JP2583583B2 - Automatic rudder control device for agricultural work machine - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention is intended to enable a farming machine such as a rice transplanter to be run substantially in parallel along a so-called crop row already planted in a field and arranged in a row. The present invention relates to a structure of an automatic steering control device that performs the following operations.

[Conventional technology]

Conventionally, for example, when planting seedlings in a field by a rice transplanter,
The rice transplanter is provided with a planting mechanism at appropriate intervals on the left and right in the direction of travel, and the planting mechanism that rotates up and down as the rice transplanter progresses is used to plant the seedling mats on the seedling mounting table in the field scene while appropriately dividing the seedling mat by the number of plants. Therefore, in the field scene, at the appropriate seedling planting interval along the traveling direction of the rice transplanter, the planting seedling locations are lined up at the same time, it is possible to be planted in multiple rows at appropriate intervals in the left and right direction to the traveling direction. It is well known.

Japanese Patent Application Laid-Open No. 62-61509 discloses a prior art of an automatic steering device that enables a rice transplanter to run in a direction substantially parallel to a seedling row already planted in a field (hereinafter referred to as a crop row). With a color video camera mounted on the rice transplanter to be imaged, an appropriate range of the crop row in the adjacent portion is imaged, and this image information is binarized to extract a region corresponding to each planted crop location. Thereafter, a straight line is approximately calculated from the row composed of the plurality of regions by a process such as Hough transformation, and a virtual line on this calculation and an arbitrary reference line such as a vertical and horizontal center line of the imaging screen and a lateral shift with respect to a reference point are calculated. It has been proposed to execute the steering control of the airframe so that the gap of the inclination shift falls within a certain allowable range.

[Problems to be solved by the invention]

Since the imaging means mounted on an agricultural working machine such as a rice transplanter as in the prior art is used to image the plant crop row while moving forward along the crop row, the following performance for the crop row is good. Therefore, for example, in the case of unmanned automatic steering control, if the operator creates a crop row curved left and right at the time of the first sample traveling, the meandering steering according to the curved curve crop row of the sample will also be performed during the subsequent unmanned copying control. Therefore, there was a problem that the straight running performance was not good.

In order to solve this problem, an attempt has been made to improve the straight running performance of the agricultural work machine by utilizing the detection value of the geomagnetic sensor in addition to the detection of the imaging means and the copying sensor described above. If there is a source, for example, an iron product such as another agricultural working machine, there is a problem that the detection value of the geomagnetic sensor is likely to be out of order.

In addition, the management machine for applying fertilizer or spraying the chemical at an appropriate time after the seedling planting operation performs the work while proceeding along the already planted crop row, and in the combine, the uncut portion and the already cut portion are combined. Since the harvesting and threshing work is performed while moving forward along the substantially linear boundary line, the same problem as described above occurs even when such an agricultural work machine is equipped with an automatic steering device.

An object of the present invention is to solve this problem.

[Means for solving the problem]

Therefore, the present invention relates to a device for automatically steering an agricultural work machine such as a rice transplanter so as to travel in a substantially parallel manner on the side along a row of crops already planted in a field, Features such as copying image capturing means for capturing an image of a crop row, and distant view image capturing means for capturing an image of a target located far away from the agricultural work machine, and features such as binarization of each image information obtained by each image capturing means. Instruction amount determining means for performing image processing on each of the extracting means and determining a steering operation instruction amount based on both the deviation of the characteristic of the crop row with respect to the reference such as straight traveling of the agricultural work machine and the deviation of the characteristic of the target object with respect to the reference. It is a thing.

[Functions and effects of the invention]

In the present invention, since the two imaging means are so-called optical detection means, there is no failure of a movable part as in a conventional contact-type sensor, and there is no distance limitation,
In addition, unlike the geomagnetic sensor, it is possible to reliably detect the crop row and the target without being affected by other magnetic sources.

The one copying imaging means is for photographing a crop row near the agricultural work machine, and by processing the image information, the direction of the crop row is parallel or inclined with respect to the traveling direction of the farm work machine. It is possible to determine the state of the farm row and the degree of approaching or moving away from the side of the farm work machine, that is, a deviation from a reference when the farm work machine runs at a predetermined distance from and parallel to the crop row.

The other distant imaging means captures an image of a target located at a position far from the agricultural work machine, for example, a distant mountain or building, and extracts features of the target, thereby obtaining the target with respect to a reference when the agricultural work machine travels straight. Can also be determined.

And about the range where the agricultural work machine moves in the field,
The deviation of the target at a distant position from the reference when the farm work machine goes straight does not change rapidly.

Therefore, when the steering operation instruction amount is determined from the deviation between the two characteristics obtained in this way, it is possible to travel so as to follow the crop row at a position close to the normal agricultural work machine and not lose the crop row. However, by considering the deviation of the result of the feature extraction by the distant view imaging means, the straightness of the agricultural work machine is also improved, and the meandering as in the case of steering only considering the deviation in the crop row is considered. Running is lost.

In other words, the shortcomings of the detection by the imaging means for copying are compensated by the detection by the imaging means for distant view, and the automatic steering in which the agricultural work machine such as the rice transplanter is run parallel to the side of the crop row, such as straight ahead, at an appropriate distance. Can be executed reliably.

〔Example〕

In the following, an embodiment applied to a rice transplanter will be described. In FIG. 1, reference numeral 1 denotes a traveling body supported by front wheels 3, 3 on the front left and right sides and rear wheels 4, 4 on the rear left and right sides. At the rear, a multi-row planting seedling device 7 including a seedling mounting table 5 and a plurality of planting mechanisms 6 is mounted via a link mechanism 8 so as to be able to move up and down.

The power of the engine 9 mounted on the upper surface of the traveling body 1 is transmitted via the clutch 10 and the transmission case 11 to the front and rear wheels 3,
4 while transmitting power to the seedling plant 7 via a PTO shaft 12 protruding from the transmission case 11. Reference numeral 13 denotes an actuator for ON / OFF of the clutch 10, 14
Is an actuator for traveling speed change, and 15 is an actuator for PTO shaft speed change.

The steering mechanism 18 is rotated via a steering handle 17 provided in front of a control seat 16 of the traveling body 1 to change the direction of the front wheels 3, 3 to the left and right. Is an L-shaped steering arm 20 mounted horizontally on a rotation fulcrum shaft 19 of the steering mechanism 18, a pair of left and right tie rods 21 and 21 connected to the steering arm 20, a hydraulic cylinder 22, It comprises a control valve 23 for steering and a pitman arm 25 capable of swinging back and forth of a steering gear box 24 for operating the steering control valve 23.

The control valve 23 is connected to the steering arm 20 backward through a ball joint, and a spool at the rear end of the control valve 23 is connected to the pitman arm 25.

A rear end of a hydraulic cylinder 22 that rotatably connects a front end to the traveling body 1 is rotatably connected to the steering arm 20 and swings in accordance with a rotation angle of the steering handle 17. The pitman arm 25 moves the spool of the control valve 23 forward and backward, sends the hydraulic pressure from the hydraulic pump 26 driven by the engine 9, moves the piston rod in the hydraulic cylinder 22 out of the way, and responds to the rotation of the steering arm 20. And change the direction of the left and right front wheels 3,3.

The hydraulic cylinder 22 is also driven by an electromagnetic solenoid type automatic steering control valve 27. At this time, the steering angle of the front wheel 3 is detected by a potentiometer 28 attached to a rotation fulcrum shaft 23. You can know by doing.

The automatic steering control valve 27 is actuated by an output signal of a steering controller 31 driven by a central control device 30 for automatic steering and traveling.
The actuator 13 for F, the actuator 14 for traveling shift, and the actuator 15 for PTO shaft shifting are also operated by the traveling controller 32 driven by the central controller 30.

The detection device 33 of the present invention includes a tracing imaging unit 34 for imaging a crop row, a distant imaging unit 35 for imaging a distant target M such as a building or a mountain far from the agricultural work machine, Feature extracting means 36 and 37 for extracting features such as binarization of image information taken by means 34 and 35, and information (data) required by processing the features by these feature extracting means 36 and 37
And an image processing controller 38 for exchanging with the central control device 30.

The two imaging units 34 and 35 are turned around an arm or a standing column protruding to the left side (or right side) of the traveling body 1 via a posture maintaining unit (not shown) such as a stepping motor. The imaging means for copying 34 is set movably, and the reference line Ko of the imaging travel is set so as to be parallel to the traveling direction on the side of the traveling body 1 and obliquely downward and forward, and the imaging means for distant view 35 It is attached to the upper end of a column or the like standing upright from the traveling machine body 1 and set so that a distant view does not deviate from its imaging range.

The imaging units 34 and 35 are area sensors having a two-dimensional plane having a horizontal and vertical spread such as an xy plane, as in a so-called video camera, when detecting an object. In a device incorporating a two-dimensional MOS image sensor or two-dimensional CCD image sensor, an image formed through a lens is sensed by each image sensor (photoelectric device) arranged in a two-dimensional array on the image forming surface. The information on the imaging screen 40 can be output as an electric signal.

These imaging means 34 and 35 may be either color or black and white.However, by using the copying imaging means 34 for imaging a crop row for color, it is possible to distinguish between a field scene and a crop row. Its characteristics can be clearly recognized.

Next, a schematic flowchart (FIG. 4) of processing executed by the detection device 33 and the central control device 30 will be described.

First, in step S1 following the start, after setting an initial value, in step S2, the operator of the agricultural work machine rotates the traveling machine 1 to travel along the side of the crop row already planted in the field scene. enter.

Next, in steps S3 and S4, the two types of imaging means 34 and 35 are simultaneously activated to start an imaging operation, and in steps S5 and S6, image data of the respective imaging means 34 and 35 is captured. FIG. 5 shows a state where the planted crop location (NAE) is imaged on the imaging screen 40, and FIG. 7 shows a state where a distant target object (M) such as a mountain is imaged.

Next, in step S7, a binarization process for extracting and discriminating the planted crop location (NAE) from another field scene is performed.

In the present embodiment, when the copying imaging means 34 is used for color, the color light of the RGB color system [red (R), green (G), and blue (B) is used as primary color light, and white is obtained by adding light. ], The characteristics of the field scene are extracted from the signals of the respective red, green and blue components, and the signal output ratio of the green (G) component to the sum (R + G + B = 1) of the signal outputs of the three color components Is determined to be a seedling (crop), and the area (N) is segmented from another portion of the imaging screen 40 (Segmentation), and a binarization process is performed.

In addition, binarization may be executed by a color difference process in which a portion where the color difference image data (GB) obtained by subtracting the blue component from the green component of the color line segment has a predetermined output or more is determined as a seedling.

Regions (N1), (N2), (N3) of a plurality of planted crop locations obtained by binarizing the image information of the copying imaging means 34
Since ‥‥ is simultaneously displayed on the imaging screen 40, in step S8, after performing a calculation for determining the coordinates of the position of each binarized planted crop location, the respective regions with respect to the reference line Ko on the imaging screen 40 are determined. The lateral shift deviation ρ of (N1), (N2), (N3) ‥‥ and the variation Δρ thereof are calculated.

That is, a reference line Ko is taken in parallel with the axis (Y axis) of the traveling body 1 in the traveling direction, the X axis is taken in the width direction of the traveling body 1, and planted at every pitch (P) along the traveling direction of the traveling body 1. The distance from each area (N1), (N2), (N3) ‥‥ of the crop location to the point where it intersects perpendicularly to the reference line Ko is defined as the lateral deviation deviation ρ. The difference between the lateral shift deviations in the adjacent areas is defined as a change amount Δρ.
For example, the lateral deviation deviation ρ1 in the area (N1) and the area (N
The change Δρ when the lateral deviation deviation ρ2 in 2) is =
ρ2−ρ1.

In FIG. 6, when the regions (N1), (N2), and (N3) ‥‥ are on the right side of the reference line Ko on the imaging screen 40, ρ> 0
Ρ <0 at the left. In addition, the imaging screen 40
Δρ> 0 when the region moves away from the reference line Ko as going upward, and ρ <
Set to 0.

On the other hand, the image information of the distant view imaging means 35 can be used for either color or monochrome. Then, in step S9, in order to extract the contour features of the target object (M) such as a mountain or a building,
The feature of the target object (M) is extracted through an edge detection process (see FIG. 8) for detecting a portion where the gray value of the image changes rapidly, a region division process for dividing the image into a region in which the value can be regarded as substantially constant, and the like. The binarization process is performed.

Then, as shown in FIG. 9, an arbitrary feature point Mo on the target (M) from the reference point O, which is the intersection of the reference line Y on the imaging screen 41 and an arbitrary horizontal axis (for example, the X axis at the lower edge).
The angle between the line segment to the reference line Y and the azimuth deviation θ is calculated (step S10).

As described above, the copying imaging unit 34 and the distant view imaging unit 35 simultaneously obtain image data, and simultaneously execute the processing from the image processing to the feature extraction, and then execute the data storage unit, the calculation unit, and the comparison unit. In the central control device 30 including the following steps, a steering operation instruction amount s0 is determined according to a control theory such as fuzzy control based on the plurality of deviation amounts ρ, Δρ, θ according to a flowchart from step S11 to step S13 below.
And outputs a steering output signal to a steering controller 31, for example, an automatic steering control valve 27 for driving the steering to rotate.

In this case, in the fuzzy control, the two imaging means 3
In order to execute the image processing of the image information obtained in 4, 35 at high speed, and to compensate for the loss of the accuracy of the detection of crop rows and the detection of distant targets, the detection values with rough accuracy, that is, vague A practical control output can be obtained from the input, and the relationship between the condition to be detected in the control and the control amount, in other words, the relationship between the input and the output can be strictly modeled and described. It is suitable for control in difficult cases.

Further, the fuzzy control is suitable for a case where a plurality of seemingly unrelated objects such as a row of crops and a distant target, such as the present invention, are used as a control condition part.

Next, fuzzy control using fuzzy inference will be described.

Generally, in control by fuzzy inference, a control algorithm is described as an ambiguous relationship between a plurality of input variables of information for control, for example, two input variables (x, y) and an output (manipulation amount) z to a control device. Is what you do.

 For example, if x is small and y is large, let z be medium.

 If x is large and y is medium, z is increased.

, The control algorithm is (if ‥‥
It is expressed by what is called a fuzzy control rule of (if-then) format. The if ‥‥ part of the rule is the antecedent part,
The part of then ‥‥ is called the consequent part.

Now, in applying this fuzzy control rule to the control of automatic steering, in the present embodiment, the data set of the lateral deviation deviation ρ of the crop row and the rate of change Δρ of the deviation is set as the antecedent part, and the automatic steering operation is performed. Nine fuzzy control rules from 1-1 to 1-9, with the command amount s as the consequent part, and a data set of the lateral deviation deviation ρ and the azimuth deviation θ as the antecedent part, the operation instruction of automatic steering The fuzzy control rule with the quantity s as the consequent part is 2-1.
Table 1 (Nos. 1 and 2) shows a total of 18 rules from 9 to 2-9. Here, the rule 1-1 is that if the label of ρ is 0 and the label of Δρ is 0, s is 0
(Go straight).

From Table 2 to Table 4, fuzzy variables taken by each input variable of each deviation ρ, Δρ, θ and output variable of the operation instruction amount s are represented by discrete fuzzy variables discretized into an integer region. Things.

The fuzzy variable region, which is an ambiguous region, is defined by giving a degree (grade) in which elements (members) of the entire set of input variables are included in the region (variation). The function that gives is called the membership function. For example, the region of the fuzzy variable of ρ is defined by giving the degree (grade) in which the elements (members) of the entire set of lateral deviations ρ are included in the region (variation).

The (value) at the top of each table indicates a domain of each variable. For example, -5 to 5 in Table 2 are values taken by the input variable ρ.

In the example, the label of the fuzzy variable in each table is
"Large right", "Small right", "0", "Small left",
There are five types, "largely left", and the degree to which the value of the variable is included in the set of these labels (fitness = membership value) is represented stepwise by an integer from 0 to 10.

In the following, ρ = 3 (unit 1 / 0.67 pixel), Δρ = 2 (unit
A method of controlling fuzzy inference in the present embodiment when 1 / 0.67 pixel) and θ = −3 (unit 1/20 rad.) Will be described.

The fuzzy control rules that are established from the set of ρ = 3 and Δρ = 1 are Table 2, Table 1 from No.1-1, No.1-4, No.1-8, and ρ = 3 Table 2 shows the fuzzy control rules that are established from the set of と and θ = -3 from No. 2-1, No. 2-2, No. 2-4,
No. 2-8.

In No. 1-1 of the fuzzy control rule, the label of ρ as the antecedent part takes “0”. At this time, the degree of conformity of the membership function at the label “0”, which can take the variable ρ = 3, is “3” from Table 2. Similarly, the label of Δρ as the antecedent part takes “0”. At this time, the fitness of the membership function at the label “0”, which can be taken by the variable Δρ = 1, is “7” from Table 3.

The smaller value “3” of the fitness values “3” and “7” of the two membership functions is referred to as “if ρ is 0 and Δρ is 0” in No. 1-1 of the fuzzy control rule. The degree of conformity to the conditions of the entire antecedent part.

Further, the label of the operation instruction amount s given by the consequent part (inference result) of “No. 1-1” of the fuzzy control rule “0 to s” is “0”. The upper limit of the fitness of the discrete membership function is set to “3” of the fitness of the entire antecedent (see FIG. 10 (a)).

After executing a total of 7 processes up to the fuzzy control rules No. 2-8 in the same manner as described above, the operation instruction amount s obtained from each rule is synthesized. The combination is obtained by superimposing the respective operation instruction amounts s and adopting the maximum value.

The final output value so is the value of the center of gravity of the set of membership functions of s thus obtained (so = -1.71)
(See FIG. 12).

Referring to the above tables, these membership functions are shown in FIG. 10 (a) for fuzzy control rule No. 1-1.
The same applies to fuzzy control rules No. 1-4.
Fig. 10 (b), Fuzzy control rule No. 1-8
FIG. 11 (c) shows the fuzzy control rule No. 2-1 shown in FIG. 11 (a).
Fig. 11 (b), Fuzzy control rule No. 2-4
FIG. 11C and fuzzy control rules No. 2-8 are shown in FIG. 11D, respectively.

In this case, for example, the degree of conformity of the operation instruction amount s of the fuzzy control rule No. 1-1 is a region surrounded by a thick line below the height “3”.

The hydraulic cylinder 22 that changes the rotation angle of the steering arm 20 in the steering mechanism by outputting an output signal to the central control device 30 and operating the electromagnetic solenoid of the steering control valve 27 in accordance with the value of the final steering operation instruction amount so The vehicle is driven to correct the steering, and automatic steering control is performed on the side of a predetermined crop row so that the traveling body 1 moves forward in parallel along the crop row.

In this case, it is determined whether or not the front wheel 3 has a steering angle that is already inclined rightward or leftward with respect to the traveling direction by a potentiometer 28 that detects the rotation angle of the steering arm 20. This can be used for convergence control that progresses in parallel.

Further, in a farming operation in a place such as a large-scale field in which a plot is arranged, the farming machine is easy to go straight because the field is rectangular in a plan view. Therefore, in the fuzzy control, the azimuth of a distant target with respect to a reference line is determined. If the setting of so-called weighting is changed so as to increase the degree of fitness of the membership function of the deviation θ, the automatic steering control using the target in the distant view as a mark can be enhanced.

In addition, in order to strongly perform automatic steering control with a target in the distant view as a mark, the operator sets a fuzzy control switch so that only the antecedent of the combination of ρ and θ in the fuzzy control rules. You may switch.

As described above, if the automatic steering control using the target in the distant view as a mark is strengthened, the straight traveling performance of the agricultural work machine is greatly improved.

On the other hand, in the place where the ridge of the field is curved to the left and right, in order to use the deviation from the detection result of the crop row strongly,
Fuzzy control switches are switched so that only the antecedent of the combination of ρ and Δρ is obtained, and so-called weighting is performed so that the membership function of the azimuth deviation θ with respect to the reference line of the distant target is reduced. , The automatic steering control using the distant target as a mark can be weakened.

If the fuzzy control is employed in the automatic steering control as in the present embodiment, it is possible to perform control using a plurality of different detection targets (in this embodiment, deviations as inputs) and various combinations of input conditions as a condition part. Furthermore, the degree of adaptation of the input can be very easily changed by the control software.

[Brief description of the drawings]

1 is a plan view of a riding type rice transplanter, FIG. 2 is a side view, and FIG. 3 is a block diagram of a steering / running automatic control device and an operation explanatory diagram including a hydraulic circuit. FIG. 4 is a flowchart, FIG. 5 is a view of an imaging screen by the imaging means for copying, FIG. 6 is a view of an imaging screen binarized, FIG. 7 is a view of an imaging screen by the imaging means for distant view, FIG. FIG. 8 is a diagram of an image obtained by extracting the feature of the target, FIG. 9 is a diagram showing the azimuth deviation of the target, and FIGS. 10 (a), (b) and (c) are diagrams of the case of the fuzzy control rule 1 FIG. 11 (a), (b), (c) and (d) are diagrams of the membership function in the case of the fuzzy control rule No. 2, and FIG. 12 is the determination of the steering operation instruction amount. It is explanatory drawing which shows a method. 1 ... traveling body, 2 ... frame, 3,4 ... wheels, 5 ...
... Seedling stand, 6 ... planting mechanism, 7 ... seedling planting equipment, 9 ... engine, 11 ... mission case, 17 ... steering handle, 20 ... steering handle, NAE ... planted crop location, 19
… Rotating shaft, 20… steering arm, 22… hydraulic cylinder, 27… automatic steering control valve, 30… central control device, 31… steering controller, 32… travel controller, 33… detection Apparatus, 34 ... Copying imaging means, 35 ... Distant view imaging means, 36, 37 ... Feature extraction means, 38 ... Image processing controller.

Claims (1)

(57) [Claims] 1. An apparatus for automatically steering an agricultural work machine such as a rice transplanter so as to travel substantially parallel to a side along a row of crops already planted in a field. It is equipped with a copying imaging means for imaging a row, and a distant view imaging means for imaging a target at a position far from the agricultural work machine, and performs feature extraction such as binarization on each image information obtained by each imaging means. Means for image processing by each means, and an instruction amount determining means for determining a steering operation instruction amount from both the deviation of the characteristic of the crop row with respect to the reference such as the straight traveling of the agricultural work machine and the deviation of the characteristic of the target object with respect to the reference. An automatic steering control device for an agricultural work machine.