JP2010079425A - Vehicle driving support device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To suppress an avoidance operation by setting a risk based on an area where a three-dimensional object crosses a road, and stop a vehicle safely before the crossing area, in driving support. <P>SOLUTION: When a three-dimensional object such as a pedestrian exists in a pedestrian crossing area, a risk function Dline and a cross risk function Dcross with respect to a road are synthesized, and a total risk function D is set so that distribution having a relatively high risk value at a road edge side appears within a predetermined distance from the vehicle and that ridge-like risk distribution having a high risk value in a road width direction appears in the pedestrian crossing region. Then, when the risk value in the ridge-like risk distribution reaches a threshold value as the vehicle travels, a warning for stop is output to a driver, and an automatic brake operates as necessary. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a driving support apparatus for a vehicle that sets a risk in the surrounding environment of the host vehicle and performs driving support of the host vehicle based on the risk.

  In recent years, in vehicles such as automobiles, the surrounding environment outside the vehicle is detected by an in-vehicle camera, a laser radar device, etc., and obstacles and three-dimensional objects such as preceding vehicles are recognized, and various controls such as warning, automatic braking, and automatic steering are performed. Technologies that avoid collisions and improve safety by executing them have been developed and put to practical use.

For example, in Patent Document 1 (Japanese Patent Application Laid-Open No. 2002-74594), assuming a sphere having a predetermined radius centered on the coordinate values of the host vehicle or an obstacle, the collision probability corresponding to the distance calculation result between the spheres is set. A technique for determining the possibility of collision between an obstacle and the host vehicle by calculating is disclosed.
Japanese Patent Laid-Open No. 2002-74594

  However, in the technique disclosed in Patent Document 1, although there is a pedestrian crossing, the risk of collision with a pedestrian can be set high by increasing the risk of pedestrians, but avoiding pedestrians on the pedestrian crossing. There is a possibility that the steering control is executed to a person having a low risk, and it is difficult to perform control such as warning based on the pedestrian crossing or stopping the vehicle in front of the pedestrian crossing.

  In this way, in the conventional technology, since the risk is set for the obstacle itself, in a situation where a pedestrian is walking at an intersection or a pedestrian crossing, avoid the situation where the possibility of a collision with the own vehicle is low There is a possibility of intruding into a pedestrian crossing by controlling, and there is a concern of giving danger to pedestrians.

  The present invention has been made in view of the above circumstances, and sets a risk based on a crossing region of a three-dimensional object that crosses a road, suppresses avoidance operations, and safely supports the vehicle in front of the crossing region. It is an object of the present invention to provide a vehicle driving support device that enables the vehicle.

  In order to achieve the above object, a driving support apparatus for a vehicle according to the present invention sets a risk in the surrounding environment of the own vehicle, and the driving support apparatus for a vehicle that performs driving support of the own vehicle based on the risk includes: A crossing three-dimensional object recognition unit for recognizing a three-dimensional object presumed to be crossing or crossing a traveling road, and a risk based on the presence of the three-dimensional object, crossing at least the traveling lane of the own vehicle in the width direction of the road And a crossing risk setting unit for setting as a distributed crossing risk.

  ADVANTAGE OF THE INVENTION According to this invention, the driving assistance which suppresses avoidance operation | movement and stops the own vehicle safely before a crossing area | region is attained by setting a risk on the basis of the crossing area | region of the solid thing which crosses a road.

  Embodiments of the present invention will be described below with reference to the drawings. 1 to 5 relate to a first embodiment of the present invention, FIG. 1 is a schematic configuration diagram of a driving support device mounted on a vehicle, FIG. 2 is an explanatory diagram showing the own vehicle and a pedestrian crossing area, and FIG. 3 is a crossing FIG. 4 is an explanatory diagram showing the risk distribution when there is no three-dimensional object in the sidewalk area, FIG. 4 is an explanatory diagram showing the risk distribution when there is a three-dimensional object in the pedestrian crossing area, and FIG. 5 is viewed from a direction orthogonal to the traveling direction. It is explanatory drawing which shows distribution of a crossing risk.

  In FIG. 1, reference numeral 1 denotes a vehicle (host vehicle) such as an automobile, and a driving support device 2 that recognizes an external driving environment and performs driving support for avoiding a collision with an obstacle is provided on the host vehicle 1. It is installed. In the present embodiment, the driving support device 2 is based on a device group for recognition of the external environment by the stereo camera 3, the stereo image recognition device 4, and the traveling environment information acquisition device 5, and information from each device. A control unit 6 including a microcomputer or the like that performs various processes for driving support is provided as a main part. The control unit 6 is connected to various devices related to driving support such as a display 21 that also serves as an alarm device, an automatic brake control device 22, and an automatic steering control device 23.

  The stereo image recognition device 4, the driving environment information acquisition device 5, the control unit 6, the automatic brake control device 22, the automatic steering control device 23, and the like are each configured as a control unit including a single or a plurality of computer systems. Data is exchanged with each other via a communication bus.

  Further, the host vehicle 1 is provided with a vehicle speed sensor 11 that detects the host vehicle speed V, a yaw rate sensor 12 that detects a yaw rate, a main switch 13 to which an ON-OFF signal for driving support control is input, and the like. The own vehicle speed V is input to the stereo image recognition device 4 and the control unit 6, the yaw rate is input to the control unit 6, and the ON / OFF signal for driving support control is input to the control unit 6.

  The stereo camera 3 is used as one of recognition sensors for recognizing the external environment, and includes a pair of (left and right) cameras using a solid-state image sensor such as a CCD or a CMOS. Each camera is attached to the front of the ceiling in the vehicle interior with a certain baseline length, and takes a stereo image of an object outside the vehicle from different viewpoints, and outputs image data to the stereo image recognition device 4.

  The stereo image recognition device 4 includes an image processing engine that processes an image captured by the stereo camera 3 at a high speed, and is configured as a processing unit that performs a recognition process based on an output result of the image processing engine. Image processing of the stereo camera 3 in the stereo image recognition device 4 is performed as follows, for example.

  That is, first, the stereo image recognition device 4 obtains distance information from a corresponding position shift amount with respect to a pair of stereo images in the traveling direction of the host vehicle 1 captured by the stereo camera 3, and generates a distance image. . 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. These data are calculated as data in a 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. White line data, guardrails existing along the road, curbstones And the like, the type of the three-dimensional object, the distance from the host vehicle 1, the center position, the speed, and the like are transmitted to the control unit 6 as obstacle information.

  In this embodiment, an example in which the stereo camera 3 is used as one of the recognition sensors will be described. However, other surrounding sensors may be used to recognize the surrounding environment using other recognition sensors such as a monocular camera and a millimeter wave radar. May be.

  The travel environment information acquisition device 5 detects objects in a wider range than the detection range of the stereo camera 3 and acquires various information such as road shape, information on the presence / absence of intersections / crosswalks, position information, and traffic jam information. . Specifically, the traveling environment information acquisition device 5 receives light and radio wave beacons from road incidental facilities and acquires various information such as traffic congestion information, weather information, and traffic regulation information for a specific area. Communicate with other vehicles around the vehicle (vehicle-to-vehicle communication), and exchange vehicle information such as vehicle type, vehicle position, vehicle speed, acceleration / deceleration state, brake operation state, blinker state, etc. It is configured as a device capable of acquiring a wide range of travel environment information by collecting information from a communication device, a positioning device such as a GPS, and a navigation device.

  Based on the vehicle speed V from the vehicle speed sensor 11, the yaw rate from the yaw rate sensor 12, the obstacle information from the stereo image recognition device 4, and the obstacle information from the traveling environment information acquisition device 5, the control unit 6 A risk level (risk) is set in the surrounding environment of the vehicle, and driving support control such as warning, braking control, steering support and the like for the driver is executed based on this risk.

  In this case, the control unit 6 stops the host vehicle 1 without giving anxiety in front of these, particularly for pedestrians, bicycles, welfare vehicles, etc. crossing the road, and to ensure safety. A support function is provided. The support function of the control unit 6 is that the risk based on the presence of a three-dimensional object that is estimated to cross or cross the road on which the host vehicle 1 travels is defined as a crossing risk distributed in the width direction of the road on which the host vehicle 1 travels. This is realized by the function as a crossing risk setting section to be set.

  The three-dimensional object estimated to cross the road on which the host vehicle 1 travels is a device group for external environment recognition, in this embodiment, a stereo camera 3, a stereo image recognition device 4, and travel environment information acquisition. Recognized by the function of the crossing three-dimensional object recognition unit by the device 5. For example, when a pedestrian crossing and a pedestrian (a three-dimensional object) are recognized by the stereo camera 3, the pedestrian is not only during the crossing but also before the start of the crossing, and the crossing risk is set. In addition, even if the pedestrian is stopped, if the position of the pedestrian is close to a pedestrian crossing or intersection, it is assumed that the pedestrian will cross the pedestrian crossing or intersection and a crossing risk is set at the pedestrian crossing or intersection. To do.

For this reason, the control unit 6 uses a risk for the road on which the host vehicle 1 travels and a risk (crossing risk) for a three-dimensional object (obstacle) that crosses the road in the traveling direction of the host vehicle 1 as risk functions Dline and Dcross, respectively. It is converted into a function, and a risk distribution in an XY coordinate system with the traveling direction of the host vehicle 1 as the X axis and the width direction as the Y axis is obtained. These risk functions Dline and Dcross are added and integrated as shown in the following equation (1) to set a total risk function D. Based on the total risk function D, an alarm output and an automatic brake control device are set. By carrying out automatic braking via 22, the host vehicle 1 can be safely stopped before a pedestrian or the like.
D = Dline + Dcross (1)

The risk function Dline set for the road is set by a function that leads to a larger risk value, for example, a function as shown in the following equation (2), as it is closer to the end position from the road center. .
Dline = exp (aR · y ⁴ ) −1 (2)
However, aR = (2 / WR) ⁴ · log (Ds + 1)
WR: Road width
Ds: Risk at the road edge (y = ± WR / 2)

The risk function Dline may be derived by a higher order function of quadratic or higher. For example, when a quadratic function is used, a predetermined gain is obtained as shown in the following equation (2 ′). It can be a risk function with Kline.
Dline = Kline · y ² (2 ′)

  Further, the crossing risk function Dcross is a function having a predetermined distribution according to the presence or absence of an obstacle and the distance from the host vehicle 1, and for example, as shown in FIG. If a predetermined area including the pedestrian 50 and the pedestrian crossing P is set as the pedestrian crossing area over the entire road width, and there is a three-dimensional object in the pedestrian crossing area, the risk value is higher than the surrounding area. Set the correct function.

In FIG. 2, the crossing risk function Dcross is calculated by the following equation (3), where xg is the distance from the center position of the host vehicle 1 to the center position of the pedestrian crossing P.
Dcross = a _cross · exp (−a _x · (x−xg) ² (3)
Where a _{cross is} a correction parameter corresponding to a three-dimensional object
a _x : Correction parameter according to the width of the pedestrian crossing

  Note that the crossing risk function Dcross is not limited to the expression (3), and a quadratic function or other functions can be substituted by limiting the application range in the X-axis direction.

The correction parameter a _{cross in the} equation (3) is a parameter that is basically set to zero or a value close to zero when a three-dimensional object does not exist. At this time, the crossing risk function Dcross becomes a value close to zero or zero. The total risk function D is represented by the risk function Dline for the road. As a result, as shown in FIG. 3, the risk distribution shown in the three-dimensional coordinate system of the X, Y, and D axes is such that the risk value on the road edge side is slightly higher than the center of the road (Y = 0). Distribution.

  On the other hand, when a three-dimensional object such as a pedestrian exists in the pedestrian crossing area, the total risk function D is a distribution obtained by combining the risk function Dline and the crossing risk function Dcross for the road, and as shown in FIG. From 1 to a predetermined distance, the risk value on the road edge side is a slightly high distribution, and a distribution having a high risk value that forms a ridge in the road width direction appears in the pedestrian crossing area ahead. Then, as shown in FIG. 5, when the risk value in the ridge-like risk distribution reaches the threshold value Rk due to the progress of the host vehicle 1, a warning to stop the driver is output, and in some cases, the automatic brake is activated. Let

  In such a case, in the conventional technique, the risk function is simply set around the position where the three-dimensional object is present. For example, the risk is set only in a predetermined range surrounding a pedestrian walking on a pedestrian crossing. Become. For this reason, there is a possibility of passing a pedestrian crossing while avoiding pedestrians, which may give a sense of danger to pedestrians.

  On the other hand, in the present driving support device 2, the crossing risk having a high ridge-like risk distribution is set over the entire width direction of the road including the traveling lane and the oncoming lane of the host vehicle 1. As a result, the driver can be alerted and the automatic brake can be actuated so that the vehicle 1 is surely stopped before entering the pedestrian crossing, ensuring safety without giving the pedestrian a sense of danger. be able to.

The position at which the host vehicle 1 is stopped by an alarm or automatic brake can be appropriately set by adjusting the center position of the risk distribution and the values of the correction parameters a _cross and a _x . For example, in the example of FIG. 2, the distance between the center position of the pedestrian crossing P and the center position of the host vehicle 1 is adopted as the distance xg in the expression (3) for calculating the crossing risk function Dcross. The center of the risk distribution is moved using the distance xg ′ between the end position on the own vehicle side of the sidewalk P and the own vehicle 1, and the position that reaches the threshold value Rk is moved to the own vehicle side as shown by the broken line in FIG. You may make it let it. The correction parameter a _cross may be a constant value, but may be variably set according to the position of the pedestrian crossing or the three-dimensional object, the crossing speed, etc., and the size of the entire risk distribution is changed by the correction parameter a _cross . or, by varying the width of the risk distribution by the correction parameter a _x, a position of the threshold value Rk can be moved toward the vehicle.

  As described above, in the present embodiment, when a three-dimensional object such as a pedestrian, a bicycle, or a welfare vehicle is assumed to be crossing or crossing the road, a crossing risk that forms a ridge in the road width direction is set in the crossing region. By doing so, it is possible to safely stop the host vehicle 1 safely without avoiding a three-dimensional object, and it is possible to realize driving support control that does not give danger to pedestrians and the like.

  In this embodiment, an example using the total risk function D obtained by synthesizing the risk function Dline for roads and the crossing risk function Dcross has been described. However, the risk function Dline for roads is omitted and only the crossing risk function Dcross is used. It is also possible to use.

  In this embodiment, an example in which the crossing risk is set over the entire road width including the traveling lane and the opposite lane of the host vehicle 1 is described. However, depending on the conditions, the crossing risk is not necessarily applied to the entire road width. It is not necessary to set, and it is also possible to set the crossing risk only for the traveling lane of the host vehicle 1.

  For example, by setting the risk centered on pedestrians and the crossing risk on the driving lane side of the host vehicle 1, even if the pedestrian is crossing or trying to cross the crosswalk from the opposite lane side, The host vehicle 1 can be stopped in front of the pedestrian crossing without passing through a portion where there are no pedestrians on the sidewalk.

  Further, the risk distribution on the road is set as the risk distribution on the traveling lane of the own vehicle 1, and the risk function Dline of the equation (2) is set to be a high risk on the opposite lane side, and then crossed to the traveling lane of the own vehicle 1 By setting the risk, when the pedestrian is crossing or trying to cross the pedestrian crossing from the own vehicle side, the avoidance operation to the opposite lane side can be suppressed and the own vehicle 1 can be stopped in front of the pedestrian crossing. .

  Next, a second embodiment of the present invention will be described. 6 and 7 relate to the second embodiment of the present invention, FIG. 6 is an explanatory diagram showing a movement vector of a three-dimensional object, and FIG. 7 is an explanatory diagram showing a temporary crosswalk.

  The second form deals with a case where the exact position of a pedestrian crossing or an intersection cannot be acquired, or a case where a pedestrian or bicycle crosses on a road without a pedestrian crossing.

  For this reason, in the second embodiment, information such as a pedestrian and a bicycle is first acquired by the stereo camera 3 and the stereo image recognition device 4, and an intersection is obtained by the travel environment information acquisition device 5 via a navigation device, GPS, or the like. And get information on pedestrian crossings. As a result, when the presence of a three-dimensional object such as a pedestrian or a bicycle is recognized in the traveling direction of the host vehicle 1, but the position of an intersection or a pedestrian crossing cannot be recognized, the movement vector of the three-dimensional object is determined from the position and speed of the three-dimensional object. Is used to estimate / determine whether the three-dimensional object crosses the road.

  For example, as shown in FIG. 6, when the three-dimensional object 51, 52 is detected on the own vehicle side and the three-dimensional object 53 is detected on the opposite lane side, the optical flow in the image captured by the stereo camera 3 is obtained. The movement vector of each solid object is calculated. When the movement vector of the three-dimensional object 50, 53 intersects with the traveling path of the host vehicle 1 and the movement vector of the three-dimensional object 51 does not intersect with the traveling path of the own vehicle 1, the three-dimensional object 51 does not cross the road. It is determined that the objects 50 and 53 cross the road, and a virtual crosswalk (temporary crosswalk) Pimg as shown in FIG. 7 is set.

  This temporary pedestrian crossing Pimg is set based on the position of the three-dimensional object closest to the host vehicle 1 when there are a plurality of three-dimensional objects determined to cross the road. For example, as illustrated in FIG. 7, when it is determined that the three-dimensional objects 50 and 53 cross the road and the three-dimensional object 53 is closer to the host vehicle 1 than the three-dimensional object 50, the X-axis coordinate position of the three-dimensional object 53 is centered. Is set in the Y-axis direction, and a crossing risk is set in the temporary crosswalk Pimg. The crossing risk function Dcross at this time uses the distance xg of the equation (2) described in the first embodiment as the distance xgimg between the host vehicle 1 and the center position of the temporary crosswalk Pimg (the position of the three-dimensional object 53). Calculated.

  In the second form, it is known that there are intersections and pedestrian crossings, but when accurate position information cannot be obtained, or pedestrians and bicycles try to cross the road forcibly even though there are no pedestrian crossings. Even in this case, by appropriately determining whether or not to cross and setting a temporary pedestrian crossing, the host vehicle 1 can be safely paused in front of a pedestrian or bicycle as in the first embodiment. This eliminates the avoidance action for pedestrians and bicycles while crossing, and does not give a sense of danger.

The schematic block diagram of the driving assistance device mounted in the vehicle according to the first embodiment of the present invention. Same as above, explanatory diagram showing own vehicle and pedestrian crossing area Same as above, explanatory diagram showing risk distribution when there is no solid object in the pedestrian crossing area Same as above, explanatory diagram showing risk distribution when there is a solid object in the pedestrian crossing area Same as above, explanatory diagram showing the distribution of crossing risk as seen from the direction perpendicular to the direction of travel Explanatory drawing which shows the movement vector of a solid object according to 2nd Embodiment of this invention. Same as above, explanatory diagram showing a temporary crosswalk

Explanation of symbols

1 Vehicle 2 Driving support device 3 Stereo camera
4 Stereo image recognition device (crossing solid object recognition unit)
5 Driving environment information acquisition device (crossing solid object recognition unit)
6 Control unit (crossing risk setting section)
Dcross Crossing risk function P Crosswalk Pimg Temporary crosswalk

Claims (5)

In a driving support device for a vehicle that sets a risk in the surrounding environment of the host vehicle and supports driving of the host vehicle based on the risk,
A crossing three-dimensional object recognition unit for recognizing a three-dimensional object presumed to cross or cross the road on which the host vehicle travels,
A vehicle driving support, comprising: a crossing risk setting unit that sets a risk based on the presence of the three-dimensional object as a crossing risk distributed across at least the traveling lane of the host vehicle in the width direction of the road. apparatus.   2. The vehicle according to claim 1, wherein, even if the three-dimensional object is in a stopped state, if the position of the three-dimensional object is close to a pedestrian crossing or an intersection, the crossing risk is set at the pedestrian crossing or the intersection. Driving assistance device.   When the three-dimensional object crossing the road in a state where the position of the pedestrian crossing or the intersection cannot be recognized, a temporary pedestrian crossing based on the position of the three-dimensional object is set, and the temporary pedestrian crossing is The vehicle driving support apparatus according to claim 1, wherein a crossing risk is set.   4. The driving support apparatus for a vehicle according to claim 3, wherein when there are a plurality of the three-dimensional objects, the temporary crosswalk is set based on the position of the three-dimensional object closest to the own vehicle.   The vehicle driving support device according to any one of claims 1 to 4, wherein the crossing risk is a risk distribution that intervenes driving support for temporarily stopping the host vehicle in front of the three-dimensional object. .

Images