JP3690099B2 - Light distribution control device for automotive headlights - Google Patents

Abstract

PROBLEM TO BE SOLVED: To irradiate the running lane of the frontside of a vehicular travel direction precisely, regardless of the road state and the position and direction for a vehicular running lane, on the light distribution control device of the head light for automobile. SOLUTION: The white line on a road is recognized from the road image information of a vehicular frontside by a white line recognition means 3 and the running lane in which own vehicle is running is estimated based on the recognized white line by a running lane estimation means 4. The road curvature of the running lane estimated by the running lane estimation means 4 is calculated by a road curvature calculation means 4A and the side shift amount the vehicular running position from the standard position of the running lane is calculated by a side shift amount calculation means 4B and the yaw angle of the own vehicle center line for the running lane is calculated by a yaw angle calculation means 4C. A optical axis angle target value is calculated base on the calculated road curvature by a optical axis angle target value calculation means 5 and the calculated optical axis angle target value is corrected based on the side shift amount and yaw angle by a optical axis angle target value correction means 5A. The optical axis angle of a head light 9 is adjusted so as to be equal to this corrected optical axis angle target value through a optical axis actuator 8 by a control means 7.

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a light distribution control device for an automobile headlight that changes the optical axis of the headlight according to the curve state of a road.
[0002]
[Prior art]
In general, a headlight for an automobile is configured to be capable of switching between a high beam and a low beam. By switching between the high beam and the low beam, the driver can select the optical axis angle of the headlight according to road conditions.
However, this high beam / low beam switching only switches the optical axis angle in the vertical direction, and does not change the optical axis angle in the horizontal direction, and the optical axis of the headlight is always in the direction of the vehicle center line. It is fixed to. For this reason, when the front road direction aimed at by the vehicle and the vehicle center line deviate from each other like a curved road, the road surface ahead of the vehicle cannot be effectively illuminated. In addition, the optical axis angle in the vertical direction can only be adjusted in two steps, high beam / low beam, and finer adjustment of the optical axis angle is desired depending on the situation. On the other hand, as the optical axis angle can be finely adjusted, manual operation is troublesome and difficult to use effectively.
[0003]
Therefore, in order to respond to such demands and problems, various proposals related to light distribution control of automobile headlights have been made. In particular, as a technique for performing lateral light distribution control, Japanese Patent Laid-Open No. 6-206491 discloses a technique for controlling the lateral optical axis angle of a headlight corresponding to a curve. In this technology, the angular velocity sensor and the vehicle speed sensor are used to detect the rotational angular velocity and the vehicle speed when the vehicle turns, and the horizontal light axis of the headlight is determined according to the curve radius calculated based on these detected values. It changes the angle.
[0004]
The technique disclosed in Japanese Patent Application Laid-Open No. 7-232589 similarly controls the lateral optical axis angle of the headlight corresponding to the curve. In this technique, an acceleration sensor and a vehicle speed sensor are used. Are used to detect the lateral acceleration and vehicle speed of the vehicle, and the lateral optical axis angle is changed in accordance with the curve radius calculated based on these detected values.
[0005]
An apparatus that performs vertical light distribution control is disclosed in JP-A-1-278848. In this apparatus, the vertical angle of the optical axis is controlled according to the distance to the preceding vehicle.
[0006]
[Problems to be solved by the invention]
By the way, as a light distribution control device that controls the optical axis angle of the headlight in the lateral direction corresponding to the curve, the curve radius (or curvature) is calculated from the angular velocity and lateral acceleration of the vehicle when traveling along the curve as described above. ) Can be used to obtain information about the curve of the vehicle that is running at that time, but information about the curve of the road ahead of the host vehicle (ie, the portion of the road to which the headlight is to be irradiated) is to be obtained. I can't. Therefore, it is conceivable to capture a road image ahead of the vehicle through image information input means such as a camera, estimate the traveling lane based on this image information, and calculate the curve radius from the estimated traveling lane information.
[0007]
In this case, when the vehicle is traveling along the traveling lane, the traveling lane can be illuminated by setting the lateral optical axis angle of the headlight according to the calculated curve radius. The vehicle does not always travel along the travel lane, and the travel lane cannot be accurately illuminated only by setting the optical axis angle calculated according to the curve radius.
[0008]
Therefore, it is desired to control the optical axis angle of the headlight so that the traveling lane can always be accurately irradiated regardless of the positional relationship of the vehicle with respect to the traveling lane.
Also, the estimation of the driving lane based on the image information is generally based on the road white line on the road image, but this is because the normal road surface has low luminance and the luminance change is small. This is because the white line has a very high luminance compared to a normal road surface, and thus it can be easily recognized by paying attention to the difference in lightness of the road.
[0009]
However, depending on the road conditions, for example, when the road surface is wet or when there are a large number of water pools, the light is easily reflected, so the luminance of the road surface increases and the luminance difference from the white line decreases. . In such a case, it becomes difficult to distinguish the white line from the normal road surface, and the recognition accuracy of the white line is lowered, and the curve radius cannot be calculated accurately. And even if the optical axis angle of the headlight is controlled based on such an inaccurate curve radius, it is not possible to accurately irradiate the driving lane, and the control becomes unstable and the driver feels visually uncomfortable. There is a risk of giving.
[0010]
Here, when white line recognition fails, it may be possible to stop the light distribution control of the headlight based on the image information and return the optical axis to the normal position. It is unavoidable to give a sense of incongruity, and particularly when the white line recognition is intermittently malfunctioning, the sense of discomfort given to the driver is large.
The present invention was devised in view of the above-described problems, and is an automobile headlight that can accurately irradiate the traveling lane ahead of the traveling direction of the vehicle regardless of the road conditions and the position and direction of the vehicle with respect to the traveling lane. An object of the present invention is to provide a light distribution control device.
[0011]
[Means for Solving the Problems]
  For this reason, in the light distribution control device for an automobile headlight according to the first aspect of the present invention,The road information ahead of the vehicle is imaged by the imaging means,From road image information ahead of the vehicleBy white line recognition meansRecognize white lines on the road,hereA travel lane in which the vehicle travels by travel lane estimation means based on the recognized white lineRoad center line of the driving lane as a relative position of the vehicleIs estimated.
[0012]
  Then, the road curvature calculation means calculates the travel lane estimated by the travel lane estimation means.Road centerlineThe curvature is calculated, and the lateral deviation calculation means calculates the estimated travel lane.Road center lineVehicle fromCenterlineThe travel lane estimated by the yaw angle calculation meansRoad center lineThe yaw angle of the vehicle center line with respect to is calculated.
  In the optical axis angle target value calculation means, the road calculated by the road curvature calculation meansCenterlineThe optical axis angle target value is calculated based on the curvature, and the optical axis angle target value correcting means,lightAxis angle target valueInLateral deviation amount calculation meansAdd the ratio of the amount of lateral deviation calculated in step 1 and the headlight irradiation distance,Yaw angle calculation meansSubtract the yaw angle calculated into correct.
[0013]
  Then, the control means adjusts the optical axis angle of the headlight to be equal to the corrected optical axis angle target value via the optical axis actuator.
  Thus, regardless of the position and orientation of the vehicle relative to the travel lane, the travel laneRoad center lineIs adjusted along the optical axis of the headlight.
  Moreover, the automotive headlight of the present invention according to claim 2 is provided.Light distribution controlIn the apparatus, when the white line recognition by the white line recognition unit is not satisfactory, the accuracy of the road curvature calculated by the road curvature calculation unit based on the white line becomes unstable. In this case, the lateral acceleration detection unit adds to the vehicle. The lateral acceleration is detected, and the traveling speed of the vehicle is detected by the traveling speed detecting means.
[0014]
  Then, the optical axis angle target value calculation means is estimated based on the lateral acceleration and the running speed respectively detected by the lateral acceleration detecting means and the running speed detecting means instead of the road curvature calculated by the road curvature calculating means. An optical axis angle target value is calculated based on the estimated road curvature.
  As a result, the continuity of control is maintained even when white line recognition is unsatisfactory, and the optical axis of the headlight can be controlled without causing the driver to feel uncomfortable.
  According to a third aspect of the present invention, there is provided a vehicle headlight light distribution control device according to the present invention, wherein when the white line recognition by the white line recognition means is unsatisfactory, the white line recognition means outputs a recognition lost signal to the optical axis angle target value calculation means. Then, the optical axis angle target value calculation means stops the correction of the optical axis angle target value.
  Thereby, the influence of the road condition is removed in setting the optical axis angle target value, and stable optical axis control is performed.
  In the vehicle headlight light distribution control device according to the fourth aspect of the present invention, the time of the optical axis angle target value is obtained by a low-pass filter interposed between the control means and the optical axis angle target value calculation means. Changes are smoothed.
  This prevents the driver from feeling uncomfortable due to unnatural fluctuations in the optical axis.
  In the vehicle headlight light distribution control device according to the fifth aspect of the present invention, when the recognition lost signal from the white line recognition means is input by the optical axis angle target value calculation means, the optical axis angle target value is calculated. Gradually approach 0.
  This prevents the driver from feeling uncomfortable due to unnatural fluctuations in the optical axis.
[0015]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 to FIG. 7 show a light distribution control device for a headlight for an automobile as an embodiment of the present invention.
As shown in FIG. 1, the present light distribution control device is a camera (image information input means) that images the road condition ahead of the vehicle in order to recognize the position of the host vehicle with respect to the travel lane (travel lane). ) 2, image information processing means (white line recognition means) 3 for recognizing right and left white line positions on the road ahead by appropriately processing image information from the camera 2, and white line position image information by the image information processing means 3 Road curvature calculation means 4A for calculating the road curvature ρ of the traveling lane of the vehicle is provided.
[0016]
In addition to the road curvature calculation means 4A that is calculated based on the white line position image information, the light distribution control device as a means for calculating the road curvature is based on the road curvature (based on the lateral acceleration G and the vehicle speed V of the vehicle). A substitute road curvature calculating means 25 for calculating an estimated road curvature ρ ′) is provided.
The road curvature calculation unit 4A includes a lateral deviation amount calculation unit 4B that calculates a lateral deviation amount ΔY from the reference position of the traveling lane of the host vehicle and a yaw angle calculation unit 4C that calculates a yaw angle β with respect to the traveling lane of the host vehicle. It is provided as a functional element in travel lane estimation means (travel lane estimation means) 4 for estimating the relative position of the travel lane (travel lane) with respect to the host vehicle.
[0017]
Further, the present light distribution control device determines the longitudinal optical axis angle target value θ based on the road curvature ρ calculated by the road curvature calculation means 4A or the estimated road curvature ρ ′ calculated by the substitute road curvature calculation means 25.₁And lateral optical axis angle target value θ₂The optical axis angle target value calculating means 5 for calculating the optical axis, the optical axis actuator 8 for moving the optical axis of the headlight 9, and the optical axis of the optical axis angle of the headlight 9 calculated by the optical axis angle target value calculating means 5. Angle target value θ₁, Θ₂And a control means (controller) 7 for controlling the optical axis actuator 8 so as to be equal to. In this embodiment, the longitudinal optical axis angle target value θ₁Is set by manual adjustment of the driver (for example, two-stage adjustment of high beam / low beam).
[0018]
In addition, image information processing means 3, travel lane estimation means 4 (road curvature calculation means 4A, lateral deviation amount calculation means 4B, yaw angle calculation means 4C), substitute road curvature calculation means 25, optical axis angle target value calculation means 5, controller 7 is configured as an electronic control unit including a CPU, an input / output interface, a ROM, a RAM, and the like.
First, calculation of the road curvature ρ of the travel lane will be described.
[0019]
In the image information processing means 3, first, as shown in FIG. 2, an original image 3A from the camera 2 is taken, a road white line is extracted from the original image 3A, and an image of the extracted road white line is viewed from vertically above. Is converted to a planar view image 3B. Here, the recognition of the white lines 12L and 12R will be described with reference to FIG. Here, the recognition of the white line 12L as the roadside line at the left end of the driving lane will be described. However, the same applies to the case where the white line 12R at the right end of the driving lane is used as a reference. I will call it.
[0020]
In the image information recognizing means 3, as shown in FIG. 3 (a), the camera 2 provided in the vehicle captures black and white image information in a range ahead of the vehicle on a flat ground, and a vertical image on the screen is obtained from this image information. Some are omitted. Then, a plurality of horizontal lines 11 are set at equal intervals on this screen.
The capture of the black and white image information is updated every minute control cycle. As shown in FIG. 3B, the left and right required white line positions on the previous screen on each horizontal line 11 are obtained. Is set as a white line search area (processing target area) 10 (here, left and right 50 pixels (dot)). The first screen uses the white line position on the straight road as the previous screen data.
[0021]
Then, as shown in FIG. 3C, the brightness of each horizontal line is differentiated from the left to the lateral direction. Moreover, the code | symbol 14 in a figure is a guardrail.
By the way, a normal road surface has a low luminance and a luminance change is also small. On the other hand, since the brightness of the white line 12 is much higher than that of the normal road surface, when the brightness of the road is differentiated in this way, the brightness change is positive at the boundary point from the normal road surface to the white line 12, and the white line 12 Thus, differential data is obtained in which the change in luminance is negative at the boundary point from to the normal road surface. An example of such differential data is shown in FIG.
[0022]
For each of the data on each horizontal line 11, the differential value peaks appear in the order of plus and minus from the left, and the interval between the peaks seems to be appropriate as the white line 12 (from the plus peak to the minus peak). 3 is extracted as a white line candidate, and normally, the midpoint M is stored as a white line candidate point 15 as shown in FIG.
[0023]
Of these white line candidate points 15, only those closest to the center of the screen are left as final candidate points.
For example, when the vehicle is on the left side, the right side in the search area 10 is a road surface with little normal luminance change, and the white line candidate point 15 closest to the normal road surface can be determined as the white line 12. Therefore, even when an object (for example, a guard rail 14) that causes noise exists on the left side of the white line 12, the white line 12 can be reliably recognized from the image information captured by the camera 2.
[0024]
Then, as shown in FIG. 3F, finally, the continuity in the vertical direction of the white line candidate points 15 in each horizontal line data is sequentially verified from the bottom of the screen.
First, the inclination between the upper and lower ends of the white line 12 on the previous screen is calculated in advance. Then, assuming that the lowest point 15A is the white line 12, it is verified whether the candidate point 15B on the horizontal line 11 that is only one above is within the range of ± 50 dots of the previous white line 12 (within the error range). .
[0025]
If the candidate point 15B falls within this range, it is set as a white line, and if it does not, the candidate point 15B is rejected, and the coordinates calculated by interpolation from the above-described inclination are regarded as the white line position. Then, the same white line 12 can be recognized by performing the same operation for each horizontal line.
Such a white line recognition operation is continuously performed in a required cycle, and the recognition of the white line 12 is updated each time.
[0026]
The white line 12R as the roadside line at the right end of the traveling lane is recognized in the same manner.
By the way, even when there is a candidate point 15 that does not fall within the error range as described above, the white line 12 can be recognized by performing interpolation calculation from the slope between the upper and lower ends of the white line 12 on the previous screen. However, when many of the candidate points 15 on the screen are rejected without falling within the error range, it cannot be said that the white line on the road surface is effectively recognized. Cannot expect accurate control. Therefore, the image information processing means 3 outputs a recognition lost signal to the optical axis angle target value calculation means 5 assuming that white line recognition is not successful when a certain number of candidate points 15 or more are rejected. . However, when either one of the left and right white lines 12L and 12R can be recognized effectively, control can be performed based on the white line 12 that is recognized effectively, and therefore no recognition lost signal is output.
[0027]
In addition, as a case where the white line 12 cannot be recognized in this way, the road surface may be wet or there may be a large number of water pools. In such a case, since the light is easily reflected, the brightness of the road surface is increased and the brightness difference between the white line 12 and the white line 12 is reduced. Therefore, it is difficult to distinguish the white line 12 from the normal road surface. Further, even when the white line 12 is broken or unclear on the way, or when there is a display or object close to white near the white line 12, the white line may not be recognized.
[0028]
The travel lane estimation means 4 converts the white lines 12R, 12L on the original image 3A recognized in each recognition period in this way into a planar view image 3B and can be estimated from the white line 12L at the left end of the travel lane._LRoad center line LC that can be estimated from the white line 12R at the right end of the driving lane_RBased on the above, the road center line LC is estimated. Based on the road center line LC, the road curvature calculating means 4A calculates the road curvature ρ of the traveling lane ahead of the vehicle.
[0029]
In the road curvature calculation means 4A, as shown in FIG. 4A, a plurality of matching arc patterns 30 having different curvatures are stored, and these matching arc patterns 30 are stored on the road center line on the planar view image 3B. Whether or not they coincide with each other by being superposed on the LC is checked.
As a collation method, for example, an arc pattern (that is, a straight line) having a curvature of 0 is sequentially superposed on the road center line LC on the planar view image 3B. Then, as shown in FIG. 4B, for example, using the least square method, the distance between the matching arc pattern 30 and the road center line LC on the horizontal line 11 in order from the bottom of the screen (between the two points). The number of dots) is squared and accumulated, and the accumulated value is compared with the accumulated value in the matching arc pattern 30 where the previous collation is performed at the place where the uppermost horizontal line 11 is accumulated.
[0030]
At this time, when the integrated value in the current matching arc pattern 30 is smaller, the matching with the next matching arc pattern 30 is performed again. On the other hand, when the integrated value in the current matching arc pattern 30 is larger, the previous matching arc pattern 30 is regarded as an arc pattern matching the road center line LC, and the curvature of this arc pattern is defined as the road curvature ρ. It is supposed to be.
[0031]
The lateral deviation amount calculation means 4B and the yaw angle calculation means 4C calculate the lateral deviation amount ΔY and the yaw angle β, respectively, based on the image information of the road center line LC and the position information of the host vehicle with respect to the road center line LC. It has become.
That is, as shown in FIG. 5, the lateral deviation amount calculation means 4 </ b> B has a lateral distance (road width) between the first detection point (LC1) that is a point on the road centerline LC closest to the vehicle 1 and the vehicle centerline. Direction, that is, the distance in the horizontal direction of the camera image) is calculated as a lateral deviation amount ΔY, and the yaw angle calculation means 4C is formed by the tangent at the first detection point (LC1) of the road centerline LC and the vehicle centerline. The angle is calculated as the yaw angle β.
[0032]
  On the other hand, the substitute road curvature calculating means 25 calculates the estimated road curvature ρ ′ as follows.
  In other words, the lateral acceleration based on the centrifugal force acting on the vehicle is G, and the vehicle at that timeTravelingIf the magnitude of the speed is V, the estimated road curvature ρ ′ of the currently running road portion is calculated by the following equation.
[0033]
ρ ′ = G / V²              ... (1)
Therefore, the substitute road curvature calculating means 25 detects the lateral acceleration G by the lateral acceleration sensor (lateral acceleration detecting means) 21 and detects the vehicle speed (traveling speed) V by the vehicle speed sensor (traveling speed detecting means) 20. Based on the detected value, the estimated road curvature ρ ′ is calculated from the equation (1).
[0034]
In the optical axis angle target value calculation means 5, the optical axis angle target value θ is calculated based on the road curvature ρ, the lateral deviation amount ΔY, the yaw angle β, and the estimated road curvature ρ ′ calculated as described above.₁, Θ₂(= Θ₂′) Is calculated. The horizontal optical axis angle target value θ₂Since this is corrected after this, when distinguishing between before and after correction, the target value before correction is θ₂Indicated as ′.
That is, as shown in FIG. 6, when the vehicle travels along the road center line LC with the road curvature ρ and tries to illuminate the road center line LC in front of the vehicle, the irradiation distance of the headlight 9 is L. , Target value θ₂Can be expressed, for example, by the following equation.
[0035]
θ₂≒ ρ × L / 2 (2)
The irradiation distance L is the longitudinal optical axis angle target value θ of the headlight 9.₁This is a distance determined by the longitudinal optical axis angle target value θ₁For example, when switching between high and low by a driver operation, the irradiation distance L also changes accordingly.
[0036]
Therefore, the optical axis angle target value calculation means 5 first performs the longitudinal optical axis angle target value θ by a driver switching operation or the like.₁Is determined, this longitudinal optical axis angle target value θ₁The lateral optical axis angle target value θ using the equation (2) from the irradiation distance L determined by the above and the road curvature ρ calculated based on the image information.₂Is to decide.
However, in actual traveling, the vehicle is not necessarily traveling on the road center line LC, and as shown in FIG. It is running at an angle. Therefore, the optical axis angle target value calculation means 5 is the optical axis angle target value θ calculated as described above.₁, Θ₂In particular, the horizontal optical axis angle target value θ₂Is corrected by the optical axis angle target value correcting means 5A.
[0037]
  The optical axis angle target value correcting means 5A converts the vehicle lateral deviation amount ΔY calculated by the lateral deviation amount calculating means 4B and the yaw angle β formed by the vehicle calculated by the yaw angle calculating means 4C with respect to the road center line LC. Based on the horizontal optical axis angle target value θ₂The correction angle dθ is determined. That is, the correction angle dθ is expressed as follows:
        dθ = β−ΔY / L (3)
  Then, the optical axis angle target value correcting means 5A converts the correction angle dθ into the horizontal optical axis angle target value θ.₂From ′DecreaseBy calculating, correction according to the lateral deviation and inclination of the vehicle with respect to the road center line LC is performed. That is, the horizontal optical axis angle target value θ₂′ Is corrected as follows:
[0038]
θ₂= Θ₂′ −dθ (4)
The optical axis angle target value calculator 5 calculates the lateral optical axis angle target value θ corrected in this way.₂And the longitudinal optical axis angle target value θ set by manual operation of the driver₁Are input to the controller 7.
By the way, when the white line recognition by the image information processing means 3 is unsatisfactory, a recognition lost signal is input from the image information processing means 3 to the optical axis angle target value calculation means 5, while the recognition lost signal is being input. Since the road curve ρ is calculated based on the information of the road center line LC with low accuracy, it cannot be said that the road curve ρ accurately represents the curvature of the road center line. Therefore, based on this road curve ρ, the lateral optical axis angle target value θ₂Even if it is calculated and output to the controller 7, there is a high possibility that the optical axis control along the road center line LC cannot be performed effectively.
[0039]
Therefore, when the recognition lost signal is input from the image information processing unit 3, the optical axis angle target value calculation unit 5 receives the road curvature calculation unit 4 </ b> A by the switching unit 5 </ b> B that is a functional element of the optical axis angle target value calculation unit 5. Lateral optical axis angle target value θ based on road curvature ρ calculated in₂From the calculation of the horizontal direction optical axis angle target value θ based on the estimated road curvature ρ ′ calculated by the substitute road curvature calculation means 25.₂Switch to the calculation of.
[0040]
Further, the correction angle dθ in the optical axis target value correction means 5A is also calculated based on the image information, so that it is lacking in accuracy while the recognition lost signal is being input, and appropriate correction can be performed. It ’s hard. For this reason, the optical axis angle target value calculation means 5 performs the horizontal optical axis angle target value θ by the optical axis angle target value correction means 5A while the recognition lost signal is input.₂The correction is canceled.
[0041]
It should be noted that due to an error between the road curvature ρ and the estimated road curvature ρ ′, the lateral optical axis angle target value θ at the time of switching from the road curvature calculation means 4A to the substitute road curvature calculation means 25.₂Can change rapidly. In such a case, the driver may feel uncomfortable, but for this, as shown in FIG. 1, a low-pass filter 6 interposed between the optical axis target value calculation means 5 and the controller 7 is used. It is dealt with by smoothing. Further, when the correction by the optical axis angle target value correcting means 5A is stopped, the lateral angle light is not increased by suddenly making the correction angle dθ close to 0, instead of suddenly setting the correction angle dθ to 0. Axial angle target value θ₂To prevent sudden changes.
[0042]
Since the light distribution control device for an automobile headlight as one embodiment of the present invention is configured as described above, the light distribution control processing is performed as shown in FIG. 7, for example.
That is, the image information processing means 3 appropriately processes the image information in front of the vehicle input from the camera 2 to recognize the left and right white line positions on the road ahead (step S10), and can recognize the white line position effectively. If the white line position cannot be effectively recognized (that is, the white line recognition lost state), the process of step S50 is performed.
[0043]
First, when the white line position can be recognized effectively, the road curvature calculation means 4A calculates the road curvature ρ based on the white line position image information, and the optical axis angle target value calculation means 5A is calculated based on the road curvature ρ. The horizontal optical axis angle target value θ₂Is calculated (step S20). Similarly, the lateral deviation amount ΔY and the yaw angle β are respectively calculated by the lateral deviation amount calculation means 4B and the yaw angle calculation means 4C based on the image information, and the optical axis angle target value correction is performed based on the lateral deviation amount ΔY and the yaw angle β. Mean optical axis angle target value θ by means 5A₂Is corrected (step S30). On the other hand, if the position of the white line cannot be recognized effectively, the estimated road curvature ρ ′ is calculated by the substitute road curvature calculation means 4A based on the lateral acceleration G of the vehicle, the driver's operation, etc., and the estimated road curvature ρ ′ is calculated. On the basis of the optical axis angle target value calculation means 5A, the horizontal optical axis angle target value θ₂Is calculated (step S50). Then, the optical axis target value of the headlight 9 is calculated as the calculated horizontal optical axis target value θ.₂And the longitudinal optical axis angle target value θ set by the driver operation₁So that the optical axis actuator 8 is operated through the controller 7 (step S40).
[0044]
Here, the horizontal optical axis angle target value θ₂In detail, the road curvature ρ needs to be calculated. In this apparatus, first, as shown in FIG. 2, the image information processing means 3 processes the original image 3A of the road ahead of the vehicle captured from the camera 2 to recognize the white road white lines 12R and 12L. Then, the travel lane estimation means 4 converts the white lines 12R, 12L on the original image 3A recognized by the image information processing means 3 into a planar view image 3B as viewed from above, and the white line 12L at the left end of the travel lane. Estimated from the road center line LC_LRoad center line LC that can be estimated from the white line 12R at the right end of the driving lane_RBased on the above, the road center line LC is estimated.
[0045]
Based on the road center line LC, the road curvature calculation means 4A calculates the road curvature ρ of the traveling lane ahead of the vehicle.
That is, in the road curvature calculation means 4A, as shown in FIG. 4A, a plurality of matching arc patterns 30 having different curvatures are stored, and these matching arc patterns 30 are stored on the road on the planar view image 3B. The images are sequentially superimposed on the center line LC. Then, as shown in FIG. 4B, for each matching arc pattern 30, the distance between the matching arc pattern 30 and the road center line LC on the horizontal line 11 in order from the bottom of the screen (dot between two points). The number is squared and accumulated, and the accumulated value is compared with the accumulated value in the collation arc pattern 30 that has been collated last time when the accumulation is performed up to the uppermost horizontal line 11. When the integrated value in the current matching arc pattern 30 is larger, the previous matching arc pattern 30 is regarded as an arc pattern that matches the road center line LC, and the curvature of this arc pattern is defined as the road curvature ρ.
[0046]
At this time, the lateral deviation amount calculation means 4B and the yaw angle calculation means 4C calculate the lateral deviation amount ΔY and the yaw angle β, respectively, based on the image information of the road center line LC and the position information of the vehicle with respect to the road center line LC. To do. That is, as shown in FIG. 5, the lateral deviation amount calculation means 4 </ b> B has a lateral distance (road width) between the first detection point (LC1) that is a point on the road centerline LC closest to the vehicle 1 and the vehicle centerline. Direction, that is, the distance in the horizontal direction of the camera image) is calculated as a lateral deviation amount ΔY, and the yaw angle calculation means 4C is formed by the tangent at the first detection point (LC1) of the road centerline LC and the vehicle centerline. The angle is calculated as the yaw angle β.
[0047]
Based on the road curvature ρ, the lateral deviation amount ΔY, and the yaw angle β calculated by the road curvature calculation unit 4A, the lateral deviation amount calculation unit 4B, and the yaw angle calculation unit 4C, the optical axis angle target value calculation unit 5 calculates the horizontal optical axis. Angle target value θ₂Is calculated.
First, the optical axis angle target value calculation means 5 has the road curvature ρ calculated by the road curvature calculation means 4A and the longitudinal optical axis angle target value θ set by the driver's manual operation.₁Based on the irradiation distance L determined by the following equation (2), the horizontal optical axis angle target value θ₂Is calculated.
[0048]
However, in actual traveling, the vehicle is not necessarily traveling on the road center line, and is deviated from the road center line or is inclined with respect to the road center line. Therefore, the optical axis angle target value correcting means 5A, which is a functional element of the optical axis angle target value calculating means 5, has a lateral deviation amount ΔY with respect to the road center line LC calculated by the lateral deviation amount calculating means 4B and the yaw angle calculating means 4C, respectively. Based on the yaw angle β, the lateral optical axis angle target value θ₂Correct.
[0049]
  In other words, the optical axis angle target value correcting means 5A is based on the lateral deviation amount ΔY and the yaw angle β and the horizontal optical axis angle target value θ according to the above equation (3).₂The correction angle dθ is determined, and the correction angle dθ is converted into the horizontal optical axis angle target value θ as shown in the above equation (4).₂FromDecreaseBy calculating the lateral optical axis angle target value θ according to the lateral deviation and inclination with respect to the road center line of the vehicle₂Make corrections.
[0050]
The controller 7 obtains the lateral optical axis angle target value θ thus obtained.₂And longitudinal optical axis angle target value θ₁The optical axis actuator 8 is actuated on the basis of the above, so that no matter how the vehicle is traveling on the traveling lane, that is, regardless of the lateral deviation or the yaw angle from the road center line, The irradiated light always illuminates the road center line in front of the vehicle.
[0051]
Also, the longitudinal optical axis angle target value θ₁For example, even when the driver operates the vertical optical axis of the headlight 9 from high to low, the vertical optical axis angle target value θ₁The horizontal optical axis angle target value θ according to the irradiation distance L calculated from₂Therefore, even in such a case, the road center line in front of the vehicle can always be illuminated.
By the way, when the road surface is wet or there are many water pools on the road surface, the light is easily reflected and the brightness of the road surface increases, and the difference in brightness from the white line decreases. And the normal road surface becomes difficult to distinguish. Further, when the white line is broken or unclear on the way, or when a display or object close to white exists near the white line, the white line cannot be recognized. In such a case, in the white line recognition processing in the image information processing means 3, many of the candidate points 15 of the white line 12 do not fall within the error range, and the white line 12 from the candidate point 15 cannot be recognized. The white line 12 is recognized by performing interpolation calculation from the slope between the upper and lower ends of the white line 12 on the screen.
[0052]
However, when many of the candidate points 15 on the screen are rejected without falling within the error range, it cannot be said that the white line on the road surface can be recognized effectively, and such an accuracy is high. Even if the road curvature ρ is calculated based on the low white line information, it cannot be said that it accurately represents the curvature of the road centerline. Therefore, based on this road curve ρ, the lateral optical axis angle target value θ₂Even if it is calculated and output to the controller 7, there is a high possibility that the optical axis control along the road center line cannot be performed effectively.
[0053]
Therefore, when a certain number or more of candidate points 15 are rejected, a recognition lost signal is output to the optical axis angle target value calculation means 5 because white line recognition is not satisfactory. However, if either one of the left and right white lines 12L and 12R can be recognized effectively, the control can be performed based on the white line 12 that has been effectively recognized, so that the recognition lost signal is not output and the left and right white lines are output. A recognition lost signal is output only when both 12L and 12R cannot be recognized.
[0054]
When the recognition lost signal is input from the image information processing means 3, the optical axis angle target value calculation means 5 is calculated by the road curvature calculation means 4 A by the switching means 5 B that is a functional element of the optical axis angle target value calculation means 5. Horizontal optical axis angle target value θ based on measured road curvature ρ₂From the calculation of the horizontal direction optical axis angle target value θ based on the estimated road curvature ρ ′ calculated by the substitute road curvature calculation means 25.₂Switch to calculation.
[0055]
This substitute road curvature calculation means 25 calculates the road curvature ρ based on the road center line LC estimated by the travel lane estimation means 4 while the road curvature calculation means 4A calculates the lateral acceleration actually acting on the vehicle 1. G and vehicle speed V are detected by the lateral acceleration sensor 21 and the vehicle speed sensor 20, respectively, and the estimated road curvature ρ 'is estimated from the above equation (1) based on the detected lateral acceleration G and vehicle speed V. Optical axis control can be performed regardless of the white line recognition state of the processing means 3. Therefore, when the white line recognition of the image information processing means 3 is unsatisfactory, the estimated road curvature ρ ′ is calculated by switching to the substitute road curvature calculation means 25, and the lateral optical axis angle target value θ is calculated from the estimated road curvature ρ ′.₂It is possible to prevent the optical axis control from becoming unstable due to the road condition.
[0056]
The substitute road curvature calculation means 25 detects the lateral acceleration G directly by the lateral acceleration sensor 21, calculates the lateral acceleration G based on the steering angle α when the driver steers the steering wheel, and calculates the calculated lateral acceleration. An estimated road curvature ρ ′ can also be calculated based on G.
That is, a steering angle sensor 22 is provided on a steering shaft (not shown) of the vehicle, the steering angle sensor 22 detects the steering angle α of the steering wheel, and the lateral acceleration G is calculated by the following equation together with the vehicle speed V detected by the vehicle speed sensor 20.
[0057]
G = [α / (n × g)] × (V²/ Wb) / (1 + a × V²(5)
Then, the estimated road curvature ρ ′ is calculated by substituting the equation (5) into the equation (1). Here, n, g, wb, and a are fixed values indicating gear ratio, gravitational acceleration, wheel base, and stability factor, respectively. It should be noted that due to an error between the road curvature ρ and the estimated road curvature ρ ′, the lateral optical axis angle target value θ at the time of switching from the road curvature calculation means 4A to the substitute road curvature calculation means 25.₂May be unnaturally fluctuated, but this can be dealt with by performing a smoothing process with a low-pass filter 6 interposed between the optical axis angle target value calculation means 5 and the controller 7. Thus, it is possible to prevent the driver from feeling uncomfortable due to an unnatural change in the optical axis.
[0058]
If the white line recognition of the image information processing means 3 is not satisfactory, the correction angle dθ in the optical axis angle target value correction means 5A is also calculated based on the image information, so that the accuracy is lacking and appropriate correction can be performed. Since it is difficult to say, the optical axis target value calculation means 5 performs the horizontal optical axis angle target value θ by the optical axis angle target value correction means 5A while the recognition lost signal is input.₂Cancel the correction. Then, when the correction by the optical axis target value correction means 5A is stopped, the correction angle dθ is not suddenly made zero, but the magnitude of the correction angle dθ is gradually brought close to zero. As a result, the horizontal optical axis angle target value θ₂In this setting, the influence of the road condition is removed, stable optical axis control can be performed, and the correction angle dθ is gradually reduced to 0, thereby preventing the driver from feeling uncomfortable due to unnatural fluctuations in the optical axis. It can be done.
[0062]
【The invention's effect】
  As described in detail above, according to the light distribution control device for an automotive headlight of the present invention according to claim 1,The imaging means captures road information ahead of the host vehicle, and based on the road image information ahead of the captured vehicle.Driving lane from white line on road recognized by white line recognition meansRoad center lineSince the road curvature is calculated based on the estimated travel lane, the curve condition of the travel lane ahead of the vehicle can be accurately grasped, and the travel lane of the vehicleRoad center lineThe position and orientation of the vehicle can also be grasped by the lateral deviation amount calculation means and the yaw angle calculation means. Therefore, by correcting the target value of the optical axis angle calculated based on the road curvature according to the lateral deviation and the yaw angle, Regardless of the position and orientation relative to the travel lane, the travel lane is alwaysRoad center lineThe optical axis angle of the headlight can be controlled so as to accurately irradiate the light.
[0063]
  According to the light distribution control device for a headlight for an automobile according to the second aspect of the present invention, when the white line recognition by the white line recognition unit is not satisfactory, the road curvature calculated by the road curvature calculation unit based on the image information is replaced. Thus, the estimated road curvature estimated based on the lateral acceleration detected by the lateral acceleration detecting means and the traveling speed detected by the traveling speed detecting means is substituted for the calculation of the optical axis angle target value. It is possible to prevent the light distribution control from becoming discontinuous and causing the driver to feel uncomfortable.
  According to the light distribution control device for an automobile headlight according to the third aspect of the present invention, the influence of the road condition can be removed in setting the optical axis angle target value, and stable optical axis control is performed. be able to.
  According to the light distribution control device for an automobile headlight according to the fourth aspect of the present invention, it is possible to prevent the driver from feeling uncomfortable due to an unnatural change in the optical axis.
  According to the fifth aspect of the present invention, the vehicle headlight light distribution control device gradually reduces the correction angle dθ to 0 to prevent the driver from feeling uncomfortable due to unnatural fluctuations in the optical axis. be able to.
[Brief description of the drawings]
FIG. 1 is a block diagram schematically showing a configuration of a light distribution control device for an automotive headlight according to an embodiment of the present invention.
FIG. 2 is a diagram illustrating image processing for travel lane recognition according to an embodiment of the present invention.
FIG. 3 is a schematic diagram illustrating travel lane recognition according to an embodiment of the present invention in the order of (a) to (f).
FIG. 4 is an explanatory diagram for explaining calculation of road curvature of a traveling lane.
FIG. 5 is an explanatory diagram for explaining calculation of a lateral shift amount and a yaw angle.
FIG. 6 is an explanatory diagram for describing calculation of a target value of an optical axis angle in the horizontal direction.
FIG. 7 is a flowchart for explaining the operation of the light distribution control device for an automobile headlight as one embodiment of the present invention.
[Explanation of symbols]
2 Camera (image information input means)
3 Image information processing means (white line recognition means)
4 Travel lane estimation means (travel lane estimation means)
4A Road curvature calculation means
4B lateral deviation amount calculation means
4C Yaw angle calculation means
5 Optical axis angle target value calculation means
5A Optical axis angle target value correction means
5B switching means
6 Low-pass filter
7 Controller (control means)
8 Optical axis actuator
9 Headlight
20 Lateral acceleration sensor (lateral acceleration detection means)
21 Vehicle speed sensor (traveling speed detection means)
22 Steering angle sensor
25 Substitute road curvature calculation means
LC road center line

Claims (5)

A light distribution control device for an automotive headlight that adjusts an optical axis angle of a headlight via an optical axis actuator according to a curve condition of a traveling lane ahead of the host vehicle,
Imaging means for imaging the road condition ahead of the vehicle;
A white line recognition means for recognizing a white line該道road from the image information captured by the imaging means,
Travel lane estimation means for estimating a road center line of the travel lane as a relative position of the travel lane on which the host vehicle travels from the white line recognized by the white line recognition means;
Road curvature calculating means for calculating the curvature of the road center line of the traveling lane estimated by the traveling lane estimating means;
And the optical axis angle target value calculating means for calculating the optical axis angle target value based on the curvature calculated in the road curvature calculating means,
Control means for controlling the optical axis actuator so that the optical axis angle of the headlight is equal to the optical axis angle target value calculated by the optical axis angle target value calculating means;
A lateral deviation amount calculating means for calculating the lateral deviation amount of car both the center line of the road center line of the traffic lane estimated by the traffic lane estimating means,
Yaw angle calculation means for calculating a yaw angle of the vehicle center line with respect to a road center line of the travel lane estimated by the travel lane estimation means;
The optical axis angle target value calculating means adds the ratio of the lateral deviation amount calculated by the lateral deviation amount calculating means to the irradiation distance of the headlight, and subtracts the yaw angle calculated by the yaw angle calculating means. And an optical axis angle target value correcting means for correcting the optical axis angle target value calculated by the optical axis angle target value calculating means. . Lateral acceleration detecting means for detecting lateral acceleration applied to the vehicle;
A running speed detecting means for detecting the running speed of the vehicle;
When the white line recognition by the white line recognition unit is unsatisfactory, the optical axis angle target value calculation unit replaces the road curvature calculated by the road curvature calculation unit and detects the lateral acceleration detected by the lateral acceleration detection unit. 2. The automobile head according to claim 1, wherein the optical axis angle target value is calculated based on an estimated road curvature estimated based on an acceleration and the traveling speed detected by the traveling speed detecting means. Light distribution control device. The white line recognition means outputs a recognition lost signal to the optical axis angle target value calculation means when the white line recognition is not successful.
The optical axis angle target value calculation means stops the correction of the optical axis angle target value by the optical axis angle target value correction means while the recognition lost signal is input.
The light distribution control device for an automotive headlight according to claim 1 or 2, A low-pass filter that is interposed between the control unit and the optical axis target value calculation unit and smoothes the temporal change of the optical axis target value is provided.
The light distribution control device for a headlight for an automobile according to any one of claims 1 to 3, wherein When the recognition lost signal is input, the optical axis target value calculation means gradually approaches the correction angle for correcting the optical axis target value to 0.
The light distribution control device for an automotive headlight according to claim 3 or 4,

Images