JPH0619541A - End detecting device for automatic travel working vehicle - Google Patents

Abstract

PURPOSE:To judge an end after traveling in a working process along a ridge without any error by performing detecting processing of a 1st border detecting means according to an area processed by a selecting means. CONSTITUTION:An area extracting means 100 extracts an area corresponding to a nonwork place among plural areas which are arrayed in a direction crossing the travel direction of the machine body in a picked-up image as to each of plural lines. The 1st border detecting means 101 detects the 1st border between the nonwork place and a work place which adjoins in front of the machine body in its travel direction according to position information on the line where the number of areas As larger than a set number in the machine body travel direction. A judging means 104 judges the end of the working process according to the position information on the 1st border in the machine body travel. A 2nd detecting means 102 detects a 2nd border between the nonwork place and an adjacent work place in the direction crossing the machine body travel direction. A selecting means 103 selects only the area positioned more on the work place side than the 2nd border position and the 1st border detecting means 101 detects the 1st border according to this area.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image pickup means for picking up an image of the front side of a machine body advancing direction, and a plurality of lines lined up in the machine body advancing direction intersect with the machine body advancing direction based on imaged image information of the image pickup means. An area extracting means for extracting an area corresponding to a non-working area among a plurality of areas arranged in a direction; and a line for which the number of the areas extracted by the area extracting means out of the plurality of lines is a set number or more. First boundary detection means for detecting a first boundary between the non-work site and a work site adjacent to the non-work site on the front side in the machine travel direction based on position information in the machine travel direction, and the first boundary detection means. End for a self-driving work vehicle provided with a judging means for judging the end of the work stroke based on position information on the screen coordinate axis of the image pickup means along the machine body traveling direction of the first boundary detected by Inspection Apparatus on.

[0002]

2. Description of the Related Art A terminal detecting device for an automatic traveling work vehicle of this type includes, for example, a plurality of work strokes set by a rice transplanter as an automatic traveling work vehicle so as to be parallel to each other in a field as a work site. The present invention is applied to a method in which seedlings are planted in rows on a whole surface of a field at every set interval while reciprocating along (hereinafter referred to as field field). Then, conventionally, in order to detect a first boundary between a field and a ridge as a non-working land adjacent to a field on the front side in the machine body traveling direction, an image capturing means such as a television camera for imaging the front side in the machine body traveling direction is used. In the captured image information, for example, each scan of the imaging screen that intersects with the aircraft traveling direction of the region extracted corresponding to the ridge based on the difference in color information of the grass and the field which occupy most of the ridge If the number on the line is equal to or larger than the set number (for example, 3/4 of the number of regions on each scanning line), the line is determined to be a ridge on the front side in the traveling direction, and is determined to be the ridge. Among the lines, the line closest to the field, that is, the line on the front side was detected as the first boundary. Then, the position on the screen coordinate axis along the traveling direction of the detected first boundary is within a predetermined interval (for example, 1/4 of the entire vertical width of the screen) from the lower end of the screen on the front side in the traveling direction, for example. If so, it is configured to determine that the end of the field process has been reached.

[0003]

However, in the above-mentioned conventional technique, the field travels along the boundary (called the second boundary) with the ridge adjacent to the field in the direction intersecting the traveling direction, that is, along the ridge. Later, in the case of detecting the first boundary with the ridge on the front side in the traveling direction, there is a possibility that the first boundary cannot be reliably detected. Specifically,
As shown in FIGS. 5 (a) and (b), considering the case of traveling along the left ridge in the traveling direction of the aircraft, in the portion where the left ridge and the front ridge are connected, the front ridge is connected. Does not exist on the entire width of the screen and is cut off on the left side. Therefore, the number of regions extracted corresponding to the ridge is smaller than the number when the ridge exists in the entire width, and in this case, the number of the regions corresponding to the ridge on the front side on each scanning line is smaller. Since it does not exceed the set number (for example, 3/4 of the number of regions in each scan line), the ridge on the front side cannot be detected depending on the set number. Therefore, it is conceivable to reduce the set number so that the ridge on the front side in the traveling direction can be detected as much as possible. In this case, if the set number is too small, as shown in FIG. If a relatively large number of regions corresponding to the left ridge are extracted in, the number of regions corresponding to the front ridge may be approximately the same as the number of regions corresponding to the left ridge. In this case, the left ridge is erroneously detected as the front ridge and, for example, the bottom edge of the screen is erroneously detected as the first boundary. Therefore, the end of the work stroke based on the detection of the first boundary is detected. You will make a mistake in your judgment.

The present invention has been made in view of the above circumstances, and an object thereof is to detect a first boundary between a ridge on the front side in the traveling direction and a field after running a work stroke along the ridge. When determining the end of the work process, the first boundary is surely detected and the end is determined without error.

[0005]

A characteristic construction of an end detecting device for an automatic traveling work vehicle according to the present invention is based on the area information extracted by the area extracting means, and the non-working area and the machine body traveling therewith. A second boundary detecting means for detecting a second boundary with the work site adjacent in the direction intersecting the direction; and the area extracted by the area extracting means, detected by the second boundary detecting means. And a sorting unit that sorts only a region located closer to the work site than a position of the second border, and the first boundary detecting unit is configured to sort the region based on the region sorted by the sorting unit. It is configured to execute the detection processing of the first boundary.

[0006]

According to the characterizing feature of the present invention, among the areas extracted corresponding to the ridges which are non-working areas, for example, the ridges and the work areas which are adjacent to the ridges in a direction intersecting with the aircraft traveling direction, for example, Only the region located on the field side with respect to the position of the second boundary with the field is selected, and the first boundary between the field and the adjacent ridge on the front side in the machine traveling direction is detected based on only this selected region. Therefore, the information on the region by the ridges adjacent to each other in the direction intersecting the traveling direction of the machine body is excluded in the detection of the first boundary. As a result, in each line lined up in the aircraft traveling direction, the weight of the information of the area due to the adjacent ridges on the front side in the aircraft traveling direction becomes large, and the aircraft is determined by whether or not the number of areas where the weight is large is the set number or more. The line corresponding to the ridge on the front side in the traveling direction is surely discriminated, and, for example, the frontmost line of the lines discriminated as these ridges is detected as the first boundary. Then, the end of the work process is determined based on the detected position information of the first boundary on the screen coordinate axis, for example, the position is within a predetermined interval from the lower end of the screen on the front side in the traveling direction. Will be done.

[0007]

Therefore, according to the present invention, the first ridge and the field on the front side in the traveling direction after the traveling of the work process along the ridge is performed.
When the boundary is detected to determine the end of the work process, the first boundary can be surely detected and the end can be determined without error.

[0008]

EXAMPLES In order to detect the end of the field stroke after planting seedlings in a field M as a work site adjacent to a ridge N as a non-work site by the present invention, a rice transplanter as an automatic work vehicle is used. An embodiment when applied to the above device will be described with reference to the drawings.

As shown in FIGS. 8 and 9, a seedling planting device 2 is provided at the rear of a machine body V in which both the front wheels 1F and the rear wheels 1R can be steered so that it can move up and down. A plurality of seedlings T are planted in a row in 6 rows by the device 2. Further, a color type image sensor S1 as an image pickup means for picking up an image of the farm field M on the front side in the traveling direction of the machine body and the ridge N adjacent thereto is provided on the front side of the machine body V.

Explaining the mounting structure of the image sensor S1, the ridge adjacent to the lateral side of the machine body V on the lateral side of the machine body V is adjacent to the tip of the support member 4 projecting toward the lateral outer side of the machine body V. N and the farm field M are provided so as to be imaged obliquely from above in the traveling direction of the machine body.
That is, in the state in which the machine body V is properly aligned with the second boundary between the ridge N and the adjacent field M in the direction intersecting with the aircraft traveling direction, that is, the ridge Nk, it corresponds to the ridge Nk. The line segment L is made to coincide with a traveling reference line La that passes through the center of the image pickup visual field of the image sensor S1 in the front-rear direction. Further, one image sensor S1 is provided on each of the left and right sides of the machine body V, and the sensor on the side to be used can be switched between the left and right.

Explaining the structure of the machine body V, as shown in FIG. 1, the output of the engine E is transmitted to each of the front wheel 1F and the rear wheel 1R via a transmission device 5, and the transmission device 5 operates. as the operation state of the speed change operation state corresponds to a preset speed, the shifting state detection potentiometer R ₃ provided, and, on the basis of the detection information of the shifting state detecting potentiometer R _3, shift It is configured to drive the electric motor 6 for use.

Further, the front wheel 1F and the rear wheel 1R are constructed so that power steering is operated individually by hydraulic cylinders 7F and 7R, respectively, and steering angle detecting potentiometers R ₁ and R interlocked with the steering operation of the wheels. _The electromagnetically operated control valves 8F, 8R for operating the hydraulic cylinders 7F, 7R are driven so that the detected steering angle by ₂ becomes the target steering angle.

Therefore, the front steering wheel 1F and the rear steering wheel 1R are parallel steering type steering systems which steer the front wheels 1F and the rear wheels 1R at the same phase and at the same angle.
It is possible to selectively use three types of steering types, that is, a four-wheel steering type in which the rear wheels 1R are operated in opposite phases and at the same angle, and a two-wheel steering type in which only the front wheels 1F are turned. However, when the steering operation is automatically performed based on the image pickup information of the image sensor S1, the two-wheel steering type is used, and when one field stroke is finished and the next field stroke is reached, the four wheels are used. The steering type and the parallel steering type are used. In FIG. 1, S2 is a distance sensor for detecting the traveling distance based on the output speed of the transmission 5.

Next, the line segment L corresponding to the ridge Nk is calculated based on the imaging information of the image sensor S1 and the end of the field stroke along the ridge Nk is detected. The control configuration will be described. As shown in FIG. 1, the image sensor S1 includes three primary color information R, G, B.
And binarizing by subtracting blue information B not including the color component of grass K from green information G including the color component of grass K that occupies the majority of the ridge N. Thus, the area Ka corresponding to the ridge N is extracted (see FIGS. 4A and 4B).

Explaining the extraction of the above-mentioned area Ka, regarding the intensities of the green information G and the blue information B of the three primary color information output from the image sensor S1, in the portion where the grass K exists, in the mud and the soil surface. Considering each of the corresponding portion and the water surface portion that reflects natural light, in the portion where the grass K exists, the intensity of the green information G is high and the intensity of the blue information B is low. Further, in the portion corresponding to the mud and the soil surface, both the green information G and the blue information B have small intensities. Further, on the water surface part, green information G and blue information B
Both of the strengths are large. Therefore, the influence of the natural light reflected on the water surface and the image information corresponding to the mud and the soil surface are removed by the magnitude of the signal level obtained by subtracting the blue information B from the green information G. That is, the grass K is obtained by extracting and binarizing a portion in which the signal level obtained by subtracting the blue information B from the green information G is equal to or higher than the set threshold value.
The information of the area Ka corresponding to can be extracted.

Then, a subtracter 9 for subtracting the blue color information B from the green color information G in the state of an analog signal, and the output of the subtractor 9 is binarized based on a preset threshold value to be set in the area Ka. A comparator 10 that outputs corresponding binary information, and an output signal of the comparator 10 has a preset pixel density (horizontal axis 32 pixels × vertical axis 32 pixels / 1
Image memory 11 stored as pixel information corresponding to (set on the screen), and a line segment L corresponding to the edge Nk is approximated to a straight line or a curve based on the information stored in the image memory 11. A control using a microcomputer that obtains information, executes traveling control based on the information, and detects the end of the field stroke based on the positional information of the ridge N located along the lateral direction of the screen on the front side in the traveling direction. Each of the devices 12 is provided.

That is, using the subtractor 9 and the comparator 10, a plurality of lines (32 lines parallel to the x-axis which is the abscissa axis of the screen) are arranged in the machine advancing direction based on the imaged image information by the image sensor S1. Direction (x-axis direction) intersecting the aircraft traveling direction
Area extracting means 100 is configured to extract an area Ka corresponding to the ridge N in which the grass K occupies most of the plurality of areas arranged in line, and the area extracting means 100 uses the control device 12 to extract the area Ka. Second boundary detection means 102 for detecting the ridge Nk (second boundary) is configured based on the extracted information of the area Ka. The second boundary detecting means 102 will be described in detail. In the area Ka, on each of the 32 lines (y coordinates = 1 to 32), the number of consecutive pixels is a predetermined number (for example, a screen). The number corresponding to 1/4 of the width, that is, 8) or more, and
Based on the information of the representative point P extracted as a point located closest to the field M side in each of these selected pixel rows Kr, only the pixel row Kr closest to the field M side is selected, The line segment L corresponding to the edge Nk is obtained.

Further, by utilizing the control device 12, the area extracting means 10 out of the plurality of lines (32 lines).
Based on the position information in the machine traveling direction (screen y axis) of a line in which the number of the areas Ka extracted by 0 is equal to or greater than a set number (for example, the number of pixels corresponding to 3/4 of the screen width), the traveling direction A ridge N located along the horizontal direction of the screen on the front side
And the first boundary Nz with the adjacent field M on the near side,
For example, the first boundary detecting means 10 for detecting the line located closest to the front among the lines having the set number or more.
1 is configured.

Further, utilizing the control device 12, the ridge Nk (second boundary) detected by the second boundary detecting means 102 in the area Ka extracted by the area extracting means 100. The selecting unit 103 is configured to select only the region Ka located on the side of the field M with respect to the position of the line segment L corresponding to. Then, the first boundary detection means 101 is configured to execute the detection processing of the first boundary Nz based on the area Ka selected by the selection means 103. Therefore,
In this case, the set number is based on the number of pixels on the screen corresponding to the horizontal width from the position of the line segment L to the screen edge in the field M side direction (less than the original number of pixels on the screen, 32). It is set to 3/4 of the number of pixels.

Further, by using the control device 12, the image sensor S1 along the airframe traveling direction of the first boundary Nz detected by the first boundary detecting means 101.
The determination means 104 is configured to determine the end of the work process (that is, the field process) based on the position information on the screen coordinate axis (y-axis). Specifically, for example, the line corresponding to the first boundary Nz is 1/4 (8) of the full width in the vertical direction of the screen (32 lines in terms of the number of lines) from the lower end position of the screen on the front side in the traveling direction. When it is located within the line), it is judged that the end has been reached.

Next, referring to the flow chart shown in FIG. 2, the operation of the control device 12 will be described, and the process of detecting the ridge Nk will be described in detail. Every time the machine body V travels a set distance, or The image pickup process by the image sensor S1 is executed every set time, and after the image pickup process,
The area extracting unit 100 extracts the area Ka corresponding to the ridge N (see FIGS. 4A and 4B). In the figure, the area Ka corresponding to the grass K is extracted not only on the side of the ridge N but also on the field M, but this is due to floating grass or the like existing on the field M.

Next, with respect to each horizontal scanning line at y = 1 to 32 on the y coordinate axis of the screen, first, the number of consecutive pixels of a predetermined number (for example, 8) or more is located closest to the field M side. The pixel column Kr is selected. Therefore, if the pixel row Kr having the number of consecutive pixels equal to or larger than the predetermined number does not exist, there is no pixel row Kr to be selected in the horizontal scanning line. The area Ka of floating grass or the like existing on the field M usually has a smaller number of pixels than the predetermined number. Therefore, the area Ka corresponding to the floating grass or the like is removed by the above-described selection processing. Next, the pixel located closest to the field M side in each of the selected pixel rows Kr is extracted as the representative point P (FIG. 4).
(See (c)).

After the above-mentioned selection processing and representative point extraction processing have been completed for all (32) horizontal scanning lines of the screen coordinate axis, on the y axis of the screen coordinate axis, that is, on the horizontal scanning line of the representative point P adjacent in the advancing direction. It is checked whether or not there is a discontinuous portion where the position at is different from the width of the screen by 1/4 (8 in number of pixels) or more. If there is no discontinuous portion, it is determined that the end of the field stroke is not approached, so the line segment L is calculated based on the extracted representative point P. In this case, the Hough transform process is used. When the discontinuous portion is present, the end detection processing described later will be performed.

The Hough transform will be described. As shown in FIG. 6, the center of the image pickup field of the image sensor S1 (x =
16, y = 16) is used as a reference line in the polar coordinate system, and a plurality of straight lines passing through each of the representative points P are set to 0 to 180 degrees with respect to the x-axis based on the following formula (i). It is calculated as a combination of an inclination θ set in advance in a plurality of steps in the range and a distance ρ from the center of the screen corresponding to the origin, that is, the center of the imaging visual field. ρ = y · sin θ + x · cos θ …… (i)

Then, for all of the representative points P, the frequency of each straight line corresponding to the combination of the parameters (ρ, θ) is counted until the value of the inclination θ set in the plurality of steps reaches 180 degrees. The process of adding the two-dimensional histogram for is repeated. When the counting of the frequencies of the straight lines for all the representative points P is completed, the slope θ and the distance ρ that are the maximum frequencies are calculated from the values added to the two-dimensional histogram.
One straight line Lx having the maximum frequency is determined by obtaining the combination of the above, and the straight line Lx is obtained as information obtained by linearly approximating the line segment L corresponding to the edge Nk on the imaging surface of the image sensor S1.

Next, a straight line Lx on the image pickup surface is stored on the image pickup surface through which the stored information of the shape and size of the image pickup field A of the image sensor S1 measured in advance on the ground surface and the straight line Lx having the maximum frequency pass. Based on the pixel positions a, b, and c (see FIG. 7), the information of the straight line L on the ground surface is converted. That is, as shown in FIG. 7, a straight line L on the ground surface that is set as a value of an inclination ψ with respect to a traveling reference line La that passes through the center of the imaging visual field A in the widthwise direction in the front-rear direction and a position δ in the widthwise direction. Will be converted into information.

In addition, the lengths l ₀ and l of the two sides at the front and rear positions (y = 0 and y = 32) of the imaging visual field A in the direction intersecting the straight line L corresponding to the ridge Nk are added. ₃₂ , the length l ₁₆ in the lateral width direction of the imaging visual field A at the center of the screen (pixel position where x = 16, y = 16) and the distance h between the front and rear sides are measured in advance, It will be stored in the control device 12.

Then, the straight line Lx on the imaging plane is
X coordinate values X ₀ and X _{32 of} pixel positions a and b (positions where y = 0 and y = 32) intersecting the x axis corresponding to two sides at the front and rear positions of the imaging visual field A, and The value x ₁₆ of the x-coordinate of the position c of the pixel where the straight line Lx intersects the x-axis passing through the center of the screen is obtained from the following formula (ii) which is a modification of the above formula (i). Xi = (ρ−Yi · sin θ) / cos θ (ii) However, Yi is 0, 16, and 32, respectively.

Then, based on the coordinate value on the x-axis obtained by the above equation (ii), the position δ in the lateral width direction with respect to the traveling reference line La is calculated from the following equations (iii) and (iv).
And the slope ψ, and the values of the calculated position δ and the slope ψ are
It is calculated as the position information of the straight line L corresponding to the ridge Nk on the ground surface.

[0030]

[Equation 1]

Next, the end detection processing will be described with reference to the flow chart shown in FIG. in this case,
For example, as shown in FIG. 5 (a), when traveling along a ridge Nk located on the left side of the aircraft V, the ridge N is connected to a ridge N that intersects the traveling direction on the front side in the traveling direction. In this case, the field M is partitioned and the result of extracting the region Ka corresponding to the ridge N is shown in FIG. In the figure, the number of consecutive pixels of the area Ka on each horizontal scanning line changes depending on the state of the grass K on the ridge N.
Not only in the field M, but also in the field M, a region Ka due to floating grass or the like is extracted. Further, similarly to the above, on each horizontal scanning line, the pixel row Kr located closest to the field M side is selected by a predetermined number (for example, 8) or more continuous pixels (in this process, The area Ka on the M due to floating grass or the like is removed), and the pixel located closest to the field M side in each of the selected pixel rows Kr is extracted as the representative point P (see FIG. 5C). Then, the position of the representative point P adjacent to the front end of the ridge N on the front side is discontinuous (x coordinate value is x.
It is shown that the size _z exists.

In the end detection process, first, only the representative points P located on the front side of the horizontal scanning line in which the discontinuity of the adjacent representative point positions exists is selected, and based on these selected representative points P, The edge Nk (second boundary) is obtained by the Hough transform process as in the above. That is, one straight line Lx is determined based on the combination of the inclination θ and the distance ρ that have the maximum frequency, and the straight line Lx is linearly approximated to the line segment L corresponding to the edge Nk on the imaging surface of the image sensor S1. Requested information (see Figure 6)

Next, the second boundary obtained by the above linear approximation
The area Ka located on the field side with respect to the position of the line segment L
Select only. Then, the area Ka
The vertical axis is the position of each horizontal scanning line of the screen coordinate axis,
Also, the horizontal axis represents the number of pixels in the area Ka in each horizontal scanning line.
h_wHistogram set to (This is the horizontal histogram
(Referred to as Lamb) is calculated (see FIG. 5D). here,
Maximum value H on the horizontal axis_WIs between the position of the line segment L and the right edge of the screen
Take the screen coordinate value of the distance. Then, the number of pixels h _wIs the formula
Water in a region Ka that is located closest to the front and that satisfies (v)
Set the flat line position to the front ridge N and the ridge N
Determined as the first boundary Nz with the adjacent field M on the front side
It (H_w/ H_W)> 3/4 ... (v)

Then, the interval v _d between the horizontal scanning line where the first boundary Nz is located and the lower end of the screen and the width V _D in the vertical direction of the screen are set.
If the ratio v _d / V _D of the following satisfies the following expression (vi), it is determined to be the termination. (V _d / V _D )> 1/4 (vi)

Here, if it is not determined to be the end, in order to obtain information for steering control along the ridge Nk, the second boundary on the image pickup screen, that is, the same as the above, is obtained. The line segment L is the image sensor S on the ground surface measured in advance.
Memory information of the shape and size of the imaging field A of No. 1 and pixel positions a, b, c on the imaging surface through which the line segment L passes (see FIG. 7).
Based on and, it is converted into information of the straight line L on the ground surface.

As described above, in the steering control for automatically traveling the vehicle body V along the ridge Nk along the traveling direction of the vehicle body, the inclination ψ of the straight line L with respect to the traveling reference line La and the position in the lateral width direction are controlled. so that both δ and 0 approach zero,
Steering operation will be done in a two-wheel steering system.

Explaining the steering control, from the values of the inclination ψ of the straight line L with respect to the traveling reference line La and the position δ in the lateral width direction, and the current steering angle φ of the front wheels 1F. The target steering angle θf of the front wheels 1F is set based on the following equation (vii), and the current steering angle φ detected by the steering angle detecting potentiometer R ₁ for the front wheels is set to the target steering angle φf. Direction θf
On the other hand, the control valve 8F of the front wheel hydraulic cylinder 7F is driven so as to be maintained within the set dead zone. θf = K ₁ δ + K ₂ ψ + K ₃ Φ (vii) Note that K ₁ , K ₂ and K ₃ are constants set in advance according to the steering characteristics.

As described above, when it is determined from the image information of the image sensor S1 that the end of the field stroke is determined, the turn is made toward the start of the next field stroke (in this case, the side opposite to the ridge N). The outline of this turn control will be explained. After the planting work by the seedling planting device 2 is interrupted, the two-wheel steering system is switched to the four-wheel steering system and the maximum cutting angle is maintained for a set time. The direction of the next field is changed by 180 degrees. At this time, while switching to the image sensor S1 on the side opposite to the image sensor S1 used in the previous stroke, while detecting the position of the seedling T already planted in the previous field stroke adjacent to the next field stroke from the image information. , The position in the lateral direction of the machine body in the next field stroke is corrected, and the turn is ended. After the end of the turn, the two-wheel steering system is restored, and the steering control in the next field stroke is restarted.

[Other Embodiments] In the above embodiment, the color type image sensor S1 is used as the image pickup means, and the blue color information B is subtracted from the green color information G and binarized based on the set threshold value. An example is shown in which the region Ka corresponding to the ridge N in which K occupies most of the area is extracted.
For example, all the three primary color information R, G, and B may be used to extract a region whose ratio is a set ratio range corresponding to the color of the grass K as the region Ka, or the image pickup means. As an example, the monochrome image sensor S1 is used to binarize the lightness signals of the grass K and other places such as the mud surface based on an appropriate threshold setting so that the area can be extracted by a simple device. The area extraction means 100
The specific configuration of can be changed in various ways.

In the above embodiment, the screen coordinate axis (y axis)
When determining the line corresponding to the first boundary Nz from among the horizontal scanning lines (y = 1 to 32) of
The number of pixels corresponding to 3/4 of the screen width from the field edge to the field M side is determined by the size of the set number, but the set number is not limited to 3/4 of the screen width, The first boundary detection means 10 depending on the situation such as 2/3 or 4/5
The configuration of 1 can be changed as appropriate.

In the above embodiment, the Hough transform is used to linearly approximate the line segment L corresponding to the second boundary (edge Nk). However, for example, the least squares method is used. The linear approximation or the curve approximation can be used to obtain information, and the specific configuration of the second boundary detecting means 102 can be variously changed.

Further, in the above embodiment, the screen coordinate axis (y of the image pickup means S1 along the direction of travel of the first boundary Nz in the machine body (y
As a criterion for determining the end of the work process (that is, the field process) based on the position information on the axis, the line corresponding to the first boundary Nz is the entire width in the vertical direction of the screen from the lower end position of the screen (in number of lines). It is determined whether or not the position is within 1/4 (8 lines) of 32 lines). However, this criterion is not limited to 1/4 (8 lines) of the full width in the vertical direction of the screen. The configuration of the determination unit 104 can be appropriately changed according to the device status such as the screen magnification of the unit S1 and the imaging angle.

Further, in the above embodiment, the present invention is applied to the rice transplanter as the automatic traveling work vehicle, and the first boundary between the field M as the work place and the ridge N as the non-work place adjacent thereto ( Detecting the end point) and the second boundary (for running control),
Although the case of turning along with the detection of the end while automatically traveling along the second boundary is illustrated, the present invention is also applicable to other automatic traveling work vehicles such as lawn mowers, At that time, the types of the work site M and the non-work site N, the configuration of each part of the traveling system of the machine body, and the like, and the first boundary between the detected work site M and the non-work site N (for detecting the end point) )
The usage pattern of the information of the second boundary (for traveling control) can be variously changed.

It should be noted that reference numerals are given in the claims for convenience of comparison with the drawings, but the present invention is not limited to the configurations of the accompanying drawings by the entry.

[Brief description of drawings]

FIG. 1 is a block diagram of a control configuration.

FIG. 2 is a flowchart of the edge detection process.

FIG. 3 is a flowchart of end detection processing.

FIG. 4 is an explanatory diagram of image processing.

FIG. 5 is an explanatory diagram of image processing.

FIG. 6 is an explanatory diagram of Hough transform.

FIG. 7 is an explanatory diagram showing the relationship between the aircraft traveling direction and the approximate straight line in the imaging field of view.

FIG. 8 is a schematic plan view of a rice transplanter.

FIG. 9 is a schematic side view of the rice transplanter.

[Explanation of symbols]

 S1 image pickup means N non-working area Ka area M working area Nz first boundary Nk second boundary 100 area extracting means 101 first boundary detecting means 102 second boundary detecting means 103 selecting means 104 judging means

Claims (1)

[Claims] 1. An image pickup means (S1) for picking up an image of the front side of the machine body advancing direction, and a plurality of lines lined up in the machine body advancing direction intersect with the machine body advancing direction based on imaged image information of the image pickup means (S1). A region extracting means (100) for extracting a region (Ka) corresponding to the non-working place (N) among a plurality of regions arranged in the direction, and the region extracting means (100) for extracting the region among the plurality of lines. The area (K
Based on the position information in the machine advancing direction of the line in which the number of a) is equal to or more than the set number, the non-working area (N) and the work area (M) adjacent to the non-working area (N) adjacent to the non-working area First boundary detection means (101) for detecting one boundary (Nz), and the imaging means (S1) along the aircraft traveling direction of the first boundary (Nz) detected by the first boundary detection means (101). ) Is a terminal detecting device for an automated traveling work vehicle, which is provided with a judging means (104) for judging the end of the work stroke based on the position information on the screen coordinate axis, and the area extracting means (100) includes: A second boundary for detecting a second boundary (Nk) between the non-working site (N) and the work site (M) adjacent to the non-working site (N) in a direction intersecting with the non-working site (N) based on the extracted area information. The detecting means (102) and the area extracting means (100) Therefore, in the extracted area (Ka), an area (Ka) located closer to the work site (M) than the position of the second boundary (Nk) detected by the second boundary detecting means (102). ) Is provided, and the first boundary detection means (101) is provided with the selection means (1).
03) An end detection device for an automated work vehicle configured to execute the detection process of the first boundary (Nz) based on the region (Ka) that has been selected by the process 03).