Ai-based self-sufficient smart farm system and execution method thereof

Abstract

An AI-based self-sufficient smart farm system according to the present invention comprises: a rainwater collection device which is installed at the upper end or on the outer surface of the smart farm system and effectively collects rainwater; a water storage tank which stores the rainwater transferred through the rainwater collection device and supplies the rainwater to the smart farm as agricultural water; a solar panel which is installed outside the smart farm system, absorbs sunlight to generate electricity, and supplies the generated electricity to an electricity storage device in the smart farm; a crop cultivation robot which performs agricultural activities in the smart farm and collects data on the growth state of each crop in the process of performing the agricultural activities; and an AI analysis device which monitors the growth state of the crop on the basis of the data collected by the cultivation robot and controls the operation of an agricultural environment control device.

Description

AI-based self-sufficient smart farm system and method of implementing the same

[0001] The present invention relates to an AI-based self-sufficient smart farm system and a method for implementing the same. More specifically, it relates to an AI-based self-sufficient smart farm system and a method for implementing the same that enables the maximum utilization of agricultural space through a vertical cultivation structure and the automatic performance of farm tasks using robots and AI technology, thereby reducing labor, increasing management efficiency, and lowering labor costs.

[0002] Traditional farming methods require large amounts of water and electricity, and efficient operation is becoming difficult due to changes in the agricultural environment and rising labor costs. In particular, agricultural systems dependent on electricity and water resources pose significant problems in regions facing climate change, resource scarcity, and infrastructure limitations.

[0003] For example, in mountainous regions, deserts, and arid zones, agricultural activities are limited due to difficulties in supplying electricity and water, making it very difficult to operate economically sustainable agriculture under these conditions. These problems lower agricultural productivity and can affect the food supply in the long term.

[0004] Existing agricultural systems largely rely on external resources. Cultivating crops requires a certain amount of water and electricity, which are essential elements for growth and harvesting.

[0005] For example, agricultural water is supplied through canals and irrigation systems, while electricity is essential for operating agricultural machinery, greenhouse temperature control equipment, lighting, and watering systems. Consequently, it is difficult to maintain agricultural productivity in environments where water and electricity are not supplied. Furthermore, these systems not only have high initial installation costs but also require substantial maintenance and management expenses, adding to the economic burden on farmers.

[0006] In particular, self-sufficient agricultural systems utilizing rainwater collection and storage and solar power generation have already proven effective in several studies and experiments, but they have not been widely adopted in actual agricultural fields due to technical limitations and cost issues.

[0007] For example, existing rainwater collection systems are difficult to install on a large scale, and solar power generation systems have limitations as they are heavily dependent on sunlight and make it difficult to utilize stored electricity efficiently. Furthermore, robotic and AI technologies required for agricultural automation are still in the early stages of commercialization; due to high initial investment costs and technical uncertainties, most small-scale farms find it difficult to apply them.

[0008] The present invention aims to provide an AI-based self-sufficient smart farm system and a method for implementing the same, which enables agriculture even in mountainous areas, deserts, and arid regions where power and water supplies are unreliable, by providing an agricultural module capable of self-sufficiency without external water and electricity supply through a rainwater collection and solar power generation system.

[0009] In addition, the present invention aims to provide an AI-based self-sufficient smart farm system and a method for implementing the same, which enables the efficient management of resources and reduces the costs of electricity and water usage required for conventional agriculture by utilizing rainwater, a natural resource, through a rainwater collection device and a water storage tank, and by generating eco-friendly electricity through solar panels and storing it in an electrical storage device.

[0010] In addition, the present invention aims to provide an AI-based self-sufficient smart farm system and a method for implementing the same, which can increase agricultural productivity by maximizing the use of agricultural space through a vertical cultivation structure, thereby enabling the cultivation of more crops in the same area.

[0011] In addition, the present invention aims to provide an AI-based self-sufficient smart farm system and a method for implementing the same, which enables reducing labor, increasing management efficiency, and lowering labor costs by automatically performing farm tasks using robots and AI technology.

[0012] In addition, the present invention aims to provide an AI-based self-sufficient smart farm system and a method for implementing the same, which enhances convenience by allowing a user to remotely monitor and manage the status of a farm through a smartphone-linked system, thereby enabling real-time control of the farm environment anytime and anywhere and immediate response in the event of a problem.

[0013]

[0014] The objects of the present invention are not limited to those mentioned above, and other unmentioned objects and advantages of the present invention may be understood from the following description and will be more clearly understood by the embodiments of the present invention. Furthermore, it will be readily apparent that the objects and advantages of the present invention can be realized by the means and combinations thereof set forth in the claims.

[0015] An AI-based self-sufficient smart farm system for achieving this purpose includes a rainwater collection device installed on the top or external surface of the smart farm system to effectively collect rainwater; a water storage tank that stores rainwater delivered through the rainwater collection device and supplies it to the smart farm as agricultural water; a solar panel installed on the exterior of the smart farm system to absorb sunlight and generate electricity, and supply the generated electricity to an electrical storage device within the smart farm; a crop cultivation robot that performs agricultural activities in the smart farm and collects data on the growth status of each crop during the process of performing said agricultural activities; and an AI analysis device that monitors the growth status of crops based on the data collected by the cultivation robot and controls the operation of an agricultural environment control device.

[0016] In one embodiment, the solar panel is formed with an angle adjustment device that adjusts the tilt and direction according to the position of the sun so that sunlight is incident nearly perpendicularly on the panel surface.

[0017] In one embodiment, the AI ​​analysis device can analyze pixels of an image collected by the cultivation robot to extract features, divide the image into a crop portion and a background portion, determine whether pests have occurred based on pixels of the crop portion, and control the cultivation robot so that an agricultural environment control device for removing pests operates according to the determination result.

[0018] In one embodiment, the AI ​​analysis device examines each pixel of the crop portion to calculate the number of normal crop colors and crop colors associated with pests, and if the number of crop colors associated with pests is greater than a specific threshold as a result of the verification, it can determine that pests are present in the area.

[0019] In addition, a method for implementing an AI-based self-sufficient smart farm system to achieve this purpose includes the steps of: collecting rainwater through a rainwater collection device installed on the top or external surface of the smart farm system and storing the rainwater in a water storage tank; absorbing sunlight through a solar panel installed outside the smart farm system and converting it into electricity to supply power to an electrical storage device within the smart farm; performing agricultural activities in the smart farm using a crop cultivation robot and collecting data on the growth status of each crop during this process; and analyzing the collected data using an AI analysis device and controlling the operation of an agricultural environment control device according to the growth status of the crops or supplying the rainwater stored in the water storage tank to the smart farm as agricultural water.

[0020] In one embodiment, the method for executing an AI-based self-sufficient smart farm system may further include the step of adjusting the tilt and direction according to the position of the solar light through an angle adjustment device formed on the solar panel so that the solar light is incident nearly perpendicularly on the surface of the panel.

[0021] In one embodiment, a method for executing an AI-based self-sufficient smart farm system may include the steps of: the AI ​​analysis device analyzing pixels of an image collected by the cultivation robot to extract features; dividing the image into a crop portion and a background portion and determining whether pests have occurred based on pixels of the crop portion; and controlling the cultivation robot so that an agricultural environment control device for removing pests operates according to the determination result.

[0022] In one embodiment, a method for executing an AI-based self-sufficient smart farm system includes the step of the AI ​​analysis device examining each pixel of the crop portion to calculate the number of normal crop colors and crop colors associated with pests, and the step of determining that pests exist in the area if the number of crop colors associated with pests in the verification result is greater than or equal to a specific threshold.

[0023]

[0024] According to the present invention as described above, by providing an agricultural module capable of self-sufficiency without external water and electricity supply through a rainwater collection and solar power generation system, there is an advantage that agriculture is possible even in mountainous areas, deserts, and arid regions where electricity and water supply are unreliable.

[0025] Furthermore, according to the present invention, the purpose is to provide an AI-based self-sufficient smart farm system and a method for implementing the same, which enables the reduction of electricity and water usage costs required for conventional agriculture and the efficient management of resources by utilizing rainwater, a natural resource, through a rainwater collection device and a water storage tank, and generating eco-friendly electricity through a solar panel and storing it in an electrical storage device.

[0026] In addition, according to the present invention, there is an advantage in that agricultural productivity can be increased by maximizing the use of agricultural space through a vertical cultivation structure, thereby enabling the cultivation of more crops in the same area.

[0027] In addition, according to the present invention, by utilizing robot and AI technologies to automatically perform farm tasks, there is an advantage of reducing labor, increasing management efficiency, and lowering labor costs.

[0028] In addition, according to the present invention, by enabling a user to remotely monitor and manage the status of a farm through a smartphone-linked system, convenience is enhanced, thereby allowing the farm environment to be controlled in real time anytime and anywhere, and enabling immediate response in the event of a problem.

[0029] FIG. 1 is a network configuration diagram for explaining an AI-based self-sufficient smart farm system according to one embodiment of the present invention.

[0030] FIG. 2 is a flowchart illustrating a method for implementing an AI-based self-sufficient smart farm system according to the present invention.

[0031] FIGS. 3 to 6 are illustrative diagrams for explaining an AI-based self-sufficient smart farm system according to an embodiment of the present invention.

[0032]

[0033] The aforementioned objectives, signatures, and advantages are described in detail below with reference to the attached drawings, thereby enabling those skilled in the art to easily implement the technical concept of the present invention. In describing the present invention, detailed descriptions of known technologies related to the present invention are omitted if it is determined that such descriptions would unnecessarily obscure the essence of the invention. Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the attached drawings. In the drawings, the same reference numerals are used to indicate the same or similar components.

[0034]

[0035] FIG. 1 is a network configuration diagram for explaining an AI-based self-sufficient smart farm system according to one embodiment of the present invention.

[0036] Referring to FIG. 1, the AI-based self-sufficient smart farm system includes a rainwater collection device (100), a water storage tank (200), a solar panel (300), an electric storage device (400), a vertical cultivation area (500), a cultivation robot (600), and an AI analysis device (700).

[0037] A rainwater collection device (100) is a device installed on the top or outer surface of a smart farm system to effectively collect rainwater. This rainwater collection device (100) is mainly installed on a roof or an outer wall, and collects rainwater in one place through a slanted surface or a rainwater collection groove, and then transmits it to a water storage tank (200) through a pipe.

[0038] The water storage tank (200) stores rainwater delivered through the rainwater collection device (100) and supplies it as agricultural water when needed. The water storage tank (200) may be equipped with a sensor for measuring the amount of water and a pump for supplying water.

[0039] A solar panel (300) is installed on the exterior of the smart farm system to convert sunlight into electrical energy. The solar panel (300) can be installed on the module roof or exterior wall, absorbs sunlight during the day to generate electricity, and the generated electricity is supplied to an electrical storage device (400) within the smart farm.

[0040] An angle adjustment device is formed in the solar panel (300), and the tilt (angle of inclination) and direction (angle of direction) of the solar panel are adjusted according to the position (altitude and azimuth) of the sun, thereby optimizing the solar energy generation efficiency so that the sunlight is incident nearly perpendicularly on the surface of the panel.

[0041] The angle control device accurately determines the position of the sun (altitude and azimuth) in real time. To achieve this, the angle control device tracks the sun's position using solar tracking sensors, light sensors, GPS data, or weather data.

[0042] More specifically, the angle adjustment device calculates the solar altitude and solar azimuth. Solar altitude is the angle at which the sun is above the horizon, and is the vertical angle between the horizon and the sun.

[0043] The azimuth of the sun is the angle the sun moves clockwise relative to the north. Azimuths generally range from 0 degrees (north) to 360 degrees (back to north), and for example, east is expressed as 90 degrees, south as 180 degrees, and west as 270 degrees.

[0044] The solar elevation angle is the angle at which the sun is positioned above the horizon, indicating how high the sun is in the sky. The elevation angle ranges from 0 degrees (sunrise and sunset) to 90 degrees (when the sun is at its zenith at noon) relative to the horizon. The angle adjustment device can calculate the solar elevation angle using [Equation 1].

[0045]

[0046] [Mathematical Formula 1]

[0047]

[0048]

[0049] : As the solar declination angle, it varies with the date and is calculated according to [Equation 2].

[0050] : Latitude of current location

[0051] : As an hour angle, calculated according to [Equation 3] as the angle the sun moves from east to west relative to noon.

[0052]

[0053] [Mathematical Formula 2]

[0054]

[0055]

[0056] : Solar declination

[0057] N: Number of days with current weather

[0058]

[0059] [Mathematical Formula 3]

[0060]

[0061]

[0062] t: current time

[0063] H: Hour angle

[0064]

[0065] After that, the angle adjustment device can calculate the sun's azimuth angle through [Equation 1].

[0066]

[0067] [Mathematical Formula 4]

[0068]

[0069]

[0070] H: Hour angle

[0071] : Solar declination

[0072] : Latitude of current location

[0073]

[0074] The angle adjustment device calculates the sun's altitude and azimuth, and then adjusts the panel's tilt (angle of inclination) and direction (angle of azimuth) so that the solar panel faces the sun directly.

[0075] At this time, slope (angle of inclination, ) is an angle indicating how much the solar panel should be tilted relative to the horizontal plane. The angle adjustment device calculates the tilt of the panel based on the elevation angle so that sunlight can be incident perpendicularly on the panel surface as in [Equation 5].

[0076]

[0077] [Mathematical Formula 5]

[0078]

[0079]

[0080] : Angle of inclination

[0081] h: Solar elevation angle

[0082]

[0083] In other words, the angle adjustment device can calculate the tilt of the solar panel through [Equation 5]. For example, if the sun's elevation angle is 45 degrees, the tilt β of the solar panel is calculated as 45° (90° - 45° = 45°). Through this, the angle adjustment device can maintain the angle of the panel close to horizontal when the sun is low, and adjust the panel closer to vertical when the sun is near its zenith.

[0084] In addition, the angle adjustment device determines the tilt (tilt angle β) and direction (azimuth angle γ) of the solar panel based on the sun's altitude and azimuth angle. To adjust the solar panel so that sunlight is incident nearly perpendicularly on the panel surface according to the sun's position, the following optimization conditions are used.

[0085] The electric storage device (400) serves to store electricity generated from solar panels and use it when needed. The electric storage device can mainly use recycled electric vehicle waste batteries or lithium-ion batteries. Through this, electricity generated during the day can be stored, allowing for a stable supply of power even in environments with insufficient sunlight, such as at night or on cloudy days.

[0086] As mentioned above, since raw materials can be obtained from waste batteries without directly mining core raw materials, the supply of battery raw materials is becoming increasingly difficult and raw material resources are finite, leading to a growing number of companies investing in battery recycling.

[0087] The vertical cultivation area (500) consists of cultivation beds stacked in multiple layers. Each cultivation bed provides space for planting crops and can be designed in a fixed or movable form.

[0088] Each of the above cultivation beds is equipped with a watering system that automatically supplies water and nutrients as needed. This watering system regulates the amount of stored water and optimizes water supply according to the crop's water requirements.

[0089] The drainage system connected to the water supply system discharges excess water to prevent it from accumulating in the cultivation beds, and the discharged water can be reused after undergoing a filtration process. This enables the efficient management and recycling of water resources.

[0090] The cultivation robot (600) moves according to the position and height of the crop formed in each vertical cultivation area (500) and can perform tasks necessary for the cultivated plants. For example, it can automatically perform pruning, harvesting, or nutritional status checks.

[0091] The arm of the cultivation robot (600) has joints that can move around multiple axes, allowing it to rotate in multiple directions or move up, down, left, and right. This multi-axis joint structure enables the cultivation robot (600) to perform tasks accurately according to the various positions and heights of the crops.

[0092] An agricultural environment control device that performs various agricultural tasks may be formed at the end of the arm of the cultivation robot (600). At this time, the cultivation robot (600) operates the agricultural environment control device according to the control of the AI ​​analysis device (700).

[0093] In addition, the cultivation robot (600) is equipped with a high-resolution camera and various sensors (temperature, humidity, soil moisture, CO2 concentration, etc.) to monitor the growth status of each crop and collect data.

[0094] The camera of the cultivation robot (600) monitors the color, size, condition, etc. of the crop in real time and transmits the data to the AI ​​analysis device (700). Accordingly, the AI ​​analysis device (700) can analyze the data received from the cultivation robot (600) and control the operation of the cultivation robot (600) based on the analyzed information.

[0095] The AI ​​analysis device (700) can determine the growth status of crops, the occurrence of pests and diseases, and the harvest time based on data collected by the cultivation robot (600), and control the operation of the agricultural environment control device formed in the cultivation robot (600).

[0096] First, the AI ​​analysis device (700) performs feature extraction on images collected by the cultivation robot (600). At this time, the AI ​​analysis device (700) analyzes information such as the color, boundary, and contour of each pixel of the image to extract features such as specific colors, shapes, and textures related to pests.

[0097] After feature extraction is completed, the AI ​​analysis device (700) divides the image into crop parts and background parts. Through this process, key parts such as leaves, stems, and fruits are identified, and parts where pests are likely to occur are set as regions of interest.

[0098] After that, the AI ​​analysis device (700) determines whether pests have occurred based on pixels in the area of ​​interest. This process is achieved through changes in pixel values, color contrast, pattern recognition, etc.

[0099] In one embodiment, the AI ​​analysis device (700) examines each pixel within the region of interest to determine how much of a normal crop color (e.g., green) and other colors (e.g., black, brown, red, etc.) are distributed.

[0100] For example, when the AI ​​analysis device (700) detects a color pixel (a color likely to be associated with pests) that contrasts with the green pixel of the crop above a specific threshold, it calculates the density of the corresponding pixel.

[0101] In another embodiment, the AI ​​analysis device (700) checks whether adjacent pixels are arranged in a pattern similar to a predetermined shape of a pest. This is because the size and shape of the pest have a certain pattern.

[0102] For example, assuming the size of the pest is approximately 10x10 pixels, the AI ​​analysis device (700) can determine that the pest is present in the area when more than 70% of the pixels within the 10x10 area are similar in color to the pest.

[0103] In another embodiment, the AI ​​analysis device (700) detects the presence of pests by grouping (clustering) pixels in an image according to similarity.

[0104] First, the AI ​​analysis device (700) generates a feature vector based on the color value and location information of each pixel to perform pixel clustering. At this time, the feature vector of each pixel includes the color value and location information of the pixel. For example, the feature vector of a pixel can be defined as [R, G, B, x, y] or [H, S, V, x, y].

[0105] Then, the AI ​​analysis device (700) calculates the Euclidean distance by comparing the pixel's characteristic vector with K cluster centers based on [Equation 6] and assigns it to the nearest cluster.

[0106]

[0107] [Mathematical Formula 6]

[0108]

[0109]

[0110] In [Equation 6], (Ri, Gi, Bi, xi, yi) is the feature vector of pixel i, and (Rj, Gj, Bj, xj, yj) is the feature vector of cluster center j. When all pixels are assigned to a cluster, the AI ​​analysis device (700) recalculates the center of each cluster. At this time, the cluster center is set as the average value of the feature vectors of all pixels belonging to that cluster. That is, the new center Cj of cluster j is calculated as in [Equation 7].

[0111]

[0112] [Mathematical Formula 7]

[0113]

[0114]

[0115] In [Equation 7], (Ri, Gi, Bi, xi, yi) is the feature vector of pixel i, and N is the number of pixels belonging to cluster j.

[0116] If the AI ​​analysis device (700) terminates clustering when the cluster center does not change, or even if it does change, the amount of change is below a preset threshold. Each cluster determined through the above process is a group of pixels clustered based on a specific color and location. Accordingly, the AI ​​analysis device (700) selects a cluster similar to a pest based on the size, shape, and color information of each cluster to determine whether it is a pest.

[0117] In one embodiment, the AI ​​analysis device (700) uses a machine learning / deep learning-based pest detection model (e.g., CNN) to finally determine whether the clustered area is an actual pest. This model takes the features of the cluster as input, calculates the probability of pest detection, and outputs the result.

[0118] In the above embodiment, the AI ​​analysis device (700) identifies the contour, size, shape, etc. of the cluster and includes them in the input vector to analyze the shape of the cluster. For example, it identifies morphological features using area, boundary length, contour features, size, and ratio.

[0119] The above area is determined using the total number of pixels included in the cluster, the boundary length is determined using the length calculated along the cluster boundary, and the contour features are analyzed for shape complexity, smoothness, and convexity using the contours formed along the cluster boundary, while the size and ratio are analyzed for the elongated or circular shape of the pest by calculating the ratio of the cluster's width to height.

[0120] For example, the AI ​​analysis device (700) can determine that there are pests in a certain area if a specific cluster has a shape similar to the color and size of a pest and occupies more than a certain proportion within the area of ​​interest.

[0121] As described above, the AI ​​analysis device (700) predicts the probability of pest detection based on the feature vector input through the model and derives the result. For example, when the output value is [0.85, 0.15], the first value (pest present) indicates a probability of 85%, and the second value (pest not present) indicates a probability of 15%.

[0122] After that, the AI ​​analysis device (700) classifies whether pests are detected based on the predicted values. Using an activation function such as a softmax function, the predicted values ​​are converted into probability values, and the class with the highest probability is determined as the final classification result.

[0123] For example, if the pest detection model is a binary classification problem with two classes, pest present and pest not present, when the output value is 0.7, this means that there is a 70% probability that pests are present, and the model classifies the cluster as pests.

[0124] The model derives the final predicted probability and, based on this, determines the likelihood that the cluster is an actual pest. For example, if the model's output value is [0.9, 0.1], it means that the cluster is a pest with a 90% probability. In this case, if it is higher than the threshold set by the user (e.g., 0.7), it is determined that it has been detected as a pest, and if it is lower, it is determined that there is no pest.

[0125] In addition, the smart farm system is equipped with lighting, and in one embodiment, lighting that aids plant growth is used by utilizing blue light with a wavelength of 400nm to 500nm and red light with a wavelength of 640nm to 700nm, which aids plant photosynthesis. In another embodiment, lighting that can be used as supplemental lighting for sunlight in facility cultivation or as a substitute for sunlight in plant factories is used. Such lighting is a product implemented within visible light, excluding the ultraviolet (UV) portion harmful to the human body and the infrared (IR) portion involved in heat, and is not harmful to the human body or plants.

[0126] Although not illustrated in Fig. 1, the user terminal can access and monitor the AI-based self-sufficient smart farm system.

[0127]

[0128] FIG. 2 is a flowchart illustrating a method for implementing an AI-based self-sufficient smart farm system according to the present invention.

[0129] Referring to FIG. 2, rainwater is collected through a rainwater collection device installed on the top or outer surface of the smart farm system and stored in a water storage tank (step S210).

[0130] Solar energy is absorbed through solar panels installed outside the smart farm system, converted into electricity, and supplied to an electrical storage device within the smart farm (step S220).

[0131] Agricultural activities are performed in a smart farm using a crop cultivation robot, and data on the growth status of each crop is collected during this process (step S230).

[0132] Data collected using an AI analysis device is analyzed, and the operation of the agricultural environment control device is controlled according to the growth status of the crop, or rainwater stored in the water storage tank is supplied to the smart farm as agricultural water (step S240).

[0133]

[0134] FIGS. 3 to 6 are illustrative diagrams for explaining an AI-based self-sufficient smart farm system according to an embodiment of the present invention.

[0135] Referring to FIGS. 3 to 6, a solar panel (300) is formed on the exterior of the AI-based self-sufficient smart farm system as shown in FIG. 3. The solar panel (300) can be installed on the module roof or outer wall, absorbs sunlight during the day to generate electricity, and the generated electricity is supplied to an electrical storage device within the smart farm.

[0136] Additionally, the vertical cultivation area (500) consists of cultivation beds stacked in multiple layers. Each cultivation bed provides space for planting crops and can be designed in a fixed or movable form.

[0137] Each of the above cultivation beds is equipped with a watering system that automatically supplies water and nutrients as needed. This watering system regulates the amount of stored water and optimizes water supply according to the crop's water requirements.

[0138] As shown in FIGS. 4 and 5, the cultivation robot (600) moves according to the position and height of the crop formed in each of the vertical cultivation areas (500) and can perform tasks necessary for the plants being cultivated. For example, it can automatically perform pruning, harvesting, or nutritional status checks.

[0139] The arm of the cultivation robot (600) has joints that can move around multiple axes, thereby allowing it to rotate in various directions or move up, down, left, and right. This multi-axis joint structure enables the cultivation robot (600) to perform tasks accurately according to the various positions and heights of the crops. Additionally, as shown in FIG. 6, temperature and humidity can also be displayed.

[0140]

[0141] Although the present invention has been described by the embodiments and drawings described above, the present invention is not limited to the above embodiments, and various modifications and variations are possible from this description by those skilled in the art to which the present invention pertains. Accordingly, the concept of the present invention should be understood only by the claims set forth below, and all equivalent or analogous variations thereof shall be considered to fall within the scope of the concept of the present invention.

Claims

1. A rainwater collection device installed on the top or external surface of a smart farm system to effectively collect rainwater; A water storage tank that stores rainwater delivered through the above-mentioned rainwater collection device and supplies it to a smart farm as agricultural water; A solar panel installed outside the smart farm system to absorb sunlight and generate electricity, and supplying the generated electricity to an electrical storage device within the smart farm; A crop cultivation robot that performs agricultural activities in the smart farm and collects data on the growth status of each crop during the process of performing the agricultural activities; Characterized by including an AI analysis device that monitors the growth status of crops based on data collected by the cultivation robot and controls the operation of an agricultural environment control device. AI-based self-sufficient smart farm system.

2. In Paragraph 1, The above solar panel is Characterized by having an angle adjustment device formed to adjust the tilt and direction according to the position of the sun so that sunlight is incident nearly perpendicularly on the panel surface. AI-based self-sufficient smart farm system.

3. In Paragraph 1, The above AI analysis device The method is characterized by analyzing pixels of an image collected by the cultivation robot to extract features, dividing the image into a crop portion and a background portion, determining whether pests have occurred based on pixels of the crop portion, and controlling the cultivation robot so that an agricultural environment control device for removing pests operates according to the determination result. AI-based self-sufficient smart farm system.

4. In Paragraph 3, The above AI analysis device Characterized by examining each pixel of the crop portion to calculate the number of normal crop colors and crop colors associated with pests, and determining that pests exist in the area if the calculated number of crop colors associated with pests is greater than or equal to a specific threshold. AI-based self-sufficient smart farm system.

5. A step of collecting rainwater through a rainwater collection device installed on the top or external surface of the smart farm system and storing the rainwater in a water storage tank; A step of absorbing sunlight through solar panels installed outside the smart farm system, converting it into electricity, and supplying power to an electrical storage device within the smart farm; A step of performing agricultural activities in a smart farm using a crop cultivation robot, and collecting data on the growth status of each crop during this process; and The method is characterized by including the step of analyzing collected data using an AI analysis device, controlling the operation of an agricultural environment control device according to the growth status of crops, or supplying rainwater stored in the water storage tank to the smart farm as agricultural water. Implementation method of an AI-based self-sufficient smart farm system.

6. In Paragraph 5, The method is characterized by including a step of adjusting the tilt and direction according to the position of the sun through an angle adjustment device formed on the solar panel, so that the sunlight is incident on the panel surface at a position close to perpendicular. Implementation method of an AI-based self-sufficient smart farm system.

7. In Paragraph 5, A step in which the AI ​​analysis device analyzes pixels of an image collected by the cultivation robot to extract features; A step of dividing the above image into a crop portion and a background portion and determining whether pests occur based on pixels of the crop portion; and Characterized by controlling the cultivation robot so that an agricultural environment control device for removing pests operates according to the above judgment result. Implementation method of an AI-based self-sufficient smart farm system.

8. In Paragraph 7, A step in which the AI ​​analysis device examines each pixel of the crop portion to calculate the number of normal crop colors and crop colors associated with pests; Characterized by including a step of determining that a pest exists in the area if the number of crop colors associated with the pest is greater than or equal to a specific threshold. Implementation method of an AI-based self-sufficient smart farm system.

Images