JP3323209B2 - Mobile car environment recognition device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for recognizing a vehicle environment based on an image input by an image input means such as a camera in a moving vehicle such as an automobile or a robot.

[0002]

2. Description of the Related Art In recent years, an apparatus for recognizing the environment of a vehicle using a vehicle-mounted camera or the like has been proposed. For example, JP
Japanese Patent Application Laid-Open No. 4-26913 discloses an environment recognition apparatus that performs such obstacle recognition using an in-vehicle camera in order to avoid an obstacle in front of a vehicle.

In an environment recognizing device for an automobile or the like, recognizing a preceding vehicle or the like plays an important role.
In other words, this recognition is performed by tracking an image (specific object image) of a preceding vehicle or the like in the image capturing frame of the image input unit.

[0004]

When the specific object image is tracked in the image pickup frame of the image input means as described above, the vibration and the shaking of the vehicle body become a serious obstacle. In particular, since the rotational vibration or shaking changes the direction of the optical axis of the image input means, the specific object image easily deviates from the imaging frame, and it is difficult to track a specific vehicle or the like by continuous image processing. Becomes For example, even in a gentle curved road, the image in the imaging frame greatly changes to the left and right due to the yaw motion of the vehicle body, so that it is necessary to continuously track a specific vehicle or the like in a continuous image including the motion. There are many difficulties.

[0005] The present invention has been made in view of such circumstances, and even when a yaw motion, a pitching motion, and a roll motion occur in a vehicle body, a specific object image can be stored in an imaging frame of an image input means. It is an object of the present invention to provide an environment recognition device for a mobile vehicle, which can track and reduce the load on an image processing system.

[0006]

SUMMARY OF THE INVENTION According to the present invention, there is provided an environment recognizing apparatus for a moving vehicle, wherein an image input means is controlled in a direction opposite to a yaw direction, a pitching direction and a roll direction based on the yaw angle, pitch angle and roll angle of the vehicle. By controlling the rotation in the direction, it is possible to correct the direction of the imaging frame in response to the yaw motion, pitching motion and roll motion of the vehicle, thereby achieving the above object.

That is, according to the first aspect of the present invention, there is provided an image input means provided in a vehicle so as to be rotatable in the yaw direction of the vehicle body, and an image input means for inputting an image for recognizing the environment of the vehicle, and the image input means being provided with a yaw. a rotation driving means for rotating the direction, predetermined
Travel distance detector that detects the travel distance of the vehicle per hour
Step , detected steering angle , and travel distance detection
A yaw angle calculating means for calculating a yaw angle of the vehicle body based on a detected value of a traveling distance by the output means; and the image input means in a direction opposite to a steering operation direction based on a calculation result of the yaw angle calculating means. And control means for controlling the rotation driving means in order to drive the rotation . Then, the control means, the angle component of the same size as the vehicle body yaw angle, so that rotationally drives the image input means in the opposite direction to the steering operation direction, and controls the rotation driving means .

According to the invention of claim 2 , the yaw direction of the vehicle body
Installed in the vehicle so that it can rotate in the
Image input for inputting an image for environment recognition of the vehicle
Means, and the image input means, the yaw direction and the pitching method
Drive means for rotating the vehicle in a predetermined direction, and a vehicle per predetermined time
Traveling distance detecting means for detecting the traveling distance of the stearin
Detection value of the steering angle and the travel by the travel distance detecting means
Yaw angle that calculates the yaw angle of the vehicle body based on the distance detection value
Calculating means and a pitch for determining a pitch angle generated in the vehicle body.
Pitch angle calculating means, yaw angle calculating means and pitch angle
Based on the calculation result of the degree calculation means,
When the image input means is rotationally driven in a direction opposite to the direction
In both cases, enter the above image in the direction opposite to the pitching direction of the vehicle.
Control the above-mentioned rotation drive means to rotate the force means
And control means for performing the control .

Further, according to the third aspect of the present invention, the yaw
It is provided on the vehicle in a state where it can rotate in the direction and the roll direction,
Image input means for inputting an image for environment recognition of the vehicle
And rotate this image input means in the yaw and roll directions
Rotating drive means to be driven and the traveling distance of the vehicle per predetermined time
Travel distance detecting means for detecting separation,
The detected value and the travel distance detected by the travel distance detecting means are detected.
Yaw angle calculating means for calculating the yaw angle of the vehicle body based on the output value
And the roll angle to find the roll angle occurring on the vehicle body
Calculating means, the yaw angle calculating means and the roll angle calculating means.
Based on the calculation result of the gear, reverse to the steering operation direction
Drive the image input means in the direction of
Rotate the image input means in the direction opposite to the body roll direction
Control means for controlling the rotation drive means for driving
It is characterized in that it comprises and.

[0010]

As described above, the image input means of the present invention is controlled by the control means based on the yaw angle, the pitch angle and the roll angle of the vehicle, via the rotary drive means, and the yaw direction of the vehicle body. Since the rotation is controlled in the directions opposite to the pitching direction and the roll direction, it is possible to correct the direction of the imaging frame in response to the yaw movement, the pitching movement and the roll movement of the vehicle body.

Therefore, according to the present invention, the specific object image can be tracked in the image capturing frame of the image input means even when the yaw motion, the pitching motion, and the roll motion occur in the vehicle body. Thus, the burden on the image processing system can be reduced.

[0012]

An embodiment of the present invention will be described below with reference to the accompanying drawings.

FIG. 1 is a block diagram showing an apparatus for recognizing the environment of a mobile vehicle according to this embodiment.

The environment recognition device for a mobile vehicle includes a camera 12
(Image input means) and camera head 14 (rotation drive means)
, A steering angle detection device 16 (steering angle detection means), a travel distance detection device 18 (travel distance detection means), and a camera platform control device 20 (control means).

The camera 12 is, as shown in FIG.
2 is rotatably supported in the yaw direction by a camera platform 14 provided on the upper surface of the front part of the camera 2 so as to input an image for environment recognition in front of the vehicle. The camera head 14 also has a rotation drive function for rotating the camera 12 in the yaw direction, and is driven and controlled by the camera head control device 20 based on the steering angle and traveling distance of the vehicle. I have.

The steering angle of the vehicle is detected by the steering angle detecting device 16 based on the displacement of the steering rod or the rotational displacement of the steering shaft.
8, detection is made based on a pulse or the like generated by rotation of the transmission output shaft.

The driving control of the camera head 14 by the camera head control device 20 is performed such that the image in the imaging frame of the camera 12 is not largely moved right and left even when the vehicle body 22 makes a yaw motion. Is a control for correcting the direction of the imaging frame in response to the yaw movement in order to enable the tracking of the specific object image. That is, the yaw angle of the vehicle body is sequentially calculated based on the steering angle and the traveling distance of the vehicle (for example, every 1/30 second corresponding to the most general image processing interval) (constituting the yaw angle calculating means). ), The camera 12 is swung in the reverse direction by an angle equal to the yaw angle, thereby canceling the lateral movement amount of the image caused by the yaw movement.

The yaw angle calculation based on the steering angle and the traveling distance is performed as follows. That is,
Since the steering angle can be considered to be constant within a short time such as 1/30 second, as shown in FIG. 3, if the vehicle travels for a traveling distance L at a constant steering angle ω,
The turning radius R at this time is R = h in a two-wheel model.
/ Ω. Here, h is the wheelbase. If the yaw angle at this time is θ, it is expressed by θ = L / R. From these two equations, the yaw angle θ is θ = L · ω /
h. However, in calculating the yaw angle, it is assumed that the traction of the tire is appropriately performed.

Next, a control example of the camera head control device 20 will be described.

When there is a preceding vehicle V in a straight running state as shown in FIG. 2, an image of the preceding vehicle V is included in the image pickup frame F of the camera 12. When trying to pass the preceding vehicle V from this state, it is important to always grasp the behavior of the preceding vehicle V. Thus, even if the preceding vehicle V itself moves left and right, the change in position as an image is small, and the preceding vehicle V does not suddenly fall out of the imaging frame F. However, it is necessary to change the lane for passing. When the vehicle is steered to the right, the vehicle body 22 yaw-moves. If no control is performed on the vehicle body 22, the imaging frame F of the camera 12 is swung to the right, and the image in the imaging frame F Moves greatly to the left. Therefore, the preceding vehicle V suddenly comes off the imaging frame F.

However, in the present embodiment, the camera pan controller 20 sequentially calculates the yaw angle θ of the vehicle body 22 based on the steering angle ω and the traveling distance L of the vehicle, and calculates an angle equal to the yaw angle θ. The camera 12 is moved to the left by the amount corresponding to the direction correction control of the imaging frame F, thereby canceling the amount of lateral movement of the image caused by the yaw movement. Even if the yaw motion is performed by the camera 22, the image of the preceding vehicle V can be tracked in the imaging frame F of the camera 12 without largely moving the image in the imaging frame F left and right.

As described in detail above, according to this embodiment,
The camera 12 is controlled by the camera head control device 20 via the camera head control device 20 based on the steering angle and the travel distance of the vehicle detected by the steering angle detection device 16 and the travel distance detection device 18. 22
The rotation is controlled in the direction opposite to the yaw direction (i.e., the steering operation direction), so that the direction of the imaging frame F can be corrected in response to the yaw movement of the vehicle body 22.

Therefore, according to this embodiment, the vehicle body 22
The specific object image of the preceding vehicle V or the like can be tracked in the imaging frame F of the camera 12 even when the yaw motion occurs. Thus, the burden on the image processing system can be reduced.

In this embodiment, for the sake of simplicity, only the correction of the direction of the imaging frame F in response to the yaw motion has been described as the control content of the camera pan controller 20. In addition, correction of the tracking direction of the imaging frame F accompanying the lateral movement of the preceding vehicle V itself, correction of the swinging direction of the imaging frame F in the turning direction in the steady turning state, and the like are performed.

If the following camera mount mechanism is provided in addition to the configuration of the above-described embodiment, the imaging frame F can be used for pitching and rolling of the vehicle body 22.
Can be corrected.

That is, as shown in FIG. 4, the camera mount mechanism supports the camera 12 so as to be rotatable in a pitching direction and a rolling direction, and a potentiometer for detecting a suspension stroke value of each of the front and rear wheels. And the like, and a pitch angle and a roll angle of the vehicle body 22 are calculated based on the detection signals from the stroke sensors 34, and the camera mount is mounted so as to eliminate the relative rotational movement between the camera 12 and the road surface. And a control device 36 composed of a microcomputer or an electronic circuit for controlling the operation of the control device 32.

The calculation up to calculating the posture correction value of the camera mount 32 from the suspension stroke detection value in the control device 36 is performed as follows. That is, as shown in FIG. 5, when the stroke values of the suspensions are Z11, Z12, Z21, and Z22, Zp = (Z10-
Z20) / Ly. However, Z10 = (Z11
+ Z12) / 2, and Z20 = (Z21 + Z22) / 2.

The roll angle Zr of the vehicle body with respect to the road surface
Can be approximated as Zr = (Z01-Z02) / Lx. However, Z01 = (Z11 + Z21) / 2, Z02 = (Z12 + Z22)
/ 2.

If the camera mount 32 is driven and controlled so that the angle between the vehicle body 22 and the optical axis of the camera 12 is the pitch angle -Zp and the roll angle -Zr, the camera 12 and the road surface can be connected regardless of the movement of the vehicle body 22. , The relative pitch angle and the relative roll angle can be kept constant.

In addition to the camera mount mechanism using an electric system, a camera mount mechanism using a hydraulic system as shown in FIG. 6 can be used. That is, while the linear operation type or rotary operation type cylinder 42 that operates according to the suspension stroke value is mounted on four wheels,
An operating hydraulic cylinder 44 that supports the camera 12 at four points is provided, and a camera mount mechanism that eliminates the relative rotational movement between the camera 12 and the road surface by connecting both cylinders by a hydraulic circuit is possible. It is.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a mobile vehicle environment recognition apparatus according to the present invention.

FIG. 2 is a view showing a vehicle front view for explaining a specific example of the operation of the embodiment.

FIG. 3 is a plan view for explaining a yaw angle calculation method by the camera pan head control device of the embodiment.

FIG. 4 is a block diagram showing a camera mount mechanism that can be provided in addition to the configuration of the embodiment.

FIG. 5 is a perspective view for explaining a method of calculating a posture correction value in the camera mount mechanism.

FIG. 6 is a perspective view showing another example of the camera mount mechanism.

[Explanation of symbols]

Reference Signs List 12 camera (image input means) 14 camera head (rotation driving means) 16 steering angle detection device (steering angle detection means) 18 travel distance detection device (travel distance detection means) 20 camera head control device (control means) 22 Body F Imaging frame V Leading vehicle

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) G05D 1/02

Claims (3)

(57) [Claims] 1. An image input means provided in a vehicle so as to be rotatable in a yaw direction of a vehicle body, and an image input means for inputting an image for recognizing an environment of the vehicle, and a rotation drive means for rotating the image input means in a yaw direction. And a travel distance detection for detecting the travel distance of the vehicle per predetermined time.
Output means, the detected value of the steering angle, and the travel distance detecting means.
A yaw angle calculating means for calculating a yaw angle of the vehicle body based on the detected value of the traveling distance, and rotating the image input means in a direction opposite to the steering operation direction based on a calculation result of the yaw angle calculating means. to be driven, and control means for controlling the rotation driving means
With it, the said control means, the angle component of the same size as the vehicle body yaw angle,
The image input means in the direction opposite to the steering operation direction
An environment recognizing device for a mobile vehicle, wherein the rotational driving means is controlled to rotationally drive the vehicle . 2. The vehicle according to claim 1, wherein the vehicle rotates in a yaw direction and a pitching direction.
It is installed on the vehicle in a state where it can be rolled, and the environment of the vehicle is recognized.
Input means for inputting an image for rotation, and rotating the image input means in the yaw direction and the pitching direction
Rotating drive means for driving the vehicle, and a travel distance detection for detecting a travel distance of the vehicle per a predetermined time.
Output means, the detected value of the steering angle, and the travel distance detecting means.
Calculates the yaw angle of the vehicle based on the detected travel distance
Angle calculating means to calculate the pitch angle occurring in the vehicle body
Means and calculation results of the yaw angle calculating means and the pitch angle calculating means
Based on the above, in the direction opposite to the steering operation direction
Rotating the image input means and pitching the vehicle
Drive the image input means in the direction opposite to the
Control means for controlling the rotation drive means.
An environment recognition device for a mobile vehicle. 3. It is rotatable in the yaw and roll directions of the vehicle body.
The vehicle performance condition set vignetting, for environmental recognition of the vehicle
Image input means for inputting an image, and rotating the image input means in the yaw direction and the roll direction.
Rotation driving means, and a travel distance detection means for detecting a travel distance of the vehicle per predetermined time.
Output means, the detected value of the steering angle, and the travel distance detecting means.
Calculates the yaw angle of the vehicle based on the detected travel distance
Angle calculation means to calculate the roll angle generated on the vehicle body
Means, calculation results of the yaw angle calculation means and the roll angle calculation means
Based on the above, in the direction opposite to the steering operation direction
Rotate the image input means and roll the vehicle
The image input means should be rotationally driven in a direction opposite to the
And a control means for controlling the rotation drive means .