KR102089423B1 - Real time drone control method and system for fertilizer application and pest control - Google Patents

Abstract

Disclosed are a real-time drone control method for fertilization and control and a real-time drone control system. According to one embodiment of the present invention, the real-time drone control method for fertilization and control comprises the steps of: spraying a material for fertilization or control held in a drone flying forward at a predetermined altitude to an area to be sprayed on farmland corresponding to the altitude in response to sending of a spraying start command; and sending a direction change command along with the spraying start command to the drone as the area to be sprayed is adjacent to the boundary of the farmland when the drone flies and sprays the material.

Description

REAL TIME DRONE CONTROL METHOD AND SYSTEM FOR FERTILIZER APPLICATION AND PEST CONTROL}

The present invention relates to a technique for controlling drones to manure crops on a farmland or to prevent pests, and more specifically, a drone that changes a flight direction by real-time recognizing a boundary ('farmland') on a farmland during flight It is related to the improvement of ease of fertilization and control work using.

According to the prior art, in order to perform fertilization and control using a drone, a process of setting a flight path in advance is required.

As an example, precedent document 1 (Korean Patent Publication No. 10-2019-0076227, 2019.07.02) regarding "automatic obstacle avoidance and drone control method for fully automatic flying agriculture" refers to obstacles, flight zones, etc. in 3D map information captured by aerial. Disclosed is a configuration for automatically extracting information and flying a drone.

In addition, Prior Art Document 2 (Japanese Patent Publication No .: 2018-111429, 2018.07.19) on "Method of spraying pesticides using drones" discloses a configuration for setting a spraying scheme of pesticides for individual spray sections from near infrared images.

In other words, the prior documents are planning in advance the route to fly the drone by preliminarily identifying the area to be sprayed in the aerial photograph taken earlier.

On the other hand, the method of setting the flight route of the drone is the manual setting method of finding the farmland to be sprayed on the orthogonal image map and adding and connecting the way-point along the shape of the terrain. Therefore, there is an automatic setting method that automatically generates Way-Points using software.

Since the orthogonal image map must be prepared in both ways, it is inevitable to produce an orthogonal image map for the new terrain when the shape of the farmland has changed or a new farmland is not present in the map.

Also, even if the farmland is already on the satellite map, the actual GPS coordinate values may not match, so it is necessary to go through the process of producing a precise orthogonal image map through low-altitude aerial photography.

As described above, in order to perform fertilization and control using a drone, it is necessary to set a flight path in advance, and in order to set a flight path, production of a precise orthogonal image map through aerial photography must be preceded, which requires a lot of time and cost. do.

Accordingly, there is a need for a drone control technology that enables real-time recognition of a farming route during flight and fertilization and control at a low cost without having to set a flight path in advance.

In an embodiment of the present invention, when a material for fertilization and control is sprayed using a drone flying over farmland, the boundary of the farmland ('farmland' or farmland) is analyzed in real time by analyzing the ground image photographed by the drone. The goal is to realize real-time control without the need to set the flight path in advance by recognizing and automatically switching the flight direction.

Embodiment of the present invention, without having to set the flight path of the drone in advance, by determining the flight direction in real time during the flight, reducing the time and cost of preparation and control operations using the drone, due to the coordinate error of the satellite map The aim is to minimize the area that is misapplied.

A real-time drone control method for fertilization and control according to an embodiment of the present invention is for controlling fertilization or control held in a drone flying forward at a predetermined altitude according to the transmission of a spray start command. Spraying a material to a spraying area on the farmland corresponding to the altitude, and as the drone sprays the material while flying, when the spraying area is adjacent to the boundary of the farmland, the drone waits for a spraying command and It may include the step of sending a direction change command together.

In addition, the real-time drone control system for fertilization and control according to an embodiment of the present invention, the material for fertilization or control held in a drone flying forward at a predetermined altitude, according to the transmission of the spray start command. , When spraying to the spraying area on the farmland corresponding to the altitude, and as the drone is flying and spraying the material, when the spraying area is adjacent to the boundary of the farmland, the drone is commanded to change direction with the spraying standby command. It may include a controller for transmitting a.

According to an embodiment of the present invention, by analyzing the ground image photographed by the drone in real time to recognize the boundary of the farmland ('farmland' or the form of the farmland) to automatically switch the flight direction, to advance the flight path of the drone in advance You can control the drone in real time without setting.

According to one embodiment of the present invention, the process of setting the flight route in advance during fertilization and control operations using a drone is omitted, and the process of producing a precise orthogonal image map through low-altitude aerial photography is also unnecessary, so it is necessary to prepare in advance. It can greatly reduce the working time and cost, and minimize the area that is misapplied due to the coordinate error of the satellite map.

1 is a block diagram showing the configuration of a real-time drone control system (hereinafter referred to as 'real-time drone control system') for fertilization and control according to an embodiment of the present invention.
2 is a view showing a process of controlling the flight and spraying of a drone according to the transmission of a spray start command in a real-time drone control system according to an embodiment of the present invention.
FIG. 3 is a diagram illustrating an example of determining a direction in which a drone will switch when adjacent to a farmland boundary in a real-time drone control system according to an embodiment of the present invention.
FIG. 4 is a diagram illustrating a process of controlling flight by switching a drone's direction when adjacent to a farmland boundary in a real-time drone control system according to an embodiment of the present invention.
FIG. 5 is a diagram illustrating a process of recovering a drone to a designated home point according to a transmission of an end-of-spray command in a real-time drone control system according to an embodiment of the present invention.
FIG. 6 is a diagram illustrating an example of sending a command to end spraying when there is a farm road in a direction in which a drone is to switch in a real-time drone control system according to an embodiment of the present invention.
7 is a view showing a process of performing fertilization and control in an opposite direction from an initial direction change point in a real-time drone control system according to an embodiment of the present invention.
8 is a flowchart illustrating a procedure of a real-time drone control method for fertilization and control according to an embodiment of the present invention.
9 is a detailed flowchart illustrating a procedure of a real-time drone control method for fertilization and control according to an embodiment of the present invention.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, various changes may be made to the embodiments, and the scope of the patent application right is not limited or limited by these embodiments. It should be understood that all modifications, equivalents, or substitutes for the embodiments are included in the scope of rights.

The terms used in the examples are used for illustrative purposes only and should not be construed as limiting. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this specification, the terms "include" or "have" are intended to indicate the presence of features, numbers, steps, actions, components, parts or combinations thereof described herein, one or more other features. It should be understood that the existence or addition possibilities of fields or numbers, steps, operations, components, parts or combinations thereof are not excluded in advance.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person skilled in the art to which the embodiment belongs. Terms such as those defined in a commonly used dictionary should be interpreted as having meanings consistent with meanings in the context of related technologies, and should not be interpreted as ideal or excessively formal meanings unless explicitly defined in the present application. Does not.

In addition, in the description with reference to the accompanying drawings, the same reference numerals are assigned to the same components regardless of reference numerals, and redundant descriptions thereof will be omitted. In describing the embodiments, when it is determined that detailed descriptions of related known technologies may unnecessarily obscure the subject matter of the embodiments, detailed descriptions thereof will be omitted.

The real-time drone control system for fertilization and control according to an embodiment of the present invention may be implemented by being included inside a drone controller or a ground control station (GCS) of a user stage.

1 is a block diagram showing the configuration of a real-time drone control system (hereinafter referred to as 'real-time drone control system') for fertilization and control according to an embodiment of the present invention.

Referring to FIG. 2, the real-time drone control system 100 according to an embodiment of the present invention may include a controller 110, an analysis unit 120, and a database 130.

The controller 110, according to the transmission of the spraying start command, the material for the fertilization or control (eg, manure or pesticide, etc.) held in the drone 101 flying forward at a predetermined altitude to the altitude. When the spraying area is applied to a corresponding spraying area on the farmland, and as the drone 101 sprays the material while flying, when the spraying area is adjacent to the boundary of the farmland, the drone 101 is commanded to change direction with the spraying standby command. To send.

Prior to sending the spray start command, the controller 110 determines the flight speed of the drone 101, the spray area and the spray amount relative to the altitude and altitude, and the home point, which is the takeoff and landing point of the drone 101 when the altitude and the spray amount are determined. And the drone 101 can be positioned so that the head points to the crop in the farmland at the home point.

For example, the controller 110 may set the altitude to approximately 2 to 3 m from the ground, and may adjust the spray amount relative to the spray area according to the altitude according to the performance of the sprayer mounted on the drone 101. In addition, the home point can be selected as the first farmland adjacent to the crops in the farmland on the outer farmland surrounding the farmland.

The drone 101, which has received the spray start command, may take off vertically from the home point and adjust the camera vertically downward to shoot a ground image. Therefore, it is possible to photograph a ground image in which the center point of the ground image is the location of the drone 101. 2 shows ground images 210 and 220 photographed by the drone 101.

Hereinafter, a process of controlling flight and spraying of the drone 101 will be described with reference to FIG. 2.

2 is a view showing a process of controlling the flight and spraying of a drone according to the transmission of a spray start command in a real-time drone control system according to an embodiment of the present invention.

Referring to FIG. 2, the ground image 210 is an image photographed by the drone 101 vertically taken off from the home point 201, and the ground image 220 is a crop 202 in the farmland in a vertically taken off state. It may be an image taken when flying forward in the direction of the arrow toward the spraying point.

When the controller 110 analyzes ground images received from the drone 101 according to a machine learning technique, classifies crop areas and farmland areas from the ground images, and takes off vertically at the home point 201 Since the takeoff location is a farm road as in the ground image 210, the drone 101 can be gradually advanced while gradually increasing the pitch value.

The spraying starting point may be a point where the central point 203 in the ground image 220 starts to be located in the crop region 202 out of the farm road while the drone 101 is flying forward in the direction of the arrow.

The controller 110 continuously observes the crop region 202 through machine learning analysis from the ground image received from the drone 101 and then causes the central point 203 to enter the crop region 202 like the ground image 220. If it is, it can be judged that it is located at the starting point of spraying.

When the drone 101 is located at the starting point of spraying, it is checked whether a predetermined altitude (approximately 2 to 3 m) has been reached, and when the altitude is reached, the drone 10 is applied to the spraying area 204 on the farmland determined according to the altitude. ) Manure for fertilization or pesticide for control can be sprayed toward the farmland. Here, the area of the spraying area 204 may be different according to the spraying performance of the sprayer mounted on the drone 101.

1, the analysis unit 120 analyzes the ground image photographed through the camera mounted on the drone 101 according to a machine learning technique, so that the spraying area is adjacent to the boundary of the farmland. It serves to determine whether or not.

For example, the analysis unit 120 analyzes the ground image photographed through the camera mounted on the drone 101 according to a machine learning technique, and when the spraying area overlaps at least a part of the outer farm road surrounding the farmland, the It can be determined that the spraying area is adjacent to the boundary of the farmland.

In another example, the analysis unit 120 extracts a crop region corresponding to the farmland from the ground image using the machine learning technique, and if there is at least one contact point between the spray region and the border of the crop region, It may be determined that the spraying area is adjacent to the boundary of the farmland.

At this time, the analysis unit 120 may identify the presence or absence of the contact point by excluding a region in which the center point of the ground image is not included among the crop regions.

When it is determined that the spraying area is adjacent to the boundary of the farmland, the analysis unit 120 is left and right based on the center point (current position of the drone) in the ground image to determine the direction in which the drone 101 will switch. The proportion (length) of the crop area in the direction can be taken into account.

Specifically, the analysis unit 120, the proportion of the crop area in the first direction based on the center point in the ground image taken at the time when it is determined that the spraying area is adjacent to the boundary of the farmland, the first Whether the crop area in the second direction opposite to the direction is greater than the proportion occupied, and the controller 110 if the proportion occupied by the crop area in the first direction is greater than the proportion occupied by the crop area in the second direction, The direction change command for determining the first direction as the direction in which the drone 101 will switch may be written.

Hereinafter, a process of determining a direction turning point and determining a direction to switch will be described with reference to FIG. 3.

FIG. 3 is a diagram illustrating an example of determining a direction in which a drone will switch when adjacent to a farmland boundary in a real-time drone control system according to an embodiment of the present invention.

Referring to the ground image 310 of FIG. 3, the analysis unit 120 continuously receives and analyzes the ground image photographed by the drone 101 flying forward while maintaining a predetermined flight speed and spray amount, and the drone 101 ), It may be identified whether there is at least one contact point between the upper side 301 of the spreading area and each rim 302 of the crop area 320.

As an example, the analysis unit 120 may have both ends of the upper side 301 of the spraying area outside the crop area 320 when the spraying area is a square with the vertical position of the drone 101 as the center (+). If it does, the application area can be considered to be adjacent to the boundary of the farmland.

If there is a contact, the analysis unit 120 determines that the spreading area is adjacent to the boundary of the farmland, and the controller 110 may transmit a direction change command along with a spraying wait command to the drone 101.

For example, the analysis unit 120 drones the left side which is the longer of the length of the line segment 321 on the left side and the length of the line segment 322 on the right side based on the center point on the line passing through the center point of the crop area 320 ( 101).

Each successive length can be measured by calculating the length of the line segment to the center point after obtaining the contact point (marked part of x) of all sides of the crop area 320 and the blue or yellow line. If they are the same, one of the two can be arbitrarily determined.

At this time, the controller 110, by the direction change command, when the direction of the drone 101 is switched to the first direction (eg, 'left'), the spraying area, the crop area in the first direction or the agent When the crop area 320 is out of the two directions, a command to end spraying may be sent to the drone 101 to recover the drone 101 to a designated home point.

In addition, when the direction of the drone 101 is changed by the direction change command, the controller 110 moves the drone 101 by the horizontal length of the spray area, and rotates the head of the drone 101 by a predetermined angle. In one state, it is controlled to fly forward, and as the drone 101 moves forward, the spray start command may be re-sent.

Hereinafter, a process of changing the direction of the drone 101 will be described with reference to FIG. 4.

FIG. 4 is a diagram illustrating a process of controlling flight by switching a drone's direction when adjacent to a farmland boundary in a real-time drone control system according to an embodiment of the present invention.

Referring to 410 and 420 of FIG. 4, the controller 110 switches the drones 101 and 401 to the left according to the direction change command and moves the head in the direction of the arrow as much as the horizontal length of the spraying area, then moves the head to the yaw axis Can be rotated by 90 degrees.

The controller 110 may retract until the base 403 of the spreading area is all out of the crop area in the ground image photographed by the drones 101 and 402 after rotation.

The controller 110 may fly the drone 402 forward while maintaining the flight speed and spray amount, and restart the stopped spraying. In addition, the analysis unit 120 may analyze the ground images continuously received from the drone 402 to monitor whether the crop area ends and the farm road appears in the forward direction, that is, whether the spraying area is adjacent to the boundary of the farmland.

After re-transmitting the spraying start command, whenever the spraying area is adjacent to the boundary of the farmland, the controller 110 changes the direction opposite to the direction in which the drone 101 just switched, and the drone 101 The direction changing command, which determines the direction to switch, may be re-sent to the drone 101 together with the spray waiting command.

Hereinafter, an example in which the direction of the drone is reversed will be described whenever the spraying area is adjacent to the boundary of the farmland with reference to FIG. 5.

FIG. 5 is a diagram illustrating a process of recovering a drone to a designated home point according to a transmission of an end-of-spray command in a real-time drone control system according to an embodiment of the present invention.

Referring to FIG. 5, the controller 110 repeatedly flys to switch in the opposite direction (left) and in the opposite direction (right) each time the drone 101 is adjacent to the boundary on the farmland while flying forward. You can fly along route 501.

At this time, the controller 110 is a drone (when the spraying area when the drone 101 is switched to the second direction (eg, 'right') at the last turning point of the flight path 501 is adjacent to the farm road 502) 101), the drone 101 can be returned to the designated home point 503 by sending a command to finish spraying.

Hereinafter, an example in which the spraying end command is transmitted without changing the direction will be described with reference to FIG. 6.

FIG. 6 is a diagram illustrating an example of sending a command to end spraying when there is a farm road in a direction in which a drone is to switch in a real-time drone control system according to an embodiment of the present invention.

Referring to FIG. 6, the controller 110 continues to spray and change direction, and when a drone 110 finds a farm road in a direction in which the direction-side spraying area is out of the crop area, sends a command to terminate the spraying to the drone 110. And the drone 110 can be recovered.

Hereinafter, an example in which fertilization and control is carried out will be described with reference to FIG. 7 when the home point is selected at an intermediate point rather than four corners of the farmland.

7 is a view showing a process of performing fertilization and control in an opposite direction from an initial direction change point in a real-time drone control system according to an embodiment of the present invention.

Referring to 710 of FIG. 7, when the controller 110 selects a home point 701 on a farm road between four corners of the farmland, when the drone 101 is adjacent to the farm road 703 at the last turning point, the drone ( 101) after returning to the first direction turning point 704, switching to the opposite direction to resume spraying while flying over the remaining area 705 of the crop area, and again, the drone 101 becomes the turning point at the last direction and the farm road 703 ), The drone 101 can be returned to the home point 701 by sending a command to terminate spraying.

According to an embodiment, after the controller 110 re-transmits the spraying start command, when the spraying area is adjacent to the boundary of the farmland, the proportion occupied by the crop area in the first direction (eg, 'left') If the proportion occupied by the crop area in the second direction ('right') is the same as within the selected range, the drone 101 does not switch at the first direction change point in the first direction ('left'). The direction change command for determining the second direction ('right') as the direction in which the drone 101 will switch may be re-transmitted to the drone 101.

For example, referring to FIG. 7, if the home point 701 is not the four corners of the crop area in the process of determining the initial switching direction, the controller 110 has the same length of the left and right lines calculated, which is not the same. There may be a case of losing, and at this time, the drone 101 is moved to the first direction turning point 704 without entering the spray termination step, and the head is also rotated in the direction at that time, and then flying in the direction opposite to the direction in which it was originally switched. In doing so, you can start spraying the rest of the crop area.

That is, according to the present invention, even if the home point 701 is placed anywhere on the farmland to start spraying, it is possible to easily perform fertilization and control operations using the drone 101 for all areas of the farmland.

According to an embodiment, the analysis unit 120 records the spreading area adjacent to the boundary of the farmland in the database 130 as terrain data for the farmland, and the controller 110 displays the spreading recorded as the terrain data. When the areas are connected, when constructing an outer farm road surrounding the farmland, a command to end spraying may be sent to the drone 101 to recover the drone 101 to a designated home point.

That is, according to the present invention, when the spraying is completed for all areas of the farmland, a command to end the spraying can be sent to the drone 101, and the spraying area adjacent to the farm road can be easily collected as terrain data of the farmland.

As described above, according to an embodiment of the present invention, when spraying a material for fertilization and control using a drone 101 flying over farmland, the ground image photographed by the drone 101 is analyzed to real-time farmland By recognizing the boundary ('farmland' or farmland type) and automatically switching the flight direction, real-time control without the need to set the flight path in advance can be implemented. In addition, according to an embodiment of the present invention, without having to set the flight path of the drone 101 in advance, by determining the flight direction in real time during the flight, preparation time and cost for fertilization and control using the drone 101 It can reduce and minimize the area that is misapplied due to the coordinate error of the satellite map.

Hereinafter, the operation flow of the real-time drone control system 100 according to embodiments of the present invention will be described in detail with reference to FIGS. 8 to 9.

8 is a flowchart illustrating a procedure of a real-time drone control method for fertilization and control according to an embodiment of the present invention.

The real-time drone control method for fertilization and control according to this embodiment may be performed by the real-time drone control system 100 described above.

Referring to FIG. 8, in step 810, the real-time drone control system 100 transmits a spray start command to a drone located at a home point installed on a boundary (ie, 'farm road') of farmland.

The drone receiving the spray start command may take off vertically from the home point and fly forward toward the crops in the farmland, and at the same time, adjust the camera vertically downward to shoot a ground image including the farmland.

In step 820, the real-time drone control system 100 sprays a material for fertilizing or controlling a spray area on the farmland while the drone is flying.

The real-time drone control system 100 analyzes ground images received from the drone 101 according to a machine learning technique, classifies crop areas and farmland areas from the ground images, and central points in the ground images (the drone's It can be confirmed that the vertical position) is located outside the farmland area and within the crop area.

The real-time drone control system 100 has a central point located in a crop area, and when the drone reaches a predetermined altitude (approximately 2 to 3 m), manure or pesticide in drone 101 is applied to a spraying area on the farmland determined according to the altitude. Spraying may be used to fertilize or control.

In step 830, the real-time drone control system 100 determines whether the spray area is adjacent to the boundary of the farmland while the drone is spraying the material while flying.

For example, the real-time drone control system 100 extracts a crop region corresponding to the farmland from the ground image using the machine learning technique, and if there is at least one contact point between the spray region and the border of the crop region, It may be determined that the spraying area is adjacent to the boundary of the farmland.

If there is no contact point between the spraying area and the border of the crop area, since no farming route has been found in the advancing direction, the real-time drone control system 100 moves to step 820 to fly to the drone while flying to the spraying area on the farmland. Substances for fertilization or control are continuously sprayed.

If it is determined in step 830 that the spraying area is adjacent to the boundary of the farmland, in step 840, the real-time drone control system 100 sends a command to switch to the drone along with the spraying standby command.

For example, the real-time drone control system 100 further determines the length of the crop region on the left and the length of the crop region on the right, based on the center point in the ground image taken when the application area is determined to be adjacent to the boundary of the farmland. It is possible to write the redirection command to determine the long side (left) as the direction in which the drone 101 is to be switched.

In step 850, the real-time drone control system 100, when the direction of the drone is changed in accordance with the direction change command, sends the spray start command back to the drone.

The drone that has received the spraying standby command pauses spraying of the material, switches to a direction designated by the redirection command, moves the horizontal length of the spraying area, and moves the drone head by a predetermined angle. It is controlled to fly forward in a rotated state, and the spray start command can be re-sent as the drone moves forward.

After re-transmitting the spray start command, whenever the spray area is adjacent to the boundary of the farmland, the real-time drone control system 100 changes the direction opposite to the direction in which the drone just switched, and the drone switches. The direction changing command for determining the direction to be performed may be re-transmitted to the drone together with the spray waiting command.

As described above, according to one embodiment of the present invention, by analyzing the ground image photographed by the drone, real-time recognition of the boundary of the farmland (the 'route' or the form of the farmland) and automatically switching the flight direction, the drone is flying in advance. You can control the drone in real time without setting a route.

9 is a detailed flowchart illustrating a procedure of a real-time drone control method for fertilization and control according to an embodiment of the present invention.

The real-time drone control method for fertilization and control according to this embodiment may be performed by the real-time drone control system 100 and the drone 101 described above.

Referring to FIG. 9, in step 901, the real-time drone control system 100 transmits a spray start command to the drone 101.

At steps 902 and 903, the drone 101, which has received the spray start command, starts flying from the home point and checks whether a predetermined altitude has been reached.

The drone 101, which has reached the predetermined altitude in step 904, is flying forward, spraying the material for fertilization or control to the spraying area on the farmland.

In step 905, the drone 101 photographs the ground image including the farmland, and transmits the photographed ground image to the real-time drone control system 100.

In steps 906 and 907, the real-time drone control system 100 analyzes the ground image continuously received from the drone 101 according to the machine learning technique, and determines whether the spraying area is adjacent to the boundary of the farmland.

If it is determined that the spraying area is adjacent to the boundary of the farmland (that is, 'farm road'), in step 908, the real-time drone control system 100 records the spraying area in the database DB as terrain data of the farmland. .

In step 909, the real-time drone control system 100 checks whether an outer farmland surrounding the farmland is configured when all the sprayed areas recorded in the database are connected.

As a result of the check in step 909, if the spreading areas recorded as the terrain data have not constituted all of the outer farm roads, that is, if there are still farm lands that have not been fertilized and controlled, real-time drone control in steps 910 and 911 The system 100 determines the direction in which the drone 101 will switch, and transmits a direction change command to the drone 101 along with a spray waiting command.

In steps 912 and 913, the drone 101 temporarily stops spraying, and then moves forward after switching to a direction designated by the redirection command.

In step 914, the real-time drone control system 100 re-transmits the start-of-spray command to the drone 101 flying forward, spraying the material for fertilization or control into the spraying area on the farmland.

As a result of the check in step 909, when the sprayed areas recorded as the terrain data form all of the outer farm road, in step 915, the real-time drone control system 100 recognizes that fertilization and control have been performed on all areas of the farmland. The drone 101 is sent to the drone 101, and the drone 101, which receives the sprinkling end instruction in step 916, is recovered to the home point.

As described above, according to an embodiment of the present invention, a process of setting a flight route in advance during fertilization and control using a drone is omitted, so a process of manufacturing a precise orthogonal image map through low-altitude aerial photography is also unnecessary. In addition, it is possible to greatly reduce the time and cost of preliminary preparation, and to minimize the area that is incorrectly distributed due to the coordinate error of the satellite map.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, or the like alone or in combination. The program instructions recorded in the medium may be specially designed and configured for the embodiments or may be known and usable by those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs, DVDs, and magnetic media such as floptical disks. -Hardware devices specifically configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter, etc., as well as machine language codes produced by a compiler. The hardware device described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

The software may include a computer program, code, instruction, or a combination of one or more of these, and configure the processing device to operate as desired, or process independently or collectively You can command the device. Software and / or data may be interpreted by a processing device, or to provide instructions or data to a processing device, of any type of machine, component, physical device, virtual equipment, computer storage medium or device. , Or may be permanently or temporarily embodied in the transmitted signal wave. The software may be distributed over networked computer systems, and stored or executed in a distributed manner. Software and data may be stored in one or more computer-readable recording media.

As described above, although the embodiments have been described with limited drawings, those skilled in the art may apply various technical modifications and variations based on the above. For example, the described techniques are performed in a different order than the described method, or components of the described system, structure, device, circuit, etc. are combined or combined in a different form from the described method, or other components or equivalents Even if replaced by or substituted by, appropriate results can be achieved.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

100: real-time drone control system
101: drone
110: controller
120: analysis unit
130: database

Claims (16)

In response to the issuance of a spraying start command, spraying a material for fertilization or control held in a drone flying forward at a predetermined altitude to a spraying area on the farmland corresponding to the altitude; And
When the drone is spraying the material while flying, when the spraying area is adjacent to the boundary of the farmland,
In the ground image photographed at the time when it is determined that the spreading area is adjacent to the boundary of the farmland, the proportion of the crop area in the first direction based on the center point is cropped in the second direction opposite to the first direction. If the area is larger than the percentage occupied,
Writing a direction change command for determining the first direction as a direction in which the drone will switch; And
Sending the direction changing command to the drone together with a spray waiting command.
Real-time drone control method for fertilization and control, including. According to claim 1,
The ground image photographed through the camera mounted on the drone is analyzed according to a machine learning technique (ML),
Determining that the spreading area is adjacent to the boundary of the farmland when the spreading area overlaps at least a portion of an outer farm road surrounding the farmland.
Real-time drone control method for fertilization and control further comprising a. According to claim 2,
Using the machine learning technique,
Extracting a crop area corresponding to the farmland from the ground image; And
If there is at least one contact point between the spreading area and the border of the crop area, determining that the spreading area is adjacent to the boundary of the farmland.
Real-time drone control method for fertilization and control further comprising a. According to claim 3,
Identifying the presence or absence of the contact point by excluding an area in which the center point of the ground image is not included among the crop areas.
Real-time drone control method for fertilization and control further comprising a. delete According to claim 1,
When the direction of the drone changes the direction of the drone to the first direction by the direction change command, when the crop area in the first direction or the crop area in the second direction is deviated,
Sending a command to end the spraying to the drone, and recovering the drone to a designated home point
Real-time drone control method for fertilization and control further comprising a. According to claim 1,
When the direction of the drone is changed by the direction change command,
Controlling the drone to move forward by the horizontal length of the spray area and to fly forward while the drone head is rotated by a predetermined angle; And
Re-transmitting the spray start command as the drone moves forward.
Real-time drone control method for fertilization and control further comprising a. The method of claim 6,
After resending the spray start command, whenever the spray area is adjacent to the farmland boundary,
Retransmitting the direction change command to determine the direction opposite to the direction in which the drone switched immediately before, and the direction in which the drone is to switch, to the drone together with a spray waiting command.
Real-time drone control method for fertilization and control further comprising a. The method of claim 6,
After re-transmitting the spraying start command, when the spraying area is adjacent to the boundary of the farmland, the proportion occupied by the crop area in the first direction, the proportion occupied by the crop area in the second direction, and the selected range If it is the same within,
Retransmitting the redirection command to the drone to determine the second direction in which the drone does not switch, in the direction in which the drone is to switch, at the first direction change point converted to the first direction.
Real-time drone control method for fertilization and control further comprising a. According to claim 1,
Recording a spraying area adjacent to the boundary of the farmland in a database as terrain data for the farmland; And
When connecting the spreading areas recorded as the terrain data, when constructing an outer farmland surrounding the farmland, sending a command to end the spreading to the drone, and recovering the drone to a designated home point
Real-time drone control method for fertilization and control further comprising a. A controller for spraying material for fertilization or control held in a drone flying forward at a predetermined altitude to a spraying area on the farmland corresponding to the altitude, according to the transmission of the spraying start command; And
When the drone is spraying the material while flying, when the spraying area is adjacent to the boundary of the farmland,
In the ground image photographed at the time when it is determined that the spreading area is adjacent to the boundary of the farmland, the proportion of the crop area in the first direction based on the center point is cropped in the second direction opposite to the first direction. Analysis unit to check whether the area is greater than the proportion
Including,
If the ratio occupied by the crop region in the first direction is larger than the ratio occupied by the crop region in the second direction,
The controller,
Write a direction change command to determine the first direction, the direction in which the drone will switch,
The direction change command is sent to the drone along with the standby command.
Real-time drone control system for fertilization and control. The method of claim 11,
The analysis unit,
When the ground image photographed through the camera mounted on the drone is analyzed according to a machine learning technique (ML), if the spraying area overlaps at least a part of an outer farm road surrounding the farmland, the spraying area is the boundary of the farmland Judging by
Real-time drone control system for fertilization and control. delete The method of claim 11,
When the direction of the drone is changed by the direction change command,
The controller,
The drone is moved to the horizontal length of the spray area, and the head of the drone is controlled to fly forward while rotating by a predetermined angle.
As the drone moves forward, it resends the command to start spraying.
Real-time drone control system for fertilization and control. The method of claim 14,
After resending the spray start command, whenever the spray area is adjacent to the farmland boundary,
The controller,
Re-transmitting the redirection command to determine the direction opposite to the direction in which the drone switched immediately before, the direction in which the drone will switch, and to the drone together with the spraying standby command.
Real-time drone control system for fertilization and control. The method of claim 11,
The analysis unit,
The spreading area adjacent to the boundary of the farmland is recorded in a database as terrain data for the farmland,
The controller,
When connecting the spreading areas recorded as the terrain data, when constructing an outer farmland surrounding the farmland, a command to end the spreading is sent to the drone to recover the drone to a designated home point.
Real-time drone control system for fertilization and control.

Images