KR102298784B1 - Artificial intelligence variable complex tillage robot for weed management of upland crop - Google Patents

Abstract

In accordance with the present invention, provided is a cultivating robot for upland crop weed management. The cultivating robot includes: a base (100) having a body (110) equipped with an AI module (50) using a planting characteristic of a field sensed by a sensor part (20) as an input layer, and using control signals of a rotary driving member (350) and a robot driving member (310) as an output layer; a first wheel (130) and a second wheel (140) located on the base (100); a main rotary (210) connected with the rotary driving member (350) provided in the base (100), located on the base (100) and enabling topsoil cultivation; and a pair of transverse rotaries (220) connected with the rotary driving member (350), located on both sides of the main rotary (210), and enabling topsoil cultivation. The AI module (50) controls the rotary driving member (350) through the control signal to control the driving of the main rotary (210) and the transverse rotaries (220), and controls the robot driving member (310) to control the driving of the first wheel (130) and the second wheel (140). Therefore, the present invention is capable of removing weeds without time constraints.

Description

Artificial intelligence variable complex tillage robot for weed management of upland crop

The present invention relates to the agricultural field, and more specifically, to a variable complex tillage robot including an artificial intelligence module and removing field weeds using a method of tilling topsoil.

In field farming, weeds are one of the important factors that reduce yield along with pests and diseases. When weeds grow over a certain amount, it is essential to remove weeds because there is a problem with growth and yield due to competition, such as taking away nutrients from crops planted in the field.

In particular, there are about 500 types of field weeds, and in particular, most weeds in agricultural land are affected by light (sunlight) and occur (light germination, dark germination). It is desirable to suppress or eliminate the occurrence.

As a method of removing weeds, there are a method of directly cutting weeds, a method of spraying chemical agents on weeds, a method of removing roots, and the like.

In recent years, the public's preference for eco-friendliness is increasing, and efforts to introduce eco-friendly farming methods are increasing in agricultural technology. Therefore, it is important to minimize the use of pesticides such as herbicides. Accordingly, methods such as direct cutting or uprooting of weeds are preferred, which require a lot of manpower.

However, due to the aging of farm households, there are not many young and aging workers, so it has been difficult to solve these problems.

In addition, safety accidents occurred frequently because sharp tools such as sickles were used to cut weeds directly or to uproot the weeds. Moreover, when the area required to remove weeds increases, it is essential to increase the working time, which places a burden on aging farms.

To solve this, various agricultural robots are being developed, but it took a long time to learn the steering ability in an aging farmhouse.

Techniques have been developed to overcome the problems of such a conventional weed removal method.

Korean Patent Application Laid-Open No. 10-2015-0124305 discloses an agricultural robot. In the prior art, it is possible to control weeds or pests, or flower thinning or cuttings by providing a scanner, and discloses a method of burning or ripening as a method of removing weeds.

However, it could not be applied to fields already planted with field crops because the planting characteristics were not taken into account.

KR 10-2015-0124305 A JP 2011-205967 A KR 10-1816557 B1 KR 10-2181930 B1 CN 211090589 U KR 10-2019-0031391 A

The present invention has been devised to solve the above problems.

Specifically, it is intended to solve the problem that the field weed removal ability changes depending on the skill level of the operator who operates the robot, and damage to field crops occurs. At this time, the generation state of field weeds is not taken into account and the field weeds having different growth conditions are removed using the same removal method, and in this case, the problem of poor field weed removal efficiency is to be solved. In particular, it is intended to propose a field weed management tillage robot that can fully function in environments with various ridge spacing and furrow depth of field crops.

In addition, it is intended to solve the problem that the robot's internal configuration is damaged due to the collision of the robot and the obstacle in front, resulting in malfunction. Of course, we would like to propose a tillage robot that can manage field weeds even in an environment without furrows or in a flat environment without furrows and furrows.

In addition, it is intended to solve the problem of climatic and time constraints because the operator controls in the same space.

In addition, it is intended to solve the problem that driving is impossible when the electricity of the provided battery runs out.

In one embodiment of the present invention for solving the above problems, the planting characteristic of the field sensed by the sensor unit 20 is an input layer, and the rotary driving member 350 and the robot driving member 310 are controlled. a base 100 provided with a main body 110 on which an Al module 50 using a signal as an output layer is mounted; a first wheel 130 and a second wheel 140 positioned on the base 100; a main rotary 210 connected to the rotary driving member 350 provided in the base 100 and located at the rear of the base 100 and capable of topsoil cultivation; and a transverse rotary 220 connected to the rotary drive member 350, wherein the AI module 50 controls the rotary drive member 350 through the control signal to control the main rotary 210 and To provide a tillage robot that controls the driving of the lateral rotary 220 and controls the driving of the first wheel 130 and the second wheel 140 by controlling the robot driving member 310 .

In addition, the planting characteristics of the field made by the AI module 50 as an input layer include field crop planting density, furrows and types of field crops, and the AI module 50 includes weed generation information, and up to obstacles in front. With the distance as an input layer, the AI module 50 controls the rotary drive member 350 by recognizing the planting density and furrows of the field crop, and the rotary drive member 350 according to the type of the field crop It is preferable to control, control the rotary driving member 350 according to the generation information of the weeds, and control the robot driving member 310 according to the distance to the obstacle to control the running and rotation of the robot.

In addition, the main rotary 210 is rotatable in the horizontal direction by the first power transmission shaft 351 connected to the rotary driving member 350, the upper part is rotatable 360 degrees, and the lateral rotary 220 ) is located on both sides of the main rotary 210 as a pair, and each of the transverse rotary 220 is transversely rotated by a second power transmission shaft 352 formed separately from the first power transmission shaft 351 . And, it is preferable that the length and angle can be changed.

In addition, it further comprises a cover part 200 for covering the upper portion of the main rotary 210 and the pair of transverse rotary 220, the main rotary 210 and the entire length of the transverse rotary 220. Accordingly, it is preferable that the length of the cover part 200 is variable.

In addition, it is preferable to further include; a photovoltaic module 170 spaced apart from the upper surface of the base 100 .

In addition, the sensor unit 20, the first sensor module 102 provided in the base 100; and a second sensor module 172 positioned to be spaced apart from the upper part of the first sensor module 102, and planting the field by the first sensor module 102 and the second sensor module 172 Characteristics, the generation information of the weeds, and the distance to the front obstacle are collected, and it is preferable that at least one of the first sensor module 102 and the second sensor module 172 is a camera.

As a method using the tillage robot 1000 according to embodiments of the present invention for solving the above problems, (a) the AI module 50 is the planting of the field detected by the sensor unit 20 collecting characteristics, the type of the field crop, the generation information of the weeds, and the distance to the front obstacle; and (b) the AI module 50 uses the pre-learned result, and controls the rotary drive member 350 according to the planting density of the field crops, and the rotary drive member 350 according to the generation information of the weeds. and controlling the driving and rotation of the tillage robot 1000 by controlling the robot driving member 310 according to the distance to the front obstacle.

The present invention obtains the following effects through solving the above problems.

First, it is possible to remove field weeds by unmanned driving, so it is possible to effectively remove field weeds regardless of control ability. Since weed generation information is used to remove field weeds, it is possible to effectively remove field weeds by providing topsoil partial tillage suitable for growth conditions. It is possible to effectively remove field weeds by recognizing furrows by AI. Of course, effective work is possible even in a flat environment without furrows or both furrows and furrows.

Second, since the robot automatically recognizes the obstacle in front by artificial intelligence and corrects the path without additional manipulation of the operator, it is possible to prevent the problem of malfunction of the robot due to collision with the obstacle.

Third, because it is possible to drive unmanned, it is possible to remove field weeds without climatic and time constraints.

Fourth, even without electricity supply, it is self-generated by the solar module and can continuously drive unmanned, so it is possible to remove field weeds for a long time.

1 is a perspective view of a tillage robot according to the present invention.
2 is a system diagram of a tillage robot according to the present invention.
3 is a flowchart illustrating a method for removing field weeds using a tillage robot according to the present invention.
4 is a view for explaining the main rotary and the transverse rotary of the tillage robot according to the present invention.

The above objects, features and other advantages of the present invention will become more apparent by describing preferred embodiments of the present invention in detail with reference to the accompanying drawings. In this process, the thickness of the lines or the size of the components shown in the drawings may be exaggerated for clarity and convenience of explanation. In addition, the terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to the intention or custom of the user or operator. Therefore, definitions of these terms should be described based on the content throughout this specification.

In addition, the described embodiments are provided by way of example for the description of the invention, and do not limit the technical scope of the present invention.

Hereinafter, a tillage robot according to a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings.

Here, "planting characteristics" is a concept including the planting density of field crops and types of field crops.

The configuration of the tillage robot 1000 according to the present invention will be described with reference to FIG. 1 .

The base 100 may provide buoyancy to the tillage robot 1000 . Accordingly, buoyancy is applied to the puddle formed in the field during the rainy season.

A body 110 is provided on the base 100 .

The body 110 preferably includes a housing 111 made of a waterproof material to protect the mounted modules and systems. The housing 111 is formed to protect the AI module 50, which is a system located on the upper part of the main body 110, from dust or foreign substances in one embodiment, and a door is provided so that an operator can easily use it when checking the system. In another embodiment not shown, the robot driving member 310 may be further positioned inside the housing 111 . The Al module 50 is mounted on the body 110 . AI module 50 controls the driving of the main rotary 210 and the pair of lateral rotary 220 by controlling the rotary drive member 350, a detailed description will be described later.

The first wheel 130 is located at the lower end of the base 100 . Facilitates angular deformation, thus facilitating movement between the tides.

The second wheel 140 is located on both sides of the base 100 from the rear than the first wheel 130 as a pair.

The second wheel 140 preferably has a wider width and diameter than the first wheel 130 . Accordingly, it is easy to operate on various types of farmland such as flat land, wetland, and sloping land.

The robot driving member 310 provides power to the first wheel 130 , the second wheel 140 , or both, and may be an internal combustion engine such as an engine or an electric motor. In the case of an electric motor, electricity charged by the solar module 170 may be used.

The rotary driving member 350 may be provided on the base 100 to provide driving force to the main rotary 210 and the lateral rotary 220 . It may be an electric motor, and the solar energy charged in the solar module 170 may be used.

The solar module 170 is spaced apart from the upper surface of the base 100 to charge electricity. The photovoltaic module 170 is formed spaced apart from the upper surface of the base 100 by a plurality of support legs 171 , but may be variously modified. The aforementioned housing 111 may be positioned in a space formed spaced apart to provide a dustproof effect to system components such as the AI module 50 .

The main rotary 210 is connected to the rotary driving member 350 and is located at the rear of the base 100 to remove field weeds by cultivating the topsoil. Specifically, by recognizing the furrow, the main rotary 210 is lifted by the rotary driving member 350 to press the field weeds, and the provided rotary blades receive the driving signal to transmit the first power connected to the rotary driving member 350 . It is rotated in the horizontal direction by the shaft 351 to impart friction to the field weeds. When there is no furrow, the main rotary 210 rotates in a horizontal direction, so that topsoil tillage and weed removal are possible.

The shape of the main rotary 210 is not limited, but the rotary blade is formed in an inverted cone shape and may be seated closely in the furrow. In this case, the AI module 50 may recognize the furrow according to the planting density and lower the main rotary 210 to till the furrow topsoil. At this time, the upper part of the main rotary 210 can rotate and vibrate 360 degrees.

4, the axis of the rotating blade of the main rotary 210 becomes smaller toward the bottom and is formed in an inverted cone shape, and the rotating blade formed in three stages connected to the shaft formed in three stages rotates in three stages as the axis is folded. It is possible to remove field weeds by changing the shape according to the shape of the furrow and tilling the topsoil in the form of a blade, a two-stage rotary blade or a single rotary blade, ie, having three different depths.

The lateral rotary 220 is a pair, connected to the rotary driving member 350, and rotated by being positioned on both sides of the main rotary 210 to remove field weeds by cultivating the topsoil. Each of the transverse rotary 220 is transversely rotated on both sides of the main rotary 210 by the second power transmission shaft 352 so that the topsoil is tilled by rotation and field weeds located in the topsoil are removed.

Specifically, the transverse rotary 220 is variable in length and angle, so that the length and angle are automatically changed according to a control signal. For example, the horizontal rotary 220 may be adjusted in length within the width of the base 100 before receiving the control signal in three pairs to prevent damage to field crops during operation. In the case of a field crop having a planting density wider than the width of the base 100, when a control signal is received, the horizontal rotary 220 is adjusted to a length longer than the width of the base 100 according to the control signal to remove field weeds. can The angle of the transverse rotary 220 is varied by elevating the end portion to which the second power transmission shaft 352 is not connected.

The main rotary 210 and the pair of lateral rotary 220 are arranged to partially overlap in the horizontal direction. 4 , the rear side of the main rotary 210 may partially overlap in the horizontal direction.

In an embodiment of the present invention, the first power transmission shaft 351 and the second power transmission shaft 352 may be separated or integrated.

In another embodiment of the present invention, the first power transmission shaft 351 and the second power transmission shaft 352 may be formed in parallel as shown in FIG. 1 , and may form an angle with each other.

The upper part of the main rotary 210 and the pair of lateral rotary 220 is covered by the cover part 200 .

The cover part 200 prevents the splashing of dust scattered when the main rotary 210 and the lateral rotary 220 rotate. In one embodiment, it may include a solar module.

The length of the cover part 200 is extendable in the transverse direction, and as the length and angle of the lateral rotary 220 vary, the length of the cover part 200 also varies.

For example, if the distance by the field crop planting density is 30cm, only the main rotary 210 will operate and the transverse rotary 220 will not work, and if it is 30cm or more, both the main rotary 210 and the transverse rotary 220 will operate , If the width of the base 100 or more, the horizontal rotary 220 will vary and increase in length, and if the maximum is 120 cm, the horizontal rotary 220 will be operated with a variable maximum length.

As the planting characteristics of the field are confirmed by the AI module 50, the horizontal rotary 220 is changed, and details will be described later.

In one embodiment, in consideration of the planting characteristics of the field, the diameter of the main rotary 210 is 30 cm or less, and the horizontal rotary 220 is preferably 45 cm or less as a set of three roundabouts having a length of 15 cm.

A system of the tillage robot 1000 according to the present invention will be described with further reference to FIG. 2 .

The system of the tillage robot 1000 according to the present invention is capable of autonomous driving including the AI module 50, and includes a control unit 10, a sensor unit 20, a communication unit 30, and a driving unit 300 for this purpose. do.

The control unit 10 is connected to the sensor unit 20 , the communication unit 30 , the driving unit 300 , and the AI module 50 , and the information sensed by the sensor unit 20 becomes an input layer of the AI module 50 . , a control signal that is an output layer is calculated using the result previously learned from the AI module 50 , and the driving unit 300 is controlled using this. Also, the controller 10 may manually control the driving unit 300 using information input through the terminal 31 wirelessly connected to the communication unit 30 .

The planting characteristic of the field is sensed by the sensor unit 20 . Specifically, the sensor unit 20 includes an image sensor 21 such as a camera, and detects any one or more of a speed sensor 22 , a GPS sensor 23 , an ultrasonic sensor 24 , and a distance measurement sensor 25 . may include more.

The sensor unit 20 further includes a first sensor module 102 provided in the base 100 and a second sensor module 172 spaced apart from the upper portion of the first sensor module 102 . This is for the image sensor 21 to sense an image of a more accurate and wide angle of view.

In addition to the planting characteristics of the field by the first sensor module 102 and the second sensor module 172 , growth information of field crops, information on the occurrence of weeds, and a distance to a front obstacle may be further collected.

In other embodiments, there may be additional sensors positioned adjacent to the main rotary 210 and the transverse rotary 220 . The additional sensor may be a camera, located close to the field crop, and may provide an additional detection result to the calculation result of the AI module 50 to increase accuracy.

The AI module 50 uses the planting characteristics of the field sensed by the sensor unit 20 as an input layer, and uses the control signals of the rotary drive member 350 and the robot drive member 310 as an output layer.

In addition, the AI module 50 controls the rotary driving member 350 to control the driving of the main rotary 210 and the pair of lateral rotary 220 . Specifically, by controlling the driving of the rotation direction and speed of the horizontal rotary 220, the variable of the horizontal rotary 220, the pressing strength of the main rotary 210, the rotation speed, etc., the topsoil is tilled to remove field weeds.

In addition, the AI module 50 controls the driving of the first wheel 130 and the second wheel 140 by controlling the robot driving member 310 . Specifically, it is possible to control the driving of the first wheel 130 and the second wheel 140 , such as moving forward, backward, and directional rotation, to move to an area where field weeds need to be removed and to rotate and move.

A detailed control method will be further described with reference to FIG. 3 .

The AI module 50 serves two functions. The first is to determine the entire boundary of the area to remove field weeds, and the second is to control it by calculating the path in real time while the tillage robot 1000 is running.

In order to perform the method according to the present invention, first, the AI module 50 determines the boundary by checking the entire image (ie, the entire region image) of the area to remove field weeds from the sensor unit 20 ( S100 ). ).

Various images and pre-determined boundaries were used as training data, and the AI module can automatically determine the boundaries when receiving an image of the entire area using artificial intelligence.

Next, the AI module 50 determines a work area using the boundary determined in step S100 and determines a path (S200).

Next, when the AI module 50 receives an image of the entire area using artificial intelligence, it identifies field crops in the image, automatically checks planting characteristics such as field crop planting density and field crop type, and uses the confirmed planting characteristics. A path of the entire region is determined (S300).

To this end, the AI module 50 learns by using various images including predetermined positions of field crops for each type as learning data. The AI module 50 determines the path of the entire area using the planting characteristics identified using the received image. In this case, multi-operation is also possible in a large area by using a plurality of tillage robots 1000 at the same time. In this case, a separate control terminal for multi-operation may be used.

Now, the tillage robot 1000 starts to operate (S400).

As the tillage robot 1000 travels, the sensor unit 20 checks the image in real time, and the AI module 50 checks the planting characteristics of the field from the checked image (S500).

The confirmed planting characteristics of the field are used as an input layer in the AI module 50, and an output layer for controlling the rotary driving member 350 and the robot driving member 310 is calculated through this (S600).

To this end, the AI module 50 learns in advance various planting characteristics and a rotary driving method and a robot driving method optimized therefor. For example, when the AI module 50 confirms that the spacing by the field crop planting density is 60 cm as a planting characteristic, the rotary driving member 350 drives both the main rotary 210 and the pair of transverse rotary 220 , A control signal for controlling the robot driving member 310 to variably control to shorten the length of the rotary 220 and to move the robot driving member 310 between rows in which field crops are grown is calculated as an output layer. When it is confirmed that the interval is 120 cm, the rotary driving member 350 drives both the main rotary 210 and the pair of transverse rotary 220, variably control to extend the length of the transverse rotary 220 to the maximum and driving the robot A control signal for controlling the robot driving member 310 is calculated as an output layer so that the member 310 moves between rows in which field crops are grown.

On the other hand, the AI module 50 may further use the generation information of weeds. For example, if it is confirmed that the AI module 50 has grown to 10% coverage as the generation information of weeds through the real-time image confirmed by the sensor unit 20, the rotary driving member 350 drives the main rotary 210 and At the same time, a control signal for controlling both of the pair of lateral rotary 220 to be driven to the maximum output is calculated as an output layer.

In another embodiment, the AI module 50 may further use the distance to the front obstacle. For example, when the AI module 50 confirms that there is an obstacle at 50 cm as the distance to the front obstacle through the real-time image confirmed by the sensor unit 20 and confirms that the distance is reduced in real time, the robot driving member 310 rotates A control signal that performs an operation is calculated as an output layer. At this time, the rotation direction refers to the path set in step S300.

Next, when it is determined that the AI module 50 is the provided route driven through the image confirmed in real time, it is determined that the driving is completed and the control unit 10 calculates an end signal as an output layer.

When the method according to the present invention is applied to an actual field, it can be operated as follows.

When field crops are raised, the first operation starts before or at the beginning of weeds. In this case, it can be applied even in the early stage of planting when transplanting field crops.

The operation of the tillage robot 1000 is repeated at intervals of 5 to 7 days depending on the condition of the corresponding lot. At this time, the operation will be repeated until the crops form a sufficient colony, that is, 5 to 7 times will be operated. A multi-run may be performed.

Of course, the operator can freely adjust the number of operations of the tillage robot 1000 according to the condition of the field or the occurrence of weeds.

In the above, the present specification has been described with reference to the embodiments shown in the drawings so that those skilled in the art can easily understand and reproduce the present invention, but these are merely exemplary, and those skilled in the art may make various modifications and equivalent other modifications from the embodiments of the present invention. It will be appreciated that embodiments are possible. Therefore, the protection scope of the present invention should be defined by the claims.

10: control
20: sensor unit
21: image sensor
22: speed sensor
23: GPS sensor
24: ultrasonic sensor
25: distance measuring sensor
30: communication department
31: terminal
50: AI module
100: base
110: body
130: first wheel
140: second wheel
170: solar module
200: cover part
210: main rotary
220: transverse rotary
300: driving member
310: robot driving member
350: rotary driving member
351: first power transmission shaft
352: second power transmission shaft

Claims (7)

Al module 50 using the planting characteristics and weed generation information sensed by the sensor unit 20 as an input layer, and the control signals of the rotary driving member 350 and the robot driving member 310 as an output layer a base 100 having a mounted body 110;
a first wheel 130 and a second wheel 140 positioned on the base 100;
a main rotary 210 connected to the rotary driving member 350 provided in the base 100 and located at the rear of the base 100 and capable of topsoil cultivation; and
and a transverse rotary 220 connected to the rotary driving member 350,
The main rotary 210,
The rotary drive member 350 and the vertical first power transmission shaft 351 connected to the lifting and rotatable in the horizontal direction,
The main rotary 210 has a rotating blade formed in an inverted conical shape,
The horizontal rotary 220,
It is located on both sides of the main rotary 210 as a pair, but partially overlaps with the main rotary 210 in the horizontal direction,
Each of the transverse rotary 220 is variably controlled in length, and is transversely rotatable by a horizontal second power transmission shaft 352,
The AI module 50 controls the driving of the main rotary 210 and the lateral rotary 220 by controlling the rotary driving member 350 through the control signal, and the robot driving member 310 to control the driving of the first wheel 130 and the second wheel 140,
The planting characteristics of the field as the input layer include the planting density of field crops, furrows and types of field crops,
The AI module 50,
Recognizing the field crop planting density and furrows, controlling the lifting, pressing strength, and rotational speed of the main rotary 210 and controlling the length variation, rotation direction and speed of the transverse rotary 220,
Further control of the variable length of the horizontal rotary 220 according to the type of the field crop,
Controlling the output of the main rotary 210 and the horizontal rotary 220 by controlling the rotary driving member 350 according to the generation information of the weeds,
Field weed management tillage robot.
The method of claim 1,
The AI module 50 further uses the distance to the front obstacle as an input layer,
The AI module 50,
Controlling the driving and rotation of the robot by controlling the robot driving member 310 according to the distance to the obstacle,
Field weed management tillage robot.
3. The method of claim 2,
The main rotary 210,
The upper part can be rotated 360 degrees,
Each of the horizontal rotary 220,
angle is changeable,
Field weed management tillage robot.
4. The method of claim 3,
Further comprising a cover part 200 covering the upper part of the main rotary 210 and the pair of lateral rotary 220,
The length of the cover part 200 is variable according to the total length of the main rotary 210 and the horizontal rotary 220,
Field weed management tillage robot.
The method of claim 1,
Further comprising; a photovoltaic module 170 spaced apart from the upper surface of the base 100;
Field weed management tillage robot.
3. The method of claim 2,
The sensor unit 20,
a first sensor module 102 provided on the base 100; and
Including a second sensor module 172 located spaced apart from the upper than the first sensor module 102,
The planting characteristics of the field, the generation information of the weeds, and the distance to the front obstacle are collected by the first sensor module 102 and the second sensor module 172,
Any one or more of the first sensor module 102 and the second sensor module 172 is a camera,
Field weed management tillage robot.
A method using the tillage robot 1000 according to any one of claims 2 to 6, comprising:
(a) collecting, by the AI module 50, the planting characteristics of the field detected by the sensor unit 20, the type of field crop, the generation information of the weeds, and the distance to the front obstacle; and
(b) the AI module 50 uses the pre-learned result to control the rotary drive member 350 according to the planting density of the field crops, and the rotary drive member 350 according to the generation information of the weeds and controlling the driving and rotation of the tillage robot 1000 by controlling the robot driving member 310 according to the distance to the front obstacle.
Way.

Images