JP2737902B2 - Driving route determination processing method for image type unmanned vehicles - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for determining a traveling route of an unmanned vehicle, and more particularly, to one of marks arranged discretely on a road surface at least a plurality of times by an imaging device. The traveling of an image type unmanned vehicle in which an image is taken, a traveling route of the unmanned vehicle is determined based on a mark in the image photographed at each time, and the traveling route is sequentially updated and travels along the updated traveling route. It relates to a route determination processing method.

[Related Art] In recent years, a traveling line drawn continuously on a road surface is picked up by an image pickup device mounted on an unmanned vehicle, and an image (visual) guidance for causing the taken image to be processed and running along the running line. Unmanned vehicles of the type have been proposed. Also, in this image guidance system, a mark with a bar code indicating operation information such as vehicle speed, stop, branch, etc. on the surface is discretely arranged on the road surface,
An image type unmanned vehicle has been proposed in which the mark is taken by an image pickup device, the bar code on the mark is image-processed, the operation information is read, and the vehicle travels based on the operation information.

Further, a light reflecting member (corner cube) that reflects light is employed as a mark discretely arranged on the road surface,
The light reflecting member is illuminated with light using a light provided on an unmanned vehicle,
There has been proposed a so-called spot mark type unmanned vehicle that determines a traveling route by imaging the reflected light reflected by the light reflecting member with an imaging device. In this spot mark type unmanned vehicle, since the light reflecting member is regarded as a point on the screen imaged by the imaging device, the light reflecting member is imaged as a point when the light reflecting member is located at a far position, but becomes large when the light reflecting member is imaged at a near position. The image can no longer be considered as a point. Therefore, a device has been devised so that the image of the reflected light is captured in the same size regardless of the distance of the light reflecting member using the special filter. However, since the image is not actually picked up as a point, the brightest point in the reflected light portion in the image is extracted and that point is determined as the center position of the light reflecting member.

[Problems to be Solved by the Invention] However, in these image-based unmanned vehicle image processing apparatuses, when determining a mark on a screen, the mark is determined by using pixel data of the entire screen, that is, pixel data of all pixels captured by the imaging apparatus. The recognition of. Therefore, when there is an object having the same color as the color of the mark on the road surface, or when there is a light reflecting object in the latter, there is a possibility that the object is erroneously recognized as a mark or a light reflecting member.

SUMMARY OF THE INVENTION An object of the present invention is to solve the above-described problems, to determine an accurate and accurate traveling route without a risk of picking up noise other than marks, and to apply to detection of abnormal traveling of unmanned vehicles during traveling. It is an object of the present invention to provide an image-based unmanned vehicle traveling route determination processing method.

SUMMARY OF THE INVENTION [Means for Solving the Problems] In order to achieve the above object, the present invention provides an unmanned vehicle with one of marks indicating the passing positions of unmanned vehicles discretely arranged on a road surface while traveling. Imaging at least a plurality of times with the imaging device, recognizing the position of the mark based on the mark in the image captured at each time, determining the traveling route of the unmanned vehicle and sequentially updating the updated traveling route, In a method of determining a traveling route of an image type unmanned vehicle that is caused to travel along a first coordinate system, a first coordinate system having an origin at a previously captured old point is set, and a mark of the mark is set in the first coordinate system. In addition to obtaining the position, a new imaging point at which the next imaging operation is to be performed is obtained on the travel route obtained based on the previously imaged mark, and then a second coordinate system having the new imaging point as the origin is set. And the mark of the first coordinate system Is obtained by performing coordinate conversion to the position in the second coordinate system, and from the converted position, the center position in the image of the mark captured at the new imaging point or the area where the mark is completely included in the image The gist of the present invention is a traveling route determination processing method of an image type unmanned vehicle that predicts by calculation and recognizes the position of a mark only in a portion corresponding to the center position or in the area and obtains a new traveling route. .

[Operation] The new imaging point at which the imaging operation is performed next is obtained by a function of the traveling route obtained based on the previously captured mark.

Then, the center position of the mark in the first coordinate system whose origin is the previously captured imaging point is converted to a second coordinate system whose original point is the new imaging point. Subsequently, the converted position is converted to a position in the image. Thus, the center position of the mark in the image when the image is captured at the new imaging point or the area that completely includes the mark when the image is captured at the new imaging point is predicted. Therefore, the mark can be recognized only in the portion corresponding to the center position or in the area, and a new traveling route can be obtained based on the mark.

[Embodiment] An embodiment of a traveling route determination processing method according to the present invention will be described below with reference to the drawings.

FIG. 2 shows a side view of the image type unmanned vehicle 1 and the unmanned vehicle 1.
A CCD (charge coupled device) as an imaging device is provided at the upper center position of the support frame 2 erected at the upper center position on the front side of the camera.
ce) A camera 3 is provided. The CCD camera 3 is set on the support frame 2 so as to capture an area 4a on the road surface 4 in front of the unmanned vehicle 1 set in advance. CCD camera 3
Shows an area 4a shown in FIG. 3 as an image 5 shown in FIG. In the present embodiment, the image 5 is constituted by 256 × 256 pixel data.

As shown in FIGS. 2 and 3, marks 6 for determining the traveling route L of the unmanned vehicle 1 are provided on the road surface 4 at predetermined intervals, and in this embodiment, are shown in FIG. As described above, the shape is circular, and a pair of pointed corners 6a are extended on both sides facing each other. The line 1 connecting the pair of corners 6a indicates the traveling direction when the unmanned vehicle 1 passes through the mark 6, and in this embodiment, indicates the direction in which the next mark 6 is located. The color of the mark 6 is configured to be white different from the color of the road surface 4. Then, the CCD camera 3 sequentially captures images while traveling on the fixed discretely arranged marks 6.

In this embodiment, the CCD camera 3 employs a camera in which the level of the pixel signal of the white mark 6 is high and the level of the pixel signal of the dark road surface 4 is low.

Next, the electrical configuration of the travel control device mounted on the unmanned vehicle 1 will be described with reference to FIG.

The microcomputer 10 is a central processing unit (hereinafter simply referred to as
CPU 11) and a program memory 12 consisting of a read-only memory (ROM) storing a control program and a CPU 11
A working memory 13 composed of a readable and rewritable memory (RAM) for temporarily storing the calculation processing results and the like, a timer 14, and the like. Then, the CPU 11 follows the control program stored in the program memory 12
The CCD camera 3 is operated to determine a traveling route L on which the unmanned vehicle 1 travels based on the image information captured and execute various arithmetic processing operations for steering control.

The CPU 11 controls the scanning of the CCD camera 3 via the input / output interface 15 and the A / D converter 16 at regular intervals based on the time measured by the timer 14, and controls the CCD.
The pixel signal from the camera 3 is converted into pixel data via the A / D converter 16 and the bus controller 17 and stored in the working memory 13. When converting the pixel signal from the CCD camera 3 from an analog value to a digital value, the A / D converter 16 determines whether or not each pixel signal is equal to or greater than a predetermined set value. Each pixel is sequentially input in such a manner that “1” is set as the pixel of the white mark 6 and “0” is set as the pixel of the dark road surface 4 when the pixel signal is less than the pixel signal. The signal is binarized and stored in the working memory 13 via the bus controller 17 as pixel data.

Therefore, the work memory 13 stores the image 5 taken by the CCD camera 3 as 256 × 256 pixel data.

In the present embodiment, the scanning control of the CCD camera 3 is performed in the horizontal direction for the sake of convenience of description, and the scanning method in which the scanning shifts from the top to the bottom of the image 5 is adopted. Of course it is good.

The binarization level controller 18 outputs to the A / D converter 16 data of a set value for the A / D converter 16 to binarize based on a control signal from the CPU 11.

The drive controller 19 controls a traveling motor 20 and a steering mechanism 21 (not shown) based on a control signal from the CPU 11. The rotation speed of the traveling motor 20 is controlled based on the control signal, and the steering angle Θs of the steering mechanism 21 is controlled based on the control signal. In this embodiment, the unmanned vehicle 1 runs at a constant speed V except for starting and stopping, and the CPU 11 controls the rotation speed of the traveling motor 20 to rotate at a constant speed. I have.

In this embodiment, the CCD camera 3 is controlled so that the imaging operation is performed at intervals of a predetermined time, and the position of one mark 6 is changed three times as shown in FIG. Image. Accordingly, as the unmanned vehicle 1 approaches the mark 6, the mark 6 in the image 5 picked up by the CCD camera 3 gradually approaches the front as shown in FIG. Further, in the present embodiment, when the mark 6 is imaged three times, the next mark is deviated from the area 4a when the next image is taken, and the next new mark 6 enters the area 4a.

 Next, the processing operation of the CPU 11 will be described.

First, the basic operation of the CPU 11 will be described.
, A recognition processing operation for performing image processing to recognize the mark 6 provided on the road surface 4 based on the image captured by the camera 3, and a traveling route of the unmanned vehicle based on the recognition result. There are a calculation processing operation for calculating and a traveling control operation for causing the unmanned vehicle 1 to travel by controlling the steering mechanism 21 based on the calculation result. The CPU 11 performs the operations in the order of the imaging processing operation, the recognition processing operation, the arithmetic processing operation, and the traveling control operation, and sets operation times T1 to T4 of each operation in advance. These operations are sequentially repeated to drive the unmanned vehicle 1. It should be noted that a period from when the imaging processing operation to the arithmetic processing operation is completed to when the traveling control operation is started (T1 + T2 + T
Regarding 3), the CPU 11 controls the traveling of the unmanned vehicle 1 by a traveling control operation based on the traveling route obtained in the previous arithmetic processing operation.

More specifically, as shown in FIG. 8, when the traveling unmanned vehicle 1 reaches the point P0 based on the traveling route L0 obtained by the previous calculation, the CPU 11 causes the CCD camera 3 to perform an imaging operation to perform an imaging process. The operation is started, the recognition processing operation and the calculation processing operation are performed, and the next new traveling route L1 is determined. Time required for it Ta
During (T1 + T2 + T3), the unmanned vehicle 1 travels from the point P0 to the point P1 based on the traveling route L0. Then, the CPU 11 starts a travel control operation based on the travel route L1 from this point P1, and controls the travel of the unmanned vehicle 1 so as to travel along the travel route L1.

When the unmanned vehicle 1 has traveled to the point P2 after a predetermined time T4 has elapsed since the start of the travel control operation based on the travel route L1, the CPU 11
Starts the next imaging processing operation. The CPU 11 determines the travel route L2 based on the image information captured at the point P2 by the time after the Ta time has elapsed (when the unmanned vehicle 1 travels along the travel route L1 to the point P3), and from the point P3, The traveling control of the unmanned vehicle 1 is performed so that the unmanned vehicle 1 travels according to the optimal traveling route L2. Thereafter, by repeating the same operation, the CPU 11 controls the traveling of the unmanned vehicle 1 based on the traveling paths L0, L1, L2,... Calculated at each time. Therefore, the CPU 11 is traveling while updating a new traveling route in the cycle T (= T1 + T2 + T3 + T4).

 Next, details of each processing operation will be described.

In FIG. 9, the unmanned vehicle 1 is the travel route L0 calculated earlier.
When the position of the CCD camera 3 is controlled to scan based on the control signal from the CPU 11 when the vehicle reaches the point P0 in a state where the traveling is controlled according to the above, the CCD camera 3 is not perpendicular to the road surface 4 but tilts at a certain angle. Since the image has been captured, the area 4a ahead is captured as an image 5 as shown in FIG. The image 5 captured by the CCD camera 3 is output to the A / D converter 16 as a pixel signal, and the A / D converter 16 converts the pixel signal of the mark 6 and the pixel signal of the road surface 4 into “ 1 ",
It is determined to be binarized to “0” and stored as pixel data in a predetermined storage area of the working memory 13.

For convenience of description, the mark 6 imaged at the point P0 is the mark 6 imaged for the first time, and is assumed to be imaged from the farthest position.

The CPU 11 performs image recognition of the mark 6 based on all pixel data stored in the working memory 13. The CPU 11 obtains the actual position G on the road surface 4 where the center position g of the portion determined to be the shape of the mark 6 in the image 5 is stored in the working memory 13 by a known method.

This calculation includes four points a, including the pointed points a and c of the pair of corners 6a of the portion determined as the mark 6 from the image 5, as shown in FIG.
Find b, c, d. These four points are represented by a vertical pixel row at the 128th pixel counted from the left in each pixel constituting the image 5.
The 128-th horizontal pixel row counted from the top is represented by x, y coordinates defining the y-axis.
In this case, since the mark 6 is the first imaged mark 6, the mark 6 is recognized by using all pixel data of the image 5 to determine the mark 6 in the image by a known method.
The four points a, b, c, d of the mark 6 are determined.

Next, the CPU 11 performs projection transformation of these four points a to d. In the projective transformation, the positions (hereinafter referred to as base points) A to D on the actual area 4a shown in FIG. As shown in FIG. 9, the center position of the unmanned vehicle 1 (accurately, the position of the CCD camera 3) is set as the origin PH, and the traveling direction of the unmanned vehicle 1 is X.
X and Y when the direction orthogonal to the X axis is the Y axis
An arithmetic process is performed to determine the coordinate positions of the base points A to D in the coordinate system.

a (xa, ya) → A (Xa, Ya) b (xb, yb) → B (Xb, Yb) c (xc, yc) → C (Xc, Yc) d (xd, yd) → D (Xd, Yd) As described above, since the CCD camera 3 does not vertically image the road surface 4, the mark 6 in the image 5 and the mark 6 in the actual area 4a are matched.

Note that this projection conversion processing operation is performed by a preset CCD.
Projection transformation, that is, coordinate transformation, is performed based on the focal length, magnification, and other specifications of the camera 3 and installation conditions, such as tilt and height. The general formula of this projective transformation is as follows.

The position coordinates of each point are x, y, the position coordinates of the base point are X, Y, the height of the camera 3 is H, the inclination of the camera 3 is Θ, the focal length of the camera is F, and the magnification constant is k.

Here, p is the distance between the center of the vehicle and the CCD camera installation point.

After performing the projective transformation, the CPU 11 uses the coordinates (X) of the intersection of the line connecting the base points A and C and the line connecting the base points B and D from the coordinates of the base points A to D.
g, Yg) as the center position G of the mark 6 and
From the coordinates (Xa, Ya) and (Xc, Yc) of the base points A and C, the inclination of the mark 6 (the inclination of the line 1 connecting the pair of corners 6a and indicating the direction in which the unmanned vehicle 1 moves) Φm is obtained. In the present embodiment, the center G is determined from the four points a to d in the image.
The middle point of the line 1 connecting the two points ac is set as the center position g of the mark 6 in the image, and the center position g is projected and transformed to the center position G.
It may be. Alternatively, a center of gravity may be obtained from the mark 6 in the image, and the center of gravity may be set as the mark center, and the center position G may be obtained by projective transformation.

Next, the inclination (posture angle) Φp1 of the unmanned vehicle 1 at the point P1 at which the traveling control is started by the traveling route L1 determined based on the image information captured at the point P0, and the coordinates (Xp1, Yp1). This calculation can be easily obtained by using the cubic function f (X) of the traveling route L0 determined in the previous calculation.

Both positions G (Xg, Yg), P1 (Xp1, Yp1) and their positions G, P1
The CPU 11 obtains a traveling route L1 that passes through both positions and is expressed by the following cubic function f (X) based on the inclinations Φm and Φp1 at

f (X) = αX ³ + βX ² + γX + δ The coefficients α, β, γ, and δ can be easily obtained by solving the following simultaneous quartic equations.

^{Yg = αXg 3 + βXg 2 +} γXg + δ Yp1 = αXp1 3 + βXp1 2 + γXp1 + δ tanΦm = 3αXg 2 + 2βXg + γ tanΦp1 = 3αXp1 2 + 2βXp1 + γ CPU11 unattended passing the mark center position G of the cubic function f (X) from the point P1 in the attitude angle Φm Travel route L1 of car 1
To be determined.

Then, when the unmanned vehicle 1 arrives at the point P1 according to the preceding traveling route L0, the CPU 11 shifts to a traveling control operation and this function f
The steering mechanism 21 is controlled based on (X). This control is the eleventh
As shown in the drawing, this is a control operation for running the unmanned vehicle 1 along the curve of the function f (X) from the point P1, and the attitude angle Φ (X) at each running position is obtained.
Determines the steering angle Θs so as to have the attitude angle Φ (X) at each time, and controls the operation of the steering mechanism 21.

The inverse tangent of the differential value of the cubic function f (X) is the attitude angle Φ (X) (= tan ^-1 {f '(X)}), and a point (Xn, f (Xn)) to the next point (Xn + 1, f (X
n + 1)), the attitude angle Φ (X) is Φ (X
It can be seen that it suffices to satisfy the condition from (n) to Φ (Xn + 1).

In this embodiment, a traveling control method for satisfying this condition is embodied as a steady turning circular traveling.

That is, the steady turning circle running is a running turning with a constant radius R when the steering angle Θs is kept constant as shown in FIG. 12, and the change amount of the attitude angle Φ (X) after ΔT seconds is Δ
Assuming that Φ, the following equation is established.

 ΔΦ = VΘs ・ ΔT / WR R = W / Θs V is the running speed, and W is the wheel base.

Then, in order to change the attitude angle by ΔΦ while moving forward by V · ΔT from both equations, the radius R (= V · Δ
T / ΔΦ) and calculate the steering angle Θs from the radius R.
(= W / R = W · ΔΦ / V · ΔT may be calculated.

Therefore, the CPU 11 calculates the steering angle Θs every ΔT seconds based on the above equation and controls the operation of the steering mechanism 21 to allow the unmanned vehicle 1 to travel along the travel path L1.

When the unmanned vehicle 1 travels according to the traveling route L1 and the traveling control operation based on the traveling route L1 is started and a time T4 has elapsed (the point at that time is P2), the CPU 11 executes the processing operation of the next imaging device, and Similarly, the CCD camera 3 is operated to image the mark 6 in the area 4a at that time. The processing operation for obtaining the cubic function f (X) of the next new travel route L2 is performed at the point P3.
Do it until you reach.

In this case, unlike the above, when all the pixel data of the image 5 captured by the CCD camera 3 is stored in the working memory 13,
As shown in FIG. 14, the CPU 11 executes the setting operation of the center position g of the mark 6 in the image 5 and the processing area e completely including the mark 6.

In this setting calculation, the point P2 is set as the origin from the X, Y coordinate system shown in FIG. 11, the traveling direction of the unmanned vehicle 1 at the point P2 is defined as the Xm axis, and the direction orthogonal to the Xm axis is defined as Ym. The coordinates are converted into an Xm, Ym coordinate system as an axis, that is, the coordinates (Xg, Yg) of the mark center position G in the X, Y coordinates are converted into coordinates (Xmg, Ymg) in the Xm, Ym coordinate system. Therefore, in this case, the old imaging point is the point P0, the new imaging point is the point P2, the first coordinate system is the X, Y coordinate system, and the second coordinate system is the Xm, Ym coordinate system.

This transformation is performed using an affine transformation, and the coordinate position of the mark center position Gm in the Xm, Ym coordinate system (X
mg, Ymg).

Xmg = (Xg−Xp2) cos (−Δm) − (Yg−Yp2) sin (−Δm) Ymg = (Xg−Xp2) sin (−Δm) + (Yg−Yp2) cos (−Δm) where Xp2, Yp2 is the coordinates of the point P2 in the X, Y coordinate system, and can be obtained from the cubic function of the traveling route L1.

Θm is Ym of the Xm, Ym coordinate system with respect to the Y axis of the X, Y coordinate system.
The rotation angle of the shaft, that is, the change amount of the attitude angle of the unmanned vehicle 1 at the point P2 with respect to the attitude angle of the unmanned vehicle 1 at the point P0, which can be obtained from the cubic function of the traveling route L1.

When the coordinate position (Xmg, Ymg) of the center position Gm of the mark 6 on the road surface 4 in the Xm, Ym coordinate system is determined, the CPU 11 next sets an image when the coordinates (Xmg, Ymg) are captured at the point P2. 5, that is, where the mark center position g is located in the image 5.

Gm (Xmg, Ymg) → g (xg, yg) This can be obtained by performing a conversion process (reverse projection conversion process) reverse to the above-mentioned projection conversion process operation.
The general formula of this inverse projection conversion formula is as follows.

When the position g (xg, yg) obtained by the inverse projection transformation is obtained, C
The PU 11 determines that the mark 6 in the image 5 is imaged around the position g (xg, yg), and then performs a setting calculation of the processing area e.

The setting calculation of the processing area e is performed by the coordinate position (Xmg, Ymg) of the mark center position Gm on the road surface 4 in the calculated Xm, Ym coordinate system.
It is calculated based on

This means that the size of the mark 6 on the road surface 4 is stored in advance as data, and a rectangular frame-shaped region Z that completely includes the mark 6 on the road surface 4 is set in advance. The data of the relative positional relationship between the center position of the mark 6 and the mark center position of the point Za at one corner of the area Z, which is always invariable relative to the center position of the mark 6, is stored in advance in the program memory.
The CPU 11 calculates the coordinates (Xmza, Ymza) of the point Za on the road surface 4 in the Xm, Ym coordinate system based on the data.

Similarly, the coordinates (Xmzd, Ymzb) (Xmzc, Ymzc) (Xmzd, Ymzd) of the points Zb, Zc, Zd at the other three corners of the area Z are obtained.

Next, the CPU 11 coordinates (x za, y za) (x zb, y zb) (x zc, y) of each point za to zd in the image 5 corresponding to each of the points Za to Zd.
zc) (xzd, yzd) is obtained by inverse projection transformation in the same manner as described above. Then, a range connecting these points becomes a processing area e, and the area e becomes a horizontally long rectangular area in which the mark 6 to be imaged is surely included. The area e becomes longer horizontally because the CCD camera 3 obliquely captures an image of the road surface 4 to cause distortion in the x-axis direction.

When the processing area e in the image 5 is set, the CPU 11 reads out only the pixel data group in the processing area e from all the pixel data groups stored in the working memory 13 and reads the mark 6 that should be in the same area e. Execute image recognition. CPU 11 is in area e
Similarly, the shape of the mark 4 is determined by a known method in the same manner as described above, and the center position g and the positions of four points a to d including the peak points a and c of the corner 6a of the mark 6 in the image 5 are determined. Ask for.

Accordingly, the processing time is greatly reduced as compared with the conventional case where the CCD camera 3 performs image recognition of the mark 6 using all pixel data groups stored in the working memory 13 by taking an image.

When the center position g and the four points a to d are determined, the CPU 11 projects and changes in the same manner to determine the inclinations Φm of the unmanned vehicle 1 at the points G, A to D and the point G. Further, the CPU 11 determines the position of the point P3 and the inclination Φ at that time based on the cubic function f (X) of the traveling route L1.
Find p3. Then, based on the obtained values, a travel route L2 composed of the following new cubic function is determined in the same manner as when the travel route L1 was determined.

At this time, the CPU 11 calculates the function f (X) of the traveling route L2 and until the time when the traveling control operation based on the new traveling route L2 is started, that is, up to the point P3 of the traveling route L1, the preceding traveling route L1. Of the unmanned vehicle 1 based on the function f (X).

Then, when the unmanned vehicle 1 reaches the point P2, the CPU 11 controls the traveling of the unmanned vehicle 1 along the traveling route L2 including the obtained new cubic function f (X).

Then, when the imaging operation of the CCD camera 3 is performed to find the next new traveling route while traveling along the traveling route L2, the CPU 11 performs the same processing operation as the method for finding the traveling route L2. Accordingly, the CPU 11 first performs a setting calculation of the processing area e, and executes image recognition in the newly set area e. In this case, when the area e is set, the area Z including the actual mark 6 on the road surface 4 is first set, and the area Z including the area Z is determined by projective transformation. Therefore, as the mark 6 approaches the unmanned vehicle 1, Even if a large image is taken, the processing area e is also enlarged so as to completely cover the mark 6,
The mark 6 can be reliably recognized. And the second
As shown in FIG. 15, since the processing area e is expanded in the range of the processing area e in which the traveling route L2 is set, the pixel data group for performing the image recognition increases and the processing time increases by that much. This is much shorter than when all the pixel data of the image 5 captured by the CCD camera 3 and stored in the work memory 13 is used as in the related art.

In addition, when the traveling route is determined based on the mark 6 shown in FIG. 15 and traveling on the same route to perform the next imaging, in the case of this embodiment,
The mark 6 deviates from the imaging area 4a of the CCD camera 3, and the next new mark 6 enters the area 4a. Since it is difficult for the CPU 11 to predict where the new mark 6 is to be imaged, the CPU 11 uses all pixel data only when the new mark 6 is first imaged as described above.
Execute image recognition.

Thereafter, the CPU 11 repeats the operation in the same manner to sequentially update the traveling route and controls the traveling according to the mark 6 on which the unmanned vehicle 1 is imaged.

As described above, in the present embodiment, when obtaining each of the traveling routes L1, L2,... Represented by the cubic function f (X), the points P0, P2 and the inclination of the points P0, P2 (the unmanned vehicle 1) Without using the attitude angle), the CCD camera 3 takes the time difference from when the image of the mark 6 is picked up to when a new travel route is determined and the travel is controlled in accordance with the determined travel route. Since the points P1 and P3 and the gradients Φp1 and Φp3 of the new traveling route determined based on the function of the route are used, the driverless vehicle 1 must pass over the mark 6 without fail.
It is possible to reliably advance in the traveling direction indicated by the pair of corners 6a on the mark 6, and the speed of the unmanned vehicle 1 can be increased.

In the present embodiment, the mark 6 can be recognized by using only the pixel data group of the processing area e that completely includes the mark 6 predicted in advance in the image captured by the CCD camera 3.
Image processing time can be shortened, and as a result, unmanned vehicles 1
Can be speeded up. In addition, since the processing area e is set and the mark 6 is recognized, even if noise that does not hinder traveling is included in the pixel data group, the noise is not included in the processing area e. Since it is not affected, highly accurate image recognition can be performed, and a highly accurate traveling route can be determined.

Further, in the present embodiment, when the mark 6 is not recognized in the recognition of the mark 6 within the set processing area e, if the mark 6 is not recognized, the unmanned vehicle 1 is traveling at an abnormal position where the mark 6 does not exist. If it is determined that the unmanned vehicle 1 is in an abnormal state and the unmanned vehicle 1 is stopped, for example, the unmanned vehicle 1 can be operated more reliably.

In addition, in this embodiment, the processing area e is set, and the image processing is executed only in the area e. However, in the spot mark type unmanned vehicle described above, the processing is performed without setting the area e. Similar effects can be expected when only the central portion of the light reflecting member corresponding to the position g is predicted and only the predicted portion is image-processed.

In the above-described embodiment, the image processing time is shortened by using the predicted center position g of the mark 6 in the image 5 to set the processing area e. However, the present invention is not limited to this. A deviation between the mark center position g and the center position of the mark 6 actually recognized from the image 5 is obtained, and the unmanned vehicle is driven safely by using the deviation difference to determine whether the vehicle is traveling normally. You may use it for the system which makes it.

Further, in the above-described embodiment, one mark 6 is imaged by the CCD camera 3 three times. However, the present invention is not limited to the number of times of imaging, and the number of times may be increased or decreased, or the imaging timing may be changed from time to time. Or may be implemented.

Further, in this embodiment, when one mark 6 is imaged three times, the next new mark 6 is set to enter the area 4a at the time of the next imaging. However, the present invention is not limited to this. After the state in which the mark 6 is not imaged continues, a new mark 6 may be imaged.

Further, in the above-described embodiment, the setting of the processing area e is performed immediately before performing the image recognition. However, after one travel route is determined and before the travel is started based on the travel route, the next imaging position is set. And a processing area may be set in advance.

In the above embodiment, the travel route is determined by a cubic function. However, the present invention is not limited to this, and the travel route may be determined by using another function.

Further, in the above embodiment, the CCD camera 3 is used as the imaging device. However, the present invention may be implemented by using another imaging device. In the above embodiment, the pixel configuration (resolution) of the image in the CCD camera 3 is set to 256. × 256 pixels, but is not limited to this, for example, 512 × 512 pixels, 1024 × 10
It may be implemented by appropriately changing the number of pixels, such as 24 pixels.

Further, in the above-described embodiment, the traveling route of the unmanned vehicle 1 is determined and the traveling control is performed in the order of the imaging processing operation, the recognition processing operation, the arithmetic processing operation, and the traveling control operation. In other words, any operation sequence may be used as long as the travel route is determined in consideration of the time difference from when the mark is imaged to when the travel control based on the travel route is started. .

Further, in the above embodiment, the traveling speed V of the unmanned vehicle 1 is fixed, but may be changed at each time. In this case, a vehicle speed detector for measuring the vehicle speed of the unmanned vehicle may be provided, and the vehicle speed and the traveling distance may be calculated based on the detector.

[Effects of the Invention] As described in detail above, according to the present invention, it is possible to predict the center position of a mark or an area where a mark is completely included when an image is captured at a new imaging point. It is sufficient to recognize the mark only in the portion or the area corresponding to, and it is possible to determine an accurate and accurate traveling route without a possibility of picking up noise other than the mark, and furthermore, to detect an abnormal traveling of the unmanned vehicle during traveling. It can be used for detection, and can be used for a system that shortens the image processing time, and has an excellent effect as a traveling route determination processing method for an image type unmanned vehicle.

[Brief description of the drawings]

FIG. 1 is an electric block circuit diagram of a traveling route determination processing device embodying the present invention, FIG. 2 is a side view of a traveling unmanned vehicle, FIG. 3 is a plan view of the same, FIG. ,
FIG. 5 is an explanatory diagram for explaining an image taken by a CCD camera, FIG. 6 is a diagram showing a relationship between the number of times of imaging and a mark, FIG. 7 is a diagram showing a change of a mark in the image, FIG. Is a diagram showing the relationship between the operation sequence of the traveling route determination processing device and the traveling position of the unmanned vehicle, FIG. 9 is a diagram showing an area captured by the CCD camera during traveling, and FIG. 10 is a diagram showing an image in that area. 11, FIG. 11 is a diagram showing a traveling route determined by the traveling route determination processing device, FIG. 12 is an explanatory diagram for explaining a steady turning circular traveling, FIG. 13 is a diagram showing the relationship between the attitude angle and the radius, 14 and 15 are diagrams for explaining the processing area, and FIG. 16 is a diagram for explaining the coordinate conversion. In the figure, 1 is an image type unmanned vehicle, 3 is a CCD camera, 4 is a road surface, 4
a is an area, 5 is an image, 6 is a mark, 6a is a corner, 10 is a microcomputer, 11 is a CPU, 12 is a program memory, 16 is an A / D converter, 18 is a binary level controller, 1
9 is a drive controller, 20 is a traveling motor, 21 is a steering mechanism, e is a processing area as an area, L, L0, L1, L2 are traveling paths, G is a mark center position, Φm, Φp1, Φp3 are attitude angles, P
0 is a point as an imaging old point, and P2 is a point as an imaging old point and a new point.

Claims (1)

(57) [Claims] An image of one of marks indicating a passing position of an unmanned vehicle discretely arranged on a road surface is taken at least a plurality of times by an image pickup device provided on the unmanned vehicle while traveling, and an image taken at each time is taken. An image-based unmanned vehicle traveling route determination process that recognizes the position of the mark based on the inside mark, determines the traveling route of the unmanned vehicle, sequentially updates the traveling route, and drives the vehicle along the updated traveling route. In the method, a first coordinate system having an origin at a previously captured imaging point is set, a position of a mark is determined in the first coordinate system, and a traveling route determined based on the previously captured mark is determined. Then, a new imaging point at which the imaging operation is performed next is determined. Next, a second coordinate system having the new imaging point as the origin is set, and the center position of the mark in the first coordinate system is determined by the first coordinate system. Determined by converting the coordinates to the position in the coordinate system 2 From the converted position, the center position in the image of the mark imaged at the imaging new point or the region where the mark is completely included in the image is predicted by calculation, and only the portion corresponding to the center position or only in the region An image-based unmanned vehicle traveling route determination processing method for recognizing a position of a mark and obtaining a new traveling route.