JP5210064B2 - Vehicle collision prevention device - Google Patents

Description

  The present invention relates to a vehicle collision prevention device for preventing a collision between a vehicle and a moving object.

  In recent years, in vehicles such as automobiles, it is possible to detect the driving environment in the outside world with an in-vehicle camera, laser radar device, etc., recognize obstacles and preceding vehicles, and execute various controls such as warning, automatic braking, and automatic steering. In addition, technologies for improving safety by preventing vehicle collision accidents have been developed and put into practical use.

For example, in Patent Document 1 (Japanese Patent Application Laid-Open No. 2004-302621), a moving object on a road connected to a road on which the vehicle travels continues to exist at a certain angle in the horizontal direction with respect to the traveling direction of the vehicle. By detecting this, a technique for determining that the possibility of collision is high is disclosed.
JP 2004-302621 A

  The technique of Patent Document 1 is intended to prevent accidents at intersections, but even on ordinary straight roads, a situation in which pedestrians and bicycles existing on the sidewalks and shoulders cross the road frequently occurs. Therefore, it is difficult to deal with such a pedestrian or bicycle crossing the road with the technique of Patent Document 1.

  Further, in the technique of Patent Document 1, since it is determined that the possibility of collision is high when the traveling speed of the host vehicle is high, there is a possibility of entering the blind spot of the driver when the speed of the moving object is high. It becomes difficult to respond. Further, even when the traveling speed of the host vehicle is higher than that of the moving object, if the host vehicle turns right or left at the intersection, the possibility that the moving object enters the driver's blind spot increases, and it is difficult to cope with the same.

  The present invention has been made in view of the above circumstances, and provides a vehicle collision prevention apparatus that can determine the possibility of a collision in consideration of the movement of a moving object around the host vehicle and improve safety. It is an object.

In order to achieve the above object, a vehicle collision prevention apparatus according to the present invention is a vehicle collision prevention apparatus for preventing a collision between a vehicle and a moving object, and is a moving object that exists around the host vehicle and moves in the width direction of the vehicle. A moving object recognizing unit that recognizes at least, and a horizontal angular position of the moving object recognized by the moving object recognizing unit with respect to the own vehicle does not change for a certain period of time, and a relative distance of the moving object with respect to the own vehicle If it is smaller, it is determined that there is a possibility of collision between the host vehicle and the moving object, and a collision risk is set for the moving object that is determined to have a collision possibility, and execution of collision avoidance control is instructed. and a collision determination unit, the collision determination unit, the moving object collision when the velocity of a moving object which is determined that there is a possibility collision by said determination is faster than the speed of the vehicle Combing is set higher than normal, and setting a value corresponding to the possibility of entering the collision risk of the moving object moving in the width direction of the vehicle in the blind spot of the driver.

  According to the present invention, since the possibility of a collision is determined based on the horizontal angular position and relative distance of a moving object around the own vehicle with respect to the own vehicle, the possibility of a collision can be reliably and reliably determined. , Can improve safety.

  Embodiments of the present invention will be described below with reference to the drawings. 1 to 5 relate to an embodiment of the present invention, FIG. 1 is a schematic configuration diagram of a collision prevention device mounted on a vehicle, FIG. 2 is an explanatory diagram showing the possibility of a collision at a pedestrian crossing, and FIG. 3 is an intersection. 4 is an explanatory diagram showing the possibility of collision with another vehicle, FIG. 4 is an explanatory diagram showing the possibility of collision with a bicycle at an intersection, and FIG. 5 is a flowchart of collision determination processing.

  In FIG. 1, reference numeral 1 denotes a vehicle such as an automobile (own vehicle). The vehicle 1 includes a collision prevention device 2 that recognizes an external traveling environment and provides driving assistance to the driver to prevent the collision of the vehicle. It is installed. In the present embodiment, the collision prevention device 2 includes a stereo camera 3, a stereo image recognition device 4, a control unit 5 including a microcomputer and the like as main parts, and an automatic brake control device 22 and an automatic steering control device 23. It is possible to make a collision between the moving object and the own vehicle by recognizing the driving environment based on the captured image of the external environment and monitoring the moving object existing around the own vehicle. Judging sex.

  In FIG. 1, the stereo camera 3 is shown for forward recognition, but the stereo camera 3 includes a wide viewing angle camera, a camera capable of rotating and scanning a predetermined range, and a forward recognition camera. Like a side recognition camera, the vehicle can be recognized not only for front recognition but also for side recognition.

  Further, the stereo image recognition device 4, the control unit 5, the automatic brake control device 22, the automatic steering control device 23, etc. are each configured as a control unit composed of a single or a plurality of computer systems, and communicate with each other via a communication bus. Exchange each other.

  The own vehicle 1 is provided with a vehicle speed sensor 11 for detecting the own vehicle speed V, a yaw rate sensor 12 for detecting the yaw rate, a main switch 13 to which an ON-OFF signal for driving support control is input, and the like. The stereo image recognition device 4 and the control unit 5 are input, the yaw rate is input to the control unit 5, and the driving support control ON-OFF signal and the like are input to the control unit 5.

  The stereo camera 3 and the stereo image recognition device 4 function as a moving object recognition unit that recognizes a moving environment outside the host vehicle 1 and recognizes a moving object around the host vehicle 1, and the stereo camera 3 as a recognition sensor. Is composed of a pair of (left and right) cameras using, for example, a solid-state imaging device such as a CCD or CMOS, and the stereo image recognition device 4 includes an image processing engine that processes an image captured by the stereo camera 3 at high speed. It is configured as a processing unit that performs recognition processing. A set of cameras constituting the stereo camera 3 is attached with a certain base line length, stereo images of objects outside the vehicle are taken from different viewpoints, and image data is input to the stereo image recognition device 4.

  Image processing in the stereo image recognition device 4 is performed as follows, for example. First, distance information is obtained from a corresponding positional shift amount for a stereo image pair in the surrounding environment outside the host vehicle 1 captured by the stereo camera 3, and a distance image is generated. Then, based on this data, a well-known grouping process is performed and compared with frames (windows) such as three-dimensional road shape data, side wall data, and three-dimensional object data stored in advance. Side wall data such as guardrails and curbs that exist along the road are extracted, and three-dimensional objects are classified and extracted into other three-dimensional objects such as two-wheeled vehicles, ordinary vehicles, large vehicles, pedestrians, and utility poles.

  Each of the recognized data is calculated by calculating the position in the coordinate system in which the own vehicle 1 is the origin, the front-rear direction of the own vehicle 1 is the X axis, and the width direction is the Y axis. In the vehicle data of a large vehicle, the length in the front-rear direction is preliminarily estimated as 3 m, 4.5 m, 10 m, etc., and the width direction uses the center position of the detected width. The current center position is calculated. In addition, when the longitudinal length of the vehicle can be obtained with high accuracy by inter-vehicle communication or the like, the above-described center position may be calculated using the length data.

  Further, in the three-dimensional object data, the relative speed with respect to the own vehicle 1 is calculated from the change in the X-axis direction and the change in the Y-axis direction of the distance from the own vehicle 1, and the vector V is taken into consideration for the speed V of the own vehicle 1 in this relative speed. Thus, the X-axis direction speed and the Y-axis direction speed of each three-dimensional object are calculated. Each information obtained in this way, that is, white line data, side data such as guardrails and curbs along the road, and solid object data (type, distance from the vehicle 1, center position coordinates, speed data, etc.) Therefore, a moving object such as a pedestrian or a light vehicle around the own vehicle and other vehicles traveling on a road connected to the road on which the own vehicle travels is recognized.

  In this embodiment, an example in which the surrounding environment is recognized based on a stereo image will be described. However, the surrounding environment may be recognized using another recognition sensor such as a monocular camera or a millimeter wave radar. Also good.

  The control unit 5 includes the vehicle speed sensor 11 to the own vehicle speed V, the yaw rate sensor 12 to the yaw rate, the stereo image recognition device 4 to the white line data, the guard rails existing along the road, side walls such as curbs, and three-dimensional object data (type Each data such as a distance, center position coordinates, speed, etc. from the own vehicle 1 is input, and the control unit 5 has a function as a collision judging unit for judging the possibility of a collision between the own vehicle and the moving object, When it is determined that there is a possibility of a collision, a control command for avoiding a collision is output to the automatic brake control device 22 or the automatic steering control device 23, and an alarm is output to the driver via the display 21, thereby assisting driving. Execute control.

  The possibility of collision between the host vehicle and the moving object is determined based on the horizontal angle position of the recognized moving object with respect to the host vehicle and the relative distance of the moving object with respect to the host vehicle. If the horizontal angular position of the moving object with respect to the host vehicle does not change for a certain period of time and the relative distance of the moving object to the host vehicle decreases with time, it is determined that there is a possibility of collision with the moving object, and collision avoidance is performed. Direct control.

  For example, as shown in FIG. 2, when a bicycle 50 that is about to cross a pedestrian crossing is recognized as a moving object in front of the host vehicle 1, the Y axis of the bicycle 50 in the XY coordinate system with the host vehicle 1 as the origin is used. The time change of the horizontal angle position indicated by the angle θ and the change of the relative distance between the host vehicle 1 and the bicycle 50 are examined. If the horizontal angular position of the bicycle 50 does not change for a certain time and the relative distance calculated from the speed of the host vehicle 1 and the absolute speed of the bicycle 50 decreases with time, the host vehicle 1 and the bicycle 50 It is determined that there is a possibility of a collision with

  The same applies to the case where the moving object is not a bicycle 50 but a pedestrian, and the collision can be prevented in advance by making a collision determination in consideration of the behavior of a pedestrian crossing the road or a light vehicle such as a bicycle. .

  Also, as shown in FIG. 3, in a situation where the host vehicle 1 is about to enter an intersection of a crossroad, when another vehicle 51 proceeds as a moving object from the road on the right side of the intersection, Until the vehicle 51 approaches, it may be difficult to recognize because it is out of the driver's field of view. Even in such a case, if the horizontal angle position of the other vehicle 51 does not change for a certain time and the relative distance between the own vehicle 1 and the other vehicle 51 becomes smaller with the passage of time, the own vehicle 1 and the other vehicle 51 Therefore, it is possible to reliably prevent a collision with a moving object in the blind spot of the driver.

  Furthermore, as shown in FIG. 4, even in a situation where the host vehicle 1 makes a right / left turn at the intersection (FIG. 4 shows a right turn), a bicycle (same for pedestrians) crossing the road at the right turn destination of the intersection 52 When there is a moving object such as, the horizontal position of the bicycle 52 with respect to the host vehicle 1 has a fixed relationship from the time when the host vehicle 1 starts making a right turn, and the relative distance between the host vehicle 1 and the bicycle 52 increases with time. Get smaller. Therefore, in such a case, it is possible to prevent an accident at the time of turning left and right at the intersection by determining that there is a possibility of collision on the assumption that the moving object has a particularly high risk of collision.

  In the present embodiment, the avoidance control based on the determination that there is a possibility of collision is performed based on the risk (risk) set in the traveling environment of the host vehicle. For example, a lane formed by a white line of a road or the like is set as the driving lane of the own vehicle, and based on each input signal described above, the risk for the white line (the driving lane of the own vehicle), in front of the own vehicle (traveling direction) Risks for existing obstacles (three-dimensional objects) are each functionalized and set as risk functions Dline and Dobstacle, and a risk distribution in XY coordinates with the vehicle 1 as the origin is obtained.

Then, these risk functions Dline and Dobstacle are added and integrated as shown in the following equation (1) to set the current total risk function D, and based on this total risk function D, an avoidance route for obstacles Is output to the automatic steering control device 23 and the automatic brake control device 22. When signals are output to the automatic brake control device 22 and the automatic steering control device 23, the signals are visually displayed on the display 21 to notify the driver.
D = Dline + Dobstacle (1)

Here, the risk function Dobstacle set for an obstacle (two-dimensional vehicle, ordinary vehicle, large vehicle, pedestrian, other three-dimensional object such as a utility pole) is set by the following equation (2).
Dobstacle = Kobstacle · exp (− ((Xobstacle−x) ² / (2 · σxobstacle ² )) − ((Yobstacle−y) ² / (2 · σyobstacle ² ))) (2)
However, Kobstacle: Gain setting value
Xobstacle: X-coordinate value of the obstacle (center position)
Yobstacle: Y-coordinate value of the obstacle (center position)
σxobstacle: Pre-set variance in the X-axis direction
σyobstacle: Pre-set target dispersion in the Y-axis direction

  The variances σxobstacle and σyobstacle are set according to the recognition accuracy and presence status of the obstacle. For example, the lower the recognition accuracy by the stereo camera 3, the larger the variance σxobstacle, σyobstacle is set, and the larger is the case where the target type is a pedestrian or a two-wheeled vehicle based on the case of a normal vehicle or a large vehicle. It may be set to a small value in the case of other three-dimensional objects. Furthermore, you may make it set according to the lap | wrap rate of the width direction of the own vehicle 1 and the target solid object.

Further, the risk function Dline set for the traveling lane (white line) of the host vehicle is a function that leads to a larger risk value as the distance from the center of the lane is closer to the end position (white line), for example, (3) It is set by a function as shown in the equation.
Dline = exp (aR · y ⁴ ) −1 (3)
However, aR = (2 / WR) ⁴ · log (Ds + 1)
WR: Lane width
Ds: Risk at the lane edge (y = ± WR / 2)
Dobj: Correction value by solid object
ax: Correction parameter in the X direction
ay: Y direction correction parameter
Xobj: X coordinate position of the three-dimensional object
Yobj: Y coordinate position of the solid object

  With respect to the total risk function D obtained from these risk functions Dline and Dobstacle, its time change is predicted and evaluated, and collision avoidance steering control and automatic brake control are performed based on the evaluation result. At that time, especially for a moving object with a high risk of collision, the gain control Kobstacle of the risk function Dobstacle is made larger than that of a normal obstacle so that the collision avoidance steering control, automatic brake control and alarm control are performed earlier. To run.

  Hereinafter, the program processing related to the collision determination executed by the collision prevention apparatus 2 will be described with reference to the flowchart of FIG.

  In this program processing, first, in step S1, when a moving object is recognized around the own vehicle, a temporal change in the horizontal angle of the moving object with respect to the own vehicle is calculated. Next, it progresses to step S2 and it is investigated whether the horizontal direction angle of a moving object is changing temporally.

  As a result, if there is no temporal change in the horizontal angle of the moving object, the process proceeds from step S2 to step S3, and it is determined that there is no possibility of collision between the host vehicle and the moving object, and the program is exited.

  On the other hand, if there is no temporal change in the horizontal angle of the moving object, the process proceeds from step S2 to step S4, where the relative speed of the moving object with respect to the host vehicle is calculated, and is the moving object approaching the host vehicle from this relative speed? Determine if you are away. As a result, when it is determined that the moving object is moving away from the host vehicle, the process proceeds from step S4 to step S5, and similarly, it is determined that there is no possibility of collision between the host vehicle and the moving object, and the program is exited.

  On the other hand, if it is determined that the moving object is approaching the host vehicle, the process proceeds from step S4 to step S6, where it is determined that there is a possibility of a collision between the host vehicle and the moving object. Avoidance control is set according to whether the vehicle is inside or outside the intersection.

  That is, in step S7, information related to the traveling environment and driving conditions of the host vehicle is acquired. In step S8, whether the host vehicle is making a right turn or a left turn in the intersection from the previously acquired information. Judge whether or not. Whether the vehicle is turning right or left is determined, for example, from the operating state of the blinker and the vehicle speed. When the blinker is operating and the vehicle speed is 0 or more, it is determined that the vehicle is turning right or left, and the blinker is not operating or is stopped at 0 vehicle speed. Determines that it is not turning right or left.

  If it is determined that the vehicle is turning right or left in step S8, the process proceeds from step S8 to step S9 to instruct execution of collision avoidance control as a particularly high risk target. On the other hand, if it is determined that the vehicle is turning right or left in step S8, the absolute speed of the moving object is checked. In step S10, it is checked whether or not the absolute speed of the moving object exceeds the vehicle speed. If the absolute speed of the moving object is higher than the own vehicle speed, in step S9 described above, the moving object is instructed to perform collision avoidance control with a particularly high risk value, and the absolute speed of the moving object is If the vehicle speed is less than or equal to the vehicle speed, the execution of collision avoidance control is instructed in step S11 with the moving object as a target having a normal risk value.

  This collision avoidance control is executed as follows, for example. That is, the temporal change of the total risk function D is predicted by predicting the temporal change of the position of each object, and the Y-axis direction at the vehicle position for each time is based on the temporal change of the total risk function D. And an objective function for each time is created based on the deviation between the lateral position of the own vehicle and the minimum point for each time and the turning control amount.

  Then, the turning control amount for each time that minimizes this objective function is calculated as the turning control amount of the own vehicle, and the maximum risk function for each route when the own vehicle moves with the turning control amount for each time. The route having the smallest value or cumulative value is set as the final avoidance route, and a control signal is output to the automatic steering control device 23 based on the turning control amount of the avoidance route. Also, the earliest time when the set avoidance route is equal to or greater than the preset maximum allowable risk value is obtained, and the braking start point and the braking start time are calculated from the braking distance based on the preset deceleration G based on this time. Then, a braking control command based on the braking start point and the braking start time is output to the automatic brake control device 22. Similarly, based on the earliest time when the set avoidance route is equal to or greater than the preset allowable risk value, the warning start point and the warning start time are calculated from the braking distance by the preset deceleration G, An alarm based on the alarm start point and the alarm start time is output to the driver via the display 21.

  As described above, in this embodiment, since the possibility of collision is determined in consideration of the movement of moving objects around the host vehicle, pedestrians and bicycles crossing the road, and the blind spot of the driver at the intersection A collision can be reliably and reliably prevented with respect to a vehicle, a pedestrian or a bicycle at the time of turning left and right at an intersection, and safety can be improved.

  In this embodiment, as the collision avoidance control, the brake control, the steering control, and the alarm control are performed. However, the present invention is not limited to this, and any one of these controls or any one of these controls. You may comprise so that two may be performed.

Schematic configuration diagram of a collision prevention device mounted on a vehicle Explanatory diagram showing the possibility of collision at a pedestrian crossing Explanatory drawing showing the possibility of collision with other vehicles at the intersection Explanatory diagram showing the possibility of collision with a bicycle at an intersection Collision judgment process flowchart

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Own vehicle 2 Collision prevention apparatus 3 Stereo camera (moving object recognition part)
4 Stereo image recognition device (moving object recognition unit)
5 Control unit (collision judgment unit)

Claims (3)

In a vehicle collision prevention apparatus for preventing a collision between a vehicle and a moving object,
A moving object recognition unit that recognizes at least a moving object that exists around the host vehicle and moves in the vehicle width direction ;
When the horizontal angular position of the moving object recognized by the moving object recognition unit with respect to the own vehicle does not change for a certain period of time and the relative distance of the moving object to the own vehicle decreases with time, the own vehicle and the above A collision determination unit that determines that there is a possibility of collision with a moving object, and sets a collision risk for the moving object that is determined to have a collision possibility, and instructs execution of collision avoidance control ,
The collision determination unit
A moving object that moves in the width direction of the vehicle by setting the collision risk of the moving object higher than usual when the speed of the moving object that is determined to be likely to collide is higher than the speed of the host vehicle. The vehicle collision prevention apparatus is characterized in that the collision risk of the vehicle is set to a value corresponding to the possibility of entering the blind spot of the driver . The vehicle collision prevention apparatus according to claim 1 , wherein the moving object is a pedestrian or a light vehicle around the host vehicle. 2. The collision prevention apparatus for a vehicle according to claim 1 , wherein the moving object is another vehicle that travels on a road connected to a road on which the host vehicle travels.

Images