JP2015205546A - Road gradient detection device, drive support device and computer program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a road gradient detection device, a drive support device and a computer program for detecting a gradient of a front side section of a road on which an automobile runs without requiring map data on which a gradient of each spot is registered.SOLUTION: An edge of an image of a vertical object is extracted as a vertical edge Ve from an image photographed by a camera and an edge of an image of a road boundary passing through the neighborhood of a lower edge of the vertical edge Ve is extracted as a road edge Re. An angle ANG of the vertical edge Ve to the road edge Re is determined, when the angle ANG is larger than a prescribed angle range defined in accordance with a position of an intersection P between the vertical edge Ve and the road edge Re, it is judged that the front side road section is uphill and, when the angle ANG is smaller than the prescribed angle range, it is judged that the front side road section is down-hill.

Description

  The present invention relates to a technique for detecting the gradient of a front section of a road on which an automobile is traveling.

  As a technique for calculating the gradient of the road on which the automobile is traveling, a technique for calculating the gradient of the road on which the vehicle is traveling using map data in which the gradient of each point on the road is registered in the navigation device that calculates the current position ( For example, Patent Document 1) and a technique (for example, Patent Document 2) that uses an inclination angle detected by an inclination sensor that detects the inclination of an automobile as the gradient of a running road are known.

  In addition, a technique for changing the optical axis of an automobile headlamp in the vertical direction according to the situation in front is also known (for example, Patent Document 3).

JP 2014-52851 A JP 2005-23916 A JP 2001-347882 A

  According to the technique for calculating the gradient of a road that is running using the map data in which the gradient of each point of the road is registered, the gradient cannot be detected for a road whose gradient is not registered in the map data. . Further, in order to detect the gradient of the road that is running, map data in which the gradient of each point on the road is registered and means for calculating the current position are required.

  On the other hand, according to the technology for calculating the gradient of the road that is running using the inclination sensor that detects the inclination of the automobile, only the gradient of the point where the current automobile is located on the road that is running can be calculated. The slope of the front section of the road on which the car is traveling cannot be calculated.

  Accordingly, an object of the present invention is to detect the gradient of the front section of the road on which the vehicle is traveling without requiring map data in which the gradient of each point on the road is registered.

  In order to achieve the above object, the present invention provides a road gradient detection device mounted on an automobile, wherein a camera for photographing the front of the automobile and a vertical object included in an image photographed by a front camera are installed. A vertical edge extracting unit that extracts an edge as a vertical edge, a road edge extracting unit that extracts an edge of an image in a road extending direction of a road boundary passing near the lower end of the vertical edge included in the image, and the vertical An angle calculating unit that calculates an angle of an angle with respect to the road edge as a detection angle, and the detection angle is formed with respect to a road boundary from which the object from which the vertical edge has been extracted is extracted from the road edge, The area where the object in front of the traveling road is installed when the actual angle, which is the angle on the front side of the traveling direction of the car, represents an obtuse angle Is determined to be an uphill, and when the detected angle indicates that the actual angle is an acute angle, the section where the object in front of the running road is installed is a downhill And a gradient determination unit for determining.

  Here, such a road gradient detection device has a vertical line passing through a point projected when the front of the automobile is flat at the position for each position on the image. Is an image of the vertical line on the image corresponding to an angle on the near side with respect to the traveling direction of the automobile, which is formed with respect to the longitudinal line of the automobile passing through the point. A storage unit that stores in advance angle information that defines an angle formed with respect to the image as a reference value of the position, and in the gradient determination unit, an angle formed by the vertical edge with respect to the road edge is the vertical The object in front of the traveling road is installed when the angle information at the intersection of the edge and the road edge is larger than a reference value defined by the angle information by a predetermined level or more. That segment is determined as is uphill, section front the object of the traveling road if more than a predetermined level smaller is installed may be determined to be downhill.

  In this case, in the gradient determination unit, a difference between an angle formed by the vertical edge with respect to the road edge and a reference value defined by the angle information of the position of the intersection of the vertical edge and the road edge. May be calculated as the magnitude of the gradient of the section where the object in front of the traveling road is installed.

  According to the road gradient detection apparatus as described above, an object installed vertically on an uphill is inclined at an obtuse angle with respect to the road surface in front, and an object installed vertically on a downhill is at an acute angle with respect to the road surface in front. Inclination is used to determine the uphill and downhill of the front section of the road, so there is no need for map data in which the slope of each point on the road is registered. Using only these images, it is possible to detect the slopes of the uphill and downhill in the front section of the road on which the vehicle is traveling.

  Here, the present invention also combines the road gradient detection device as described above and the vehicle when the vehicle is approaching a starting point of a section determined by the road gradient detection device to be an uphill. There is also provided a driving support device including an optical axis control unit that changes the vertical angle of the optical axis of the headlamp to a higher angle.

  In addition, the present invention also provides a driving force of the vehicle when the vehicle reaches the starting point of the section that the road gradient detection device as described above and the road gradient detection device determine to be an uphill. There is also provided a driving support device including a driving force control unit that increases the driving force.

  As described above, according to the present invention, it is possible to detect the gradient of the front section of the road on which the vehicle is traveling without requiring map data in which the gradient of each point on the road is registered.

The structure of the driving assistance device which concerns on embodiment of this invention is shown. It is a figure which shows the content of the vertical angle table which concerns on embodiment of this invention. It is a flowchart which shows the road inclination angle calculation process which concerns on embodiment of this invention. It is a figure which shows the process example of the road inclination angle calculation process which concerns on embodiment of this invention. It is a figure which shows the process example of the road inclination angle calculation process which concerns on embodiment of this invention. It is a flowchart which shows the slope driving assistance process which concerns on embodiment of this invention.

Hereinafter, embodiments of the present invention will be described.
FIG. 1 a shows a configuration of the driving support apparatus 1 according to the present embodiment.
The driving support device 1 is a device mounted on a vehicle, and includes an ECU 2 that controls the drive system and other parts of the vehicle, and the light of a vehicle headlamp that is provided so that the vertical angle of the optical axis can be adjusted. It is connected to the optical axis direction changing actuator 3 for changing the axial direction.

  The driving support apparatus 1 includes a camera 11 that captures the front of the automobile, an image memory 12 that stores an image captured by the camera 11, a control unit 13, and a storage device 14, as shown in FIG. 1b.

  However, such a driving support device 1 may be configured using a computer including a CPU, a memory, and the like. In this case, the control unit 13 is realized by the computer executing a predetermined computer program. Is.

Here, a vertical angle table is stored in the storage device 14 in advance.
Hereinafter, the contents of the vertical angle table will be described.
As shown in FIG. 2a, the vertical angle table is perpendicular to each of image space coordinates (Ix, Iy) which are coordinates on a coordinate system Ix-Iy that defines the coordinates of each pixel on the image captured by the camera 11. This is a registration of the front-to-back projection angle θ.

Here, the vertical versus longitudinal projection angle θ registered with respect to the image space coordinates (Ix, Iy) is an angle obtained as follows.
That is, as shown in FIG. 2b, with the XY plane as the flat ground, the Z direction as the height direction, and a vertical object extending in the Z direction at each XY coordinate of the XY plane, the Y direction on the XY plane as the forward direction Assume the image of FIG. 2c obtained when shooting is performed with the camera 11 of the car that has been placed.

Then, in this case, the angle θ between the extending direction of the image of the vertical object in the image of FIG. 2c and the image direction in the image of the Y direction line on the XY plane passing through the lower end of the vertical object. Is the vertical versus forward projection angle θ registered with respect to the image space coordinates (Ixi, Iyj) at the lower end of the vertical object. However, the angle θ to be registered as the vertical versus front-and-rear projection angle θ in the image space coordinates (Ixi, Iyj) on the right side from the center in the horizontal direction of the image is measured clockwise, and the image space coordinates ( The angle θ that is registered as the vertical vs. front-rear projection angle θ in Ixi, Iyj) is measured counterclockwise.
Actually, the angle θ registered in each image space coordinate (Ixi, Iyj) is the projection characteristic of the camera 11 used on the image in the real space with respect to the image space coordinate (Ixi, Iyj) of each pixel of the image. You may make it obtain | require by calculation according to.

  Also, instead of the vertical angle table, each image space coordinate (Ixi, Iyj) and the direction in the image of the vertical object passing through each image space coordinate (Ixi, Iyj) are in the direction of the line passing through the lower end of the vertical object. Alternatively, a calculation formula representing the relationship with the angle θ (vertical vs. front-rear projection angle θ) formed may be obtained and stored in the storage device 14.

Now, referring back to FIG. 1, the control unit 13 uses such a vertical angle table to calculate the gradient of the front section of the traveling road based on the image captured by the camera 11.
Hereinafter, the road gradient calculation process performed by the control unit 13 to calculate the gradient of the front section of the traveling road will be described.
FIG. 3 shows the procedure of the road gradient calculation process.
As shown in the figure, the control unit 13 determines whether or not the vehicle is traveling straight by acquiring information on the steering angle of the vehicle from the ECU 2 (step 302).
If the vehicle is traveling straight (step 302), the latest image captured by the camera 11 is acquired from the image memory 12 (step 304), and the electric poles, guardrail columns, etc. The edge of the vertical object image is extracted as the vertical edge Ve (step 306).

Then, the following processing is performed for each of the extracted vertical edges Ve (steps 308, 328, and 330).
That is, first, an edge of an image of a road boundary such as a road boundary line or a road boundary step that is on the same side as the vertical edge Ve and passes near the lower end of the vertical edge Ve is extracted as a road edge Re (step 310). ). Here, if the lower edge of the vertical edge Ve is on the right side of the center of the image in the left-right direction, the image of the road boundary that is on the same side as the vertical edge Ve is If the lower end of the vertical edge Ve is on the left side of the center in the left-right direction of the image, it indicates the image of the road boundary on the left side of the center of the image in the left-right direction. Here, the road edge Re to be extracted may be one that passes near the lower end of the vertical edge Ve when the edge is extended. When the extracted vertical edge Ve and the extracted road edge Re do not intersect, the edges are extended so that they intersect.

  Then, an angle ANG formed by the direction of the vertical edge Ve with respect to the direction of the road edge is calculated (step 312). Here, the angle ANG is measured clockwise when the intersection P between the vertical edge Ve and the road edge Re is on the right side of the center in the left-right direction of the image, and is measured counterclockwise when it is on the left side.

  Then, the vertical front / rear projection angle θ registered with respect to the image space coordinates (Ixi, Iyj) of the intersection P of the vertical edge Ve and the road edge Re is acquired from the vertical angle table (step 314). Also, the real space Y coordinate Yd corresponding to the image space coordinates (Ixi, Iyj) of the intersection P of the vertical edge Ve and the road edge Re, that is, the vertical edge obtained on the assumption that the front of the car is horizontal to the car The distance in the front-rear direction to a point in the real space that is reflected at the position of the lower end of Ve (before the extension when vertical edge Ve is extended) is set as determination distance Yd (step 316).

  If the angle ANG> vertical-vertical projection angle θ + Mgn is set with Mgn as a fixed margin in advance (step 318), the point on the traveling road that has advanced the determination distance Yd is an uphill. In addition, the degree of ascending slope of the point on the traveling road that has advanced the determination distance forward is determined so that the larger | angle ANG−vertical-vertical projection angle θ | As the gradient information, the fact that the vehicle is uphill and the degree of the uphill are stored in the storage device 14 (step 324).

  If angle ANG <vertical vs. front / rear projection angle θ−Mgn (step 320), it is determined that the point on the traveling road that has advanced the determination distance Yd forward is a downhill, and | angle ANG−vertical pair The degree of the downward slope of the point on the traveling road that has advanced the determination distance forward is determined so as to increase as the front / rear projection angle θ | increases. Is stored in the storage device 14 (step 326).

  If the vertical versus front projection angle θ−Mgn ≦ angle ANG ≦ vertical versus front projection angle θ + Mgn (steps 318 and 320), the point on the traveling road that has advanced the determination distance Yd is a flat road. Is stored in the storage device 14 as point gradient information of the determination distance Yd (step 322).

  Then, when the processing of steps 308-326 is completed for all the extracted vertical edges Ve as described above (step 328), the processing returns to step 302.

Heretofore, the road gradient calculation process performed by the control unit 13 to calculate the gradient of the front section of the running road has been described.
Hereinafter, a processing example of such road gradient calculation processing will be described.
FIG. 4a1 shows an image taken by the camera 11 of the automobile when the automobile is running straight and the front of the running road is flat.
In the road gradient calculation process, for example, as shown in FIG. 4a2, the edge of the right utility pole image is extracted as the vertical edge Ve from the image shown in FIG. 4a2, and the step of the road boundary passing near the lower end of the vertical edge Ve is detected. The edge of the image is extracted as the road edge Re.

  Since the front side of the running road is flat, in the real world, the utility pole from which the vertical edge Ve is extracted stands upright with respect to the road surface. Further, since the automobile is traveling straight, the road edge Re coincides with the direction in the image corresponding to the longitudinal direction of the automobile.

  Therefore, the angle ANG of the vertical edge Ve with respect to the road edge Re is the vertical versus forward projection angle θn registered in the vertical table with respect to the image space coordinates (Ixi, Iyj) of the intersection P of the vertical edge Ve and the road edge Re. Almost matches. Therefore, vertical vs. front / rear projection angle θ−Mgn ≦ angle ANG ≦ vertical vs. front / rear projection angle θ + Mgn, which is the same as the front / rear direction distance to the point where the utility pole from which vertical edge Ve is extracted by road gradient calculation processing It is determined that the traveling road is flat at a distance and a point away from the automobile.

Next, FIG. 5a1 shows an image photographed by the camera 11 of the automobile when the automobile is traveling straight and the front of the running road is an uphill.
In the road gradient calculation processing, for example, as shown in FIG. 5a2, the edge of the left utility pole image is extracted as the vertical edge Ve from the image shown in FIG. 5a1, and the road boundary line image passing through the vicinity of the lower end of the vertical edge Ve. Are extracted as road edges Re.

  Since the road ahead is an uphill, as shown in FIG. 5a3, in the real world, the utility pole from which the vertical edge Ve is extracted is inclined at an obtuse angle with respect to the road surface as viewed from the uphill. In addition, since the car is going straight, the direction of the road edge Re coincides with the direction in the image corresponding to the front and rear direction of the car while traveling uphill, and when the uphill is photographed from before the uphill The direction is closer to the vertical than the direction in the image corresponding to the longitudinal direction of the automobile (the direction away from the direction of the vertical edge Ve).

  Therefore, the angle ANG of the vertical edge Ve with respect to the road edge Re is based on the vertical versus forward projection angle θn registered in the vertical table with respect to the image space coordinates (Ixi, Iyj) of the intersection P between the vertical edge Ve and the road edge Re. The angle ANG is greater than the vertical projection angle θ + Mgn, and the road gradient calculation process is the same distance as the front-rear direction distance to the point where the utility pole from which the vertical edge Ve is extracted is installed. It is determined that the traveling road is uphill at a certain point.

Next, FIG. 5b1 represents an image taken by the camera 11 of the automobile when the automobile is traveling straight and the road ahead is a downhill.
In the road gradient calculation process, for example, the edge of the right power pole image is extracted as the vertical edge Ve as shown in FIG. 5b2 from the image shown in FIG. 5b1, and the step of the road boundary passing near the lower end of the vertical edge Ve is extracted. The edge of the image is extracted as the road edge Re.

  Since the road ahead is a downhill, as shown in FIG. 5b3, in the real world, the utility pole from which the vertical edge Ve is extracted is inclined at an acute angle with respect to the road surface as viewed from the downhill. In addition, since the car is going straight, the direction of the road edge Re coincides with the direction in the image corresponding to the front and rear direction of the car while traveling downhill, and when the downhill is photographed from before the downhill The direction is closer to the horizontal than the direction in the image corresponding to the longitudinal direction of the automobile (the direction approaching the direction of the vertical edge Ve).

  Therefore, the angle ANG of the vertical edge Ve with respect to the road edge Re is based on the vertical versus forward projection angle θn registered in the vertical table with respect to the image space coordinates (Ixi, Iyj) of the intersection P between the vertical edge Ve and the road edge Re. The angle ANG is smaller than the vertical projection angle θ-Mgn, and the road gradient calculation process is the same distance as the front-rear direction distance to the point where the utility pole from which the vertical edge Ve is extracted is installed. It is determined that the traveling road is downhill at the point.

The example of the road gradient calculation process has been described above.
As described above, according to the road gradient calculation process, the vehicle is running using only the image in front of the vehicle taken by the camera 11 without requiring map data in which the gradient of each point on the road is registered. The slope of the front section of the road can be detected.

Next, the hill driving support process performed by the control unit 13 using the point gradient information calculated by the road gradient calculation process as described above and stored in the storage device 14 will be described.
FIG. 6 shows the procedure of the slope driving support process.
As shown in the figure, in the hill driving support process, during the period when the ECU 2 of the automobile is executing the auto cruise, that is, the automatic constant speed control (step 602), the vehicle is going to the starting point of the uphill slope of the automobile. The arrival (step 604) and the arrival of the car at the uphill end point (step 606) are monitored.

  When the vehicle reaches the starting point of the uphill (step 604), the ECU 2 increases the driving force of the vehicle by a predetermined amount (step 608). The driving force is increased by increasing the accelerator opening, increasing the accelerator opening, and increasing the transmission gear ratio.

  On the other hand, if the vehicle reaches the uphill end point (step 606), the ECU 2 increases the braking force of the vehicle by a predetermined amount (step 610). The driving force is reduced by reducing the accelerator opening, increasing the transmission gear ratio, or applying braking by a brake.

  Here, as described above, when the vehicle reaches the starting point of the uphill, the vehicle driving force is increased by a predetermined amount, so that the vehicle speed is determined by the automatic cruise of the ECU 2 after the vehicle speeds down and the vehicle speed decreases. Rather than increasing the vehicle speed, it is possible to suppress the vehicle speed from lowering by going uphill. This also makes it possible to suppress the occurrence of traffic congestion due to so-called sag.

  Further, when the vehicle reaches the uphill end point, the braking force of the vehicle is increased by a predetermined amount, so that the vehicle speed is increased at a constant speed by the automatic cruise of the ECU 2 after the vehicle speed increases from the uphill to the flat road or downhill. It is possible to prevent the vehicle speed itself from increasing from an uphill to a flat road or downhill.

  Further, in the hill driving support process, during the period when the headlamp is lit (step 612), the vehicle approaches (step 614) within a predetermined distance (for example, 7 m) to the starting point of the uphill ahead of the vehicle. ), Approaching within a predetermined distance to the downhill starting point ahead of the vehicle (step 616), and passing the uphill starting point or downhill starting point (step 618).

  When the uphill starting point is approached within a predetermined distance (step 614), the optical axis direction change actuator 3 is controlled so that the vertical angle of the optical axis of the headlamp is higher than the standard angle. The angle is changed (step 620). The angle to be changed may be a predetermined angle or may be an angle according to the degree of ascending slope of the uphill ahead.

  If the start point of the downhill is approached within a predetermined distance (step 616), the optical axis direction changing actuator 3 is controlled so that the vertical angle of the optical axis of the headlamp is lower than the standard angle. The angle is changed (step 622). The angle to be changed may be a predetermined angle, or may be an angle according to the degree of the downward slope of the forward downhill.

If the uphill start point or the downhill start point is passed (step 626), the vertical angle of the optical axis of the headlamp is returned to the standard angle (step 624).
By adjusting the vertical angle of the optical axis of the headlamp in this way, it is possible to illuminate the uphill and the downhill starting ahead with the headlamp to the front. Moreover, the glare given to the driver | operator of the downhill oncoming vehicle which starts ahead can be suppressed.

  In the above-described hill driving support procedure, the control unit 13 determines the uphill start point, the uphill end point, the downhill start point, the uphill uphill grade, and the downhill uphill grade. It is stored in the storage device 14 and calculated from the point gradient information for each determination distance.

  The embodiment of the present invention has been described above.

  DESCRIPTION OF SYMBOLS 1 ... Driving assistance device, 2 ... ECU, 3 ... Optical axis direction change actuator, 11 ... Camera, 12 ... Image memory, 13 ... Control part, 14 ... Memory | storage device.

Claims (8)

A road gradient detection device mounted on an automobile,
A camera for photographing the front of the car;
A vertical edge extraction unit that extracts a vertical edge of a vertically installed object included in an image captured by the previous camera as a vertical edge;
A road edge extraction unit that extracts an edge of an image in a road extending direction of a road boundary passing near the lower end of the vertical edge included in the image as a road edge;
An angle calculation unit that calculates an angle formed by the vertical edge with respect to the road edge as a detection angle;
The detected angle represents that the actual angle, which is an angle on the front side of the traveling direction of the automobile, formed with respect to the road boundary from which the object from which the vertical edge is extracted is extracted from the road edge is an obtuse angle. The section where the object in front of the traveling road is installed is an uphill, and the detected angle indicates that the actual angle is an acute angle, the traveling road A road gradient detection apparatus comprising: a gradient determination unit that determines that a section in which the object ahead of the vehicle is installed is a downhill. The road gradient detection device according to claim 1,
For each position on the image, a vertical line passing through a point projected when the front of the automobile is flat at the position is formed with respect to a longitudinal line of the automobile passing through the point, Angle information that defines, as a reference value for the position, an angle formed by the image of the vertical line on the image with respect to the image of the line in the front-rear direction, corresponding to the angle on the near side with respect to the vehicle traveling direction Has a storage unit that stores in advance,
The slope determination unit travels when the angle formed by the vertical edge with respect to the road edge is larger than a reference value defined by the angle information of the position of the intersection of the vertical edge and the road edge by a predetermined level or more. It is determined that the section where the object ahead of the running road is installed is an uphill, and the section where the object ahead of the running road is installed when it is smaller than a predetermined level A road gradient detection device, characterized in that it is determined to be a downhill. The road gradient detection device according to claim 2,
The gradient determination unit is calculating a difference between an angle formed by the vertical edge with respect to the road edge and a reference value defined by the angle information of a position of an intersection of the vertical edge and the road edge. A road gradient detection device, wherein the road gradient detection device calculates the gradient magnitude of a section where the object in front of the road is installed. A driving support device comprising the road gradient detection device according to claim 1, 2 or 3,
When the vehicle is approaching the start point of the section determined by the road gradient detection device as being uphill, the vertical angle of the optical axis of the headlight of the vehicle is changed to a higher angle. A driving support apparatus having an optical axis control unit. A driving support device comprising the road gradient detection device according to claim 1, 2 or 3,
A driving support device, comprising: a driving force control unit that increases the driving force of the automobile when the automobile reaches a starting point of a section determined to be an uphill by the road gradient detection device. A computer program that is read and executed by a computer installed in an automobile,
In the computer,
A vertical edge extracting step of extracting, as a vertical edge, a vertical edge of a vertically installed object included in an image photographed by a camera photographing the front of the automobile;
A road edge extraction step for extracting, as a road edge, an edge of an image in a road extending direction of a road boundary passing near the lower end of the vertical edge included in the image;
An angle calculating step of calculating an angle formed by the vertical edge with respect to the road edge as a detection angle;
The detected angle represents that the actual angle, which is an angle on the front side of the traveling direction of the automobile, formed with respect to the road boundary from which the object from which the vertical edge is extracted is extracted from the road edge is an obtuse angle. The section where the object in front of the traveling road is installed is an uphill, and the detected angle indicates that the actual angle is an acute angle, A computer program for executing a gradient determination step for determining that a section in which the object in front of a road is installed is a downhill. A computer program according to claim 6,
For each position on the image, the computer has a vertical line passing through a point projected when the front of the automobile is flat at the position to a longitudinal line of the automobile passing through the point. The angle formed by the image of the vertical line on the image and the image of the line in the front-rear direction corresponding to the angle on the near side with respect to the traveling direction of the automobile is a reference value of the position. The angle information defined as
In the gradient determining step, when the angle formed by the vertical edge with respect to the road edge is greater than a reference value defined by the angle information of the position of the intersection of the vertical edge and the road edge, the vehicle travels. It is determined that the section where the object ahead of the running road is installed is an uphill, and the section where the object ahead of the running road is installed when it is smaller than a predetermined level A computer program characterized in that it is determined to be a downhill. A computer program according to claim 7,
In the gradient determination step, the difference between the angle formed by the vertical edge with respect to the road edge and the reference value defined by the angle information of the position of the intersection of the vertical edge and the road edge is calculated. A computer program for calculating the magnitude of a gradient of a section where the object in front of a road is installed.