KR20220035797A - SMART PEST CONTROL DEVICE USING LiDAR THAT MINIMIZES THE EFFECT OF CHEMICALS AND OPERATION METHOD THEREOF - Google Patents

Abstract

The present invention relates to a smart control device capable of selective control through LiDAR, and to fruit tree recognition and presence/absence determination technology using LiDAR that minimizes the impact of chemical solutions. The smart pest control device includes a recognition unit that determines the shape and presence of fruit trees through LiDAR, and a shape data processing unit that compares and calculates forward data at time t and rear data at time t+i based on the recognition unit data. and a control unit that generates a control signal for controlling the injection device based on the shape data processing unit.

Description

Smart pest control device and method using LiDAR that minimizes the impact of chemicals {SMART PEST CONTROL DEVICE USING LiDAR THAT MINIMIZES THE EFFECT OF CHEMICALS AND OPERATION METHOD THEREOF}

The present invention relates to fruit tree recognition and presence/absence determination technology using LiDAR that minimizes the impact of chemical solutions.

To spray pesticides on crops planted in a house or field, pesticides contained in a medicine container were pumped up and sprayed while the worker moved directly along the furrow while holding a medicine handle connected to a hose. However, recently, pesticides were sprayed on the medicine container, pump, connecting hose, and both sides. With the spread of pesticide sprayers that doubled as agricultural power transporters equipped with pesticides, it became possible to apply pesticides while driving along the furrows of the field more comfortably.

However, since the worker must follow each person and perform appropriate operation operations until the pesticide spraying operation is completely completed, not only is a minimum amount of manpower required, but it is still difficult to avoid inhalation of a portion of the toxic pesticide, which is only a minor difference in the pesticide spraying process. There is a risk of damage to the worker's health.

In addition, the effect of fog, that is, water (fog) in particle form can be recognized, which causes the problem that it is very difficult to recognize the environment in a foggy state due to the effect of fog.

The purpose of the present invention is to provide a smart pest control device and method capable of controlling pest control using LiDAR, which can minimize the impact of sprayed and scattered chemical solutions when the smart control apparatus performs pest control work.

A smart pest control device according to an embodiment of the present invention includes a recognition unit that recognizes the forward data of the nth fruit number and the backward data of the n-1th fruit number at time t through LiDAR, and the nth fruit number at time t. A shape data processing unit that generates synthetic fruit tree shape data based on the forward data of and the backward data of the nth fruit tree at time t+i, and a spray device for controlling the nth fruit tree based on the results of the shape data processing unit. It includes a control unit that generates a control signal to control.

In one embodiment of the present invention, the recognition unit recognizes the n-th forward data and the n+1-th forward data at the time t+i.

In one embodiment of the present invention, when the LiDAR irradiates a light source with j channels, the front data is data of j/3 channels in the front part based on the vertical direction of the driving direction, and the rear data is This is the data of the j/3 channels in the rear part based on the vertical direction of the driving direction.

In one embodiment of the present invention, the shape data processing unit performs an AND operation on the front data and the rear data.

In one embodiment of the present invention, the injection device includes at least one flow rate control unit that adjusts the injection amount, at least one pressure control unit that controls the injection pressure, and at least one that individually controls the opening and closing of a plurality of injection holes. Includes a nozzle control unit.

In one embodiment of the present invention, the injection device is located rearward of the LiDAR.

In the method of operating a smart pest control device according to an embodiment of the present invention, the recognition unit recognizes the forward data of the nth fruit number and the rear data of the n-1th fruit number at time t through LiDAR, shape data processing Additionally, generating synthetic fruit shape data based on the forward data of the nth fruit number at time t and the backward data of the nth fruit number at time t+i, and the control unit generating synthetic fruit shape data based on the results of the shape data processing unit. It includes generating a control signal for controlling a spraying device for controlling the nth fruit tree.

In one embodiment of the present invention, recognizing forward data for the n-th fruit number at the time t, backward data for the n-th fruit number and forward data for the n+1-th fruit number at the time t+i. Recognizing data and comparing forward data for the nth fruit number and backward data for the nth fruit number.

In one embodiment of the present invention, the shape data processing unit at the time t+i performs an AND operation on the forward data and the backward data of the nth derivative.

The smart pest control system according to an embodiment of the present invention includes a smart pest control device and a recognition device that recognizes the direction of progress of the smart pest control device, and the smart pest control device is configured to determine the nth number of plants at time t through LiDAR. A recognition unit that recognizes forward data and backward data of the n-1th fruit number, and synthetic fruit shape data based on the forward data of the nth fruit number at time t and the backward data of the nth fruit number at time t+i. It includes a shape data processing unit that generates and a control unit that generates a control signal for controlling an injection device for controlling the nth orchard based on the results of the shape data processing unit.

The smart pest control device capable of controlling pest control through LiDAR of the present invention can minimize the impact of sprayed and scattered chemical solutions when the smart pest control device performs pest control work.

Figure 1 is a block diagram of a smart pest control device according to an embodiment of the present invention.
Figure 2 is a block diagram of a control unit according to an embodiment of the present invention.
Figure 3 is a diagram showing the recognition area of LiDAR according to an embodiment of the present invention.
4 and 5A to 5D are diagrams showing a LiDAR data recognition method according to an embodiment of the present invention.
Figure 6 is a flowchart showing the operation method of a smart pest control device according to an embodiment of the present invention.
Figure 7 is a flowchart showing the operation method of the recognition unit and the shape data processing unit according to an embodiment of the present invention.
Figure 8 is a flowchart showing a method of operating a control unit according to an embodiment of the present invention.
Figure 9 is a block diagram of a smart pest control device according to another embodiment of the present invention.
Figure 10 is a diagram showing a smart pest control system according to an embodiment of the present invention.

Hereinafter, embodiments disclosed in this specification will be described in detail with reference to the drawings. Unless otherwise specified in the text, similar reference numbers in the drawings indicate similar components. The exemplary embodiments described above in the detailed description, drawings, and claims are not intended to be limiting, and other embodiments may be used, and other changes may be made without departing from the spirit or scope of the technology disclosed herein. Those skilled in the art will be able to arrange, configure, combine, and design the components of the present disclosure, i.e., the components generally described herein and shown in the drawings, into various different configurations, all of which are clearly designed to be used in the present disclosure. It will be easy to understand that it forms part of .

Since the description of the disclosed technology is only an example for structural or functional explanation, the scope of rights of the disclosed technology should not be construed as limited by the examples described in the text. In other words, since the embodiments can be modified in various ways and can take various forms, the scope of rights of the disclosed technology should be understood to include equivalents that can realize the technical idea.

All terms used herein, unless otherwise defined, have the same meaning as commonly understood by a person of ordinary skill in the field to which the disclosed technology pertains. Terms defined in commonly used dictionaries should be interpreted as consistent with the meaning they have in the context of the related technology, and cannot be interpreted as having an ideal or excessively formal meaning unless clearly defined in the present application.

Figure 1 is a block diagram of a smart pest control device 10 according to an embodiment of the present invention.

Figure 2 is a block diagram of the control unit 300 according to an embodiment of the present invention.

Referring to FIGS. 1 and 2 , the smart pest control device 10 includes a recognition unit 100, a shape data processing unit 200, and a control unit 300. The recognition unit 100 recognizes the forward data of the nth number and the backward data of the n-1th number at time t through LiDAR. At this time, n may be a natural number of 2 or more.

LiDAR is a detection sensor that irradiates an object with a laser (e.g., infrared, visible light, etc.) with a shorter wavelength than electromagnetic waves and receives the light reflected from the object to determine the distance, direction, altitude, speed, etc. of the object. means. For example, LiDAR can recognize the shape and presence of fruit trees using a wavelength of 850 nm to 950 nm.

LiDAR may include a light source that transmits a laser and a receiver that receives reflected light. These LiDARs have higher azimuth resolution and distance resolution than radar sensors.

LiDAR may be attached to the rear of the smart pest control device 10. LiDAR can obtain distance information for each point inside a rice field or field. LiDAR is mounted vertically facing the sky and can recognize environmental information on the left and right based on a width of ±15 degrees.

The recognition unit 100 may recognize forward data for the nth fruit number at time t. Additionally, at time t+i, backward data for the n-th fruit number and forward data for the n+1-th fruit number can be recognized.

When LiDAR investigates with j channels, the forward data is data of j/3 channels in the driving direction based on the vertical direction of the driving direction, and the rear data is data of j/3 channels in the reverse direction of the driving direction based on the vertical direction of the driving direction. It may be 3-channel data.

In one embodiment, j channels may be 16. In this case, LiDAR can survey ±15 degrees with 16 channels at 2-degree intervals. In another embodiment, j channels may be 32. In this case, LiDAR can survey 32 channels at ±15 degrees at 1-degree intervals.

For example, when irradiating a laser through 16 channels at ±15 degrees at 2-degree intervals, the forward data is data from 8 channels in the driving direction based on the vertical direction of the driving direction, and the backward data is data perpendicular to the driving direction. Based on the direction, it is 8 channels of data in the reverse direction of the driving direction.

In one embodiment, the recognition unit 100 recognizes the forward data for the n-1th fruit tree at time t when the n-1th fruit tree and the n-th fruit tree are in accordance with the moving direction of the smart pest control device, and t At time +i, backward data for the n-1th fruit number and forward data for the nth fruit number are recognized.

In one embodiment, in the case of the first fruit tree, there is no prior pest control work and there is no invasive effect from the chemical solution, so the process of recognizing forward data can be omitted.

In one embodiment, the recognition unit 100 may additionally recognize side data of the nth integer at time t+p. Time t+p is between time t and time t+i, and may be 0<p<i. At this time, if the driving speed is constant, the t+p time may be an intermediate value between the t time and the t+i time. When LiDAR has j channels, side data of fruit trees can be recognized through j/3 channels on the side, based on the vertical direction of the driving direction. In this case, the shape data processing unit 200 can increase the accuracy of data recognition by additionally acquiring side data as well as forward and backward data.

The shape data processing unit 200 compares and calculates the forward data at time t and the backward data at time t+i, based on the data of the recognition unit 100. In one embodiment, the operation method may use AND operation. However, the present invention is not limited to this, and video composite image data can also be used.

By comparing the front data at time t and the rear data at time t+i, the shape data processing unit 200 can minimize the influence of the scattered chemical solution. For the same fruit tree, at time t, forward data less affected by spray can be obtained, and at time t+i, backward data affected by spray generated by the previous fruit tree can be obtained. The forward data and backward data may be image data of a fruit tree.

At this time, since the location where the chemical solution is sprayed is located behind LiDAR, if control is performed based only on rear data without forward data, recognition of fruit trees may be inaccurate due to the influence of the previously sprayed chemical solution. Therefore, the present invention obtains forward data for the same fruit tree, compares forward and backward data, and controls the smart pest control device 10 based on information detected in real time, thereby minimizing the impact of the chemical solution.

For example, suppose that after spraying a chemical solution on a first fruit tree, the chemical solution must be sprayed on a second fruit tree. When spraying a chemical solution to the second fruit tree, a problem may arise where it is difficult to recognize the rear data of the second fruit tree due to the influence of the chemical solution sprayed on the first fruit tree. Therefore, forward data for the second fruit tree is acquired at time t, backward data for the second fruit tree is acquired at time t+i, and the forward data and backward data for the second fruit tree are compared and calculated to obtain the second fruit tree. By controlling the spraying of the chemical solution, the impact of the chemical solution due to previous overwatering can be minimized.

The control unit 300 generates a control signal for controlling a spraying device for controlling the nth fruit tree based on the shape data processing unit 200. The control unit 300 controls the injection device only when front data and rear data for the same fruit tree overlap based on the results of the shape data processing unit. The control unit 300 generates a signal that controls at least one of the injection amount, injection pressure, and opening and closing of the injection hole of the injection liquid injected into the fruit tree.

The control unit 300 may include at least one flow rate control unit 320, a pressure control unit 340, and a nozzle control unit 360.

At least one flow control unit 320 may control the injection amount of the injection device. At least one pressure control unit 340 may adjust the injection pressure of the injection device.

At least one nozzle control unit 360 may individually control the opening and closing of a plurality of injection nozzles. For example, if the spraying direction should be upward, the top injection port can be opened, and if spraying can be done only at the bottom, only the bottom injection port can be opened.

For example, if the height of the apple tree is high, the pressure is increased to spray the spray farther and the spray nozzle is opened according to the height. In addition, when the straight line distance is long, a pressure control signal to increase the pressure to the pressure control unit and a flow control signal to increase the flow rate to the flow control unit are transmitted to increase the pressure and spray a large flow rate.

In one embodiment, as a result of the calculation of the shape data processing unit 200, the spray nozzle is closed and no pesticide is sprayed in the first area without fruit trees, and the pesticide is not sprayed in the second area including at least one fruit tree with fruit trees according to the reference information. Based on this, the spray nozzle can be opened to spray an appropriate amount of pesticide.

In one embodiment, the smart pest control device 10 may communicate with a user terminal.

Figure 3 is a diagram showing the recognition area of LiDAR according to an embodiment of the present invention.

Referring to Figure 3, the smart pest control device can determine the shape and presence of fruit trees in the recognition area through LiDAR. LiDAR is mounted vertically facing the sky and can recognize environmental information on the left and right based on a width of ±15 degrees. LiDAR can be mounted vertically to view the sky to manage tall fruit trees. In one embodiment, ±15 degrees may irradiate laser through 16 channels at 2-degree intervals.

If the recognition unit determines that there is a fruit tree within the recognition area through LiDAR, the shape data processing unit compares and calculates the front data and the rear data. As a result of the comparison operation, if it is determined that the fruit number recognized in the forward data at time t and the backward data at time t+i are the same fruit number, the control unit sprays a preset amount of injection liquid and closes the injection port of the injection nozzle. Then, move to the next fruit tree.

For example, when the preset spray amount of one apple tree is 0.1L and the preset spray amount of one pear tree is 0.5L, 0.1L of spray liquid is sprayed on one apple tree in the apple tree area and then sprayed on the next apple tree. Close the nozzle until. Afterwards, the nozzle is opened to spray 0.1L of the spray liquid on the next apple tree, and the same process as the previous one is repeated for all apple trees.

4 and 5A to 5D are diagrams showing a LiDAR data recognition method according to an embodiment of the present invention.

Referring to FIGS. 4 and 5A to 5D , the recognition unit of the smart pest control device may recognize forward data for the nth fruit tree at time t. At this time, backward data for the n-1th fruit number can be recognized, but it is not shown in the drawing. Additionally, at time t+i, backward data for the n-th fruit number and forward data for the n+1-th fruit number can be recognized.

For example, the smart control device at time t recognizes forward data for the first fruit tree, and the smart control device at time t+i recognizes backward data for the first fruit tree and performs comparison operations. At this time, at time t+i, the smart control device recognizes forward data about the second fruit tree along with backward data about the first fruit tree. The moving direction of the smart pest control device may be from the first fruit tree to the second fruit tree.

The shape data processing unit provides the operation result for the first fruit number to the control unit. If the control unit determines that the calculation results are the same, it generates a control signal to control the injection device.

Figure 5a is a view from the top when LiDAR irradiates a light source. In one embodiment, LiDAR can illuminate 16 channels each on the left and right sides. FIG. 5B is a view from above of rear data corresponding to each j/3 channel on the left and right sides of the LiDAR irradiated in FIG. 5A. FIG. 5C is a diagram showing the result of reflection of the LiDAR light source emitted in FIG. 5A as seen in the driving direction. FIG. 5D is a view showing the result of reflection of the LiDAR light source irradiated in FIG. 5B viewed from the driving direction. In one embodiment, the smart pest control device may include six injection ports on the left side corresponding to L1 to L6 and six injection ports on the right side corresponding to R1 to R6, as shown in FIGS. 5C and 5D. Each height of the fruit tree corresponds to an injection hole, and the controller can individually control the opening and closing of the corresponding injection hole according to each height, as described above.

As shown in Figure 5d, when only the posterior data of the nth fruit tree is measured at time t+i, a problem may occur in that it is difficult to recognize the fruit tree due to noise caused by preceding pest control work. Therefore, the present invention can recognize the fruit number by removing noise by recognizing the forward data for the nth fruit number at time t and comparing it with the backward data of the nth fruit number at time t+i.

Figure 6 is a flowchart showing the operation method of a smart pest control device according to an embodiment of the present invention.

Figure 7 is a flowchart showing the operation method of the recognition unit and the shape data processing unit according to an embodiment of the present invention.

Figure 8 is a flowchart showing a method of operating a control unit according to an embodiment of the present invention.

Referring to Figures 6 to 8, the operating method of the smart pest control device is a step (S10) in which the recognition unit recognizes the front data of the nth fruit number and the rear data of the n-1th fruit number at time t through LiDAR, shape A step (S30) in which the data processing unit compares and calculates the forward data of the n-th plus number at time t and the backward data of the n-th plus number at time t+i, based on the data of the recognition unit, and the control unit performs a comparison operation in the shape data processing unit. and generating a control signal for controlling the injection device based on the step (S50).

A step (S10) in which the recognition unit recognizes the n-th forward data and the n-1th backward data at time t through LiDAR, and the shape data processing unit recognizes the first data at time t based on the data of the recognition unit. The step of comparing and calculating the forward data of the n-th fruit number and the backward data of the n-th fruit number at time t+i (S30) includes the step of recognizing the forward data of the n-th fruit number at time t (S11), the step of recognizing the forward data of the n-th fruit number at time t+i. A step (S13) of recognizing backward data for the nth fruit number and forward data for the n+1th fruit number, and performing a comparison operation (e.g., AND operation) between the recognized forward data and backward data for the same fruit number. Includes step S21.

For example, when the first fruit tree and the second fruit tree are located along the moving direction of the smart pest control device, forward data for the first fruit tree is recognized at time t. At time t+i, backward data for the first fruit tree and forward data for the second fruit tree are recognized. If the backward data for the first fruit number is recognized at time t+i, a comparison operation can be performed with the forward data for the first fruit number stored at the previous time t.

The step (S50) in which the control unit generates a control signal for controlling the injection device based on the shape data processing unit may generate a control signal for controlling the injection device for an object that overlaps the front data and the rear data.

The control unit includes at least one flow control unit controlling the injection amount (S51), at least one pressure control unit controlling the injection pressure (S53), and at least one nozzle control unit individually controlling the opening and closing of the plurality of injection holes. It may include step S55.

In the step (S51) in which at least one flow rate control unit adjusts the injection amount, the injection amount of each injection device can be adjusted independently. For example, if the first fruit tree needs a large amount of chemical solution, the control unit may provide a signal to the flow control unit to provide a large amount of chemical solution. The control unit controls the spray amount based on the characteristics of the fruit trees, so pesticide use can be minimized and costs can be reduced.

In the step (S53) in which at least one pressure control unit adjusts the injection pressure, the injection pressure of each injection device can be adjusted independently.

The step (S55) in which at least one nozzle control unit individually controls the opening and closing of the plurality of injection ports can independently control the opening and closing of the injection ports of each injection device.

For example, if the height of the apple tree is high, the pressure is increased so that the spray liquid is sprayed far and the spray hole at the top is opened and closed, and the spray hole is opened according to the height. In addition, when the straight line distance is long, a pressure control signal to increase the pressure to the pressure control unit and a flow control signal to increase the flow rate to the flow control unit are transmitted to increase the pressure and spray a large flow rate.

The smart control device capable of selective control through LiDAR of the present invention can solve the difficulties that arise when recognizing fruit trees using only rear data by comparing and calculating front data and rear data for the same fruit tree. there is.

Figure 9 is a block diagram of a smart pest control device 20 according to another embodiment of the present invention.

Referring to FIG. 9, the smart pest control device 20 may include a plurality of sensors 50, a recognition unit 100, a shape data processing unit 200, and a control unit 300. The recognition unit 100 and the shape data processing unit 200 are the same as the recognition unit 100 and the shape data processing unit 200 shown in FIG. 1, and the control unit 300 includes a time control unit ( 380) may be further included.

Through a plurality of sensors 50 and LiDAR, at least one of the fruit tree's height, color, shape, size, height, leaf distribution, disease and pest information, pesticide mixing ratio information according to disease and pests, and PLS information can be obtained. For example, if it is determined that there is a pest based on the pest information, information on pesticides by type, amount of pesticide, and mixing ratio of each pesticide stored in advance for the pest can be provided in order to remove or prevent the pest.

The time control unit 380 calculates the height of the fruit tree and the distance between the fruit tree and the smart control device 20 through a plurality of sensors 50 and LiDAR, and determines the stopping time of the smart control device 20 required to spray the chemical solution. You can control it. For example, for fruit trees that require 0.5L of chemical solution, the smart pest control device 20 can be stopped for 1 minute, and for fruit trees that require 1L of chemical solution, the smart pest control device 20 can be stopped for 2 minutes. The time control unit 380 can control the smart pest control device 20 to stop for the time required for spraying so that a preset chemical solution can be sprayed.

Figure 10 is a diagram showing a smart pest control system according to an embodiment of the present invention.

Referring to FIG. 10, the smart pest control system may include a recognition device 80 and a smart pest control device 30. The recognition device 80 recognizes the moving direction of the smart pest control device 30. The recognition device 80 may be located on the upper side of the smart pest control device 30.

The recognition device 80 may include LiDAR. LiDAR can be sprayed into j channels. The recognition device 80 is an IP camera, HD-SDI camera, analog camera, fire detection color camera, thermal imaging camera, HD (1920x5080, HD5080p) camera at SD (720x486, NTSC) resolution, IP zoom speed camera, or CCTV. Camera, ultrasonic sensor, laser sensor, infrared and thermal sensor, proximity sensor, steadycam, gyro sensor, gesture sensor, barometric pressure sensor, magnetic sensor, acceleration sensor, grip sensor, color sensor, biometric sensor, illuminance sensor, and UV ( It may further include at least one of the ultra violet) sensors. The smart pest control system 1000 can further increase the recognition accuracy of fruit trees through the recognition device 80.

The smart pest control device 30 includes a recognition unit 100, a shape data processing unit 200, and a control unit 300. The recognition unit 100 determines the shape and presence of fruit trees through LiDAR. The smart pest control device 30 may be the same as the smart pest control device 10 disclosed in FIG. 1.

The smart control device capable of selective control through LiDAR of the present invention can minimize the impact of sprayed and scattered chemical solutions when the smart control device performs control work.

10, 20, 30: Smart pest control device
50: sensor
80: Recognition device
100: recognition unit
200: Shape data processing unit
300: control unit
320: Spray nozzle control unit
340: flow control unit
360: pressure control unit
380: time control unit
1000: Smart pest control system

Claims (10)

A recognition unit that recognizes the forward data of the nth number and the backward data of the n-1th number at time t through LiDAR;
a shape data processing unit that generates synthetic fruit shape data based on the forward data of the nth fruit number at time t and the backward data of the nth fruit number at time t+i; and
A smart pest control device comprising a control unit that generates a control signal to control a spraying device for controlling the nth orchard based on the results of the shape data processing unit. According to paragraph 1,
The recognition unit is a smart pest control device that recognizes backward data of the n-th fruit number and forward data of the n+1-th fruit number at the time t+i. According to paragraph 2,
When the LiDAR irradiates light sources through j channels,
The forward data is data of j/3 channels at the front based on the vertical direction of the driving direction,
The rear data is a smart pest control device that is data of j/3 channels at the rear based on the vertical direction of the driving direction. According to paragraph 1,
The shape data processing unit is a smart pest control device that performs an AND operation on the front data and the rear data. According to paragraph 1,
The injection device is,
At least one flow rate control unit that adjusts the injection amount;
At least one pressure control unit that adjusts the injection pressure; and
A smart pest control device including at least one nozzle control unit that individually controls the opening and closing of a plurality of injection nozzles. According to clause 5,
The spray device is a smart pest control device located behind the LiDAR. In the operation method of the smart pest control device,
A step where the recognition unit recognizes the forward data of the nth number and the backward data of the n-1th number at time t through LiDAR;
A shape data processing unit generating synthetic fruit shape data based on forward data of the nth fruit number at time t and backward data of the nth fruit number at time t+i; and
A method of operating a smart pest control device comprising the step of a control unit generating a control signal to control a spraying device for controlling the nth orchard based on the results of the shape data processing unit. In clause 7,
The step of recognizing and comparing the forward data and the backward data includes:
Recognizing forward data for the nth number at time t;
Recognizing backward data for the n-th fruit number and forward data for the n+1-th fruit number at the time t+i; and
A method of operating a smart pest control device comprising comparing forward data for the nth fruit tree and backward data for the nth fruit tree. According to clause 8,
The shape data processing unit at the time t+i performs an AND operation on the front data and the rear data of the nth number. smart pest control device;
It includes a recognition device that recognizes the direction of progress of the smart pest control device,
The smart pest control device is,
A recognition unit that recognizes the forward data of the nth number and the backward data of the n-1th number at time t through LiDAR;
a shape data processing unit that generates synthetic fruit shape data based on the forward data of the nth fruit number at time t and the backward data of the nth fruit number at time t+i; and
A smart control system comprising a control unit that generates a control signal to control a spraying device for controlling the nth fruit tree based on the results of the shape data processing unit.

Images