. As described with reference to FIG. 2, when the positions of the end points E1 and E2 of the obstacle are determined, the positions of the lines L1 and L2 are determined, and thus the line-of-sight range can be calculated.

  Based on the calculated line-of-sight range, the passable speed calculation unit 37 passes through the intersection without colliding (contacting) with a moving object approaching from outside the line-of-sight range (for example, another vehicle B shown in FIG. 2). The vehicle speed (referred to as “passable speed”) VT is calculated.

  On the other hand, the host vehicle information acquisition unit 33 acquires the output signal of the vehicle state sensor 12 at the predetermined time interval, and acquires the current speed VN of the host vehicle based on the output signal.

  The collision possibility determination unit 41 compares the passable speed VT with the current speed VN of the host vehicle. If the passable speed VT is less than or equal to the vehicle speed VN (VT ≦ VN), the driver must drive with the correct understanding of the intersection situation even if there is a moving object entering from outside the line-of-sight range. Thus, it is determined that traveling can be continued while avoiding contact with the moving object. On the other hand, if the passable speed VT is greater than the current speed VN of the vehicle (VT> VN), the host vehicle will maintain the current travel when there is a moving object entering from outside the line-of-sight range. Since it indicates that there is a possibility of contact with the moving object at the intersection, it is determined that there is a possibility of collision.

  Preferably, a stoppable speed calculation unit 39 is provided. The stoppable speed calculation unit 39 calculates a maximum speed (referred to as stoppable speed) VS at which the host vehicle can stop before entering the intersection. In this case, the collision possibility determination unit 41 compares the stoppable speed VS with the current speed VN of the vehicle. If the stoppable speed VS is greater than the vehicle speed VN (VS> VN), even if the passable speed VT is greater than the vehicle speed VN, the host vehicle can travel with the driver correctly understanding the intersection situation. Since it is possible to stop before reaching the intersection, it is determined that the collision with the moving object can be avoided by the action of the driver himself. If the stoppable speed VS is equal to or lower than the vehicle speed VN (VS ≦ VN), it is difficult to stop the host vehicle before reaching the intersection due to the driver's own action, so there is a possibility of a collision. Is determined.

  If it is determined that there is a possibility of collision, the collision avoidance control unit 43 activates the collision avoidance support device 15 and activates control for collision avoidance. Any appropriate control can be included in the control for collision avoidance. In one embodiment, notification is performed as control for collision avoidance, and the collision avoidance assistance device 15 is a notification device.

  The notification device 15 can be realized by any known appropriate technique. For example, any one or more of a tactile transmission device, a visual transmission device, and an audio transmission device may be used. Can be realized. The tactile transmission device is, for example, a seat belt device, a steering control device, or the like, and generates a predetermined tension on the seat belt, for example, in response to a control signal output from the collision avoidance control unit 43, so The driver is notified by applying a perceivable tightening force to the steering wheel or by generating vibration on the steering wheel that can be perceived by the driver.

  The visual transmission device is, for example, a display device or the like, and displays predetermined alarm information or blinks or lights a predetermined warning light in accordance with a control signal from the collision avoidance control unit 43. To inform. The auditory transmission device is a speaker or the like, for example, and notifies the driver by outputting a predetermined alarm sound or sound according to a control signal from the collision avoidance control unit 43.

  In another embodiment, the collision avoidance assistance device 15 is a brake actuator. The collision avoidance control unit 43 sends a control signal to the brake actuator as a control for collision avoidance, and operates the brake of the vehicle via the brake actuator. In this case, the brake can be actuated so that the deceleration required for the vehicle to stop before the intersection is reached.

  Note that either one of the above notification and brake control may be performed, or both may be performed.

  Hereinafter, more detailed operations of the obstacle determination unit 31, the line-of-sight range calculation unit 35, the passable speed calculation unit 37, and the stoppable speed calculation unit 39 will be described.

  First, the processing content by the obstacle determination part 31 is demonstrated. A reference numeral 120 in FIG. 3A indicates an example of an image captured by the imaging device that is the external sensor 11. For example, in the situation where obstacles D1 and D2 as described with reference to FIG. 2 exist on both the left and right sides of the road 101 of the host vehicle, when the imaging device captures the front of the vehicle, the obstacles D1 and D2 are captured images. Images are taken on the left and right sides of 120 respectively. The end points E1 and E2 in FIG. 2 represent the entire edges (121 and 122 in the figure) of the obstacles D1 and D2, respectively, as shown in the captured image. A point O indicates a vanishing point that is a point where parallel lines intersect in a so-called perspective, and represents the traveling direction of the host vehicle. For convenience, the xy coordinate system is set for the captured image with the vanishing point as the origin O as shown in the figure.

  The obstacle determination unit 31 uses each point (each pixel) of the captured image acquired this time using captured images (for example, an image captured this time and an image captured last time) acquired continuously in time. For, the optical flow is calculated. As is well known, the optical flow represents the motion of an object in a temporally continuous image as a vector, and the magnitude of the optical flow can be expressed by the magnitude of the motion vector of the object. it can. A known method can be used to calculate the optical flow. In this embodiment, the optical flow of each point on the image is calculated and averaged for each column (x direction).

  The optical flow is preferably calculated after previously removing distortion at the peripheral portion of the image caused by imaging with a wide-angle camera by image processing.

  FIG. 3B shows an example of the result of the optical flow V thus obtained. The x value on the horizontal axis is associated with the image of (a), and the left end of the captured image 120 is represented by xL and the right end is represented by xR. The vertical axis represents the magnitude of the optical flow V (the magnitude of the motion vector). Here, when the direction of the x component of the optical flow V is the positive direction of x, the magnitude of the optical flow V is expressed as a positive value, and the direction of the x component of the optical flow V is the negative direction of x. In some cases, the magnitude of the optical flow V is expressed as a negative value. Therefore, since the x component of the optical flow V at each point on the image constituting the obstacle D1 imaged on the left side faces the negative direction of x, it takes a negative value and is imaged on the right side. Since the x component of the optical flow V at each point on the image constituting the obstacle D2 is oriented in the positive direction of x, it takes a positive value.

  Since the host vehicle A is moving, the obstacle appears to approach the host vehicle A in the captured images acquired continuously in time. Therefore, the size of the optical flow of the obstacle is large. On the other hand, since the background (the area between the obstacles D1 and D2) exists far away, the size of the optical flow is small. Therefore, as shown in (b), the magnitude of the optical flow V obtained for the captured image suddenly changes with the x value corresponding to the right end 121 on the image of the obstacle D1, that is, x1, as well as the obstacle D2. The x value corresponding to the left end 122 on the image, that is, x2 changes suddenly.

  The obstacle determination unit 31 examines the magnitude of the optical flow calculated in this way, determines whether or not this has changed by a predetermined value or more, and if any x value has changed by a predetermined value or more, the obstacle is determined. Is determined to exist.

  Next, the obstacle determination unit 31 performs the behavior of the optical flow V in the x value area smaller than x1 (that is, the area between x1 and xL) and the x value area larger than x2 (that is, x2 and xR). The positions of the end points E1 and E2 of the obstacles D1 and D2 are calculated on the basis of the behavior of the optical flow V in the intermediate region. This method will be specifically described.

  First, with reference to FIG. 4 and FIG. 5, the characteristics of the optical flow depending on the arrangement of the obstacle will be described.

  In FIG. 4A, as an example, the obstacle 131 is positioned at any one of the positions P1, P2, and P3, and the side surface 133 of the obstacle 131 is parallel to the traveling direction (Z-axis direction) of the host vehicle A. Assume that it exists. The distance from the Z axis to the side surface 133 of the obstacle is represented by d. Reference numeral 111 indicates a range imaged by the external sensor 11 (not shown) as described above.

  FIG. 3B shows the case where the obstacle 131 calculated at the position P1, the position P2 and the position P3 calculated by the method described in FIG. The magnitude of the optical flow V (expressed here in absolute value) with respect to the distance from the vanishing point O (expressed in absolute value of x value) is shown. Here, the solid line is an optical flow part obtained from an actual captured image (in the example of FIG. 3, only the optical flow part between x1 and xL is cut out), and this solid line is an extended line. Is represented by a dotted line. The magnitude | V | of these optical flows is proportional to the magnitude | x | of the distance from the vanishing point O. Further, since the side surface 133 of the obstacle 131 is located in parallel with the traveling direction of the host vehicle A, when these optical flows are extended, the obstacle 133 converges to the vanishing point O as shown by the dotted line.

  The difference between the optical flows at the positions P1 to P3 is the slope of the straight line representing the magnitude | V | of the optical flow. In the figure, the optical flows at the positions P1 to P3 are represented by angles α1 to α3. Yes. As the distance d is shorter, the obstacle is present near the host vehicle, so the optical flow magnitude | V | Thus, the slope of the straight line representing the optical flow | V | is determined according to the distance d, and the slope increases as the distance d increases (that is, the magnitude of the change in the optical flow with respect to the change in the x value is greater). growing).

  Next, referring to FIG. 5, it is assumed in (a) that an obstacle 136 exists at any one of positions P4, P2 and P5 as an example. The position P2 is the same as that in FIG. Unlike FIG. 4, the side surface 137 when the obstacle 136 exists at the positions P4 to P5 is inclined with respect to the traveling direction (Z axis) of the vehicle A. This inclination is represented by θ. In the figure, the inclination θ when the obstacle 136 exists at the position P4 is shown. The inclination θ when the obstacle 136 is present at the position P5 has the same inclination θ as that at the position P4, but the direction is different (in this case, it is expressed by a negative value −θ).

  4 (b), similarly to FIG. 4 (b), the distance from the vanishing point O (when the obstacle 136 exists at the position P4, when it exists at the position P2, and when it exists at the position P5) The magnitude of the optical flow V (expressed in absolute value) with respect to (expressed in absolute value of x value) is shown. The optical flow of the obstacle 136 when it exists at the position P2 is the same as that shown in FIG. 4B, and the extension line converges to the vanishing point O.

  The magnitude | V | of these optical flows is proportional to the magnitude | x | of the distance from the vanishing point O, as in FIG. However, when the obstacle 136 exists at the positions P4 to P5, the side surface 137 of the obstacle 136 has an inclination θ with respect to the traveling direction of the host vehicle A, so that the optical flow is extended. However, it does not converge to the vanishing point O. A straight line intercept of the optical flow at the position P4 is represented by β4, and a straight line intercept of the optical flow at the position P5 is represented by β5.

  As described above, the difference between the optical flows at the positions P4, P2, and P5 is a straight line intercept representing the magnitude | V | of the optical flow. The value of the intercept approaches the origin (vanishing point) O as the magnitude of the inclination θ decreases. Here, when it is tilted in the clockwise direction with respect to the Z axis as in the position P4, the intercept becomes a positive value as shown by β4, but when it is tilted in the counterclockwise direction, the intercept is It becomes a negative value as indicated by β5. Thus, the intercept of the straight line representing the optical flow | V | is determined according to the magnitude and direction of the inclination θ, and the value of the intercept approaches zero as the magnitude of the inclination θ decreases.

  As described above, since it has been found that there is a correlation between the arrangement of the obstacle and the optical flow, the inclination θ and the distance of the obstacle are obtained from the optical flow obtained from the captured image using this correlation. What is necessary is just to obtain | require d. Here, the optical flow is calculated from the captured image, and the portion of the optical flow corresponding to the obstacle (in the example of FIG. 3B, for example, the optical flow in the region of x1 and xL) is approximated by a straight line.

  In FIG. 6, an example of the optical flow approximated in this way is indicated by a straight line 141 (represented by a solid line). The straight line 141 is extended as indicated by a dotted line, and an intercept β that is an intersection of the extended line and the vertical axis is calculated. Since the value of the intercept β and the inclination θ of the obstacle are correlated as described above, the inclination θ of the corresponding obstacle is calculated in advance through simulation or the like for each value of the intercept β. It can be stored in a storage device as a map. By referring to the map based on the value of the intercept β calculated this time, the inclination θ of the corresponding obstacle can be obtained.

  Next, the straight line 143 is drawn to the vanishing point O from the end point 142 of the straight line 141, that is, the point 142 on the straight line 141 corresponding to the left end xL of the captured image, and the inclination α of the straight line 143 is calculated. In this embodiment, as shown in FIG. 3, since the obstacle is imaged up to the left end xL of the captured image, the point 142 is on the straight line 141. When no obstacle is imaged at the left end xL, the straight line 141 is extended toward the left end xL of the captured image, and the straight line 143 is directed from the point on the extension line of the straight line 141 corresponding to xL toward the vanishing point O. Just pull. This straight line 143 shows the optical flow when the obstacle is not inclined with respect to the Z axis. As described above, since the value of the slope α of the straight line 143 and the distance d of the obstacle are correlated, the corresponding obstacle distance d is calculated in advance through simulation or the like for each value of the slope α. Then, it can be stored in the storage device as a map. By referring to the map based on the value of the inclination α calculated this time, the distance d of the corresponding obstacle can be obtained.

  Instead of the intercept β, an angle γ of the straight line 141 with respect to the straight line 143 may be used. The straight line 143 is a reference straight line indicating the optical flow when the obstacle is not inclined with respect to the Z axis. As is apparent from FIG. 5, the angle γ with respect to the reference line changes as well as the value of the intercept β depending on the inclination of the obstacle with respect to the Z axis. In this case, similarly to the intercept β, the angle γ is a positive value when the obstacle is tilted clockwise with respect to the Z axis, and the angle γ is a negative value when the obstacle is tilted counterclockwise. It becomes.

  Thus, as shown in FIG. 7, the distance d and the inclination θ of the obstacle D1 were calculated (note that the obstacle D1 in FIG. 2 is not shown to have an inclination, but here as an example The case where the obstacle D1 has an inclination is described). Next, a method for obtaining the position of the end point E1 of the obstacle D1 will be described. As a coordinate system in the real space, the position where the external sensor 11 of the own vehicle A is mounted is the origin O, the X axis is taken in the vehicle width direction of the own vehicle A, and the Z axis is taken in the traveling direction. For the end point E1, the x-coordinate value on the captured image is found to be x1. On the other hand, in the captured image, a position in the real space corresponding to a point (column) having an x value of x1 is determined in advance, and this is represented by a line 151.

  The line 153 is determined by the distance d and the inclination θ. By obtaining the intersection of the line 151 and the line 153, the position of the end point E1 (the position in the real space) can be calculated. Here, the end point E1 is represented by a coordinate system in real space, and is represented based on an X coordinate value corresponding to x1 and a Z coordinate value representing a distance in the Z-axis direction from the host vehicle.

  The end point E2 of the obstacle D2 can be calculated by the same method. The positions of the end points E1 and E2 of the obstacle calculated by the obstacle determination unit 31 are passed to the line-of-sight range calculation unit 35.

  In this embodiment, as described with reference to FIG. 3, the optical flow is calculated for each point of the captured image. However, as described above, the position of the end point of the obstacle that blocks the line of sight is calculated. Since the purpose is to determine, the optical flow may be calculated only for each point in a predetermined area above the captured image. For example, the optical flow can be calculated only for each point in the upper half area of the captured image. By doing so, it is not necessary to calculate for an object with a low height from the ground surface, such as a curb, which does not have a risk of obstructing the field of view, thereby reducing the calculation load and improving the processing speed. it can.

  Next, the line-of-sight range calculation unit 35 and the passable speed calculation unit 37 will be described. Referring to FIG. 8, a diagram similar to FIG. 2 is shown, and as described above, the positions of the end points E1 and E2 of the obstacles D1 and D2 are known. In this figure, for convenience, the position of the external sensor 11 of the host vehicle A is used as the origin, the y axis is taken in the traveling direction of the host vehicle A, and the x axis is taken in the direction perpendicular to the y axis. Here, the y-axis corresponds to the Z-axis in FIG. 7, and therefore the positions of the end points E1 and E2 are represented by x and y coordinate values in this figure.

  The position of the driver of the own vehicle A with respect to the origin is known in advance. The line-of-sight range calculation unit 35 calculates a line L1 that connects the position of the driver and the position of the end point E1, and calculates a line L2 that connects the position of the driver and the position of the end point E2, and is surrounded by these lines. The calculated area is calculated as the line-of-sight range. Alternatively, the position of the external sensor 11 may be used instead of the driver's position. Further, in this example, the case where there are obstacles on both the left and right sides of the host vehicle A has been described. However, when there are obstacles on only one side, The set range can be set as the line-of-sight range. For example, the imaging range (FIG. 2) may be set as the line-of-sight range.

The passable speed calculation unit 37 passes through an intersection without contacting the moving object even when the moving object enters from outside the line-of-sight range when the host vehicle A travels at a constant acceleration in the current traveling direction. The possible passing speed (that is, the minimum speed required to pass the intersection before the moving object) VT is calculated according to the following equation (1).

Here, the figure shows a traveling route 105 when the host vehicle A travels on the road 101 while maintaining the current traveling direction. y2 is the own vehicle A is a distance to passing completely through the intersection, the road 103 (hereinafter, referred to as the cross road) and the distance y1 to enter the intersection between the width y _W crossroad 103, it can be calculated by adding the vehicle length y _a vehicle a. The distance y1 to enter the intersection can be obtained by any appropriate method. For example, the navigation device is mounted on the vehicle A, and the distance to the intersection is obtained from the map information provided in the navigation device. Can do. Or you may acquire the distance to this intersection by detecting an intersection from the image imaged via the external sensor 11 via arbitrary appropriate image processing.

For the width y _w crossroad 103 can be acquired from the road information included in the map information navigation apparatus. Alternatively, the width may be calculated by detecting the cross road 103 from the image captured through the external sensor 11 through any appropriate image processing.

  T is the time from when the moving object enters the line-of-sight range until it reaches the traveling route 105 of the host vehicle A, assuming that there is a moving object outside the line-of-sight range. It is calculated based on the size of the range and the speed of the moving object.

  Here, in this embodiment, as shown in the figure, as a moving object, another vehicle B1 entering the line-of-sight range from the left direction, another vehicle B2 entering the line-of-sight range from the right direction, and the line-of-sight from the left direction. Assume that there is a bicycle C1 entering the range and a bicycle C2 entering the line-of-sight range from the right direction.

  As for the other vehicle B1, it is assumed that the vehicle travels in a lane far from the host vehicle, and the right side surface of the vehicle B1 can be assumed as a position crossing the line L1, that is, an intersection road on the line L1. A point 161 near the center of the road width 103 is selected, and a distance dc in the x-axis direction from the point 161 to the traveling path 105 of the host vehicle A (specifically, a line extending the left side surface of the host vehicle A) Is calculated as a numerical value representing the size of the line-of-sight range.

  Regarding the other vehicle B2, it is assumed that the vehicle is traveling in the lane closer to the own vehicle, and the left side surface of the vehicle B2 can be assumed as a position crossing the line L2, that is, an intersection road on the line L2. A point 163 of a predetermined distance (for example, 1 m) is selected from the road end 109 of 103, and from the point 163 to the traveling route 105 of the host vehicle A (specifically, a line extending the right side surface of the host vehicle A) Is calculated as a numerical value indicating the size of the line-of-sight range.

  As for the bicycle C1, it is assumed that the bicycle C1 is traveling near the road end 109 of the intersection road 103, and a point where the bicycle C1 can be assumed as a position crossing the line L1, that is, a point 165 on the line L1 immediately after the road end 109 And the distance dc in the x-axis direction from the point 165 to the traveling route 105 of the host vehicle A is calculated as a numerical value representing the size of the line-of-sight range. Similarly, for the bicycle C2, a point 167 immediately on the road end 109 on the line L2 is selected, and the distance dc in the x-axis direction from the point 167 to the traveling route 105 of the host vehicle A is set as the size of the line-of-sight range. Is calculated as a numerical value representing

  A predetermined value is used for the speeds of the other vehicles B1 and B2. Here, the predetermined value may be determined according to the road type. For example, the legal speed of the intersection road 103 is acquired from the road information included in the map information of the navigation device, and a value obtained by adding the margin value α to the legal speed can be used as the predetermined value. Alternatively, the current average vehicle speed of the intersection road 103 may be calculated in real time, and a value obtained by adding a margin value to the average vehicle speed may be used as the predetermined value. For this real-time calculation, for example, a probe system in which each traveling vehicle becomes a sensor (probe) and probe information related to the traveling state measured by the vehicle is collected in a predetermined server (center) can be used. Good. As the probe information, for example, the speed of each vehicle on the road is collected for each point of the road, and the average value of the collected values is averaged, whereby the average vehicle speed near the intersection of the road 103 can be calculated. The host vehicle A can be equipped with a device (or a navigation device) having a communication function with the server, obtain an average vehicle speed from the server, and use this as the predetermined value. Alternatively, if one or more crossing vehicles passing through the intersection can be detected by the external sensor 11, the speed may be directly calculated and the maximum value or the average value may be used as the predetermined value.

  A predetermined value is set for the speeds of the bicycles C1 and C2, and a predetermined speed is set for the predetermined value. For example, the assumed speed can be obtained by measuring the average speed of the bicycle in advance through experiments or the like.

  Thus, for each of the assumed moving objects B1, B2, C1, and C2, the arrival time T can be calculated by dividing the corresponding distance dc by the corresponding speed.

  Further, a is a value set in advance as an acceleration that can be generated when the vehicle A passes through an intersection. This can be set in advance through experiments or the like.

  Thus, the passable speed VT is calculated for each of the assumed moving objects B1, B2, C1, and C2. Among these calculated passable speeds, the passable speed VT having the highest value is selected and used for the subsequent collision possibility determination process.

  Note that the assumed types of moving objects and the positions of the points 161 to 167 used for the calculation of the distance dc are merely examples, and are not necessarily limited. For example, when the cross road 103 is one-way, only cars and bicycles entering the line-of-sight range from either the left or right may be considered. Further, if the intersection road 103 is a road that does not travel by bicycle, only the automobile may be considered.

Next, the stoppable speed calculation unit 39 will be described. The stoppable speed calculation unit 39 calculates a stoppable speed VS indicating the maximum speed at which the host vehicle A can stop before entering the intersection according to the following equation (2).

Here, as described above, y1 is the distance from the current position of the host vehicle A to entering the intersection. G _BRK is a deceleration (acceleration represented by a negative value) that can normally occur when the _host vehicle A stops, and can be set in advance through experiments or the like.

  Hereinafter, with reference to specific calculation examples shown in FIGS. 9 to 11, the collision possibility determination unit 41 using the possible passage speed VT, the current speed VN of the host vehicle, and the stoppable speed VS is possible. A sex determination method will be described.

  FIG. 9A shows numerical values used in the examples, and it is assumed that the same coordinate axes (not shown) as those in FIG. 8 are set. Since the positions in the x direction of the end points E1 and E2 of the obstacles D1 and D2 with respect to the host vehicle A are relatively far, the line-of-sight is relatively good. (B) plots the values of the passable speed VT and the stoppable speed VS for each value of the distance y1 under the situation as shown in (a). The distance y1 indicates the distance until the vehicle A enters the intersection as described above. In this example, the value of the distance y1 takes a zero value at the point where the vehicle enters the intersection, and the absolute value increases as it goes from the intersection to the current position of the host vehicle A (that is, toward the negative direction of the y-axis). Take a negative value. For example, y1 being “−2” m means that the distance until the vehicle A enters the intersection is 2 meters.

Here, regarding the predetermined values of the equations (1) and (2) for calculating the passable speed VT and the stoppable speed VS, G _BRK is set to 0.3G, and a is set to 0.2G. The assumed arrival time T for each of the moving objects is calculated as a speed of 60 km / h in the case of a car and 20 km / h in the case of a bicycle. As described above, when there are a plurality of assumed moving objects, a plurality of passable speeds are calculated, and the highest value among them is selected as the passable speed VT and plotted in the figure. is doing.

  Further, FIG. 10 is a graph in which the speed VN of the host vehicle is plotted with a black circle for each case represented by C1 to C5 on the graph of FIG. 9B. In cases C3 to C5, a straight line extends from the black circle, and this represents a case where the speed of the host vehicle A has changed from the current speed VN due to deceleration or acceleration.

  First, a black circle indicated by C1 indicates a situation (first case) where the speed VN of the host vehicle is higher than the passable speed VT and the stoppable speed VS. In such a situation, although safety confirmation is required, if the vehicle travels at the current speed or accelerates from the current speed, the moving object finishes passing through the intersection before reaching the intersection. be able to. Therefore, it is determined that the possibility of collision is low, and collision avoidance control is not performed.

  A black circle indicated by C2 indicates a situation (second case) in which the speed VN of the host vehicle is larger than the stoppable speed VS and smaller than the passable speed VT. Such a situation shows a situation in which the host vehicle cannot stop before reaching the intersection, and the moving object cannot pass through the intersection before reaching the intersection. Therefore, it is determined that there is a high possibility of collision, and some control for collision avoidance is performed.

  Black circles and lines indicated by C3 indicate that the current speed VN (black circle) of the vehicle is smaller than the passable speed VT and the stoppable speed VS, and the accelerator pedal is not depressed, for example, and the vehicle approaches the intersection. It shows the situation of slowing down. Such a situation indicates a situation in which the vehicle is not accelerating and can stop before reaching the intersection. Therefore, it is determined that the possibility of collision is low, and collision avoidance control is not performed.

  The black circle and line indicated by C4 indicate that the current speed VN (black circle) of the vehicle is smaller than the passable speed VT and the stoppable speed VS, and the vehicle is accelerating, for example, when the accelerator pedal is currently depressed. As the vehicle approaches the intersection, it indicates a situation where it will exceed the passable speed VT and then exceed the stoppable speed VS. That is, the value dTF4 of the distance y1 corresponding to the intersection of the line indicated by C4 and the passable speed VT is smaller than the value dSF4 of the distance y1 corresponding to the intersection of the line indicated by C4 and the stoppable speed VS (dTF4 <dSF4). In such a situation, although the vehicle is accelerating, it has reached the passable speed VT under a stoppable condition (in a range below the stoppable speed VS). Therefore, since acceleration is being performed correctly, it is determined that the possibility of collision is low, and collision avoidance control is not performed.

  The black circle and line indicated by C5 indicate that the current speed VN (black circle) of the vehicle is smaller than the passable speed VT and the stoppable speed VS, and the vehicle is accelerating, for example, when the accelerator pedal is currently depressed. As the vehicle approaches the intersection, it indicates a situation where it will exceed the stoppable speed VS and then exceed the passable speed VT. That is, the value dTF5 of the distance y1 corresponding to the intersection of the line indicated by C5 and the passable speed VT is larger than the value dSF5 of the distance y1 corresponding to the intersection of the line indicated by C5 and the stoppable speed VS (dTF5> dSF5). In such a situation, the vehicle is accelerating, but before reaching the passable speed VT, the vehicle cannot stop. Therefore, it is determined that the collision possibility is high, and some collision avoidance control is performed.

  On the other hand, FIG. 11A shows numerical values used in other examples, and the positions of the end points E1 and E2 of the obstacles D1 and D2 are relatively close to the own vehicle A, so that the prospect is It represents a bad situation. (B) plots the values of the passable speed VT and the stoppable speed VS for each value of the distance y1 under the situation as shown in (a). Here, the predetermined values of the equations (1) and (2) for calculating the passable speed VT and the stoppable speed VS are the same as those described in FIG. The same applies to the selection method of the passable speed.

  As is clear from FIG. 9B, the behavior of the stoppable speed VS is the same, but the behavior of the passable speed VT is different because the line-of-sight range is narrow. The collision avoidance control is performed in the second case C2 and the fifth case C5 as shown in FIG. The second case C2 is a case where the current speed VN of the vehicle is between the passable speed VT and the stoppable speed VS. In the case of FIG. 11B, since the line-of-sight range is narrow, the range between the passable speed VT and the stoppable speed VS is considerably large. Therefore, the speed VN of the host vehicle falls within this range. The possibility increases, and the collision avoidance control is more easily activated compared to the situation with good visibility in FIG.

  Further, in the fifth case C5, the current speed VN of the vehicle is smaller than the passable speed VT and the stoppable speed VS, and is accelerated to increase the stoppable speed VS before exceeding the passable speed VT. Exceeds the situation. Since the line of sight is poor, the range between the passable speed VT and the stoppable speed VS is considerably large. Even if the range is accelerated, it is likely that the stoppable speed is exceeded before the passable speed is exceeded. Therefore, also in this case, the collision avoidance control is easily activated.

  Thus, based on the passable speed VT and the speed VN of the host vehicle, it can be determined whether the host vehicle can finish passing the intersection before the moving object reaches the intersection. As described above, the passable speed VT is calculated based on the size of the line-of-sight range (represented by the distance dc in the example of FIG. 8), so that the host vehicle contacts the moving object. It is possible to improve the determination accuracy of whether or not the vehicle can pass through the intersection without any problems. Accordingly, excessive control of collision avoidance control can be suppressed at a more accurate timing without disturbing the driver. Further, by calculating the stoppable speed VS, it is possible to determine whether or not the vehicle can be stopped before the intersection, and unnecessary collision avoidance control can be suppressed. Furthermore, by considering whether or not the vehicle is currently accelerated, the possibility of collision with a moving object when accelerated can be determined with better accuracy.

  12 and 13 are flowcharts of the operation of apparatus 10 for collision avoidance according to one embodiment of the present invention. This process can be performed at predetermined time intervals.

  In step S11, an image captured by the external sensor (imaging device) 11 is acquired. In step S12, as described above, the optical flow of each point of the image is calculated based on the image captured this time and, for example, the image captured last time. In step S13, as described with reference to FIG. 3, if a point whose size changes to a predetermined value or more is extracted in the optical flow, it is determined that an obstacle exists.

  In step S14, the distance d and the inclination θ of the obstacle are calculated by the method described with reference to FIG. 6 based on the optical flow calculated in step S12 and the change point extracted in step S13. To do. In step S15, the end point of the obstacle is determined by the method described with reference to FIG.

  In step S16, the passable speed VT and the stoppable speed VS are calculated according to the above-described equations (1) and (2). As described above, when a plurality of passable speeds are calculated, the highest passable speed VT among them is used in subsequent processing. On the other hand, in step S17, the current speed VN of the host vehicle is acquired via the vehicle state sensor 12.

  Steps S21 to S31 are processes for determining the possibility of collision described with reference to FIG.

  In step S21, the passable speed VT and the vehicle speed VN are compared. If the vehicle speed VN is equal to or higher than the passable speed VT, the first case C1 described above is shown. In this case, the collision avoidance control is not performed and the process is exited.

  If the vehicle speed VN is smaller than the passable speed VT in step S21, the stoppable speed VS and the vehicle speed VN are compared in step S22. If the vehicle speed VN is equal to or higher than the stoppable speed VS, the second case C2 described above is shown that is neither stopable nor passable. Since this indicates that there is a possibility of collision, the process proceeds to step S31, and control for collision avoidance is activated. As described above, as control for collision avoidance, notification may be performed or brake control may be executed.

  In step S22, if the vehicle speed VN is smaller than the stoppable speed VS, the process proceeds to step S23 to determine whether or not the accelerator pedal is depressed. This can be detected by any known technique by a sensor provided on the accelerator pedal. If the accelerator pedal is not depressed, this indicates the above-described third case C3 in which the current speed VN is a speed that can be stopped and is not accelerated. In this case, the collision avoidance control is not activated and the process is exited.

  If the accelerator pedal is depressed in step S23, the process proceeds to step S24, and the future speed VF of the vehicle is estimated from the opening of the accelerator pedal. For example, the acceleration is calculated by any known method based on the opening of the accelerator pedal, and the vehicle speed approaches the intersection based on the current speed VN and the acceleration (that is, the distance y1 changes). It can be estimated how it will transition. For example, in the case of an internal combustion engine, the opening degree of the throttle valve may be used instead of the opening degree of the accelerator pedal. In this way, a vehicle speed estimation line as shown in the fourth and fifth cases C4 and C5 of FIG. 10 can be acquired.

  In step S25, as described with reference to FIG. 10, the first distance value dTF corresponding to the intersection of the curve indicating the passable speed VT and the line estimated for the vehicle speed of the host vehicle and the stoppable speed VS are set. The second distance value dSF corresponding to the intersection of the curve shown and the line estimated for the vehicle speed of the host vehicle is calculated.

  In step S26, the calculated first distance value dTF and second distance value dSF are compared. If the first distance value dTF is equal to or smaller than the second distance value dSF, the above-described fourth case C4 is accelerating but has reached the passable speed under a condition where it can be stopped. Since this indicates that there is no possibility of collision, in this case, the collision avoidance control is not activated and the process is exited.

  In step S26, if the first distance value dTF is larger than the second distance value dSF, the fifth case C5 is accelerated, but cannot stop before reaching the passable speed. Indicates. In this case, the process proceeds to step S31, and collision avoidance control is activated.

  As described above, specific embodiments of the present invention have been described. However, the present invention is not limited to these embodiments.

DESCRIPTION OF SYMBOLS 10 Collision avoidance apparatus 11 External field sensor 12 Vehicle state sensor 13 Processing apparatus 15 Collision avoidance assistance apparatus

Claims (3)

A device for avoiding a collision mounted on a vehicle is
An obstacle detection means disposed at the front of the vehicle for detecting obstacles around the vehicle;
Means for calculating a line-of-sight range from which the vehicle can be seen forward based on the position of the detected obstacle;
The vehicle can pass through the intersection of the traveling path of the vehicle and the road approaching the moving object without colliding with the moving object approaching the vehicle from outside the line-of-sight range. Means for calculating a passable speed based on the calculated line-of-sight range;
Means for invoking control for collision avoidance when the current speed of the vehicle is smaller than the passable speed;
A collision avoidance device. Means for calculating a maximum speed at which the vehicle can stop before entering the intersection;
When the current speed of the vehicle is lower than the stoppable speed, even if the current speed is lower than the passable speed, the activation of the control for avoiding the collision is suppressed.
The collision avoidance device according to claim 1. The obstacle detection means is an imaging device arranged at the front part of the vehicle, calculates an optical flow for an image acquired by the imaging device, and on the image where the magnitude of the optical flow changes by a predetermined value or more. Based on the point, determine the endpoint of the obstacle,
The line-of-sight range is calculated based on the position of the end point of the obstacle.
The collision avoidance device according to claim 1 or 2.

Images