JP2002104113A - Rear monitoring device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To simply detect a lane where a vehicle is running in detecting the vehicle approaching from the rear by a camera mounted backward on the vehicle. SOLUTION: A steering angle θ and a driver's own vehicle speed V are stored for the past designated time, whereby the shape of a road is estimated from the locus of running of the own vehicle, an area is set from the approximate shape of each running lane, and approach of some other vehicle and its running lane area are detected from the respective optical flows. As for the running lane, the position of a lane marker very near the own vehicle is detected to estimate the running lane, or from the storage of the lane marker, a more accurate running lane is calculated. Three-dimensional distortion is absorbed by detecting a gradient of a road to further accurately estimate the shape of the road.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to, for example, capturing an image behind a host vehicle with a camera, detecting other vehicles behind the host vehicle from the image, and judging danger in some cases.
The present invention relates to a rear monitoring device for a vehicle that issues a warning.

[0002]

2. Description of the Related Art An example of such a backward monitoring device is described in Japanese Patent Application Laid-Open No. 2000-20686. In this rear monitoring device, a video camera is attached to the rear part of the own vehicle toward the rear side, and each point of a captured image obtained during traveling of the own vehicle converges at infinity or disappearance. A point (generally called FOE: Focus of Expansion) is obtained, a feature point of an image recognized as an image of another vehicle is extracted from the captured image, and a temporal movement of the feature point is represented by a movement vector, that is, an optical vector. It is determined as a flow, and depending on whether this optical flow is diverging from the point at infinity or vanishing point or converging at the point at infinity or vanishing point, the other traveling in the adjacent lane is determined. This is to calculate the relative speed, relative distance, or relative positional relationship of the vehicle or another following vehicle approaching the own vehicle. Then, the danger of these other vehicles is determined, and if it is determined that there is danger, a warning is issued.

[0003]

By the way, in a configuration in which another vehicle is detected by the rear monitoring device as described above and a warning is issued when it is determined that there is a danger, the other vehicle approaching the own vehicle is The warning method differs depending on whether the vehicle is traveling in the same lane as the own vehicle or in the adjacent lane. For example, for an approaching other vehicle traveling in an adjacent lane, a warning is required only when the own vehicle is about to change lanes to that lane, but no warning is required for other vehicles. In order to achieve this, it is necessary to correctly detect the traveling lane of another approaching vehicle, and therefore, it is necessary to detect the shape of a lane marker (white line) that divides the traveling lane. For this reason, the conventional rearward monitoring device obtains a boundary, that is, an edge between the image of the lane marker and an image of other roads, backgrounds, and the like by using the luminance difference between the image and the shape of the lane marker. ing.

However, this lane marker detection process covers a wide range of the captured image, and the detected edge image also includes edge components of other vehicles and objects in the screen. , A complicated recognition process is required, and the time required for the arithmetic processing becomes long. In addition, as the rear monitoring device, the other vehicle detection process based on the optical flow described above must be performed, and the calculation process is complicated, so that the total time required for the calculation process is enormous.
There is a problem that the content becomes unsuitable for use as a warning device.

In order to simplify the processing contents, a configuration is also conceivable in which the lane marker detection processing is not performed and only the other vehicle detection processing is performed by setting a predetermined area in the image as each lane area. Then, for example, when traveling on a curved road, a deviation occurs between the assumed lane area and the actual lane area, the correct lane detection of other vehicles cannot be performed, and the accurate warning processing as described above is performed. Can not.

SUMMARY OF THE INVENTION The present invention has been developed to solve these problems, and performs complicated arithmetic processing by estimating the state of the road behind the host vehicle from the past behavior of the host vehicle. It is another object of the present invention to provide a rearward monitoring device capable of detecting the traveling lane of another vehicle behind the host vehicle.

[0007]

According to a first aspect of the present invention, there is provided a rearward monitoring apparatus, comprising: an image pickup means for picking up an image behind a host vehicle; A rear road condition estimating means for estimating the state of the road behind the own vehicle from the vehicle, and a predetermined image corresponding to the road condition behind the own vehicle estimated by the rear road condition estimating device in the image taken by the image taking means. Another vehicle detecting means for setting an area and detecting another vehicle in the predetermined area is provided.

According to a second aspect of the present invention, in the rearward monitoring device according to the first aspect, the road state estimating means determines the turning state detection amount and the own vehicle speed as the past behavior of the own vehicle. Are used. Further, the rear monitoring device according to claim 3 of the present invention, in the invention according to claim 1 or 2, sets a predetermined lane marker detection area in the vicinity of the own vehicle in the image taken by the imaging means, The vehicle is provided with a lane marker detecting means for detecting the position of the lane marker in the lane marker detection area, and the rear road state estimating means uses the position of the lane marker detected by the lane marker detecting means for the own vehicle to detect the position of the lane marker behind the own vehicle. It is characterized by estimating the state of the road.

According to a fourth aspect of the present invention, there is provided a rearward monitoring apparatus according to the first to third aspects, wherein the rear road state estimating means includes a past predetermined lane marker detected by the own vehicle lane marker detecting means. It is characterized in that the state of the road behind the host vehicle is estimated using the record of the time lane marker position. The rearward monitoring device according to a fifth aspect of the present invention, in the first to fourth aspects, further includes a rear road gradient detecting unit that detects a road gradient behind the host vehicle, and the rear road state estimating unit includes: The state of the road behind the host vehicle is estimated using the road gradient behind the host vehicle detected by the rear road gradient detecting means.

According to a sixth aspect of the present invention, there is provided the rearward monitoring apparatus according to the first to fifth aspects, wherein the rear road state estimating means removes a change in the behavior of the own vehicle at a predetermined frequency or higher. A frequency band cutoff filter is provided. Also, in the rearward monitoring device according to claim 7 of the present invention, in the invention of claim 1 to 6, the rear road state estimating means includes a predetermined frequency range cut-off for removing a change in own vehicle behavior in a predetermined frequency range. A filter is provided.

[0011]

According to the candidate monitoring apparatus of the first aspect of the present invention, the state of the road behind the host vehicle is estimated from the past behavior of the host vehicle, and the captured image of the host vehicle rear is taken. In this configuration, a predetermined area is set according to the estimated road condition behind the host vehicle, and another vehicle is detected in the predetermined area. Therefore, the lane marker can be detected over a wide range of the captured image. In addition, it is possible to detect the approximate shape of the traveling lane and the traveling lane in which the other vehicle is traveling, so that the time required for the arithmetic processing can be reduced.

Further, according to the rear monitoring device of the present invention, since the turning state detection amount and the own vehicle speed are used as the past behavior of the own vehicle, for example, steering is performed as the turning state detection amount. If the angle, the yaw rate, the lateral acceleration, and the like are applied, the behavior of the own vehicle can be detected easily and inexpensively, and the road condition behind the own vehicle can be easily estimated from these.

According to the rearward monitoring device of the third aspect of the present invention, a predetermined lane marker detection area is set near the own vehicle in the captured image, and the position of the lane marker is determined in the lane marker detection area. Detecting and using the position of the lane marker to estimate the state of the road behind the host vehicle, the road condition behind the host vehicle and the shape of each lane can be accurately detected, and only that much. The traveling lane in which the other vehicle is traveling can be accurately detected.

Further, according to the rearward monitoring apparatus of the present invention, the configuration of estimating the state of the road behind the own vehicle using the detected record of the position of the lane marker in the past predetermined time. Therefore, the road condition behind the host vehicle and the shape of each traveling lane can be detected more accurately, and the traveling lane on which the other vehicle is traveling can be more accurately detected. .

According to the rearward monitoring device of the present invention, the road gradient behind the host vehicle is detected, and the state of the road behind the host vehicle is detected using the road gradient behind the host vehicle. Because of the configuration for estimating, it is possible to accurately detect the road condition behind the own vehicle, and it is possible to accurately detect the shape of each traveling lane and the traveling lane in which another vehicle is traveling by that much. .

Further, according to the rearward monitoring device of the present invention, since a change in the behavior of the host vehicle at a predetermined frequency or higher is removed by a high frequency cutoff filter, a minute vehicle due to a correction steering or the like is removed. By removing the behavior change component, it is possible to stably estimate the road state behind the host vehicle, and it is possible to accurately detect the traveling lane in which the other vehicle is running.

Further, according to the rearward monitoring device of the present invention, since the change in the vehicle behavior in the predetermined frequency range is removed by the predetermined frequency cutoff filter, the vehicle behavior generated by the lane change is eliminated. If the frequency range of the change is removed by the predetermined frequency cutoff filter, the influence of the error accompanying the lane change of the own vehicle can be removed, and the road condition behind the own vehicle can be stably estimated. It becomes possible to accurately detect the traveling lane in which the other vehicle is traveling.

[0018]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic configuration block diagram of an alarm device using the rear monitoring device of the present invention. The rear monitoring device 1 includes a vehicle behavior estimating unit 1a that estimates the behavior of the host vehicle from a steering angle θ detected by a steering angle sensor 2 attached to a steering wheel and a vehicle speed V detected by a vehicle speed sensor 3. Have. Also, the rear monitoring device 1
Reads an image of the host vehicle rear image captured by a vehicle rear camera 4 such as a CCD camera mounted on the rear of the vehicle, stores the host vehicle behavior from the vehicle behavior estimating unit 1a, and stores the image of the road behind the host vehicle. It is provided with another vehicle detection section 1b for estimating the state and detecting the other vehicle and its traveling lane from the road state behind the own vehicle. Also, the rear monitoring device 1
Reads the operation state of the turn signal device 5 provided in the vehicle, determines the danger from the other vehicle detected by the other vehicle detection unit 1b and the traveling lane, and determines that there is danger. In this case, an alarm generator 1c for driving an alarm device 6 provided in the vehicle is provided.

The rear monitoring device 1 includes an arithmetic processing device such as a microcomputer, and the vehicle behavior estimating section 1a, the other vehicle detecting section 1b, and the alarm generating section 1c are configured by arithmetic processing. FIG. 2 is a flowchart of the arithmetic processing performed in the arithmetic processing device of the rearward monitoring device 1. In the arithmetic processing device, the timer is interrupted at a predetermined sampling time (in this case, 100 msec.). In this flowchart, no particular communication step is provided. For example, information obtained in the flowchart is stored in the storage device as needed, and necessary information is read from the storage device as needed. In addition, the respective devices also communicate with each other, and necessary information is always read from the device that is mainly in charge of control, and the transmitted information is stored in the storage device as needed.

In step S1 of this arithmetic processing, according to the individual arithmetic processing performed in the step, the whole screen information behind the vehicle taken by the vehicle rear camera 4 is read and input to an array composed of digital fields and the like. . In the present embodiment, for example, as shown in FIG. 3, the lower right corner of the entire image is set as the origin and the ordinate (vertical coordinate) is set as the x coordinate, and the abscissa (horizontal coordinate) is set as the y coordinate.
Information such as hue, brightness, and luminance is stored for each pixel.

Next, the process proceeds to step S2, in which the steering angle θ detected by the steering angle sensor 2 is read. Next, the process proceeds to step S3, in which the steering angle θ read in step S2 is smoothed according to the individual calculation process performed in step S3. This smoothing of the steering angle θ is for removing the influence of a minute correction steering. Since such correction steering is an operation in a relatively high frequency range, a change in the steering angle over a predetermined frequency range is removed. A low-pass filter may be used. The frequency range to be removed is about 2 Hz or more.

Next, the process proceeds to step S4, where the vehicle speed V detected by the vehicle speed sensor 3 is read. Next, step S
Then, in accordance with the individual arithmetic processing performed in the same step, the smoothed and stored steering angle data and the stored own vehicle speed data are used to eliminate the road behind the own vehicle. Estimate a point (infinity point). Specifically, the vanishing point is estimated as follows. That is, the storage device of the rear monitoring device 1 stores a past predetermined time (n × Δ
T), the smoothed steering angle θ _k (k = 0, 1,
.., N) and the own vehicle speed V _k (k = 0, 1,..., N) are stored. That is, in the storage device, n pieces of data have been accumulated in the past at the predetermined sampling time ΔT,
Here, it is shown that the stored data k times before are θ _k and V _k . Then, as shown in FIG. 4, an X axis is set behind the host vehicle, and a Y axis is set laterally.
The state (shape) of the rear road on which the host vehicle has traveled is estimated from the stored steering angle θ _k and host vehicle speed V _k .

[0023]

(Equation 1)

Here, K is a predetermined value indicating the ratio of the yaw angle to the steering angle θ, and m is an integer from 0 to n.
From the above formulas (1) to (3), it is possible to estimate the road condition where the vehicle has traveled at the (X, Y) coordinates in FIG. The coordinate values (X _m , Y
_m ) indicates the trajectory of the vehicle. next,
Equation (1) is used to determine m such that the X coordinate value exceeds a predetermined value (for example, 200 m), and vanishing point coordinates (X _VP , Y _VP ) are calculated from Equations 1 and 2. Next, this coordinate value is expressed by the following 4
The inverse perspective transformation is performed on the image plane by Equations (6) to (6) to calculate the position ( _xVP , _yVP ) of the vanishing point VP in the image plane shown in FIG.

X = X · f / L (4) y = Y · f / L (5) L = (X ² + Y ² ) ^1/2 (6) where f Is the focal length of the camera 4 behind the vehicle. Next, the process proceeds to step S6, in which the steering angle data and the own vehicle speed data are stored for the next and subsequent processes in accordance with the individual calculation processes performed in the step.

Next, the process proceeds to step S7, where the rear road area is extracted in accordance with the individual arithmetic processing performed in the step. That is, as shown in FIG.
The points A to D set in the image in advance are connected to the vanishing point VP calculated in step 5, and the point B and the vanishing point V
A triangular area surrounded by a line segment connecting P and a line segment connecting point C and the vanishing point VP is defined as a host vehicle traveling lane region, and similarly, a line segment connecting point A and the vanishing point VP, and B Point and vanishing point VP
Is defined as a lane area adjacent to the right side of the lane in which the host vehicle is traveling, and similarly, point C and vanishing point VP
And an area surrounded by a line connecting the point D and the vanishing point VP is defined as a lane area adjacent to the left side of the lane in which the host vehicle is running. The lane area set in this manner does not completely match the actual lane area since no lane marker is detected, but the width of the lane is determined to some extent, and the vehicle rear camera 4 Since the positions of the lane markers in the image picked up by are also determined to some extent, the points A to D at which the lane markers will appear
And the vanishing point VP, approximately each lane area can be extracted.

Next, the processing shifts to step S8, and in accordance with the individual calculation processing performed in the step, the above-mentioned step S8 is performed.
The optical flow is detected for each lane region extracted in step 7. The detection of the optical flow is performed, for example, as shown in FIG.
As shown in FIG. 7, each vehicle occupation area is divided into minute areas of, for example, 5 pixels square, and it is detected in which direction each minute area in the previous rear image is moving in the current rear image. .

Next, the process proceeds to step S9, and the position and the relative speed of another vehicle are detected in accordance with the individual calculation processing performed in the step. Specifically, as shown in FIG. 5, among the optical flows detected in step S8, the optical flows moving downward in the image are:
That is, since the object approaches the own vehicle, in this case, the other vehicle, the position is obtained, and the distance L to the other vehicle is obtained by performing a perspective inverse transformation from the position, and further, from the length of the optical flow, Relative speed Vr of other vehicle
Can be requested. In the case of FIG. 5, an optical flow (F section) below the image is detected in the right adjacent lane surrounded by the points A, B and the vanishing point VP, and this is another vehicle approaching.

Next, the process proceeds to step S10, where the status of the turn signal operation by the driver is read in accordance with the individual arithmetic processing performed in the step. Specifically, the driver's intention to change lanes is detected based on which direction the turn signal switch is operated. Next, the process proceeds to step S11, and it is determined whether there is an approaching vehicle according to the individual calculation processing performed in the step. If there is an approaching vehicle, the process proceeds to step S12.
If not, the program returns to the main program. Specifically, in accordance with the position and relative speed of the other vehicle detected in step S9, the turn signal operation status read in step S10 is used to determine, for example, the F portion of the right adjacent lane detected in FIG. The driver attempts to change lanes to the right lane with respect to the vehicle, and determines whether the other vehicle is sufficiently fast or close to determine the presence of an approaching vehicle. The approach state may be determined by determining whether or not a margin time Tp obtained by dividing the distance L to the other vehicle by the relative speed Vr is equal to or shorter than a predetermined time (for example, 3 seconds).

In step S12, a predetermined alarm issuance process is performed on the alarm device 6 in accordance with the individual arithmetic processing performed in step S12, and then the process returns to the main program. In this arithmetic processing, as described above, the approximate shape of the lane and the traveling lane position of another vehicle can be detected without performing complicated lane marker detection, so that the time required for the arithmetic processing can be reduced. it can. Also,
Using the steering angle θ as the turning state detection amount, it is possible to easily and inexpensively detect the own vehicle behavior from the steering angle θ and the own vehicle speed V, and to easily estimate the road state behind the own vehicle from these. it can. In addition, by removing a small vehicle behavior change component due to a correction rudder or the like by a low-pass filter, it is possible to stably estimate a road state behind the own vehicle, and to reduce a traveling lane by which another vehicle is traveling. Accurate detection is possible. In the extraction of the lane area, for example, if the road shape is simply approximated by an x, y polynomial or the like, the lane area can be extracted with higher accuracy.

As described above, the vehicle rear camera 4 and FIG.
Step S1 of the calculation processing constitutes the imaging means of the present invention, and similarly, the steering angle sensor 2, the vehicle speed sensor 3, and steps S2 to S6 of the calculation processing of FIG. Step S7 of the calculation processing of FIG.
Step S9 constitutes another vehicle detection means, and step S3 of the calculation processing in FIG. 2 constitutes a high frequency cutoff filter.

Next, a second embodiment of the backward monitoring device of the present invention will be described with reference to FIGS. FIG. 6 showing a schematic configuration of the rear monitoring device is similar to FIG. 1 of the first embodiment, and the same components are denoted by the same reference numerals. In this embodiment, a nearby white line (lane marker) detecting unit 1d is added between the other vehicle detecting unit 1b and the vehicle rear camera 4. This neighborhood white line detection unit 1d
Is for detecting a lane marker in the immediate vicinity of the vehicle and detecting the lane shape more accurately.

This backward monitoring device 1 is also actually configured by arithmetic processing performed by an arithmetic processing device. FIG.
It is a flowchart of the calculation processing. In this calculation processing, there are many steps equivalent to the calculation processing of FIG. 2 of the first embodiment, but all the steps will be described to make it easier to understand the newly added steps. This arithmetic processing is also performed for a predetermined sampling time (in this case, 10
0 msec.), A timer interrupt process is performed. In this flowchart, no particular communication step is provided. For example, information obtained in the flowchart is stored in the storage device as needed, and necessary information is read from the storage device as needed. In addition, the respective devices also communicate with each other, and necessary information is always read from the device that is mainly in charge of control, and the transmitted information is stored in the storage device as needed.

In step S21 of the arithmetic processing, as in the first embodiment, the full screen information of the rear of the vehicle captured by the vehicle rear camera 4 is read in accordance with the individual arithmetic processing performed in the same step. Input to an array consisting of digital fields. Next, the process proceeds to step S22, where the steering angle θ detected by the steering angle sensor 2 is read in the same manner as in the first embodiment.

Next, the flow shifts to step S23, where the first
As in the case of the embodiment, the steering angle θ read in step S22 according to the individual arithmetic processing performed in the same step.
Is performed. Next, the process proceeds to step S24,
The lane change component is removed from the steering angle smoothed in step S23 according to the individual arithmetic processing performed in the step. This steering angle lane change component removal processing removes a change in the steering angle due to a lane change.
A change in the steering angle of about 5 to 2 Hz is removed by a band-pass filter.

Next, the flow shifts to step S25, where the first
As in the embodiment, the own vehicle speed V detected by the vehicle speed sensor 3 is read. Next, in step S26, the smoothed steering angle data and the stored own-vehicle speed data are used to calculate the position of the rear of the own vehicle in accordance with the individual arithmetic processing performed in the same step. Estimate the shape (state) of the road. Specifically, coordinate values (X _m , Y _m ) representing the trajectory of the own vehicle are obtained in accordance with the formulas 1 to 3 described in the first embodiment, and the coordinate values (X _m , Y _m ) are respectively described in the first embodiment. 4-6 by performing inverse perspective transformation by equation image on the road shape coordinates (x _m, y _m) is calculated (see FIG. 9).

Next, the flow shifts to step S27, where the first
As in the case of the embodiment, the steering angle data and the own vehicle speed data are stored in memory for the subsequent processing according to the individual calculation processing performed in the same step. Next, the process proceeds to step S28, where a white line (lane marker) near the own vehicle is detected according to the individual arithmetic processing performed in the step.
Specifically, as shown in FIG. 8, a lane marker detection area G is set at the lower end of the own vehicle rear image, that is, at a portion closest to the own vehicle. Detected one by one (points B 'and C' in the figure). Generally, the lane marker image of the white line has high luminance and can be easily detected. The lane marker cannot be detected in the portion where the broken lane marker is interrupted. In this case, the previously detected lane marker position is used.

Next, the process proceeds to step S29, and the rear road area is extracted according to the individual arithmetic processing performed in the step. Also in this embodiment, since the same image coordinates as those in the first embodiment are used, the y-coordinate values y of the lane marker points, B 'points, and C' points detected in step S28 are used.
Using B ′ and yC ′, the y-coordinate values yA ′ and yD ′ of the virtual lane marker points A ′ and D ′ of the adjacent lane are calculated by the following equations:
It is calculated from Equation 8.

YA ′ = yB ′ − (yC′−yB ′) (7) yD ′ = yC ′ + (yC′−yB ′) (8) Next, as shown in FIG. , using the road shape coordinates on the image calculated in the step S26 (x _m, y _m), lane marker coordinate value through each of the point a 'to D' point (x
A ′ _m , yA ′ _m ) to (xD ′ _m , yD ′ _m ) are as follows:
１３13 are set. Note that x ₀ , y ₀ , x _n ,
y _n, respectively, x-coordinate values when m is 0 or n, it indicates a y coordinate values.

XA ′ _m = xB ′ _m = xC ′ _m = xD ′ _m = x _m (9) yA ′ _m = y _m + (yA′−y ₀ ) · (x _m −x _n ) / (X ₀ −x _n ) (10) yB ′ _m = y _m + (yB′−y ₀ ) · (x _m −x _n ) / (x ₀ −x _n ) (11) yC _{'m = y m + (yC'} -y 0) · (x m -x n) / (x 0 -x n) ......... (12) yD' m = y m + (yD'-y 0) · (X _m −x _n ) / (x ₀ −x _n ) (13) Next, the process proceeds to step S30, and the respective lanes extracted in step S29 according to the individual calculation processing performed in the step. Optical flow detection is performed for each area.
The detection of the optical flow itself is the same as that of the first embodiment.

Next, the flow shifts to step S31, where the first
Similar to the embodiment, the position and the relative speed of the other vehicle are detected according to the individual calculation processing performed in the same step. Next, the process proceeds to step S32, in the same manner as in the first embodiment, according to the individual arithmetic processing performed in the same step.
Read the status of turn signal operation by the driver.

Next, the flow shifts to step S33, where the first
In the same manner as in the embodiment, it is determined whether there is an approaching vehicle according to an individual calculation process performed in the same step. If there is an approaching vehicle, the process proceeds to step S34. Return to. In step S34, similarly to the first embodiment, the alarm device 6 according to the individual arithmetic processing performed in step S34.
Then, after performing a predetermined alarm issuing process, the process returns to the main program.

In this arithmetic processing, as described above, the approximate shape of the lane and the traveling lane position of another vehicle can be detected without performing complicated lane marker detection, so that the time required for the arithmetic processing is reduced. can do. Also,
Using the steering angle θ as the turning state detection amount, it is possible to easily and inexpensively detect the own vehicle behavior from the steering angle θ and the own vehicle speed V, and to easily estimate the road state behind the own vehicle from these. it can. In addition, since the lane marker detection area is set near the own vehicle in the rear image, the position of the lane marker is detected in the lane marker detection area, and the position of the lane marker is used to estimate the shape of the lane behind the own vehicle, It is possible to accurately detect the shape of each traveling lane behind the host vehicle, and accurately detect the traveling lane in which the other vehicle is traveling. In addition, by removing a small vehicle behavior change component due to a correction rudder or the like by a low-pass filter, it is possible to stably estimate a road state behind the own vehicle, and to reduce a traveling lane by which another vehicle is traveling. Accurate detection is possible.
In addition, since the frequency range of the change in the behavior of the own vehicle caused by the lane change is removed by the bandpass filter, the influence of the error due to the change of the lane of the own vehicle is removed, and the road state behind the own vehicle is stably estimated. Therefore, it is possible to accurately detect the traveling lane in which the other vehicle is traveling.

As described above, the vehicle rear camera 4 and FIG.
Step S21 of the calculation processing of the present invention constitutes the imaging means of the present invention, and similarly, the steering angle sensor 2, the vehicle speed sensor 3, and steps S22 to S27 of the calculation processing of FIG. Step S28 of the calculation processing in FIG. 7 constitutes a lane marker detecting means near the own vehicle, steps S29 to S31 of the calculation processing in FIG. 7 constitute another vehicle detection means, and step S23 in the calculation processing in FIG.
Constitutes a high frequency cutoff filter, and step S24 of the arithmetic processing in FIG. 7 constitutes a predetermined frequency cutoff filter.

Next, a third embodiment of the backward monitoring device of the present invention will be described with reference to FIGS. FIG. 10 showing a schematic configuration of the rear monitoring device is similar to FIG. 6 of the second embodiment, and the same components are denoted by the same reference numerals. In this embodiment, a gradient sensor 7 that detects a gradient of a road is input to the vehicle behavior estimation unit 1a.
Is added. As the gradient sensor 7, for example, JP-A-9-133699 or JP-A-7-17212
No. 8 is applicable.

This backward monitoring device 1 is also actually configured by arithmetic processing performed by an arithmetic processing device. FIG.
Is a flowchart of the calculation process. This arithmetic processing is performed in accordance with FIG. 2 of the first embodiment or FIG. 7 of the second embodiment.
Although there are a number of steps equivalent to the calculation processing of the above, all steps will be described to make it easier to understand the newly added steps.

In this calculation process, a timer interrupt process is performed at every predetermined sampling time (in this case, 100 msec.). In this flowchart, no particular communication step is provided. For example, information obtained in the flowchart is stored in the storage device as needed, and necessary information is read from the storage device as needed. In addition, the respective devices also communicate with each other, and necessary information is always read from the device that is mainly in charge of control, and the transmitted information is stored in the storage device as needed.

In step S41 of this calculation processing, as in the first and second embodiments, the full screen information of the rear of the vehicle captured by the vehicle rear camera 4 is obtained according to the individual calculation processing performed in the same step. Is read and input into an array composed of digital fields and the like. Next, step S4
Then, as in the first and second embodiments, the steering angle θ detected by the steering angle sensor 2 is read.

Then, the flow shifts to step S43, where the steering angle θ read in step S42 is smoothed according to the individual arithmetic processing performed in the same step as in the first and second embodiments. I do. Next, the process proceeds to step S44, where the lane change component is removed from the steering angle smoothed in step S43 according to the individual arithmetic processing performed in the same step as in the second embodiment.

Next, the flow shifts to step S45, where the own vehicle speed V detected by the vehicle speed sensor 3 is read in the same manner as in the first and second embodiments. Next, the process proceeds to step S46, where the road gradient α detected by the gradient sensor 7 is read. Next, in step S47, the smoothed steering angle data, the stored own vehicle speed data, and the stored road gradient data are stored in accordance with the individual arithmetic processing performed in step S47. To estimate the shape (state) of the road behind the own vehicle in three dimensions. Specifically, the X-axis and Y-axis in FIG. 4 described in the first embodiment are used.
In addition to the axis, the Z axis is set in the vertical direction, and in addition to the expressions 1 to 3 described in the first embodiment, according to the following expression 14,
Coordinate values (X _m , Y _m ,
Z _m ) is determined, and each of the values 4 to 7 described in the first embodiment is obtained.
In addition to the expression 6, the inverse perspective transformation by the following expression 15 is performed.
The road shape coordinate values (x _m , y _m , z _m ) on the image are calculated.

[0051]

(Equation 2)

Z = Z · f / L (15) Next, the processing shifts to step S48, and in accordance with the individual arithmetic processing performed in the step, the following processing is performed.
The steering angle data, own vehicle speed data, and road gradient data are stored in memory. Next, the process proceeds to step S49 to detect a white line (lane marker) near the own vehicle according to the individual calculation process performed in the same step as in the second embodiment.

[0053] and then proceeds to step S50, the position of the lane marker detected by said step S49, the words the point B ', C' to memory y coordinate value yB _'0, yC' ₀ points. Next, the process proceeds to step S51, and the rear road area is extracted according to the individual calculation processing performed in the step S51. Also in this embodiment, since the same image coordinates as those in the first embodiment are used, the latest y coordinate values y of the lane marker points, B ′ points, and C ′ points detected in step S49 are used.
Using B ′ ₀ and yC ′ ₀ , the y-coordinate values yA ′ ₀ and yD ′ ₀ of the virtual lane marker points A ′ and D ′ in the adjacent lane are obtained from the following equations (16) and (17).

YA ′ ₀ = yB ′ ₀ − (yC ′ ₀ −yB ′ ₀ ) (16) yD ′ ₀ = yC ′ ₀ + (yC ′ ₀ −yB ′ ₀ ) (17) Next 12, the lane marker coordinates passing through each of the points A ′ to D ′ using the road shape coordinate values (x _m , y _m , z _m ) on the image calculated in step S47. value _{(xA 'm, yA' m} , zA 'm) ~ (xD' m,
yD ' _m , zD' _m ) are set according to the following equations 18 to 23.

[0055] _{xA 'm = xB' m =} xC 'm = xD' m = x m ......... (18) yA 'm = y m + (yA' 0 -y 0) · (x m -x n) / (X ₀ −x _n ) (19) yB ′ _m = y _m + (yB ′ ₀ −y ₀ ) · (x _m −x _n ) / (x ₀ −x _n ) (20) ) yC ′ _m = y _m + (yC ′ ₀ −y ₀ ) · (x _m −x _n ) / (x ₀ −x _n ) (21) yD ′ _m = y _m + (yD ′ ₀ −) _{y 0) · (x m -x} n) / (x 0 -x n) ......... (22) zA 'm = zB' m = zC 'm = zD' m = z m ......... (23) next In step S52, the optical flow is detected for each lane region extracted in step S51 according to the individual calculation process performed in step S52.
The detection of the optical flow itself is the same as that of the first embodiment.

Next, the flow shifts to step S53, where the position and the relative speed of another vehicle are detected in accordance with the individual calculation processing performed in the same step as in the first and second embodiments. Next, the process proceeds to step S54, where the first and second
As in the embodiment, the state of the turn signal operation by the driver is read in accordance with the individual calculation processing performed in the same step.

Then, the flow shifts to step S55, where it is determined whether or not there is an approaching vehicle according to the individual calculation processing performed in the same step as in the first and second embodiments. If there is, the process proceeds to step S56, and if not, the process returns to the main program.
In step S56, as in the first and second embodiments, a predetermined alarm issuance process is performed on the alarm device 6 according to the individual calculation process performed in the step, and the process returns to the main program. .

In this arithmetic processing, as described above, the approximate shape of the lane and the traveling lane position of another vehicle can be detected without performing complicated lane marker detection, so that the time required for the arithmetic processing is reduced. can do. Also,
Using the steering angle θ as the turning state detection amount, it is possible to easily and inexpensively detect the own vehicle behavior from the steering angle θ and the own vehicle speed V, and to easily estimate the road state behind the own vehicle from these. it can. In addition, since the lane marker detection area is set near the own vehicle in the rear image, the position of the lane marker is detected in the lane marker detection area, and the position of the lane marker is used to estimate the shape of the lane behind the own vehicle, The shape of each traveling lane behind the host vehicle can be accurately detected, and the traveling lane in which another vehicle is traveling can be accurately detected. In addition, since the configuration of the road behind the host vehicle is estimated by using the detected position of the lane marker in the past predetermined time, the shape of each traveling lane behind the host vehicle can be detected more accurately. This makes it possible to more accurately detect the traveling lane in which the other vehicle is traveling. In addition, since the configuration of estimating the state of the road behind the host vehicle using the detected road gradient is used, the road state behind the host vehicle can be accurately detected, and the shape of each traveling lane and the other It is possible to accurately detect the traveling lane in which the vehicle is traveling. In addition, by removing a small vehicle behavior change component due to a correction rudder or the like by a low-pass filter, it is possible to stably estimate a road state behind the own vehicle, and to reduce a traveling lane by which another vehicle is traveling. Accurate detection is possible. In addition, since the frequency range of the change in the behavior of the own vehicle caused by the lane change is removed by the bandpass filter, the influence of the error due to the change of the lane of the own vehicle is removed, and the road state behind the own vehicle is stably estimated. Therefore, it is possible to accurately detect the traveling lane in which the other vehicle is traveling.

As described above, the vehicle rear camera 4 and FIG.
Step S41 of the calculation process 1 constitutes the imaging means of the present invention. Similarly, the gradient sensor 7 and step S46 of the calculation process of FIG. 11 constitute the road gradient detection means, and the steering angle sensor 2, the vehicle speed sensor 3 and Steps S42 to S48 of the calculation processing of FIG. 11 constitute the rear road state estimation means, Steps S49 and S50 of the calculation processing of FIG. 11 constitute the lane marker detection means near the own vehicle,
Steps S51 to S53 of the calculation processing in FIG. 11 constitute another vehicle detection means, step S43 of the calculation processing in FIG. 11 forms a high frequency cutoff filter, and step S44 of the calculation processing in FIG. It constitutes a cutoff filter.

In each of the above embodiments, the steering angle sensor is used as the turning state detecting means. However, instead of or in addition to this, a yaw rate sensor for detecting the yaw rate of the host vehicle, or a lateral acceleration of the host vehicle, is used. A lateral acceleration sensor for detection may be used. In addition, although a gradient sensor is used as a means for detecting a road gradient, a road gradient is detected by combining, for example, a throttle opening for estimating engine output and a longitudinal acceleration sensor for detecting acceleration / deceleration of a vehicle body. It is also possible.

In each of the above embodiments, a microcomputer is used for each arithmetic processing unit. However, various logic circuits can be used instead.

[Brief description of the drawings]

FIG. 1 is a system configuration diagram showing a first embodiment of a backward monitoring device of the present invention.

FIG. 2 is a flowchart of a calculation process performed by the backward monitoring device of FIG. 1;

FIG. 3 is an explanatory diagram of the calculation processing of FIG. 2;

FIG. 4 is an explanatory diagram of the calculation processing of FIG. 2;

FIG. 5 is an explanatory diagram of the calculation processing of FIG. 2;

FIG. 6 is a system configuration diagram showing a second embodiment of the backward monitoring device of the present invention.

FIG. 7 is a flowchart of a calculation process performed by the backward monitoring device of FIG. 6;

FIG. 8 is an explanatory diagram of the calculation processing of FIG. 7;

FIG. 9 is an explanatory diagram of the calculation processing of FIG. 7;

FIG. 10 is a system configuration diagram showing a third embodiment of the backward monitoring device of the present invention.

FIG. 11 is a flowchart of a calculation process performed by the backward monitoring device of FIG. 10;

FIG. 12 is an explanatory diagram of the calculation processing of FIG. 11;

[Explanation of symbols]

 1 is a rear monitoring device 2 is a steering angle sensor 3 is a vehicle speed sensor 4 is a vehicle rear camera 5 is a turn signal device 6 is an alarm device 7 is a gradient sensor

Claims (7)

[Claims] 1. An image capturing means for capturing an image behind a host vehicle, a rear road state estimating means for estimating a state of a road behind the host vehicle from past behavior of the host vehicle, and an image captured by the image capturing means Within, set a predetermined area according to the road state behind the host vehicle estimated by the rear road state estimation means,
A rear monitoring device, comprising: another vehicle detecting means for detecting another vehicle in the predetermined area. 2. The rear monitoring apparatus according to claim 1, wherein the road state estimating means uses a detected turning state and a vehicle speed as the past behavior of the vehicle. 3. An image captured by the image capturing means,
A predetermined lane marker detection area is set in the vicinity of the own vehicle; and a lane marker near the own vehicle that detects the position of the lane marker in the lane marker detection area is provided. The rearward monitoring device according to claim 1, wherein a state of a road behind the host vehicle is estimated using the position of the lane marker obtained. 4. The rear road state estimating means estimates a state of a road behind the own vehicle by using a record of a position of a lane marker in the past predetermined time detected by the lane marker detecting means near the own vehicle. The rearward monitoring device according to any one of claims 1 to 3, wherein 5. A rear road gradient detecting means for detecting a road gradient behind the host vehicle, wherein the rear road state estimating means comprises:
The rear monitoring apparatus according to claim 1, wherein a state of a road behind the host vehicle is estimated using a road gradient behind the host vehicle detected by the rear road gradient detecting unit. 6. The rear road according to claim 1, wherein the rear road state estimating unit includes a high frequency cutoff filter that removes a change in the behavior of the host vehicle at a predetermined frequency or higher. Monitoring device. 7. The rear monitoring apparatus according to claim 1, wherein the rear road state estimating means includes a predetermined frequency band cutoff filter that removes a change in vehicle behavior in a predetermined frequency band. .