JPH01235506A - Travel control device of image sensing type automatically traveling work car - Google Patents

Abstract

PURPOSE:To suitably make possible car body to entry a beginning end part of next working process, by providing a control changing means alternatively changing a steering control means and turning control means to a working state according to a detected information of a second boundary detecting means. CONSTITUTION:A first boundary L1 is detected by a first boundary detecting means according to an image information of first image sensor Sa. A second boundary L2 detected by a second boundary detecting means according to an image information of second image sensor Sb. A car body V is steering controlled to automatically travel along the first boundary L1 by a steering control means according to a detected information of the first boundary detecting means. The car body V is transferred to a beginning end part of next working process by a turning control means and the steering control means and the turning control means are alternatively changed by a control changing means according to a detected information of the second boundary detecting means.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention provides an image capturing means for capturing an image of a work area in front of the vehicle in two-dimensional directions, and based on the image information of the image capturing means, a first boundary detection means for detecting a first boundary between an untreated work area and a treated work area; and a first boundary detection means for detecting a first boundary between an untreated work area and a treated work area; a second boundary detection means for detecting a second boundary between the vehicle body and the vehicle body; a steering control means for controlling the direction of the vehicle, a turn control means for moving the vehicle body to a starting end of the next working stroke, and a control operation of the turn controlling means as the vehicle body reaches a terminal end of each working stroke; The steering control means and the turn control means are controlled based on the detection information of the second boundary detection means so that the steering control means performs a control operation as the vehicle body reaches a starting end of the next working stroke. The present invention relates to an imaging type travel control device for an automatic traveling work vehicle, which is provided with a control switching means for selectively switching the control means to an activated state.

[Conventional technology]

The imaging type travel control device for this type of autonomous work vehicle mentioned above is:
Based on image information obtained by imaging the work area in front of the vehicle in two-dimensional directions, the first boundary in the width direction of the vehicle body and the second boundary in the longitudinal direction of the vehicle body are detected. The vehicle body travel is automatically controlled so that it automatically travels along the first boundary, and as it reaches the end of each work stroke, it automatically moves to the start of the next work stroke. It is something.

By the way, in a turn to move the vehicle body to the starting end of the next work stroke, the position of the vehicle body with respect to the first boundary and the second boundary changes significantly, so it is difficult to perform steering control based on imaged information even during the turn. Have difficulty. Therefore, when fishing on a boat, the user makes a turn by steering the boat to reproduce a turn pattern that has been set and stored in advance.

Conventionally, as the vehicle body reaches the second boundary detected by the second boundary detection means, the steering control means is switched to the turn control means, and the reproduction of the preset and stored turn pattern is completed. Accordingly, the turn control means is switched to the steering control means.

[Problem to be solved by the invention]

In the conventional configuration described above, since steering control is not performed during a turn, there is a risk that the actual traveling trajectory may deviate from the trajectory assumed in the preset and stored turn pattern due to, for example, slipping.

If the actual traveling trajectory deviates from the assumed trajectory, the position of the vehicle body relative to the first boundary will deviate significantly from the appropriate setting state at the time of completion of the turn, and the vehicle will enter the next work process. There is a risk that the first boundary may be out of the imaging field of view of the imaging means, or only a portion of the first boundary may be within the imaging field of view.

If the first boundary is detected when the first boundary is out of the field of view of the imaging means or only part of the first boundary is within the field of view, the position of the first boundary may be detected incorrectly and the vehicle body There is a possibility that the vehicle will not be able to automatically travel along the first boundary properly.

The present invention has been made in view of the above-mentioned circumstances, and its first purpose is to properly move the vehicle body to the next work stroke even if the traveling trajectory of the vehicle body deviates from a preset trajectory during a turn. The purpose is to enable the entry into the starting end.

A second purpose is to enable quick and automatic correction of the deviation from the first boundary even when the amount of deviation from the first boundary is large when approaching the starting end of the work stroke.

[Means to solve the problem]

The first characteristic configuration of the imaging type travel control device for an automatic traveling work vehicle according to the present invention is as follows.

That is, the control switching means determines, based on the detection information of the second boundary detection means, that the second boundary is located at a position on the near side of the set position in the longitudinal direction of the vehicle body within the imaging field of view of the imaging means. Accordingly, the turn control means is configured to be switched to the steering control means.

Further, the second characteristic configuration is as follows.

That is, the first boundary detection means is configured to detect an amount of displacement of the position of the vehicle body in the width direction of the vehicle body with respect to the first boundary or an amount of displacement of the inclination of the vehicle body with respect to the length direction of the first boundary. and when the amount of deviation of the vehicle body with respect to the first boundary detected by the first boundary detection means at the start of the control operation is larger than a set value, the steering control means controls the amount of deviation. The present invention is configured to perform steering control with a larger steering amount than when the amount of deviation is small until the amount of deviation becomes smaller than the set residue.

[For production]

Since the first boundary and the second boundary are located so as to intersect with each other, the first boundary at the starting end of the work stroke is located further away than the second boundary within the imaging field of view.

Therefore, in the first characteristic configuration, as the second boundary becomes located in a position on the near side of the set position in the longitudinal direction of the vehicle body within the imaging field of the imaging means, the turn control means changes to the steering control means. Switch to .

Further, in the second characteristic configuration, when the control operation of the steering control means is started, that is, the second boundary is located at a position on the near side of the set position in the longitudinal direction of the vehicle body within the imaging field of the imaging means. Therefore, if the amount of deviation of the vehicle body position with respect to the first boundary detected at the time of switching from the turn control means to the steering control means is large, the steering amount is larger than when the deviation amount is small. It controls the steering.

〔Effect of the invention〕

Therefore, according to the first characteristic configuration, for example, the turn is completed on the near side of the set turn end position, and the second
Even if the boundary is located in front of the set position in the longitudinal direction of the vehicle body within the imaging field of view of the imaging means, the operator may enter the starting end of the work process without seeing the first boundary or enter the imaging field of view. Since it is possible to avoid the possibility of entering the starting end of the work process with the length of the first boundary being extremely short, after the first boundary can be detected accurately,
Steering control can be started.

Furthermore, according to the second characteristic configuration, since the steering amount is automatically adjusted depending on the magnitude of the deviation from the first boundary, regardless of the deviation from the first boundary at the time of entering the next work process, The deviation with respect to the first boundary can be quickly corrected so that the vehicle body is in a proper state with respect to the first boundary.

Also, from the time of approach to the starting end of the work stroke, automatic driving is started in a state where the deviation of the vehicle body position with respect to the first boundary between the untreated work area and the treated work area in the vehicle body width direction is small. I was able to do it.

〔Example〕

Embodiments of the present invention will be described below based on the drawings.

As shown in Fig. 5, a rectangular unmown land (B) as an untreated work area is formed surrounded by a cut land (C) as a treated work area, and the unmown land ( The part from the - side to the opposite side of B) is set as one work process, and,
A plurality of parallel working strokes are set on the unmowed land (B) side adjacent to one working stroke at intervals corresponding to the working width of a working vehicle for mowing lawns, which will be described later.

As shown in FIGS. 5 and 6, in each work process, the unmoved area (B) and the already mowed area (
In order to be able to image the first boundary (L1) with the vehicle body (V), regardless of whether the first boundary (L1) with the vehicle body (V) is located on the left or right side of the vehicle body (V), the first boundary (L1) on the front side of the vehicle is Image sensors (S1) serving as imaging means for capturing images of locations corresponding to ) in two-dimensional directions are provided on each of the left and right sides of the front portion of the vehicle body (V).

That is, in each work process, the vehicle body (V) moves to the first boundary (L) between the uncut land (B) and the mowed land (C) in the width direction of the vehicle body.
1), the first boundary (L1
) The steering is controlled based on the positional information of the first boundary (L1) detected based on the imaging information of the image sensor (S1) located on the side of ) side that is not located on the image sensor (S
The end of one work process is determined based on the positional information of the second boundary (L2) between the unmoved area (B) and the mowed area (C) in the longitudinal direction of the vehicle body, which is detected based on the imaging information in 1). As the opposite side of the uncut land (B) is reached, the 18
This will result in a 0 degree turn.

Then, the uncut land (B) adjacent to the cut land (C) is
By repeatedly traveling back and forth in the section from the negative side to the opposite side, lawn mowing work in a predetermined range can be automatically performed in a so-called reciprocating mode.

In Fig. 6, (1) is the front wheel, (2) is the rear wheel, and (3) is the mower.

However, the front wheels (1) and the rear wheels (2) are configured in a so-called four-wheel steering type and four-wheel drive type so that both function as steering wheels and driving wheels. There is.

In addition, in each of the work steps, the vehicle is driven using a two-wheel steering system in which only the front wheels (1) are steered, and when turning for the next work step, the front and rear wheels (1) are operated.
1) and (2) are made to run using a four-wheel steering system in which the vehicle is steered in opposite phases and a parallel steering system in which the vehicle is steered in the same phase.

To explain the imaging 1f(A) of the image sensor (S,), as shown in FIG. 7, the vehicle body (V) is in a proper state with respect to the first boundary (L1). , the first boundary (L1) coincides with a reference line (La) passing through the widthwise center of the imaging field of view (8) on the ground surface of the image sensor (S1) in a direction along the vehicle body traveling direction. It is set so that it is in the state.

Note that the first boundary (L1) switches between being located on the left side of the vehicle body (V) and being located on the right side of the vehicle body (V) during the outbound and return trips of the work vehicle.

Therefore, the vehicle body (V) turns and the running direction is 180 degrees.
Each time the first boundary is inverted, the image sensor (Sa) used to image the first boundary (Sa) is switched to the left or right.

As will be described in detail later, the second boundary (L2) is detected using the unused image sensor (S1).

In the following description, the image sensor that is in the state of capturing the first boundary (L1) will be referred to as the first image sensor (Sa), and the image sensor that is in the state of capturing the second boundary (L2) will be referred to as the first image sensor. 2 image sensor (Sb).

However, in the figure, for convenience, the image sensor on the right side is expressed as the first image sensor (Sa), and the image sensor on the left side is expressed as the second image sensor (sb).
These first image sensors (S
a) and the second image sensor (Sb) are the vehicle body (V
) will be replaced every turn.

To explain the control configuration for automatically driving the vehicle body (V), as shown in FIG. Then, the first boundary (L1) and the second boundary (
L2) A control device (10) using a microcomputer is provided, which detects the respective position information and controls the traveling of the vehicle body (V) based on the detected boundary position information.

That is, a first boundary detection means (100) detects the first boundary (1,,) based on the imaging information of the first image sensor (Sa) using the control device (10);
a second boundary detection means (1) that detects the second boundary (L2) based on imaging information of the second image sensor (Sb);
01), a steering control means (10) that performs steering control so that the vehicle body (V) automatically travels along the first boundary (L,) based on the detection information of the first boundary detection means (100);
2), a turn control means (103) for moving the vehicle body (-) to the starting end of the next working stroke, and a turn control means (102) based on the detection information of the second boundary detection means (101). ) and the turn control means (103)
control switching means (
104) will be configured.

However, the first boundary (L1) is a preset distance until the vehicle body reaches a stroke end determination start distance (βa) (see Fig. 5), which is preset based on the length of one working stroke. It will be repeatedly detected each time the first boundary (shi) is detected, and while the first boundary (shi) is being detected, the second boundary will be detected repeatedly.
The boundary (L2) is not detected. That is, the second boundary (L2) is repeatedly detected at every set distance after the vehicle body (V) reaches the end-of-stroke determination start distance (!a).

The length of the set distance for repeatedly detecting the second boundary (L2) is determined by the image sensor (S1).
Within the imaging field of view (A), the second boundary (L2) at the current imaging point is farther away than the reference line (Lb) (see Figure 8) passing through the center of the imaging field of view (A) in the front and back direction. In some cases, even if the vehicle body (V) travels to the next imaging point, the second boundary (L) at the next imaging point
2) is set so that it is located within the imaging field of view (A).

In Fig. 1, (4) is the hydraulic cylinder for steering the front wheels (1), (5) is the hydraulic cylinder for steering the rear wheels (2), and (6) is the output of the engine (B). This is a hydraulic continuously variable transmission device that drives the front and rear wheels (1) and (2) by changing the speed of the vehicle. (7) is a speed change motor, (8) is a control valve for the front wheel hydraulic cylinder (4), (9) is a control valve for the rear wheel hydraulic cylinder (5), and (S2) is a control valve for the front wheel hydraulic cylinder (5). ) is a distance sensor for detecting the travel distance of the vehicle body (V) based on the output rotation speed of the vehicle body (V), (R1) is a potentiometer for detecting the steering angle for the front wheels, and (R2) is a steering angle detection for the rear wheels. The potentiometer (R3) is a vehicle speed detection potentiometer that indirectly detects the vehicle speed based on the operating state of the transmission (6).

Next, the operation of the control device (10) will be explained based on the flowchart shown in FIG.

However, which of the left and right image sensors (S1) should be used to image the first boundary (L1) must be set in advance before the start of travel, and the left and right image sensors must be switched each time the vehicle turns. become.

First, before the start of travel, work data setting processing such as the length of the work stroke and the number of work strokes to be traveled is performed, and after the work data setting processing is performed, the vehicle is started to travel at the set speed. .

After the start of travel, each time the vehicle travels a set distance, the first boundary (L,) is detected by boundary detection processing, which will be described later, based on the imaging information of the first image sensor (Sa), and its position is detected. Steering control will be performed based on the information.

However, as will be described in detail later, in the steering control based on the detection information of the first boundary (L1), only the front wheels (1) are steered.

Next, from the state of determining whether the vehicle body (V) has approached the end of the working stroke and detecting the first boundary (L1),
In order to switch to the state of detecting the second boundary (L2), the traveling distance (I2) of the vehicle body (V) is determined based on the length of each work stroke based on the detection information of the distance sensor (S2). It is determined whether a set stroke end determination start distance (βaH, see FIG. 5) has been reached.

The travel distance (I is the travel end determination start distance (1a)
, the vehicle body (
In order to detect the second boundary (L2) on the side of the work stroke end while moving the machine V) straight toward the work stroke end,
A stroke end determination process, which will be described later, is executed to determine whether or not the stroke end has been detected.

If the end of the stroke is not detected, the traveling distance is adjusted so that the vehicle body (V) can be moved to the starting end of the next working stroke even if the end of the stroke cannot be detected. (p) has reached a preset forced turn distance (j!b) (see Figure 5) based on the length of the working process, and determines whether the forced turn distance i(j'b ) or until the end of the stroke is detected, the stroke end determination process is repeated.

When the end of the stroke is detected, or the forced turn distance n(
J! When b) is reached, it is determined whether or not the work has been completed based on the number of strokes traveled, etc., and if the work has been completed, the travel is stopped and the entire process is completed.

If the work is not completed, the front and rear wheels (1), (
In a four-wheel steering system in which the vehicle body (V) is steered in opposite phases, the vehicle body (V) is maintained at the maximum steering angle for a set time required for the direction of the vehicle body (V) to reverse 180 degrees.
V) is turned 180 degrees to the next working stroke side to start the turn.

After inverting the vehicle body (V) by 180 degrees, the four-wheel steering type is switched to the parallel steering type, and the vehicle body (V) is rotated to the first position in the next work process.
The second boundary (L2) at the starting end of the work stroke is detected by executing the stroke end determination process while moving the parallel width toward the direction in which the boundary (L1) is located.

Then, based on the processing result of the stroke end determination process, it is determined whether or not the second boundary (L2), that is, the stroke start end has been detected, and as the stroke start end is detected, the second boundary (
When the detection of the first boundary (L1) is finished, the detection of the first boundary (L1) is started, and based on the detection information, the vehicle body (V) is
Correct the horizontal position of.

To explain the correction of the approach position, until the stroke approach position becomes appropriate, the position in the vehicle width direction with respect to the first boundary (L1) is within the set dead zone in the parallel width shifting state using the parallel steering method. You will have to operate the steering until the

After the approach position correction is completed, after updating the stroke data such as the number of strokes traveled, the detection of the first boundary (L,) and the steering control based on the detected information are restarted, and the next work is started. This will start automatic driving on the journey.

That is, after detecting the second boundary (L2) on the end side of the work stroke or reaching the forced turn distance (1'b),
The process from the start of a 180 degree turn of the vehicle body (V) to the detection of the second boundary (shi 2) on the starting end side of the work stroke corresponds to the turn control means (103), and the process After detecting the starting end, the approach position in the width direction with respect to the first boundary (L1) is corrected in a parallel steering manner, and the first boundary (L,) is moved until the end of the stroke # is detected.
The process of repeatedly controlling the steering based on the detected information corresponds to the steering control means (102).

Further, the process of switching between steering control and turn control based on the detection result of the second boundary (L2) at each of the work stroke # end and the work stroke start end by the stroke end determination process is a control switching means. (104).

Next, each means will be explained in detail.

First, the boundary detection process for detecting the first boundary (L1) will be described in detail based on the flowchart shown in FIG.

However, in the boundary detection described below, since the mowed land (C) appears brighter than the uncut land (B), it corresponds to the first boundary (L1) or the second boundary (L2). The difference in brightness between the mowed land (C) and the unmown land (B) in a part is larger than the difference in brightness within the mowed land (C) and the unmown land (B). It is designed to take advantage of the following:

Further, the second boundary (L2) is the same boundary between the uncut land (B) and the cut land (C) except that it crosses the first boundary (L1), and this First boundary (L1)
As described below, the second
The position information of the boundary (L2) is also detected.

That is, each time the vehicle body (V) travels a set distance,
Imaging information from the image sensor (S1) is manually input to the control device (10), and is converted into a density value in which the roughness of each pixel arranged in a two-dimensional direction is quantized at a setting stage. , the differential value of each pixel arranged in the two-dimensional direction is determined based on the density value of each of the eight neighboring pixels adjacent to each pixel.

However, when detecting the position of the first boundary (L1), the first boundary (L1) is located in the uncut area (B) in the width direction of the vehicle body.
) and the mown field (C), the differential value in the X-axis direction, which is the direction along the width direction of the vehicle body, is calculated, and when detecting the position of the second boundary (L2). In this case, the differential value in the y-axis direction, which is the direction along the longitudinal direction of the vehicle body, is calculated.

Then, pixels for which the obtained differential value of each pixel is larger than a set threshold value are extracted, and the image information is binarized.

After extracting pixels with large brightness changes through binarization processing, use Hough transform to find multiple straight lines that pass through the extracted pixels and have slopes set in multiple stages. One straight line that has the highest frequency among the plurality of straight lines is selected as the first boundary (L,) or the second boundary (L2).
It will be extracted as a straight line corresponding to .

To explain, in the Hough transform, the imaging field of view (
Using the X-axis passing through the center of A) as the reference line in the polar coordinate system, draw multiple straight lines passing through the extracted pixels in a range of 0 degrees to 180 degrees with respect to the X-axis based on the following formula (i). It is determined as a combination of the inclination (θ) set in advance in multiple stages and the distance (ρ) from the origin, that is, the center of the screen (see FIG. 9).

ρ=x'cosθ+y esinθ ・−・・・・(i
) Then, for one pixel, the process of adding two-dimensional histograms for counting the frequency of each straight line is repeated until the value of the slope (θ) set in the plurality of steps reaches 180 degrees. , the frequency of multiple types of straight lines passing through all extracted pixels is counted for each extracted pixel.

After counting the frequency of straight lines for all extracted pixels,
From the values added to the two-dimensional histogram, by finding the combination of the slope (θ) and the distance (ρ) that gives the maximum frequency, one straight line (LX) that gives the maximum frequency is obtained.
(see FIG. 9), and its straight line (LX) is determined as a straight line corresponding to the boundary on the imaging surface of the image sensor (S1).

Next, the straight line (L
The stored information on the shape and size of A) and the position of the pixel (a, b, c) on the imaging plane through which the straight line (Lx) with the maximum frequency passes
) (see Figure 9), it is converted into straight line information on the ground surface.

That is, when detecting the first boundary (L,),
As shown in FIG. 7, the inclination (ψ) with respect to the reference line (La) passing through the center of the width direction of the imaging field of view (A) in the front-rear direction
This is converted into information about a straight line on the ground surface, which is set as the value of and the position (δ) in the width direction.

To explain, as shown in FIG. 7, the lengths (β1), (β3°) of the two front and rear sides of the imaging field of view (A) in the direction intersecting the first boundary (L1), and the field of view The length (βIs) in the width direction of the imaging field of view (A) at the position of the pixel that images the center (x = 16. y = o) and the distance (h) between the two front and rear sides are actually measured in advance. Then, it is stored in the control device (10).

and the position (a, b) of the pixel where the straight line (LX) on the imaging plane intersects the X axis corresponding to the front and back two sides of the imaging field of view (A) (position where y=16. y=-16) The value of the X coordinate (Xl, X32) of and the straight line (Lx)
The value of the X coordinate of the pixel (X
+s) and (ii) below, which is a modification of equation (i) above.
Obtain from the formula.

However, Yi is 16.0. -16 will be substituted.

Then, based on the coordinate 1 axis on the X axis determined by the above equation (ii), the slope (ψ) with respect to the reference line (La) and the width direction are calculated from the following equations (iii) and (iv). Find the distance (δ) at
δ) is calculated as the positional information of the corrected straight line on the ground surface, that is, the straight line corresponding to the first boundary (L1).

Therefore, in the steering control described later, the reference line (La
) and the distance in the width direction (δ) are both brought closer to zero.

However, when detecting the second boundary (L2), the eighth boundary
As shown in the figure, the imaging field of view (A) with respect to the reference line (Lb) passing through the center of the imaging field of view (A) in the front and back direction in the width direction.
) at both left and right ends (P, ).

(P2) from the vehicle body (V) to the second boundary (L2).
The method is to obtain position information corresponding to the distance to.

To explain further, since the second boundary (L2) is detected as a straight line heading in the X-axis direction, the position (a) where the straight line with the highest frequency (Lx) passes through both the left and right ends of the screen,
(b), the imaging field of view (A
) Reference line (Lb) passing through the center in the front-rear direction l: vs. t 6
The distances n(P, ) and (P2) cD of both the left and right i parts are determined.

Therefore, although the details will be described later, in the stroke end determination process, the left and right distances (PI) with respect to the reference line (Lb),
Based on the value of (P2), it is determined whether the vehicle body (V) has reached the end or start of the working stroke.

That is, the boundary detection process described above is performed by the first boundary position detection means (100) for detecting the position of the first boundary (L,) between the untreated work area and the treated work area in the vehicle width direction.
It will correspond to Further, the first boundary position detection means (100) is used to detect the second boundary (L2) and to determine whether or not the stroke end has been reached based on the detection information, which will be described later. The determination process corresponds to the second boundary detection means (101).

Next, the stroke end determination process will be explained based on the flowchart shown in FIG.

That is, as the travel distance (A) reaches the travel end determination start path fi (A1), the process is started, and first, the travel distance is determined as a set position in the longitudinal direction of the vehicle body within the imaging field of view (A). The value of the reference value (S) for edge determination is initially set to zero, which corresponds to a state in which the second boundary (L2) is located at the center of the imaging field of view (A) in the front-rear direction.

After initializing the reference value (S) to zero, the imaging information of the first image sensor (Sa) is switched to the imaging information of the second image sensor (sb), and the first boundary detection means ( 100) to create the second boundary (L2)
Detect.

Next, the position information of the detected second boundary (L2), that is, the distance (PI) at both left and right ends with respect to the reference line (Lb) passing through the center of the imaging field of view (A) in the front and back direction, (P2
) is the first set value (10
by determining whether the second boundary (L2) is smaller than the reference line (Lb).
) is larger than a set allowable value.

To explain, if the slope of the second boundary (L2) with respect to the reference line (Lb) is greater than the set tolerance value, the straight line corresponding to the second boundary (L2) will be in the imaging field of view. (A), the position where the straight line intersects the left and right sides of the imaging field of view (A) becomes the upper and lower sides from the left and right sides, and the left and right sides with respect to the reference line (Lb). Distance (P1).

(R2) may not be able to be detected accurately, the reference line (L
If the slope of the second boundary (shi2) with respect to b) is larger than the set tolerance value, it is determined that it is a false detection and the position information of the detected second boundary is ignored. be.

If the absolute value of the value obtained by subtracting the values of both distances (PI) and (R2) is smaller than the first set value, that is,
If the slope of the second boundary (L2) with respect to the reference line (Lb) is smaller than the set tolerance, both distances (P
, ), (R2) is added to the second setting value (2
(set to 0).

In other words, at the next imaging point after the vehicle body (V) has traveled a set distance from the current imaging point, will the position of the second boundary (L2) detected this time be out of the imaging field of view (A)? It is to determine whether or not.

If the sum of the values of both distances (p1) and (R2) is smaller than the second set value, then both distances (P,
) and (R2) is smaller than the reference value (S).

In other words, it is determined whether the position of the second boundary (L2) detected this time is located at a position moved closer to the vehicle body than the position of the second boundary (L2) detected last time.

If the sum of the values of both distances (P, ) and (R2) is not smaller than the standard value (S), the standard value (S)
) is updated to a value obtained by adding the values of both distances (PI) and (R2), that is, a value corresponding to the currently detected position.

Therefore, when the reference value (S) is initially set to zero, if the sum of the values of both paths fi(p, ) and (R2) is smaller than the reference value (S), teeth,
The second boundary (L2) detected this time is closer to the vehicle body than the reference line (Lb) passing through the center of the imaging field of view (A) in the longitudinal direction, and when driving to the next imaging point, the second boundary (L,) It can be determined that there is a risk that the object may fall outside the field of view (A). In addition, in the case where the reference value (S) has been updated to the value obtained by adding the values of both distances (P, ) and (R2), the values of both distances (P, ) and (R2) are added. If the value is smaller than the reference value (S), the next detection position is closer to the vehicle body than the current detection position, and the second boundary (L2) is detected properly. It can be determined.

In other words, as the position of the detected second boundary (L2) becomes closer to the front side than the reference value (S), the vehicle body (V)
It is determined that the #1 end or the start end of the work stroke has been reached.

Note that if the sum of the values of the two roads n(P, ), (R2) is not smaller than the second set value, the two distances (PI), (R2) are adjusted while maintaining the straight-ahead state. until the sum of the values becomes smaller than the second set value, or
This end-of-stroke determination process is repeated until the travel distance (1) reaches the forced turn distance (1b).

To explain the steering control, the imaging field of view (
The first line with respect to the reference line (La) passing through the center of the width direction of A)
The position (δ) of the boundary (L1) in the width direction, the inclination (ψ) with respect to the length direction, and the current steering angle (θ) of the front wheel (1)
), the front wheel (1
), and the steering angle (θ) of the front wheel (1) detected by the front wheel steering angle detection potentiometer (R1) is set as the target steering angle (θf). direction angle (
The control valve (8) of the steering hydraulic cylinder (4) is driven so that the steering hydraulic cylinder (4) is within the set dead zone with respect to θf).

θf=to, 2 to δ−I−, 3 to ψ10, θ...
・(V) [Separate flow side] In the above embodiment, the approach position at the starting end of the work stroke was corrected by parallel width adjustment using a parallel steering type. The steering control may be performed using a two-wheel steering system with the amount of steering increased.

That is, as shown in Fig. 10, when setting the work data and starting traveling, first, the position deviation hole gain flag is set (set to "1") to set the steering amount to a large amount. , the first boundary (L,) is detected, and the absolute value of the position (δ) in the width direction with respect to the reference line (La) as its position information is determined as a set value (d) for determining the magnitude of the deviation amount. ), and if it is smaller than the set value (d), the position deviation hole gain flag is reset (set to "0"), and then the position deviation hole gain flag is reset (set to "0"). It is determined whether the gain flag is in the set state or not.

When the position deviation hole gain flag is set, the above (V
) is transformed as shown in the following equation (V), and the target steering is adjusted so that the steering amount is twice that of the case where the position deviation hole gain flag is in the reset state. The angle (θf) is set.

However, when the position deviation hole gain flag is in a reset state, the target steering angle (θf) is set based on the above equation (V).

θf = 2, 2 for δ+, 3 for ψ+...vi) After setting the target steering angle (θf) to the large and urinary angle based on the position deviation hole gain flag. Similar to the steering control, the front wheels (1) are steered to a set target steering angle (θf), and the end of stroke is determined by the end of stroke determination process; Then, as the end of the stroke is detected, it is determined whether or not the work has been completed, and if the work has been completed, the vehicle is stopped running and the entire process is completed.

If the work is not completed, similarly to the above embodiment, after executing the turn control to turn the machine 180 degrees and move it to the starting end of the next work stroke, detect the first boundary (L,) and its detection. The steering control process is repeated while setting the eye Ll direction angle (θf) to the distance between the eyes and the toilet based on the information.

Note that the position (δ) in the width direction with respect to the first boundary (L1) is
), the target steering angle (θ[) may be set and changed in two levels depending on the magnitude of the inclination (ψ). Alternatively, it may be set in two stages, large and small, based on both. Furthermore, if the position (δ) or the inclination (ψ) is larger than the set value, the target steering angle (θ) determined by the above equation (V)
The value of f) may be increased by a set ratio, and the specific configuration for increasing the amount of steering can be modified in various ways.

Further, in the above embodiment, the position of the second boundary (L2) detected next after traveling the set distance is the second boundary (L2) detected this time.
By determining whether the position of the second boundary (L2) at the next imaging point is closer to the vehicle body than the position of L2), it is determined whether the position of the second boundary (L2) at the next imaging point is at the position estimated based on the imaging information at the current imaging point. Although we have exemplified a case in which a determination is made as to whether or not the process has reached the end of the stroke or the start of the stroke, the configuration may also be configured to check the difference between the current detected position and the next detected position. The specific configuration for doing so can be modified in various ways.

Further, in the above embodiment, the steering control means (102) is a two-wheel steering type in which only the front wheels (1) are steered. The position (δ) may be corrected by parallel steering, and the inclination (ψ) may be corrected by four-wheel steering, and the specific configuration can be modified in various ways. Similarly, the specific configuration of the turn control means (103) can also be changed in various ways. For example, there are modes in which the direction of travel is not reversed and the vehicle travels back and forth repeatedly for each stroke, and
It may also be implemented in a form in which the vehicle travels in a so-called circular travel mode in which the direction is changed by 0 degrees.

Further, in the above embodiment, the present invention is applied to a work vehicle for mowing lawns, and each of the first boundary detection means (100) and the second boundary detection means (101) is connected to an unmown area (B). Already mowed land (C)
Although the case is illustrated in which straight lines on the ground surface corresponding to the first boundary (L1) and the second boundary (L2) are found, each of the boundaries (L1).

The specific configuration for detecting (L2) can be changed in various ways. Furthermore, the boundaries to be detected and the specific form of the work area can be changed in various ways.

Incidentally, although reference numerals are written in the claims section for convenient comparison with the drawings, the present invention is not limited to the structure shown in the accompanying drawings.

[Brief explanation of the drawing]

The drawings show an embodiment of the image two travel control device for an automatic traveling work vehicle according to the present invention, in which FIG. 1 is a block diagram of the control configuration, FIG. 2 is a flowchart of travel control, and FIG. 3 is a flowchart of boundary detection processing. Flowchart, Figure 4 is a flowchart of stroke end detection processing, Figure 5 is a plan view of the work area, Figure 6 is a side view of the imaging field of view, and Figure 7 is the relationship between the boundary in the longitudinal direction of the vehicle body and the imaging field of view. FIG. 8 is a plan view showing the relationship between the boundary in the longitudinal direction of the vehicle body and the imaging field of view, FIG. 9 is a schematic diagram of the Hough transform, and FIG. 10 is a flowchart of the actual running control. (S,)...imaging means, (A)...imaging field of view, (B)...untreated work area, (C)...
... Treated work area, (V) ... Vehicle body, (L
1)...First boundary, (L,)...Second
Boundary, (100)...First boundary detection means, (1
01)...Second boundary detection means, (102)...
... Steering control means, (103) ... Turn control means, (104) ... Control switching means.

Claims (1)

[Claims] 1. Imaging means (S_1) that images the work area on the front side of the vehicle in two-dimensional directions, and unprocessed images in the width direction of the vehicle body based on the imaging information of the imaging means (S_1). Work area (B)
a first boundary detection means (100) for detecting a first boundary (L_1) between
Based on the imaging information of 1), the second boundary (L_2) between the untreated work area (B) and the treated work area (C) in the longitudinal direction of the vehicle body is
), the vehicle body (V) detects the first boundary (L_
1), a steering control means (102) for controlling the steering based on the detection information of the first boundary detection means (100), and a steering control means (102) for controlling the steering of the vehicle body (V) based on the detection information of the first boundary detection means (100), so that the vehicle body (V) automatically travels along A turn control means (103) for moving the vehicle body (V) to the starting end; and a control operation of the turn control means (103) as the vehicle body (V) reaches the terminal end of each work stroke; The steering control means (102) is controlled based on the detection information of the second boundary detection means (101) so that the steering control means (102) performs a control operation as the starting end of the next work stroke is reached. ) and the turn control means (103
) and control switching means (104) for selectively switching the operating state to the operating state. Based on the detection information of the means (101), the second boundary (L_2) is detected by the imaging means (S_1).
As the vehicle is located in the front side of the set position in the longitudinal direction of the vehicle within the imaging field of view (A), the steering control means (102) changes from the turn control means (103).
) An imaging type travel control device for an autonomous driving work vehicle configured to switch to 2. The first boundary detection means (100) detects the amount of displacement of the vehicle body (V) in the vehicle width direction with respect to the first boundary (L_1) or the displacement amount of the vehicle body (V) with respect to the length direction of the first boundary (L_1). The steering control means (102) is configured to detect the amount of deviation in the inclination of the vehicle body (V), and the steering control means (102) detects the first boundary detected by the first boundary detection means (100) at the start of the control operation. The vehicle body (V
) is larger than the set value, the steering control is performed with a larger steering amount than when the deviation amount is small until the deviation amount becomes smaller than the set value. The imaging type travel control device for an automatic traveling work vehicle according to claim 1.