JP7616191B2 - Paddy field farming machine - Google Patents

Description

The present invention relates to paddy field working machines such as rice transplanters.

A rice transplanter capable of variable fertilization is known (Patent Document 1).

By the way, since the fertilizer reduction rate is not uniformly determined based on the fertility of the field, but rather is determined based on the relative values within each rice field, the first thing that happens when entering the rice field is a process called variable fertilization teaching, in which the average depth and fertility of the field is calculated, and the fertilizer reduction rate is determined from these values, taking into account the variations in the depth and fertility of the field. When doing so, it is preferable to carry out variable teaching in a place that is not on the periphery or in a corner of the rice field.

On the other hand, research and development of robotic rice transplanters has been underway in recent years. For this reason, in order to first obtain information about the shape and position of the field, path generation teaching is often carried out by manually traveling around the perimeter of the field.

JP 2021-023154 A

As mentioned above, it is known that the location or position for performing variable fertilization teaching is more appropriate for estimating the average value for a field when it is slightly inside the field rather than on the outer perimeter or edge.

However, the reality is that the conventional techniques described above do not take into account the timing of execution of path generation teaching and variable fertilization teaching.

The present invention takes into consideration the problems inherent in the past and aims to provide a paddy field farming machine that can appropriately time the execution of path generation teaching and variable fertilization teaching.

The purpose is to provide.

The first aspect of the present invention is
A paddy field working machine including a control unit and a variable fertilizer application device, the control unit automatically traveling in a field along a predetermined route to perform field work,
The predetermined route is generated by manually traveling along at least a portion of the perimeter of the field through route generation teaching,
The control unit can automatically execute variable fertilization teaching for the variable fertilization device,
During the route generation teaching by the manual traveling, the control unit does not perform the variable fertilization teaching,
The variable fertilization teaching is performed using a route for variable fertilization other than the generated predetermined route after the route generation teaching is completed and before automatic traveling along the generated predetermined route is performed;
The path for variable fertilization is determined using the result of the path generation teaching;
The field is a quadrilateral field, and the path generation teaching is performed using three sides of the field;
After the path generation teaching is completed, variable fertilization automatic teaching is performed while traveling diagonally from the corner (1d) of the completed location to the diagonal corner (1b) of the field as the outward route (I4),
A free run is performed on the return path (I5) from the diagonal corner (1b) to the diagonal corner (1d),
The paddy field working machine is characterized in that the outward path (I4) and the return path (I5) are offset from each other and do not run along the same route.
The second aspect of the present invention is
The predetermined route along which the automatic travel is performed is a plurality of straight-line routes parallel to the final side of the route generation teaching performed along three sides,
The automatic traveling is started from a path (I4') adjacent to an inner path adjacent to a third side, which is the final side, among the plurality of linear paths,
A point on the side of the corner (1d) of the path adjacent to the inside of the third side, which is the final side of the path generation teaching, of the linear path is set as point (P1);
A point on the corner (1d) side of a path adjacent to a path (I4') adjacent to the path adjacent to the inside of the third side is defined as point (P4);
A point on the corner (1b) side of the straight path adjacent to the first side of the path generation teaching is set as point (P3);
A point on the corner (1b) side of a path adjacent to the path adjacent to the first side on the further inside is defined as point (P2);
Variable fertilization automatic teaching is performed on the outward journey from point (P1) to point (P2),
This is the first paddy field working machine of the present invention, which, after turning, runs idle on the return journey connecting point (P3) to point (P4), returns to point (P4), turns, and moves to a path (I4') adjacent to the path adjacent to the inside of the third side.
The first invention related to the present invention is
A paddy field working machine including a control unit and a variable fertilizer application device, the control unit automatically traveling in a field along a predetermined route to perform field work,
The predetermined route is generated by manually traveling along at least a portion of the perimeter of the field through route generation teaching,
The control unit can automatically execute variable fertilization teaching for the variable fertilization device,
This is a paddy field working machine in which, while the path generation teaching is being performed by manual driving, the control unit does not perform the variable fertilization teaching.

The second invention related to the present invention is
The variable fertilization teaching is performed using a route for variable fertilization other than the generated predetermined route after the route generation teaching is completed and before automatic traveling along the generated predetermined route is performed;
The path for variable fertilization is determined using the results of the path generation teaching in a paddy field working machine of a first invention related to the present invention.

The third invention related to the present invention is
The field is a quadrilateral field, and the path generation teaching is performed using three sides of the field;
The control unit causes the machine to run freely from the entrance/exit of the field along the remaining side that is not used for the path generation teaching, and from the corner it reaches, executes the path generation teaching using the three sides, and from the point where it returns to the entrance/exit, causes the machine to perform the variable fertilization teaching while running diagonally across the field, and then causes the machine to automatically run from the corner it reaches on the diagonal using the generated predetermined path, which is a paddy field working machine of a second invention related to the present invention .

The fourth invention related to the present invention is
The automatic variable fertilization teaching is performed during traveling along a predetermined part of the generated predetermined route,
This is a paddy field working machine of the first invention related to the present invention , which also has a function for an operator to manually teach variable fertilization at any location in the field.

The fifth invention related to the present invention is
This is a paddy field working machine of a fourth invention related to the present invention, in which if the variable fertilization teaching is performed manually while the path generation teaching is being performed, the result data of the variable fertilization teaching is reset.

The sixth invention related to the present invention is
the predetermined route includes a round trip route in a central part of the field and an inner circumferential route running around the round trip route,
In a paddy field working machine of a fifth invention related to the present invention , the predetermined portion of the route along which the automatic variable fertilization teaching is performed is a portion of the round trip route.

The seventh invention related to the present invention is
The automatic variable fertilization teaching is performed in the second stroke of the reciprocating path,
This is a paddy field working machine according to the sixth invention related to the present invention , which is used only once from the start to the end of rice planting.

The first aspect of the present invention provides an excellent effect that variable fertilization teaching is not performed at the edge of a field when the robot is automatically traveling. Furthermore, since variable fertilization teaching can use routes other than the route set by route generation teaching, it is possible to select a route that increases the accuracy of measurements such as the average fertility of the field. Furthermore, since the outbound and return routes do not overlap in variable fertilization teaching, it is possible to prevent the field from being damaged.
According to the second invention, in addition to the effects of the first invention, after variable fertilization teaching is completed, the robot can immediately move to a straight line reciprocating automatic travel from that location.
The first invention related to the present invention has the excellent effect of preventing variable fertilization teaching from being performed at the edge of a field, etc., when the robot is automatically traveling.

According to a second aspect of the present invention , variable fertilization teaching can use routes other than those set by route generation teaching, making it possible to select a route that increases the accuracy of measurements such as the average fertility of a field.

According to a third aspect of the present invention , it is possible to prevent field damage caused by variable fertilization teaching.

According to a fourth aspect of the present invention , variable teaching can be performed during automatic reciprocating travel, which increases convenience.

According to a fifth aspect of the present invention , when variable fertilization teaching is performed during path generation teaching, such data can be reset since there is a high possibility that the location is at the edge of the field.

According to a sixth aspect of the present invention , the route along which variable fertilization teaching is performed is an automatic reciprocating route, and variable fertilization teaching can be prevented from being performed at the edge of a field.

According to the seventh invention related to the present invention , variable fertilization teaching is performed in the second stroke of the round trip path, thereby preventing variable fertilization teaching at the edge of the field and reducing the process of conventional fertilization.

FIG. 1 is a side view of a rice transplanter, which is an example of a paddy field working machine according to an embodiment of the present invention. Configuration diagram of the control unit of the rice transplanter A drawing explaining the route (part 1) of the rice transplanter in the field A drawing explaining the route of the rice transplanter in the field (part 2) A drawing explaining the route of the rice transplanter in the field (part 3) Another explanatory drawing of the travel path (part 3) of the rice transplanter in the field A drawing explaining the running route (part 4) of the rice transplanter in the field A drawing explaining the conventional route of a rice transplanter in a field A drawing explaining the route the rice transplanter takes in the field A drawing explaining the travel route of the rice transplanter in a deformed field.

The following describes in detail the embodiment of the present invention with reference to the drawings.

Figure 1 is a left side view of a paddy field working machine according to an embodiment of the present invention, and Figure 2 is a block diagram of the paddy field working machine according to an embodiment of the present invention. In the following, forward/backward, left/right, and up/down are defined based on the traveling direction of the paddy field working machine.

The paddy field working machine of this embodiment is a so-called robot working machine, and its vehicle body 100 is a rice planting machine that travels on a traveling device 220 having a pair of left and right front wheels 221 and rear wheels 222 in response to manual or automatic steering operation of a steering device 240, levels the field with a ground leveling device 260 having a ground leveling float 261, plants seedlings in the field with a seedling planting device 230, and fertilizes the field with a variable fertilizer applicator 250.

The traveling device 220, the seedling planting device 230, the variable fertilizer applicator 250, and the soil leveling device 260 are driven by the power of the engine 210, which is transmitted via the main transmission 300, which is an HST.

The controller 500 is a computer that controls the vehicle body 100, engine 210, running device 220, seedling planting device 230, steering device 240, main transmission 300, etc. as described above. The memory unit 500a is a means for storing data related to the travel route, etc. The vehicle body 100 is also equipped with a GNSS device that can receive communications from the GNSS 600. Also, 400 is a fertility sensor that detects the fertility, depth, temperature, etc. of the field.

Below, we will explain how to perform path generation teaching to generate a predetermined path for the robot to travel automatically in a field, and then how to use that generated path to travel automatically, using a rectangular field as an example.

In FIG. 3, the path generation teaching starts from the entrance 1a of the field 1, and proceeds along the first path I1 while manually steering to plant seedlings using the seedling planting device 230. When the first corner 1b is reached, the direction is changed to the right, and the second path I2 is proceeded, and when the second corner 1c is reached, the direction is changed again to the right, and the third path I3 is proceeded to the third corner 1d. Through this manual travel along the three sides, the control unit 500 acquires the shape and position information of the field 1 and stores it in the memory unit 500a. In this way, the path generation teaching is completed.

The routes generated in this way are an inner route one step inward from the outer routes I1, I2, and I3 (dotted line in Figure 7; also called the robot headland work route), and multiple parallel straight reciprocating routes I4, I5, and I6 (Figure 3) in the center of the field 1, which are further inward from the inner route, and turning routes connecting these routes, etc.

After that, as a general rule, the vehicle 100 that has completed the path generation teaching skips the inner path just inside I3 (Figure 3) and moves to the first step I4 (Figure 3) of the round trip path, where it travels automatically while planting, then turns and moves to the second step I5 of the round trip path, where it travels automatically while planting.

After repeating this process, it moves to the inner peripheral path immediately inside the first step I1 (Figure 3) of the outer peripheral path, then moves to the inner peripheral path immediately inside the second step I2 (Figure 3) of the outer peripheral path, and then moves to the inner peripheral path immediately inside the third step I3 (Figure 3) of the outer peripheral path, traveling to corner 1d and ending the automatic traveling.

Finally, by manually steering, the machine travels along the remaining side 1e of the four sides of the rectangular field 1 that was not traveled during the path generation teaching, and returns to the entrance/exit 1a.

Next, we will explain variable fertilization teaching using the fertility sensor 400.

<A case where variable fertilization teaching is performed during automatic driving of a predetermined route obtained by route generation teaching driving>
The variable fertilization device 250 uses data such as the fertility, depth, and temperature of the field 1 obtained by the variable fertilization teaching to automatically perform variable fertilization thereafter, and performs variable fertilization teaching as a preparation for this while automatically traveling in the first step I4 (FIG. 3) of the linear reciprocating path in FIG. 1. In other words, the fertility of the location is sensed in this step I4 (FIG. 3).

Estimating the overall fertility of field 1 is more accurate when variable fertilization teaching is performed closer to the inside of field 1 than near the ridges. The variable fertilization teaching described above uses process I4 (Figure 3), which is located further inside field 1 than the inner route immediately inside the outer route, making it appropriate for implementing variable fertilization teaching.

Until the variable fertilization teaching is completed, the conventional amount of fertilizer is applied during the seedling planting work. Therefore, it is not desirable for the process using the conventional amount of fertilizer to be too long. For example, as in I6 in Figure 3, if variable fertilization teaching is not performed until quite close to the center of field 1, the fertilization process using the conventional amount of fertilizer will be too long. However, as another example, process I5 (Figure 3), which is one step inside process I4 (Figure 3), could also be used. Alternatively, variable fertilization teaching could be performed using two processes, process I4 (Figure 3) and process I5 (Figure 3).

The control unit 500 is programmed in advance to determine which part of the automatic straight-line reciprocating route such variable fertilization teaching will be performed on. This is possible because the route is generated by the route generation teaching as described above.

Furthermore, during step I4 (Figure 3), it is also possible to omit the portion near the ridges of field 1 and sense the fertility of as much of the central part of field 1 as possible. This is possible because the control unit 500 has route information. Also, it goes without saying that fertility sensing is not performed during automatic turning, because the turning location is close to the ridges of field 1.

It is possible to execute variable fertilization teaching manually, but if manual variable fertilization teaching is performed while path generation teaching is being performed, the result data of that manual variable fertilization teaching will be reset. This is because variable fertilization teaching will be performed in a location close to the ridges of the field.

It is also advisable to conduct variable fertilization teaching only once, rather than multiple times, because this makes it difficult to apply conventional fertilizer amounts.

<A case where variable fertilization teaching is performed on a route other than the specified route obtained by route generation teaching driving>
A case will be described where variable fertilization manual teaching is performed from the entrance/exit prior to route generation teaching that starts from the entrance/exit.

As shown in Figure 4, manual variable fertilization teaching is performed starting from the entrance/exit 1a. That is, first, the vehicle body 100 is manually run diagonally across the field 1 (dotted line I1 in Figure 4), and variable fertilization manual teaching is performed on the return path to the entrance/exit 1a along the solid line I2 (Figure 4). In this way, the vehicle returns to the entrance/exit 1a, and from there, the path generation teaching is performed as described above.

b. After the route generation teaching that starts from the entrance/exit is completed, manual or automatic teaching of variable fertilization is performed from the entrance/exit before automatic travel of the generated predetermined round-trip automatic route.

(Part 1) As shown in Figure 5, as described above, path generation teaching is performed from the entrance/exit 1a along paths I1, I2, and I3 to reach corner 1d. This generates a specified automatic round trip path and stores it in the control unit 500. Then, after the path generation teaching is completed, variable fertilization automatic teaching is performed on the outgoing path I4 (Figure 5) diagonally from the corner 1d on the opposite side of the entrance/exit 1a to the diagonal corner 1b. In this case, since the specified path has already been formed, the control unit 500 can calculate the corner 1b that is the destination of the diagonal outgoing path I4 (Figure 5), making it possible to automatically and accurately perform variable fertilization automatic teaching. Of course, it may also be run diagonally manually.

In Fig. 5, the automatic variable fertilization teaching is performed on the outbound route I4, and the vehicle runs idle on the return route I5. By not running the same route on the outbound route I4 and the return route I5, even with each other, it is possible to prevent damage to the field.

More specifically, as shown in FIG. 6, if the point on the inside of one stroke of corner 1d at the end point of the path generation teaching is P1, the point on the corner 1b side of the inner path immediately inside the first outer stroke of the path generation teaching in the round trip path is P2, the point on the inner path immediately adjacent thereto is P3, and the point adjacent to point P1 is point P4, variable fertilization automatic teaching is performed on the stroke connecting points P1 and P2, and after turning, the robot is allowed to idle on the stroke connecting points P3 and P4, returning to point P4.

Then, by turning as shown by the arrow and moving to automatic round trip path I4' and traveling there, one stroke of the inner path is opened and normal round trip automatic straight traveling becomes possible. In this way, turning from point P2 to P3 and turning from point P4 to the round trip path becomes smoother.

(Part 2) As shown in Figure 7, first, the robot is made to run free to the corner 1d on the opposite side of the entrance/exit 1a (route I1), and then normal route generation teaching is performed from there, I2, I3, and I4, until the robot reaches the entrance/exit 1a. After the route generation teaching is completed, manual or automatic variable fertilization teaching is performed from the entrance/exit 1a before the robot is made to run automatically round trip along the generated specified route. This makes it possible to omit the return route of the variable fertilization teaching.

Specifically, as shown in FIG. 7, the vehicle body 100 is manually run free in the opposite direction from the entrance/exit 1a (I1), and then the vehicle manually plants around the perimeter for three strokes while teaching the path generation (I2, I3, I4). The vehicle then returns to the entrance/exit 1a, from which point variable fertilization teaching is performed. Because the path has been generated, teaching is possible automatically. It can also be done manually.

That is, the control unit 500 drives the vehicle body 100, which has returned to the entrance/exit 1a, toward the diagonal corner 1c. In detail, the vehicle body 100 drives from point P1 next to the entrance/exit 1a toward point P2 on the corner 1c side of the immediately inner reciprocating stroke I7 of the immediately inner reciprocating stroke I6 of the inner peripheral stroke immediately inside the first stroke I2 of the route generation teaching, while performing automatic variable fertilization teaching. Note that since the route generation teaching has already been completed, the control unit can calculate the target P2 of the destination of the route I5, making such automatic variable fertilization teaching possible automatically.

After reaching P2 on the diagonal line, it can turn from that position and move to automatic reciprocating path I6, opening up one path around the inner circumference (the robot headland work path shown by the dashed dotted line) and enabling automatic reciprocating travel.

Therefore, with this method, only one route is required for variable fertilization teaching, which causes less disturbance to the field.

In the above example, variable fertilization teaching was performed automatically, but it can also be performed manually.

Below, we explain the conventional manual, variable fertilization teaching method for non-automatic robots, and an example of adapting it to an automatic robot.

Figure 8 shows the operation of the conventional, specified variable fertilization teaching. In other words, automatic measurement of variable fertilization teaching begins in the third stroke, including the idle run. Conventional fertilization (no reduced fertilization) is performed from the start of rice planting until the end of measurement. After automatic measurement is completed, variable fertilization is performed. If the "Start manual measurement" button is not touched, automatic measurement is performed as is. Variable teaching is performed only once between the start and end of rice planting.

Figure 9 shows an example of adapting a method of predetermined variable fertilization teaching, which is not used for conventional robots that run automatically, to a robot that runs automatically.

In other words, during the path generation teaching process, which teaches the shape and size of the field, the count is reset to 0 when the machine direction is changed. Variable fertilization teaching is not performed during this path generation teaching. Automatic measurement for variable teaching begins in the second process, which includes a dry run. From the start of rice planting to the end of measurement, conventional fertilization (no reduced fertilization) is used. After automatic measurement is complete, variable fertilization is performed. If the "Start manual measurement" button is not touched, automatic measurement will be performed as is.

I will explain this in more detail.

When teaching the robotic rice transplanter to generate a route, it first plants three strokes around the perimeter. In this case, if the three sides of the field are roughly the same distance apart, the conventional automatic teaching of the specified variable fertilization described above would start with three strokes, and variable fertilization teaching that avoids the targeted edge of the field would not be possible.

Therefore, in the predetermined conventional variable fertilization second stroke detection process, the previous stroke distance and the current stroke distance are set (reset) to 0 when the steering wheel is turned, thereby avoiding early start of automatic teaching of variable fertilization.

We will explain this in more detail.

In variable fertilization teaching, the field soil depth, average SVF and standard deviation are calculated in the third step of the work. There are two types of variable fertilization teaching work: automatic teaching, which is performed by automatically detecting in the third step, and manual teaching, where the worker starts and ends teaching at a location of their choice. This automatic teaching is performed as follows. Note that automatic teaching is performed only once between the start and end of rice planting.

1. 2nd stroke detection The work stroke is judged by the left rear wheel rotation sensor. During automatic ascent/descent mode, if the steering is straight ahead and an input from the left rear wheel rotation sensor is detected, an increment is made. If the steering is straight ahead and the reverse switch is detected and an input from the left rear wheel rotation sensor is detected, the pulse ldspls_dc_KT for stroke number detection is decremented.

1-1. Before detecting the second stroke When the steering is turned, if the robot is in teaching mode, ldspls_KT is reset to 0. When the steering is turned, if ldspls_dc_KT is 700 (approximately 7 m) or more, ldspls_dc_KT_sav = ldspls_dc_KT is saved as the distance of the previous stroke, and if |ldspls_dc_KT_sav - ldspls_dc_KT| < 300 (difference is within approximately 3 m), the straight-line distances are deemed to be almost the same and the second stroke is determined to be complete. After that, ldspls_dc_KT is reset to 0.

1-2. After detecting the second step When ldspls_dc_KT>300, it is determined to be the third step and teaching begins. The "Start manual measurement" button on the tablet changes to "Acquiring data." (Start of teaching) The "Planting in progress" display on the work status screen on the tablet changes to "Measuring reference data." When ldspls_dc_KT>ldspls_dc_KT_sav-300, it is determined to be in front of the ridge and teaching ends.

The "Data acquisition" button on the tablet will change to "Start manual measurement." (Teaching ends) The "Reference data measurement in progress" display on the tablet's work status screen will change to "Rice planting in progress."

2. Confirmation process during actual work Confirm the second process detection and teaching during actual work in the following procedure.

Lower the planting section and move straight for about 10 m. Turn the steering wheel to turn and align the adjacent row. Lower the planting section again and move straight for 10 m. Turn the steering wheel to turn and align the adjacent row. Lower the planting section again and move straight for about 3 m. The "Start manual measurement" button on the tablet changes to "Acquiring data". (Start of teaching) The "Planting in progress" display on the work status screen on the tablet changes to "Measuring reference data". After that, move straight for about 4 m and teaching will end. The "Acquiring data" button on the tablet changes to "Start manual measurement". (End of teaching) The "Measuring reference data" display on the work status screen on the tablet changes to "Planting in progress".

Furthermore, we will explain it from a different angle.

When the three sides of a field are roughly the same distance apart, such as a square field, the first time the steering wheel is turned after the path generation teaching is complete, it is determined to be the second stroke, and there is a problem that the stroke determined to be the second stroke may differ depending on the shape of the field.

Therefore, the second stroke detection will not be performed the first time the steering wheel is turned after the path generation teaching is completed.

This allows the second step to be detected (automatic teaching) in the same process regardless of the field shape.

Next, after the path generation teaching, the robotic rice transplanter skips one process and turns to plant. As a result, the process in which automatic teaching is carried out ends up being further inside the field than normal variable fertilization work, which creates the problem of reducing the area that can be variable fertilized. Note that fertilization is carried out using conventional methods until variable teaching is complete.

Therefore, when the steering wheel is turned, the previous travel distance is stored as before, and the first steering operation after the route generation teaching is completed is determined to be the second travel.

This allows for variable automatic teaching on the first run.

Next, in the case of a deformed rice field as shown in Figure 10, the robot skips one path, so the difference between the previous path distance and the path distance traveled in this automatic teaching becomes large. In the case of a field shape in which the path distance gradually becomes shorter, there is a high possibility that the robot will not reach the normal automatic teaching completion distance (previous path - 3 m) when it reaches the edge.

Therefore, we will add steering wheel turning operation as an OR condition to the conditions for ending automatic measurement.

This allows the steering wheel to be turned even in irregularly shaped rice fields, completing automatic teaching as soon as the machine turns at the edge of the field.

Next, in the case of irregularly shaped rice fields, the robot skips one path, so there is a large discrepancy between the previous path distance and the path distance traveled in this automatic teaching. In the case of field shapes where the path distance gradually becomes shorter, there is a high possibility that the robot will not reach the normal automatic teaching completion distance (previous path - 3 m) when it reaches the edge.

Therefore, an OR condition will be added to the conditions for ending automatic measurement when the planting part is raised.

This allows the planting area to be raised even in irregularly shaped rice fields, completing automatic teaching when the machine turns at the edge of the field.

Next, in the case of irregularly shaped rice fields, the robot skips one path, so there is a large discrepancy between the previous path distance and the path distance traveled in this automatic teaching. In the case of field shapes where the path distance gradually becomes shorter, there is a high possibility that the robot will not reach the normal automatic teaching completion distance (previous path - 3 m) when it reaches the edge.

Therefore, the condition for ending automatic measurement will be added as an OR condition when the rotor clutch is turned off.

This allows the rotor clutch to be disengaged even in irregularly shaped rice fields, i.e. the automatic teaching is completed when the machine turns at the edge of the field.

Next, when teaching variable fertilization, you want to obtain the standard deviation and average value of the data for the entire field, so it is said that manual teaching should be done diagonally so that as much overall data as possible can be obtained.

The conventional method is to manually teach the tractor from the entrance to the opposite side, then (as with conventional work) drive the tractor free on the headland while keeping the marker out so as not to damage the field, and then turn around to perform planting work.

However, current robot teaching (path generation) must be done while planting from the entrance/exit side, meaning conventional methods cannot be applied.

In addition, the method of performing manual variable fertilization teaching before starting route generation teaching requires manual back and forth from the entrance/exit to the opposite side, which places a burden on the worker.

Therefore, when the route generation teaching is completed, the outbound route (see I4 in Figure 5) for performing variable fertilization teaching is determined.

This means that when teaching the path generation, a path for teaching variable fertilization is also generated, and teaching for diagonal running can also be performed automatically. In this case, since the robot is taught while planting on the three sides other than the ridge on the seedling supply side, there is no risk of generating a path outside the field or teaching variable fertilization while traveling an abnormal path.

Next, as mentioned above, when teaching variable fertilization, it is necessary to obtain the standard deviation and average value of the data for the entire field, so it is said that manual teaching should be done diagonally so that as much overall data as possible can be obtained.

The conventional method is to manually teach the tractor from the entrance to the opposite side, then (as with conventional work) drive the tractor free on the headland while keeping the marker out so as not to damage the field, and then turn around to perform planting work.

However, route generation teaching (route generation) must be done while planting from the entrance/exit side, so conventional methods cannot be applied.

In addition, the method of performing manual variable fertilization teaching before starting route generation teaching requires manual back and forth from the entrance/exit to the opposite side, which places a burden on the worker.

Therefore, when the path generation teaching is completed, a return route (see I5 in Figure 5) is generated that returns to the initial working position from the completion of variable fertilization teaching.

As a result, when route generation teaching is performed, a route for teaching variable fertilization is also generated, and diagonal running teaching can be performed automatically. In this case, route generation teaching is performed while planting on the three sides other than the ridge on the seedling supply side, so there is no risk of generating a route outside the field or performing variable fertilization teaching on an abnormal route.

Next, as described above, when a variable fertilization teaching route is generated, if automatic steering is started in a state in which automatic steering can be started after route generation teaching is completed and variable fertilization teaching is not completed, driving will be performed along the above-mentioned outbound route of variable fertilization teaching (see I4 in Figure 5). In this way, variable fertilization teaching driving is performed automatically.

In addition, in the above example, after the variable fertilization teaching is completed by completing travel along route I4, the vehicle travels along route (I5), which is the return route from the completion of the generated variable fertilization teaching back to the initial working position, as described above.

In this way, after the variable fertilization teaching run is completed, it can automatically return to the initial working position.

If variable fertilization teaching has already been completed, there is no need to use automatic steering to go through the teaching process again. Therefore, if automatic steering is started in a state in which automatic steering can be started after path generation teaching is completed and variable fertilization teaching has been completed, automatic driving can be performed by following the driving path of a conventional robotic rice transplanter.

Next, we will explain the driving locations in more detail, with some overlap.

As described above, when performing variable fertilization teaching as shown in Fig. 6, the start point of the variable fertilization teaching outward route is set to the end point P1 of the final route of the headland work route. This makes it possible to determine the variable fertilization teaching outward route. In Fig. 6, the end point of the variable fertilization teaching outward route is set to the end point P2 on the opposite side to the seedling supply side of the work route one before entering the headland work route. This makes it possible to accurately determine the variable fertilization teaching outward route.

In this way, the outbound route for variable fertilization teaching is the route connecting points P1 and P2.

Next, for the return route, in Fig. 6, the start point of the return route of the variable fertilization teaching is set to the end point P3 of the first headland work route. The return route of the variable fertilization teaching can be determined. Also, the end point of the return route of the variable fertilization teaching is set to the start point P4 on the seedling supply side of the next work route from the work start route. This allows the return route of the variable fertilization teaching to be determined.

In this way, the return path of the variable fertilization teaching is the route connecting points P3 and P4. By making the outward and return paths different in this way, the following drawbacks can be prevented:

In other words, if the outbound and return routes overlap, the machine will travel over the same spot twice, damaging the field. Also, when turning from the outbound route to the return route, aligning the machine to face the same start and end points will cause difficulties in turning the machine. Furthermore, since the machine ultimately needs to be aligned with the previous work start position, if the previous work start position is used as a route point, it will be difficult to turn from the end point of the return route.

Next, the method described above reduces the number of routes that need to be manually driven, but it means going back and forth diagonally through the middle of the field, which is not desirable because it damages the field.

As a result, as shown in Figure 7, an outbound route (I5) is generated for carrying out variable fertilization teaching when the route generation teaching is completed. In this way, only one diagonal run is required, and there is less risk of damage.

Regarding the position of the travel route, the starting point of the variable fertilization teaching route is set to the starting point P1 of the first headland work route (see Figure 7). This allows the outbound route for variable fertilization teaching to be determined.

The end point of the variable fertilization teaching route is set to end point P2, which is on the opposite side to the seedling supply side of the next work route of the new work start route. This allows the outbound route of the variable fertilization teaching to be determined.

The variable fertilization teaching route will be the route connecting points P1 and P2. There will be no need for a return route.

As shown in Figure 7, when a route is generated for such variable fertilization teaching, the route for starting work is not a work route that skips one route from the final teaching route as in the past, but a work route that skips one route from the initial teaching route.

This allows the robot to automatically travel from an adjacent work path after performing variable fertilization teaching in a single diagonal run. From then on, it can perform the same operations as conventional robots, and even for the final manual planting process, it can generate a path that goes straight toward the entrance while planting (see Figure 7).

If the variable fertilization teaching outward route and the automatic driving operation start route overlap, the machine will travel over the same place twice, damaging the field. Also, when turning from the variable fertilization teaching outward route to the operation start position, aligning the machine with the same start and end points will cause strain on the machine's turning.

Therefore, as shown in Figure 7, it is desirable to generate the variable fertilization teaching path and the work start path for the automatic robot driving so that they do not intersect with each other.

Next, we'll look at another example.

Normally, "start planting" for variable fertilization and "start teaching" for the robotic rice transplanter are managed separately. Looking at the functionality of the robotic rice transplanter, the tablet app manages it by starting to measure work data from the point at which it recognizes that the robot's "teaching start" has begun, in order to plot the robot's travel path on the tablet. Separately, looking at the functionality of variable fertilization, when the "start planting" operation is performed using the tablet or on the robot itself, it begins to measure work data for each field, assuming that variable fertilization work will now be performed in a new field.

In the case of a model that has both the functions of a robotic rice transplanter and variable fertilization, even if the robot's "teaching start" is performed, data for each field will not be recorded unless the "planting start" operation for variable fertilization is also performed. However, if there are two "start" operations, there is a risk that the operator will forget which operation to perform.

Therefore, in the configuration of a robotic rice transplanter with variable fertilization, if the robotic rice transplanter's "automatic mode" is in the teaching (field shape teaching) state when it is not between the "start of rice planting" and "end of rice planting" stages of variable fertilization, that is, when variable fertilization work has not started (data is not being collected by the tablet), it will be forcibly switched to the "start of rice planting" state.

By doing this, the start of robot teaching can be considered as the start of work in a new field, and rice planting with variable fertilization can be started, reducing tedious work and preventing forgetting to operate the robot.

In other words, the start condition for the machine to "start planting" is when one of the following conditions is met.

(Pattern 1)
When all of the following conditions are satisfied:

1. Variable mode judgement is this unit mode (1)
2. Press the green button on the monitor (start planting)

(Pattern 2)
When all of the following conditions are satisfied:

1. Field data collection completion judgment is waiting (0)
2. Automatic mode teaching (3)
In the above case, it may seem best to start robot teaching when the "start planting" operation is performed to prevent forgetting to operate the device, but the "start planting" operation for variable fertilization is the timing for starting data measurement for each field, so while it does not matter when it is performed to a certain extent, the "teaching" of the robot must be started from a location of your choice.

Therefore, the robot is configured so that teaching does not start automatically even if the "start rice planting" operation is executed.

This prevents robot teaching from starting at an unintended time.

In addition, if the signal is restricted, the tablet cannot recognize the "start planting" signal. In the above case, the tablet is configured to accept the "start planting" signal from the unit even when in "unit mode" where it does not accept operations from the unit.

As a result, even when the mode is set to the main unit mode, which only accepts operations from the main unit, the tablet can start teaching the robot and measure work data for each field.

Next, we'll look at another example.

In rice transplanters that have a function to obtain position information from a GNSS sensor and electrically adjust the amount of fertilizer applied, and a sensor that detects the number of times the seedling tank has moved sideways, the amount of fertilizer applied is adjusted using the difference between the measured value calculated from the number of times the seedling tank has moved sideways, the set distance between plants, and the number of times the tank has been moved sideways, and the distance obtained by the GNSS sensor as the slip amount. This allows for the application of an appropriate amount of fertilizer, which is environmentally friendly.

In rice transplanters that have the ability to obtain position information from GNSS sensors and electrically adjust the amount of fertilizer applied, when the travel distance measured by the GNSS sensor reaches a reference distance, the amount of fertilizer applied is converted from the roll rotation speed, and if there is a difference from the target amount of fertilizer applied, the amount of fertilizer applied is adjusted in proportion to the difference. This allows the target amount of fertilizer to be achieved.

The dial for settings is below the steering wheel, allowing it to be placed in an easy-to-operate location. Alternatively, the monitor for settings is in front of the steering wheel, allowing it to be placed in an easy-to-operate location.

The present invention can realize a paddy field operating machine that can appropriately time the execution of path generation teaching and variable fertilization teaching, making it ideal for rice transplanters and the like.

1. Field
100 Vehicle body 230 Seedling planting device 250 Variable fertilizer application device 400 Fertility sensor etc. 500 Control unit (controller)
500a Storage unit 600 GNSS
I1 to I7 driving route

Claims (2)

A paddy field working machine including a control unit and a variable fertilizer application device, the control unit automatically traveling in a field along a predetermined route to perform field work,
The predetermined route is generated by manually traveling along at least a portion of the perimeter of the field through route generation teaching,
The control unit can automatically execute variable fertilization teaching for the variable fertilization device,
During the route generation teaching by the manual traveling, the control unit does not perform the variable fertilization teaching,
The variable fertilization teaching is performed using a route for variable fertilization other than the generated predetermined route after the route generation teaching is completed and before automatic traveling along the generated predetermined route is performed;
The path for variable fertilization is determined using the result of the path generation teaching ;
The field is a quadrilateral field, and the path generation teaching is performed using three sides of the field;
After the path generation teaching is completed, variable fertilization automatic teaching is performed while traveling diagonally from the corner (1d) of the completed location to the diagonal corner (1b) of the field as the outward route (I4),
A free run is performed on the return path (I5) from the diagonal corner (1b) to the diagonal corner (1d),
The paddy field working machine is characterized in that the outward path (I4) and the return path (I5) are offset from each other so that they do not run along the same route . The predetermined route along which the automatic travel is performed is a plurality of straight-line routes parallel to the final side of the route generation teaching performed along three sides,
The automatic traveling is started from a path (I4') adjacent to an inner path adjacent to a third side, which is the final side, among the plurality of linear paths,
A point on the side of the corner (1d) of the path adjacent to the inside of the third side, which is the final side of the path generation teaching, of the linear path is set as point (P1);
A point on the corner (1d) side of a path adjacent to a path (I4') adjacent to the path adjacent to the inside of the third side is defined as point (P4);
A point on the corner (1b) side of the straight path adjacent to the first side of the path generation teaching is set as point (P3);
A point on the corner (1b) side of a path adjacent to the path adjacent to the first side on the further inside is defined as point (P2);
Variable fertilization automatic teaching is performed on the outward journey from point (P1) to point (P2),
The paddy field working machine according to claim 1, which, after turning, is run free on the return journey connecting point (P3) to point (P4), and after returning to point (P4), turns and moves to a path (I4') adjacent to the path adjacent to the inside of the third side .

Images