JPH0535885B2 - - Google Patents

Description

[Detailed Description of the Invention] Purpose of the Invention (Field of Industrial Application) The present invention relates to a method for driving an unmanned vehicle, and more specifically, a driving line indicating the driving route of the unmanned vehicle is photographed by an imaging device, The present invention relates to a traveling method based on the traveling speed of an unmanned vehicle with image expansion, in which the unmanned vehicle travels along a traveling route determined by processing an image of a photographed traveling line.

(Prior Art and Problems) In recent years, in this type of image-based unmanned vehicle, a safe and rational way to run the unmanned vehicle along the traveling line has been developed by image processing the traveling line that indicates the traveling route taken by the imaging device. Various methods have been proposed for determining a travel route.

(Problem to be Solved by the Invention) However, in this type of image-based unmanned vehicle, image processing is performed using all the image data of the road surface at a position in front of the unmanned vehicle that is always predetermined and obtained by the imaging device. The driving route was determined. As a result, one and the same travel route is determined whether the unmanned vehicle is traveling at low speed or high speed.

Therefore, even if the running line indicating the running route is the same, depending on the running speed at that time, that is,
It was not possible to determine a driving route that matched the driving speed and drive, which caused problems in improving safety, efficiency, and driving feeling.

The purpose of the present invention is to solve the above-mentioned problems, determine a driving route according to the driving speed, and enable an unmanned vehicle to travel safely and efficiently.
It is an object of the present invention to provide a driving method based on the driving speed in an image-based unmanned vehicle that can improve the driving feeling.

Structure of the Invention (Means for Solving Problems) In order to achieve the above object, the present invention uses an imaging device installed in an unmanned vehicle to image a driving line drawn on the road surface indicating the driving route of the unmanned vehicle. Then, select multiple points in the longitudinal direction of the travel line in the image, projectively transform each point in the selected image to the actual position on the road surface, and then projectively transform each point. Alternatively, in a driving method in which the driving route of the unmanned vehicle is determined by determining a function of a curve that smoothly passes between the proximal position and the current position of the unmanned vehicle on the road surface, the driving method determines the driving route of the unmanned vehicle. As the traveling speed of the unmanned vehicle increases, the image area for extracting data for determining the traveling route is changed to an area where images are taken at a position farther away from the unmanned vehicle, and based on the traveling speed. The main idea is a driving method based on the driving speed of an image-based unmanned car that determines the driving route of the unmanned car based on the image data in the selected image area and the current position of the unmanned car and controls the steering angle. It is something to do.

(Operation) By projectively transforming a plurality of points selected from the driving line in the image, the positional relationships of the plurality of points on the image are converted to the positional relationships on the actual driving road surface. . The driving route is usually determined by a human because the faster the driving speed is, the more the image area for extracting the data for determining the driving route is changed to an area that captures images at a position farther away than the unmanned vehicle. A driving route is determined in the same manner as when a driving route is determined by looking at a farther position at higher speeds and a closer position at lower speeds. As a result, the unmanned vehicle will travel along a travel route that matches its travel speed.

(Embodiment) Hereinafter, an embodiment of an unmanned vehicle travel control device embodying the image-based unmanned vehicle travel route determination method of the present invention will be described with reference to the drawings.

In FIG. 1, a support frame 2 is erected at the upper center position on the front side of the unmanned vehicle 1, and an image pickup device is installed at the upper center position of the frame 2.
A CCD (charge coupled device) camera 3 is provided. The CCD camera 3 is set on the support frame 2 so as to photograph an area 4a on the road surface 4 in front of the unmanned vehicle 1.

On the road surface 4, as shown in FIG. 2, a travel line 5 indicating the travel route of the unmanned vehicle 1 is drawn with a constant line width, and in this embodiment, it is drawn with a white paint different from the color of the road surface 4. It is. Then, the CCD camera 3 photographs the traveling line 5 having a constant line width.

Therefore, the CCD camera 3 captures the area 4a shown in FIG. 4 as the image 6 shown in FIG. 5, and the image 6 of the area 4a is used in this embodiment.
It consists of 256 x 256 pixels. and,
A signal (hereinafter referred to as a pixel signal) captured by the CCD camera 3 that captures the white running line 5 has a high output level, while a pixel signal that captures the road surface 4 has a low output level. In addition, in this example, CCD
Although illumination lamps 7 that illuminate the road surface 4 are installed below both sides of the camera 3 in order to facilitate image pickup by the camera 3, it is also possible to carry out the process without using the illumination lamps 7.

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

The microcomputer 10 includes a central processing unit (hereinafter referred to as CPU) 11, a program memory 12 consisting of a read-only memory (ROM) that stores control programs, and a read/write memory that temporarily stores pixel data and other arithmetic processing results of the CPU 11. It is composed of a working memory 13 consisting of a replaceable memory (RAM), a timer 14, and the like. And the CPU 11 is the program memory 1
Travel line 7 according to the control program stored in 2.
The system determines the state of the vehicle and executes various calculation processing operations for driving such as steering control.

The CPU 11 scans and controls 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 also scans and controls the pixel signals from the CCD camera 3. The pixel data is converted into pixel data via the A/D converter 16 and the bus controller 17 and stored in the working memory 13.

When the A/D converter 16 converts the pixel signal from the CCD camera 3 from an analog value to a digital value,
It is determined whether each pixel signal is equal to or greater than a predetermined set value. Then, the A/D converter 16 sets the pixel to "1" for the white traveling line 5 part when the pixel signal is greater than the set value, and on the other hand, when the pixel signal is less than the set value, the dark colored part of the road surface 4. The bus controller 17 binarizes each pixel signal that is sequentially input so that the pixel is set to "0" and converts it into pixel data.
The data is stored in the working memory 13 via the .

Therefore, the working memory 13 includes the CCD camera 3.
The image 6 taken by the user will be stored as 256×256 pixel data.

The binarization level controller 18 is connected to the CPU 1
1 based on the control signal from the A/D converter 1.
The setting value data for 6 to be binarized is sent to the same A/D.
The signal is output to a converter 16. The drive controller 19 controls the driving motor 20 and steering mechanism 21 for driving the unmanned vehicle 1 based on control signals from the CPU 11 as well. The drive motor 20 determines the traveling speed V of the unmanned vehicle 1, and the steering mechanism 21 determines the steering angle Θ of the unmanned vehicle 1. The vehicle speed sensor 22 is a sensor that detects the traveling speed V of the unmanned vehicle 1, and in this embodiment, detects the rotational speed of the drive motor 20 and outputs the detection signal to the CPU 11 via the input/output interface 15. The CPU 11 calculates the running speed V at that time based on this detection signal. In this embodiment, the traveling speed V of the unmanned vehicle 1 is determined based on an external control signal input to the CPU 11 via the input/output interface 15.

In addition, in this embodiment, for convenience of explanation, the scanning control of the CCD camera 3 is performed in the horizontal direction (y-axis direction) with respect to the screen.
Although a scanning method is adopted in which the scanning moves from the top of the screen to the bottom (in the x-axis direction), it goes without saying that other scanning methods may be used.

Next, the processing operation of the CPU 11 configured as described above will be explained.

Now, based on the control signal from CPU11, the CCD
When the camera 3 is scan-controlled, since the camera 3 is not perpendicular to the road surface 3 but is taking an image at a certain angle, an image of the front area 4a shown in FIG. 6 as shown in FIG. 7 is obtained. 6. The image 6 captured by the CCD camera 3 is converted into an A/D signal as a pixel signal.
The A/D converter 16 determines whether each pixel signal is a pixel signal for the traveling line 5 or a pixel signal for the road surface 4 and processed as pixel data via the bus controller 17. data is transferred to the memory 13 for use.

When the pixel data is transferred to the working memory 13, the CPU 11 determines the running speed V of the unmanned vehicle 1 at that time.
is determined based on the detection signal from the vehicle speed sensor 22. The CPU 11 checks whether the running speed V of the unmanned vehicle 1 is equal to or higher than a preset reference speed, and determines that the vehicle is running at high speed if it is higher than or equal to the reference speed.

And now, unmanned vehicle 1 is traveling at low speed,
When the traveling speed V is determined to be less than the reference speed, the CPU 11 calculates the slope of the traveling line 5 at a point near the unmanned vehicle 1 on the traveling line 5 based on the pixel data stored in the working memory 13.
A processing operation is executed to obtain the differential coefficient (differential value) at that point when expressed by a certain curve, that is, the attitude angle Φ of the centerless vehicle 1.

First, as shown in FIG. 7, check lines L1a and L1b are set as image areas at the bottom of the image 6 (position close to the unmanned vehicle 1), and the check lines L1a and L1b are set as image areas.
1a and L1b, the pixel data of one horizontal pixel column corresponding to the lines L1a and L1b are read out, and the pixel data of the running line 5 and the pixel data of the road surface 4 on each line L1a and L1b are determined, and the center of the running line 5 that intersects the line L1a is determined. After determining the position C1a, the center position C1b of the traveling line 5 intersecting the check line L1b is similarly determined from the pixel data group of the traveling line 5 located on the check line L1b.

These center positions C1a and C1b are determined by the x-axis being the 128th vertical pixel column counting from the left in each pixel that makes up the image 6, and the y-axis being the 128th horizontal pixel column counting from the top. It is calculated using the x and y coordinates.

The CPU 11 projectively transforms both the center positions C1a and C1b, that is, the center positions C1a and C obtained from the image 6.
1b is located at which position Q1a, Q1b on the actual area 4a as the selection point. This is because the CCD camera 3 does not image the road surface 4 perpendicularly as described above, so the driving line 5 in image 6 is
This is a process of matching the difference between the line and the actual driving line 6 in the area 4a. Note that this projective transformation processing operation is performed using the CCD camera 3 that has been set in advance.
Projective transformation is performed based on setting conditions such as focal length, inclination, and height. The general formula for this projective transformation is as follows.

Let the positional coordinates of the selected point be x, y, the positional coordinates of the base point be X, Y, the height of the CCD camera 3 be H, the inclination of the CCD camera 3 be θ, and the constant that determines the magnification of the corresponding position be F.

X=x+F・tanθ/1−(x/F)tanθ・
H/F Y=y・secθ/1−(x/F)tanθ・H/
F Note that the center position Q1a obtained by this projective transformation is
In Q1b, the direction perpendicular to the traveling direction passing through the center of the unmanned vehicle 1 is the Y axis, the line dividing the area 4a into left and right halves is the X axis, and the intersection of both axes (origin) is the direction perpendicular to the direction of travel of the unmanned vehicle 1.
Current position (more precisely, the position of CCD camera 3) HP
It is calculated using the X and Y coordinates.

The CPU 11 has a running line 5 that runs from the center positions Q1a and Q1b to the check lines L1a and L1b.
Calculate the slope (differential coefficient of running line 5) Φ1,
The slope Φ1 is stored in the working memory 13.

That is, if the coordinate positions of center positions Q1a and Q1b are (X1a, Y1a) and (X1b, Y1b), respectively,
The slope Φ1 can be determined as follows.

ΔX=X1a−X1b ΔY=Y1a−Y1b Φ1=tan ⁻¹ (ΔY/ΔX) Next, the CPU 11 passes the current position HP of the unmanned vehicle 1 and the center position Q1b on the travel line 5, and at the center position Q1b. The differential coefficient F′(X1b) is ΔY/ΔX(=
The function F(x) of the cubic curve L is tanΦ1) = aX ³ +
Find bX ³ +cX+d.

In this calculation, in this example, the current position is
The differential coefficient F'(0) is set to be 0 at HP. When the function F(x) of the cubic curve L1 is determined based on the above conditions, the CPU 11 converts this function F(X) from the current position HP of the unmanned vehicle 1 to the center position Q1b on the travel line 5 in the area 4a. This is determined as the driving route.

Once the travel route is determined, the CPU 11 performs control processing to control the steering mechanism 21 based on this function F(X). This process starts from the current position HP using the function F(X)
The unmanned vehicle 1 travels along the cubic curve L1 of the line 5.
In the processing operation for moving to the center position Q1b, the attitude angle Φ at the current running position
is determined from the function F(X), the steering angle Θ is determined and the steering mechanism 21 is controlled so that the unmanned vehicle 1 has the attitude angle Φ at each time.

In the case of this embodiment, steady turning running is adopted to perform this steering control, and this running is performed at a steering angle Θs as shown in FIGS. 9 and 10.
When is held constant, it refers to traveling with a constant radius R, and if the change in attitude angle Φ after ΔT seconds is ΔΦ, the traveling speed is V, and the wheel base of the unmanned vehicle 1 is D, the following formula holds true.

ΔΦ=V・Θs・ΔT/D R=D/Θs Then, from both equations, in order to change the attitude angle Φ by ΔΦ while advancing by V・ΔT, the radius R (=V・ΔT/ΔΦ ), and from its radius R, the steering angle Θs (=D/R=D・ΔΦ/V・
ΔT). Therefore, if the CPU 11 calculates the steering angle Θs for each ΔT based on the above formula and controls the operation of the steering mechanism 21, the unmanned vehicle 1 can be caused to travel along the cubic curve L1.

The details are described in, for example, Japanese Patent Application No. 60-283851 (Japanese Unexamined Patent Publication No. 62-140109) and Japanese Patent Application No. 60-281399 (Japanese Unexamined Patent Application No. 62-139012) filed by the applicant earlier. It will be easily understood if you refer to the content given below.

On the other hand, in determining whether the traveling speed V of the unmanned vehicle 1 is high or low, if the traveling speed V is determined to be equal to or higher than the reference speed, the CPU 11 selects a traveling line 5 based on the pixel data stored in the working memory 13. A processing operation is executed to obtain the same inclination as described above, that is, the differential coefficient at a point far away from the unmanned vehicle 1 above.

First, as shown in FIG. 7, check lines L2a and L2b are set as image areas on the upper side of the image 6 (at a position far from the unmanned vehicle 1), and a horizontal pixel row corresponding to the check lines L2a and L2b is set. The pixel data is read out and processing operations similar to those described above are performed to determine the center positions C2a and C2b of the travel line 5 that intersect with the check lines L2a and L2b, the positions Q2a and Q2b on the actual area 4a, and between the check lines L2a and L2b. The differential coefficient Φ2 of the running line 5 is determined.

Next, the CPU 11 passes through the current position HP of the unmanned vehicle 1 and the center position Q2b on the traveling line 5, and calculates the differential coefficient F'(X2b) at the center position Q2b.
is ΔY/ΔX (=tanΦ2), and the current position HP
Find the function F(X)=aX ³ +bX ² +cX+d of the cubic curve L2 such that the differential coefficient F'(0)=0. Then, the CPU 11 converts the obtained function F(X) into the unmanned vehicle 1.
Driving line 5 in area 4a from current position HP
The driving route is determined to be the upper center position Q2b.

Once the travel route is determined, the CPU 11 performs control processing to control the steering mechanism 21 based on this function F(X).

As described above, in this embodiment, when driving at low speed, the check lines L1a and L1b are set at the bottom of the image 6 (position close to the unmanned vehicle 1), and the check lines L
1a, L1b, the driving route is determined based on the position Q1b on the driving line 5 near the unmanned vehicle 1 and the differential coefficient Φ1, and when traveling at high speed, the driving route is determined at the upper side of the image 6 (position far from the unmanned vehicle 1). Check lines L2a and L2b are set at the position Q2b and Φ2 on the traveling line 5 which are far from the unmanned vehicle 1 based on the check lines L2a and L2b.
Since the driving route is determined based on
The driving route of the unmanned vehicle is determined in the same manner as a human normally determines a driving route by looking at a farther position at higher speeds and a closer position at lower speeds. Therefore, the unmanned vehicle will travel along a travel route that matches the travel speed, making it possible to drive the unmanned vehicle safely and efficiently with rational steering control, and improving the driving feeling. be able to.

In addition, in this embodiment, the CCD camera 3
When finding the function F(X) of the cubic curves L1 and L2 to be placed on the traveling line 5 taken by ), so unmanned vehicle 1 traveling along cubic curve L
is surely on positions Q1b and Q2b, and the same position Q
1b, Q2b, the attitude angle Φ of unmanned vehicle 1
will match the differential coefficients Φ1 and Φ2, and the vehicle will ride on the running line 5 smoothly. Therefore, the travel route determined based on the pixel data of the travel line 5 captured by the CCD camera 3 after riding on the travel line 5 can be moved without error.

Furthermore, in this embodiment, since the cubic curve L is used such that the attitude angle Φ is zero at the current position HP, the steering angle Θ is zero when starting the unmanned vehicle 1, so it is not necessary to change the steering angle suddenly. This allows for smooth steering operation.

Note that this invention is not limited to the embodiment described above; for example, in the embodiment described above, the function F(X) from the current position to the positions Q1b and Q2b is expressed as a cubic curve L1,
L2, but the point is that any function (for example, an n-th order function) may be used as long as the positions Q1b and Q2b are selected according to the traveling speed V, and the respective traveling routes are determined based on the positions. . Of course, in addition to setting the check line and finding the center positions C1a, C1b, C2a, and C2b as in the above embodiment to find the driving route, in the case of high speed, as shown in FIG. The pixel data of a certain area Z1 of the part, on the other hand, if the speed is low, image 6
Using the pixel data of a certain area Z2 in the lower part of the center, it is used to determine, for example, the center of gravity positions G1, G of the traveling line 5 in the respective areas Z1, Z2.
2 may be determined and the travel route may be determined based on the center of gravity positions G1 and G2.

In addition, in the above embodiment, the interval between each check line,
For example, the interval between the check lines L1a and L1b is not particularly limited, but the interval may be changed as appropriate as long as the differential coefficient at that position can be calculated.

In addition, in the above embodiment, after projective transformation, each position Q
The differential coefficients Φ1 and Φ2 of 1b and Q2b were obtained, and the differential coefficients at each position C1b and C2b in the image 6 were obtained and the obtained differential coefficients were projectively transformed to obtain the differential coefficients at each position Q1b and Q2b in the area 4a. The differential coefficients Φ1 and Φ2 may also be determined.

Further, in the above embodiment, two check lines are set for high speed and low speed, but the number may be increased to correspond to each speed. Of course, as the speed increases, a check line on the upper side (farthest from the unmanned vehicle) will be selected as the check line on the image 6. In this case, increasing the number allows for more accurate driving.

In addition, in the above embodiment, the traveling route (cubic curve) for riding on the traveling line 5 is determined by the function F(x)=aX ³ +bX ²
+ c You may. For example, in the case of this embodiment, since the image 6 is 256×256, there are 256 positions Q1b. Then, if the differential coefficient Φ1 is set in steps of 5 degrees, it becomes (180°/5°-1)・256=8960, which means that 8960 types of cubic curves L1 need to be prepared in advance, that is, made into a table. . Therefore, calculation processing time is reduced.

Furthermore, in the embodiment, the vehicle speed sensor 22 selects the region in the image 6 based on the current traveling speed V, but the unmanned vehicle 1 selects the region in the image 6 based on the traveling speed determined based on an external command signal. Of course, the area within 6 may be determined.

Effects of the Invention As detailed above, according to the present invention, the driving route of an unmanned vehicle can be determined according to the driving speed, the unmanned vehicle can be driven safely and efficiently, and the driving feeling can be improved. It has excellent effects that can be achieved.

[Brief explanation of drawings]

Figure 1 is a side view of an unmanned vehicle that embodies this invention, Figure 2 is a plan view of the same, Figure 3 is an electric block circuit diagram of the control device installed in the unmanned vehicle, and Figure 4 is an image captured by a CCD camera. Diagram showing areas to be
The figure is an explanatory diagram to explain the image taken by the CCD camera, Figure 6 is a diagram to explain the differential coefficient of the running line at each check line in the area, and Figure 7 is an explanatory diagram at each check line in the image. Figure 8 is an explanatory diagram to explain the running line, Figure 8 is an explanatory diagram to find the differential coefficient, Figure 9 is an explanatory diagram to explain steady turning travel, and Figure 10 is the relationship between attitude angle and radius. 11 are explanatory diagrams showing another example of the present invention. In the figure, 1 is an unmanned vehicle and 3 is an imaging device.
CCD camera, 5 is the running line, 6 is the image, 10
is a microcomputer, 11 is a central processing unit (CPU), 12 is a program memory, 13 is a working memory, 16 is an A/D converter, 19 is a drive controller, 20 is a drive motor, 21 is a steering mechanism, and 22 is a vehicle speed Sensor, L1a, L1b, L2
a, L2b are check lines as image areas;
Φ1, Φ2 are differential coefficients, Φ is attitude angle, Q1a, Q1
b, Q2a, and Q2b are positions.

Claims (1)

[Claims] 1. An imaging device installed in an unmanned vehicle captures an image of a driving line drawn on the road surface indicating the driving route of the unmanned vehicle, and a plurality of driving lines are detected in the longitudinal direction of the driving line in the image. After selecting points and projectively transforming each point in the selected image to the actual position on the road surface, smooth the projectively transformed points or their proximal positions and the current position of the unmanned vehicle on the road surface. In the driving method in which the driving route of the unmanned vehicle is determined by determining a function of a curve passing through the unmanned vehicle, an image area for extracting data for determining the driving route from an image taken by the imaging device is provided. As the traveling speed of the unmanned vehicle increases, the imaged area is changed to a position further away from the unmanned vehicle, and the image data in the image area selected based on the traveling speed is combined with the current position of the unmanned vehicle. A driving method based on the traveling speed of an image-based unmanned vehicle that determines the traveling route from the current position of the unmanned vehicle to the selected point based on the image and controls the steering angle. 2. An imaging device that is installed in an unmanned vehicle and captures an image of a running line that indicates the running route of the unmanned vehicle, and a vehicle speed sensor that detects the running speed of the unmanned vehicle, and regarding the running line in the image captured by the imaging device. , one selection point to be selected on the same line is selected at a position farther from the unmanned vehicle if the traveling speed detected by the vehicle speed sensor at that time is fast, and on the other hand, if it is slow, one selection point is selected at a position close to the unmanned vehicle, and then Find the differential coefficient of the travel line at the selected selection point, use it as the attitude angle of the unmanned vehicle at the selected point, find a curve using the obtained differential coefficient and the current position of the unmanned vehicle, and draw that curve from the current position of the unmanned vehicle to the selected point. A traveling method based on a traveling speed in an image-based unmanned vehicle according to claim 1, wherein the driving route is determined as a traveling route.