JP2014139756A - Drive assist device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a drive assist device that performs an appropriate drive assist in an early stage when an obstacle is likely to stop.SOLUTION: The drive assist device includes: obstacle detection means 11 that detects a position of an obstacle around an own vehicle; vector calculation means 31 that monitors the position of the obstacle, and calculates a relative movement vector of the obstacle to the own vehicle; drive assist means 12 that, when it is determined on the basis of the relative movement vector that the obstacle is likely to collide with the own vehicle, performs a drive assist control in accordance with an amount of time needed for the obstacle to collide with the vehicle; and drive assist change means 33 that, when it is determined that the obstacle is not likely to collide with the own vehicle on the basis of the relative movement vector and the obstacle passes ahead of the own vehicle, causes the drive assist means to perform a drive assist control different from that in a case where the obstacle is likely to collide with the own vehicle.

Description

  The present invention relates to a driving support device that detects an obstacle and performs control for alerting or reducing damage.

  There is known a driving support device that detects an obstacle by a radar or a camera and warns a driver or performs automatic braking or the like. In such a driving assistance device, various driving assistances based on TTC (Time To Collision) are mainly performed when there is a high possibility that the lateral position of an obstacle detected in a predetermined range in front of the own vehicle is in contact with the own vehicle. It is common to do.

  However, even when the host vehicle is traveling straight, the obstacle may move closer to the host vehicle due to the obstacle moving in a direction approaching the host vehicle. For such a situation, it is considered to predict the moving direction and the collision direction of the obstacle from the situation around the own vehicle (see, for example, Patent Documents 1 and 2).

  In Patent Document 1, based on the current behavior (neighboring vehicle status) of the surrounding vehicle, the behavior of the corresponding neighboring vehicle (predicted neighboring vehicle status) is predicted, and a countermeasure (notification) is referred to by referring to the vehicle behavior risk determination table. Content) and a surrounding vehicle monitoring device for determining a risk avoidance method is disclosed. Further, Patent Document 2 discloses a collision prediction device that acquires a predicted collision location from a traveling track of an own vehicle and a moving track of an obstacle and predicts which portion of the own vehicle will collide.

JP 2010-72969 A JP 2008-195293 A

  However, the techniques described in Patent Documents 1 and 2 have a problem that the act of stopping or suddenly stopping an obstacle is not considered. People with weak traffic such as pedestrians and bicycles often stop on the spot when they find a vehicle.

  Fig.1 (a) is an example of the figure explaining the relationship between the own vehicle and the pedestrian who stops. A pedestrian crosses in front of the vehicle. There is a sufficient distance from the own vehicle to the pedestrian, and when the pedestrian crosses the road, the own vehicle basically does not provide driving assistance such as warning the driver.

  However, in the case of a traffic weak person, without passing, there is a possibility that he / she will be surprised at the own vehicle and stop in front of the own vehicle. FIG.1 (b) is an example of the figure which shows the relative motion of the pedestrian with respect to the own vehicle. The solid line indicates the movement when passing, and the dotted line indicates the movement when stopping. It is possible for a pedestrian who moves (stops) like a dotted line to operate immediately before the collision, such as automatic braking for the purpose of reducing collision damage. Driving assistance by may not be in time.

  Therefore, it is preferable to provide appropriate driving assistance when there is a possibility of stopping, but conventionally, it has not been considered to give an early warning to an obstacle that may stop.

  An object of this invention is to provide the driving assistance apparatus which performs appropriate driving assistance at an early stage, when an obstruction may stop.

  The present invention provides obstacle detection means for detecting the position of an obstacle around the host vehicle, vector calculation means for monitoring the position and calculating a relative movement vector of the obstacle with respect to the host vehicle, and the relative movement vector. If it is determined that there is a possibility that the obstacle will collide with the host vehicle, driving assistance means for performing driving assistance control according to the time until the collision, and the obstacle based on the relative movement vector Driving that causes the driving support means to perform driving support control different from the case where it is determined that there is no possibility of colliding with the vehicle and the obstacle passes in front of the host vehicle and there is a possibility of collision. And a support changing means.

  When there is a possibility that an obstacle may stop, it is possible to provide a driving support device that performs appropriate driving support at an early stage.

It is an example of the figure explaining the relationship between the own vehicle and the pedestrian who stops. It is an example of the flowchart figure which shows the schematic operation | movement procedure of the driving assistance device of a present Example. It is an example of the schematic block diagram of a driving assistance device. It is an example of the figure explaining the distance L detected by a millimeter wave radar sensor, the relative velocity V, and the horizontal position x. It is an example of the functional block diagram of collision judgment ECU. It is an example of the figure explaining the calculation of a movement vector. It is an example of the figure demonstrated and illustrated the relationship between some relative movement vectors and the own vehicle. It is an example of the flowchart figure which shows the operation | movement procedure of a driving assistance device. It is an example of the figure which shows the position of the own vehicle and an oncoming vehicle. (Example 2) which is an example of the flowchart figure which shows the operation | movement procedure of a driving assistance device. (Example 3) which is an example of the flowchart figure which shows the operation | movement procedure of a driving assistance device. It is an example of the figure explaining calculation of TTC.

  Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings. However, the technical scope of the present invention is not limited to this embodiment.

  FIG. 2 is an example of a flowchart illustrating a schematic operation procedure of the driving support apparatus according to the present embodiment.

  The driving support device predicts a collision site (hereinafter referred to as a collision prediction site) when an obstacle collides with the host vehicle based on detection results of a radar, a camera, or the like (S10).

  The driving support device determines whether or not there is a high possibility of a collision with the host vehicle based on the predicted collision predicted part (S20).

  When there is a high possibility that the vehicle will collide with the host vehicle (Yes in S20), the driving support device activates various functions of PCS (Pre-Crash Safety System) (S30). That is, PCS functions (alarm, PBA: Pre-Crash Brake Assist, PSB: Pre-Crash Seat Belt, PB: Pre-Crash Break, steering avoidance based on TTC, etc., as well as driving assistance for obstacles ahead of the host vehicle Assistance, automatic steering avoidance, etc.) are used to provide appropriate driving assistance.

  If the possibility of collision with the host vehicle is not high (No in S20), the driving support device determines whether or not an obstacle is predicted to cross the front of the host vehicle (S40). That is, it is determined whether there is an obstacle that may stop in the middle of crossing the front. In the past, there was a case where the alarm could not be in time by suddenly stopping.

  When it is determined that the obstacle crosses the front of the host vehicle (Yes in S40), the driving support device performs warning, PBA, and steering avoidance support, but does not perform PB, PSB, and automatic steering avoidance (S50). . That is, although assistance is provided when a warning is given or a driver performs an operation for avoidance, the driver is not intervened to provide assistance.

  By such control, the driver notices an obstacle that may stop due to an alarm, and when the obstacle stops, the driver can surely avoid the obstacle with assistance. Moreover, since there is no intervention control when the obstacle does not stop, it can be prevented from feeling troublesome.

[Configuration example]
FIG. 3 shows an example of a schematic configuration diagram of the driving support device. The driving support device 100 includes a sensor unit 11, a collision determination ECU (Electronic Control Unit) 12, a brake ECU 13, an electric power steering ECU 15, a seat belt ECU 16, a brake ACT 14, a meter / buzzer 17, a steering ACT 18, and a seat belt 19. The sensor unit 11, each ECU, an actuator, and the like are communicably connected by an in-vehicle LAN such as a CAN (Controller Area Network) or a dedicated line.

  Each ECU is an information processing apparatus equipped with a microcomputer, a power source, a wire harness interface, and the like. The microcomputer has a known configuration including a CPU, a ROM, a RAM, an I / O, a CAN communication device, and the like.

(Sensor part)
Although the millimeter wave radar sensor 21 and the stereo camera 22 are illustrated as the sensor unit 11, it is sufficient that at least one of them is included. A camera is relatively suitable for detecting pedestrians. The millimeter wave radar sensor 21 is disposed at the center of the front of the vehicle such as the front grille of the vehicle, and emits a millimeter wave at a predetermined angle (for example, 10 degrees left and right with the front as the center). Receives millimeter waves reflected by objects in range. The millimeter wave radar sensor 21 is, for example, an FMCW (Frequency Modulated Continuous Wave) radar or a pulse radar.

  FIG. 4 is an example of a diagram illustrating the distance L, the relative velocity V, and the lateral position x (hereinafter referred to as target information) detected by the sensor unit 11. The detection principle of the target information by the millimeter wave radar sensor 21 is well known and will be described briefly. The millimeter wave radar sensor 21 generates a beat signal for each reception antenna by mixing the transmission signal and the reception signal with a mixer. The time from when the transmission signal is transmitted until the reception signal is received is proportional to the distance to the object, and the frequency of the beat signal is shifted by the relative speed. Therefore, distance and relative speed can be obtained by FFT analysis of the beat signal, for example.

  Further, the horizontal position x (azimuth θ) of the obstacle includes MUSIC (Multiple Signal Classification) analysis, Capon analysis, DBF (Digital Beam Forming) processing, or monopulse method. Also good. In the figure, there is one obstacle, but even when there are a plurality of obstacles, the target information of each target can be obtained. The millimeter wave radar device outputs the distance L, the relative speed v, and the lateral position x to the collision determination ECU 12 periodically (for example, 100 milliseconds).

  For example, the stereo camera 22 is arranged on the rearview mirror with the optical axis facing the front of the vehicle. The stereo camera 22 has two CCD cameras or two CMOS cameras that are spaced apart by a predetermined distance. The stereo camera 22 performs pre-processing for removing lens distortion, optical axis deviation, focal length deviation, imaging element distortion, and the like on the frames imaged by each camera using calibration data prepared in advance. As a result, the frames of the two cameras have only a difference corresponding to parallax.

Since a method for detecting target information by the stereo camera 22 is also known, it will be briefly described. The stereo camera 22 evaluates the correlation between the left and right image data by a technique such as block matching, and calculates the parallax generated in the pixels where the same object is photographed. Thereby, the shift amount is obtained for every pixel for all the pixels. Assuming that the number of shifted pixels is n, the focal length of the lens is f, the distance between the optical axes (between two cameras) is m, and the pixel pitch is d, the distance L to the imaging object can be obtained from the following equation.
L = (f × m) / (n × d)
If the focal length is used by knowing the distance, for example, the three-dimensional coordinates for each pixel in the coordinate system with the camera as the origin can be obtained. The stereo camera 22 scans the image data in the vertical direction and determines that there is an obstacle other than the road surface in the pixel whose height changes (increases). Since pixels with the same distance among the pixels with obstacles can be estimated to be the pixels showing the same obstacle, these pixels are grouped to be obstacles such as pedestrians and preceding vehicles. presume.

  After that, the stereo camera 22 uses the HOG (Histograms of Oriented Gradients), Joint HOG, CPF (Cooccurrence Probability Features), and CoHOG (Co-occurrence Histograms of Oriented Gradients) to the estimated pedestrians and preceding vehicles. Create and recognize pedestrians and vehicles by pattern matching. For example, the HOG calculates the gradient strength and gradient direction in the vertical and horizontal directions for all the pixels of the image data, and adds the gradient strength to the bin corresponding to the gradient direction of each pixel in non-overlapping cells of n × n pixels. As a result, a nine-dimensional feature vector is obtained for each cell. Further, an n cell × n cell block is created, and the feature vectors in the block are connected while being shifted one cell at a time, and normalized. Thereby, a feature vector having a number of dimensions of the number of blocks × n × n × 9 is obtained. For example, if a feature vector of a teacher image is learned by an SVM (Support Vector Machine), a pedestrian can be detected by inputting the feature vector of the running image into the SVM.

  Since the stereo camera 22 repeatedly captures a predetermined number (30 to 60) of images per second, the three-dimensional coordinates of each object can be obtained for each frame. Therefore, the relative speed V is obtained from a change in the distance L between the frames. The lateral position x of the obstacle is clear from the three-dimensional coordinates. The stereo camera 22 may be a monocular camera. In this case, distance information can be acquired by an optical flow or the like.

  In this way, the millimeter wave radar sensor 21 and the stereo camera 22 can obtain equivalent information. However, the millimeter wave radar sensor 21 has high accuracy in distance and relative velocity, and the stereo camera 22 has low accuracy in distance and relative velocity and high accuracy in azimuth. Therefore, in the actual vehicle, the vehicle control is performed mainly using high-precision information of each sensor, and the reliability of the information is determined by comparing two pieces of information.

[Collision judgment ECU]
Returning to FIG. 3, the collision determination ECU 12 determines whether or not it collides with an obstacle by two methods. (i) One is a method of determining the possibility of a collision with a preceding vehicle ahead of the host vehicle, and a known method can be used. (ii) The other is a method of calculating a relative movement vector of the obstacle with respect to the host vehicle and determining whether or not there is a possibility of collision with the host vehicle.

(i) The collision determination ECU calculates TTC and determines the possibility of collision based on the distance, relative speed, and lateral position. For example, TTC is calculated as follows.
TTC = distance / relative speed For each detected object, the collision determination ECU 12 determines the object having the highest possibility of collision from the TTC and the lateral position (in the objects whose lateral position is closer to the vehicle than the predetermined value). The object having the smallest TTC is specified, and the possibility of collision with the object is determined. For example, when the smallest TTC is equal to or smaller than the threshold and the lateral position is within a range overlapping with the vehicle width of the host vehicle, it is determined that there is a possibility of collision with the obstacle. The collision determination ECU outputs a PCS operation request to the brake ECU 13, the electric power steering ECU 15, the seat belt ECU 16, and the like (hereinafter sometimes referred to as a brake ECU) according to the TTC. To do.
・ When there is a high possibility of collision: warning sound or warning lamp display
-If collision is inevitable: PSB, PB, PBA, steering avoidance support, automatic steering avoidance,
(ii) The relative movement vector of the obstacle is calculated from the target information, and whether or not the obstacle collides with the front or side surface of the own vehicle (whether or not the movement vector and the front line or side line of the own vehicle intersect). ). Further, when an obstacle collides with the host vehicle, a collision prediction part indicating whether the obstacle collides with the front line or the side line and further collides with the front part or the side part is predicted. When the vehicle collides with the side of the host vehicle or collides with the front of the host vehicle, the PCS is operated by the brake ECU or the like according to the time until the collision, the collision probability, and the collision predicted part. On the other hand, when it is estimated that the vehicle will cross the front of the vehicle, the brake ECU or the like is requested to operate a part of the PCS not including intervention control according to the time until the collision.
・ When there is a high possibility of collision: warning sound or warning lamp display
-When collision is unavoidable: PBA, steering avoidance assistance An example of the contents of each PCS driving assistance is as follows.
Warning: Alerting by sound and visual information PB: Automatic sudden braking to avoid collision even if the driver does not depress the brake pedal PBA: PSB that assists in pressurization when the driver depresses the brake pedal: Automatic steering avoidance that automatically winds up the seat belt 19: Steering avoidance support that steers the steering in a direction that avoids obstacles even if the driver does not steer: Increase the steering amount of the driver These are examples. Thus, for example, driving support such as control for extending the headrest forward may be performed, and specific control contents for realizing driving support are not limited.

[Each ECU that performs PCS]
The brake ECU 13 is connected to the brake ACT 14 and the meter buzzer 17. The brake ECU 13 transmits an actuator operation request to at least one of the brake ACT 14 and the meter buzzer 17.

  The brake ACT 14 includes a pressure reducing control valve (normally closed valve), a holding control valve (normally opened valve), and a hydraulic pressure for controlling the wheel cylinder pressure for each wheel in a hydraulic path communicating with the wheel cylinder of each wheel. A pump and a pump motor for generating are arranged. When automatic braking is not performed, the brake ECU 13 keeps the pressure reducing control valve closed and the holding control valve opened. When the driver operates the brake pedal, the working fluid passes through the holding valve and is supplied to the wheel cylinder. By opening and closing these valves, automatic braking, maintaining braking force, and reducing braking force are possible.

  The brake ECU 13 acquires an operation request from the collision determination ECU 12 and controls PB, PBA, and the like. In some cases, a light braking is applied to the control content of the PB to give a warning or a gentle brake may be included. Such a PB is included in the warning in this embodiment.

  The meter buzzer 17 sounds various display devices on the meter panel and a buzzer sound according to instructions from the brake ECU 13. The meter panel warns the driver that there is a risk of collision with an object by lighting or blinking an icon on a liquid crystal display or displaying a message. In addition, the driver is warned by turning on or blinking a warning lamp. The meter buzzer 17 warns the driver by sounding a warning sound or outputting a voice message that there is a possibility of a collision.

  The electric power steering ECU 15 is connected to the steering ACT 18. The steering ACT 18 includes a motor that rotates the steering shaft, a speed reduction mechanism, and the like, and rotationally drives the steering shaft without the driver's steering or with the driver's steering. A torque sensor for detecting the steering torque of the driver, a rotation angle sensor, a lock mechanism, and the like are disposed on the steering shaft.

  When the electric power steering ECU 15 obtains an operation request from the collision determination ECU 12, the electric power steering ECU 15 controls the steering ACT 18 to perform at least one of automatic steering avoidance and steering avoidance assistance. In the case of automatic steering avoidance, the operation request includes a steering amount in a direction to avoid an obstacle. In the case of steering avoidance assistance, the electric power steering ECU 15 adds a steering force or a steering amount to the steering direction detected by the torque sensor to make it larger than the steering amount of the driver.

  The seat belt ECU 16 is connected to the seat belt 19. The seat belt 19 can be wound up by the seat belt ECU 16 by a motor or a clutch that winds up webbing corresponding to PCS. When the seat belt ECU 16 obtains an operation request from the collision determination ECU 12, the seat belt ECU 16 performs control (PSB) to wind up the seat belt 19.

[Calculation of movement vector, etc.]
FIG. 5 shows an example of a functional block diagram of the collision determination ECU 12. The movement vector calculation unit 31 calculates a relative movement vector of the obstacle. Since the obstacle ahead of the host vehicle is determined to collide by the method (i) above, the target obstacle in the present embodiment is an obstacle whose lateral position is left or right with respect to the vehicle width of the host vehicle. In addition, as obstacles, weak people such as pedestrians, bicycles, and electric motorcycles are assumed, but are not limited thereto.

  The collision site prediction unit 32 determines whether or not there is a possibility of collision with either the front surface or the side surface of the host vehicle. In the case of a collision, it is also possible to estimate at which position on the front and side of the host vehicle the collision occurs. The collision probability calculation unit 34 calculates the collision probability of the obstacle determined to collide. The collision probability is obtained, for example, from the number of times that the collision is determined with respect to the number of times the movement vector is calculated. The front crossing determination unit 35 identifies an obstacle that crosses the front of the host vehicle among the obstacles determined not to collide with the front surface or the side surface of the host vehicle. The operation requesting unit 33 makes an operation request using all the functions of the PCS when it is predicted to collide with the front or side surface of the host vehicle, and when it is estimated to cross the front of the host vehicle, the PSB of the PCS, automatic The PCS operation request without steering avoidance and PB is performed.

  FIG. 6A is an example of a diagram illustrating the calculation of the movement vector. A pedestrian is moving in a direction crossing a road ahead of the host vehicle. When the movement vector with respect to the road surface of the pedestrian is P and the movement vector with respect to the road surface of the own vehicle is C, the relative movement vector of the pedestrian viewed from the own vehicle is PC. The movement vector calculation unit 31 obtains a pedestrian's relative movement vector corresponding to PC. Since the relative speed of the target information does not include the moving direction, for example, each time the sensor unit 11 detects pedestrian target information, the position of the pedestrian (for example, the vehicle's position) is detected from the lateral position and distance. The center in the vehicle width direction is the origin). A pedestrian's relative movement vector (relative movement direction) can be obtained by linearly approximating a plurality of past positions. Alternatively, the movement vector P of the pedestrian may be obtained from a plurality of past horizontal positions or the optical flow of the pedestrian, and the relative movement vector may be calculated from the movement vector C of the own vehicle.

FIG. 6B is an example of a diagram illustrating determination of a collision between the host vehicle and a pedestrian. The collision site prediction unit 32 determines whether or not the relative movement vector intersects the front line or the side line of the host vehicle. The formula that represents the front line. For example, the following equation is known.
y = 0-a / 2 <x <a / 2 However, the expression a represents the vehicle width side line is known as the following expression, for example.
x = -a / 2 -b <y <0 where b is the vehicle length. Therefore, the collision site prediction unit 32 indicates that the straight line of the relative movement vector (extending the relative movement vector in the direction of the vehicle) is the front line or the side surface. It can be easily determined whether or not the line intersects.

For a pedestrian for which it is determined whether or not to collide, the collision probability calculation unit 34 calculates a collision probability.
Collision probability = number of times determined to collide n / number of times N judged whether or not to collide
And the front crossing determination part 35 determines whether it crosses the front of the own vehicle with respect to the pedestrian determined not to collide. If the x coordinate of the intersection of the relative movement vector and the straight line with y = 0 is larger than a / 2, it can be seen that it crosses the front, and the y coordinate of the intersection of the relative movement vector and the straight line with x = −a / 2 is If it is smaller than -b, it will be understood that the pedestrian passes behind the host vehicle after passing the host vehicle.

  In addition, you may ignore the pedestrian who is determined to pass sufficiently ahead of the host vehicle. This is because even if the pedestrian stops, the collision determination ECU 12 has time to sound an alarm, and the driver can respond with sufficient margin. In this case, there is no collision. Therefore, when the x coordinate of the intersection of the relative movement vector and the straight line of y = 0 is larger than a / 2 and within a predetermined value (for example, 2 to 3 times the vehicle width, it may be increased or decreased depending on the vehicle speed). It is determined that the pedestrian crosses the front.

FIG. 7 is an example of a diagram illustrating and explaining the relationship between several relative movement vectors and the host vehicle.
-It is determined that the relative movement vector 1 does not collide.
The relative movement vector 2 is determined to collide with the side surface.
The relative movement vector 3 is determined to collide with the front surface.
The relative movement vector 4 is determined not to collide and is determined to cross the front.
The relative movement vector 5 is determined not to collide and is determined not to cross the front.

Therefore, the operation requesting unit 33 performs driving assistance (collision damage reduction) using all the functions of the PCS based on the TTC and the collision probability for the relative movement vectors 2 and 3. As for the collision probability, for example, it is determined that the collision occurs when the collision probability tends to increase.
・ When there is a high possibility of collision: warning sound or warning lamp display
-If collision is inevitable: PSB, PB, PBA, steering avoidance support, automatic steering avoidance,
For the relative movement vector 4, driving support using a part of the functions of the PCS is performed based on TTC. In this case, the host vehicle and the pedestrian (relative movement vector 4) do not collide, but the TTC can be obtained mechanically. More accurate TTC will be described in Example 4.
-TTC is less than threshold: Alarm sound or warning lamp display
-TTC is above threshold: PBA, steering avoidance assistance By doing so, warning sounds can be given to pedestrians crossing the front, so if the pedestrian stops in front of the vehicle, the driver has a margin Yes. Further, PBA works when the brake pedal is depressed, and steering avoidance support works when steered, so avoidance is easy. Further, since there is no intervention control, the driver does not feel bothered if the pedestrian does not stop in front of the host vehicle.

[Operation procedure]
FIG. 8 is an example of a flowchart illustrating an operation procedure of the driving support apparatus 100 according to the present embodiment. The procedure of FIG. 8 is repeatedly executed while the vehicle is traveling.

  The sensor unit 11 periodically detects an obstacle and outputs the distance, relative speed, and lateral position to the collision determination ECU 12 (S110).

  The movement vector calculation unit 31 calculates the relative movement vector of the obstacle from the distance, the relative speed, and the lateral position (S120).

  The collision site prediction unit 32 predicts a collision prediction site with the host vehicle based on the relative movement vector (S130).

  When the collision prediction part is a side surface, the operation requesting unit 33 controls operation start timings such as warning, PBA, PSB, steering avoidance support, and automatic steering avoidance according to the TTC and the collision probability. In addition, for PB, the start timing is controlled (S140).

  When the collision prediction part is the front surface, the operation requesting unit 33 controls the operation start timing such as warning, PB, PBA, PSB, steering avoidance support, automatic steering avoidance, etc. according to TTC and collision probability (S150).

  When it is determined that the collision site prediction unit 32 does not collide, the front crossing determination unit 35 determines whether or not the obstacle crosses the front (S160).

  When it determines with not crossing the front (No of S160), the operation | movement request | requirement part 33 does not operate PCS (S180).

  When it determines with crossing ahead (Yes of S160), the action | operation request | requirement part 33 controls the action | operation start timing of a warning, PBA, and steering avoidance assistance according to TTC (S170). PB, PSB and automatic steering avoidance are not activated.

  In S170, PB may not be completely prohibited, but temporary braking as a warning (sensory braking) and gentle braking for relaxing the degree of approach to an obstacle may be permitted. This is because they are closer to warnings than interventions, so it is difficult to inhibit driving.

  As described above, the driving support device 100 according to the present embodiment can perform appropriate driving support at an early stage when there is a possibility that an obstacle may stop.

  In this embodiment, a driving support device 100 corresponding to an oncoming vehicle will be described. The functional block diagram is the same as that in the first embodiment.

  FIG. 9 is an example of a diagram illustrating positions of the host vehicle and the oncoming vehicle. Since the oncoming vehicle is traveling along a curve, it does not come into contact with the host vehicle, but may be determined to cross the front of the host vehicle based on the relative movement vector. For this reason, the driving assistance apparatus 100 performs warning, PBA, and steering avoidance assistance according to the relationship with TTC.

  Therefore, in this embodiment, it is determined that the vehicle crosses the front of the host vehicle by limiting to obstacles with small movement in the front-rear direction. Since the oncoming vehicle has a higher speed in the front-rear direction (y direction) than the pedestrian, it is possible to eliminate the PCS driving assistance for the oncoming vehicle by limiting the obstacle to an obstacle with a small movement in the front-rear direction.

The speed in the y direction with respect to the road surface of the oncoming vehicle can be obtained from the relative speed and the speed of the host vehicle. Since it is a relative speed, the speed in the direction approaching the host vehicle (y direction) is known, not the speed in the traveling direction of the oncoming vehicle. For example, the movement vector calculation unit 31 subtracts the speed of the host vehicle from the relative speed to obtain the speed in the y direction with respect to the road surface of the oncoming vehicle.
Actual vehicle speed of obstacle (y-direction component) = relative speed−speed of own vehicle FIG. 10 is an example of a flowchart showing an operation procedure of the driving support device 100 of the present embodiment. FIG. 10 is substantially the same as FIG. 8, but after step S150, it is determined whether or not the speed of the obstacle in the front-rear direction is equal to or less than a threshold value (S162).

  When the speed of the obstacle in the front-rear direction is equal to or lower than the threshold value (Yes in S162), the operation request unit 33 controls the operation start timing of alarm, PBA, and steering avoidance support according to TTC (S150). In addition, bodily sensation braking and slow braking may be permitted.

  According to the present embodiment, it is possible to suppress the driving assistance device 100 from providing warning, PBA, and steering avoidance assistance to the oncoming vehicle.

  In the second embodiment, the intervention control is prohibited and the functions of a part of the PCS are activated by limiting the obstacle to a small movement in the front-rear direction. However, in this embodiment, it is detected that the obstacle is a traffic weak person. The driving support apparatus 100 that activates some functions of the PCS will be described.

  Although the functional block diagram may be the same as that in the first embodiment, the stereo camera 22 recognizes a traffic weak person such as a pedestrian or a bicycle from the image data. For example, the front crossing determination unit 35 effectively maintains the determination of crossing the front only when the obstacle is a traffic weak person.

  FIG. 11 is an example of a flowchart illustrating an operation procedure of the driving support apparatus 100 according to the present embodiment. FIG. 11 is substantially the same as FIG. 8, but after step S150, it is determined whether the obstacle is a traffic weak person (S164).

  When the obstacle is a traffic weak person (Yes in S164), the operation request unit 33 controls the operation start timing of warning, PBA, and steering avoidance support according to TTC (S150). In addition, bodily sensation braking and slow braking may be permitted.

  According to the present embodiment, the driving support device 100 can perform warning light, PBA, and steering avoidance support only for the traffic weak.

  In Examples 1 to 3, TTC was mechanically obtained, and a part of PCS whose driving timing was controlled for obstacles crossing the front of the host vehicle based on TTC was supported. However, since this obstacle does not collide with the host vehicle, the TTC may not be accurate. Therefore, in the present embodiment, a more preferable TTC calculation method for an obstacle crossing the front of the host vehicle will be described.

  FIG. 12 is an example of a diagram illustrating the calculation of TTC. The intersection of the relative movement vector and the straight line of y = 0 is defined as a virtual collision lateral position Cr. Then, if the TTC until the pedestrian reaches this virtual collision lateral position Cr is calculated, the time until the obstacle approaches to the side of the host vehicle can be known, so that it becomes appropriate as an index for operating the PCS. I can expect.

The relative velocity V of the obstacle is detected by the sensor unit 11, but the distance L ′ between the virtual collision lateral position Cr and the obstacle is not clear. Therefore, for example, the collision site prediction unit 32 calculates the distance L ′ using the azimuth θ (lateral position) and the cosine theorem.
L ′ ² = distance L ² + Cr ² −2 · distance L · Cr · COS (90 + θ)
TTC = L '/ V
Note that the moving speed to the virtual collision lateral position Cr and the relative speed V do not exactly match, but may match because the speed of the pedestrian is slower than the vehicle speed. Further, the moving speed from the relative movement vector to the virtual collision lateral position Cr may be calculated.

  Therefore, according to the present embodiment, even when an obstacle crosses the front, an accurate TTC is calculated, so that warning, PBA, and steering avoidance support can be performed at an appropriate timing.

11 Sensor part 12 Collision judgment ECU
13 Brake ECU
15 Electric power steering ECU
16 Seat belt ECU
DESCRIPTION OF SYMBOLS 31 Movement vector calculating part 32 Collision site | part prediction part 33 Action request | requirement part 34 Collision probability calculation part 35 Forward crossing determination part 100 Driving assistance apparatus

Claims (7)

Obstacle detection means for detecting the position of obstacles around the host vehicle;
Vector calculation means for monitoring the position and calculating a relative movement vector of the obstacle with respect to the vehicle;
When it is determined that the obstacle may collide with the host vehicle based on the relative movement vector, driving assistance means that performs driving assistance control according to the time until the collision;
When the obstacle does not collide with the host vehicle based on the relative movement vector and the obstacle is determined to pass in front of the host vehicle, the driving is different from the case where it is determined that the obstacle may collide. Driving support change means for causing the driving support means to perform support control; and
A driving support device comprising: The driving support means is capable of intervention control for controlling the vehicle so as to avoid a collision with the obstacle without a driver's operation,
When it is determined that there is no possibility that the obstacle collides with the host vehicle based on the relative movement vector and the obstacle passes in front of the host vehicle, the driving support change unit performs intervention control on the driving support unit. To perform driving support control other than
The driving support apparatus according to claim 1. The obstacle detection means detects relative speed information with the obstacle,
Based on the relative movement vector, the obstacle is not likely to collide with the host vehicle, the obstacle is determined to pass in front of the host vehicle, and the obstacle is obtained from the speed of the host vehicle and the relative speed information. When the approaching speed of the vehicle with respect to the road surface is equal to or less than a threshold value, the driving support changing means causes the driving support means to perform different driving support control from that determined as the possibility of collision. The driving support apparatus according to claim 1 or 2. Traffic weak detection means for detecting that the obstacle is a traffic weak for a pedestrian or a bicycle occupant,
Driving assistance change means when it is determined that the obstacle does not collide with the host vehicle based on the relative movement vector, the obstacle passes in front of the host vehicle, and the obstacle is a traffic weak person. 3. The driving support device according to claim 1, wherein the driving support unit performs driving support control different from that when it is determined that there is a possibility of collision. An intersection calculation means for obtaining an intersection of the relative movement vector and a front line that linearly extends the front of the host vehicle in a horizontal direction;
The driving support change means obtains a distance between the intersection and the obstacle based on the intersection and the position,
The relative speed information with respect to the obstacle detected by the obstacle detection means, and the distance to calculate the time until the obstacle reaches the intersection,
Depending on the calculated time, the driving support means performs driving support control different from the case where it is determined that there is a possibility of a collision,
The driving support apparatus according to any one of claims 1 to 4, wherein When the above-described intervention control enables automatic braking control to decelerate the host vehicle by applying a braking force without a driver's operation,
When it is determined that the obstacle does not collide with the host vehicle based on the relative movement vector, and the obstacle passes through the front of the host vehicle, the driving support change unit is configured to temporarily decelerate by the automatic braking control. Or, although slow deceleration is permitted, rapid deceleration for avoiding a collision by the automatic braking control is prohibited.
The driving support apparatus according to claim 2, wherein: The driving support control includes an alarm output, a pre-crash brake assist that pressurizes and assists the driver's depression of the brake pedal, a pre-crash brake that reduces the damage caused by the collision with the obstacle, and an obstacle even if the driver does not steer Automatic steering avoidance to steer the vehicle so as to avoid the vehicle, and steering avoidance assistance to increase the steering amount when the driver steers,
When it is determined that the obstacle does not collide with the host vehicle based on the relative movement vector and the obstacle passes through the front of the host vehicle, the driving support change unit automatically steers the driving support unit. Driving assistance other than avoidance and pre-crash braking,
The driving support apparatus according to any one of claims 1 to 6, wherein

Images