Computer, method and computer program for learning to determine operating conditions for multiple plant and animal cultivation devices
By implementing a machine learning system that calculates reward values from physical quantities in animal and plant breeding devices, the system can automatically determine optimal operating conditions, enhancing breeding efficiency and outcomes.
Patent Information
- Application Number
- JP2023194869
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2023-11-16
- Publication Date
- 2025-05-14
- Estimated Expiration
- 2039-11-11
AI Technical Summary
The selection of operating conditions for animal and plant breeding devices is currently manual and lacks the efficiency and optimization that machine learning could provide.
A computer system that utilizes machine learning to determine optimal operating conditions for multiple animal and plant breeding devices by acquiring physical quantities, calculating reward values based on these quantities, and iteratively learning the most suitable operating conditions.
This approach enables more efficient and optimized plant and animal breeding by automatically selecting the best operating conditions, thereby improving growth outcomes and reducing manual intervention.
Smart Images

Figure 0007675975000001 
Figure 0007675975000002 
Figure 0007675975000003
Abstract
Description
[Technical field]
[0001] The present invention relates to a computer, a method and a computer program product that learns to determine operating conditions for a plurality of animal or plant growing devices. [Background technology]
[0002] In the animal / plant cultivating device, animals and plants are cultivated based on predetermined operating conditions of the device. In this regard, the selection of appropriate operating conditions of the device is controlled by a person or a computer. Summary of the Invention [Problem to be solved by the invention]
[0003] However, the operating conditions of the animal / plant cultivating device are not selected by machine learning.
[0004] By enabling machine learning to select the operating conditions for the animal / plant cultivation device, plant / animal cultivation can be achieved more efficiently. [Means for solving the problem]
[0005] In machine learning, a reward value associated with each operating condition of the plant or animal growing device is calculated.
[0006] According to an embodiment of the present invention, a computer that learns multiple animal / plant cultivation device operating conditions may include a physical quantity acquisition unit that acquires physical quantities of animals and plants grown based on the multiple animal / plant cultivation device operating conditions, a reward calculation unit that calculates a reward value associated with at least one of the animal / plant cultivation device operating conditions based on the physical quantities, and a learning unit that learns appropriate animal / plant cultivation device operating conditions based on the reward value by repeating operations of the physical quantity acquisition unit and the reward calculation unit.
[0007] In a computer according to an embodiment of the present invention, the multiple animal / plant cultivation device operating conditions are related to a driving power amount and a value related to the amount of light generated based on the driving power amount and / or a distribution of values related to the amount of light, the physical quantity acquired by the physical quantity acquisition unit includes at least one of the mass and an image of the animal or plant, and the reward calculation unit can calculate the reward value related to the multiple animal / plant cultivation device operating conditions based on at least one of the mass and an image of the animal or plant.
[0008] In the computer according to the embodiment of the present invention, the amount of light generated based on the amount of driving power is defined by the photosynthetic photon flux density (PPFD).
[0009] The computer according to an embodiment of the present invention may include a selection unit that selects at least one animal or plant growing device operating condition from the plurality of animal or plant growing device operating conditions based on the reward value.
[0010] The plant / animal growing apparatus according to the embodiment of the present invention is a structure.
[0011] A method for learning multiple animal and plant cultivation device operating conditions according to an embodiment of the present invention may include a step of acquiring physical quantities of animals and plants grown based on the multiple animal and plant cultivation device operating conditions, a step of calculating a reward value associated with at least one of the animal and plant cultivation device operating conditions based on the physical quantities, and a step of learning appropriate animal and plant cultivation device operating conditions based on the reward value by repeating operations of the physical quantity acquisition unit and the reward calculation unit.
[0012] A computer program for learning multiple animal / plant cultivation device operating conditions by a computer according to an embodiment of the present invention can execute the steps of acquiring physical quantities of animals and plants grown based on the multiple animal / plant cultivation device operating conditions, calculating a reward value associated with at least one of the animal / plant cultivation device operating conditions based on the physical quantities, and learning appropriate animal / plant cultivation device operating conditions based on the reward value by repeating the operation of the physical quantity acquisition unit and the reward calculation unit.
[0013] A storage medium having a computer program recorded thereon according to an embodiment of the present invention. [Brief description of the drawings]
[0014] [Figure 1] 1 shows a configuration of a system according to an embodiment of the present disclosure. [Diagram 2] 2 shows a configuration of a server device according to the present disclosure. [Diagram 3] 1 shows a processing flow for realizing machine learning in a server device of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0015] [System configuration and overview] 1 shows a configuration of a system according to an embodiment of the present disclosure. The system 100 includes a server device 105 and an animal / plant growing device 110. The server device 105 communicates with the animal / plant growing device 110 via a wired line or a wireless line. In this embodiment, the server device 105 is installed outside the animal / plant growing device 110, but may be installed inside the animal / plant growing device 110. Alternatively, the server device 105 and the animal / plant growing device 110 may be integrated.
[0016] The animal / plant growing device 110 provides an environment for growing animals and plants. For example, the animal / plant growing device 110 can measure, control, and / or manage physical quantities such as the amount of light, temperature, humidity, smell, sound, vibration, and air volume. The animal / plant growing device 110 can also measure physical quantities such as the size, weight, and appearance of the grown animals and plants. Note that the above examples of physical quantities are merely examples and are not limiting.
[0017] The animal / plant growing device 110 controls the amount of light hitting the animals / plants by controlling the lighting device installed in the device 110 based on information measured by a light amount measuring device or a power measuring device that measures the driving power of the lighting device. The animal / plant growing device 110 may also control the shading of a window whose transmittance can be controlled and / or control the opening and closing of curtains placed near the window based on information measured by the light amount measuring device to manage the amount of light hitting the animals / plants. When the object to be grown in the animal / plant growing device 110 is a plant, the animal / plant growing device 110 can optimize the amount of photosynthesis of the plant by managing the amount of light. With regard to light, the animal / plant growing device 110 may control other devices to manage brightness, illuminance, chromaticity, and the like in addition to the amount of light.
[0018] The animal / plant growing device 110 controls the temperature by controlling the air conditioner and heater based on the temperature measured by the temperature measuring device.
[0019] The animal / plant growing apparatus 110 controls the temperature by controlling a dehumidifier and / or a humidifier based on the humidity measured by the humidity measuring device.
[0020] The animal / plant growing apparatus 110 controls an odor output device such as an air freshener output device based on the level of odor measured by the odor measuring device, to manage the odor.
[0021] The animal / plant cultivating device 110 uses a vibration generating device to vibrate the animal / plant cultivating device 110 itself and / or some or all of the target animals or plants placed within the animal / plant cultivating device 110.
[0022] The animal / plant growing device 110 can manage the volume of air by controlling the wind generating device based on the volume of air measured by the air volume measuring device.
[0023] The animal / plant growing device 110 uses an imaging device to acquire still images and / or videos of the grown animals and plants. This allows the animal / plant growing device 110 to acquire the external appearance of the animals and plants. The animal / plant growing device 110 can also specify the size based on the still images and / or videos. For example, the still images and / or videos may include a reference object (e.g., a ruler, an object of a specific size, etc.) to be compared with the grown animals and plants, allowing the animal / plant growing device 110 to measure and / or estimate the size of the grown animals and plants.
[0024] The animal / plant growing apparatus 110 obtains the mass of the grown animals or plants using a mass measuring device.
[0025] The server device 105 communicates with the animal / plant growing device 110 and can receive physical quantities acquired by the device 110. The server device 105 transmits information required for the animal / plant growing device 110 to manage the physical quantities (e.g., absolute or relative amounts of light to be reduced) to the animal / plant growing device 110. The animal / plant growing device 110 manages the physical quantities based on the received information. In another embodiment, the server device 105 may control the animal / plant growing device 110 based on the physical quantities.
[0026] The server device 105 manages the operating conditions of the animal / plant growing device 110. Operating conditions are prepared for each animal or plant. The operating conditions are conditions related to physical quantities necessary for growing animals or plants, and include some or all of the physical quantities such as the amount of light, temperature, humidity, smell, sound, vibration, and air volume. The server device 105 transmits the operating conditions to the animal / plant growing device 110, and the animal / plant growing device 110 grows the animals or plants based on the operating conditions. The operating conditions may be conditions that change over time. For example, when the operating conditions are conditions for plants, the operating conditions may be set based on the germination period t 1 The temperature suitable for germination is set at t 1 The post-emergence period t2 For example, if the operating conditions are for animals, the incubation period t 1 The temperature suitable for incubation is set at t 1 The post-hatching period t 2 At t , the temperature is set to be suitable for animal growth, and at 2 The egg-laying period t 3 The temperature suitable for laying eggs is set at t for the germination stage. The operating conditions may also be set for each stage of growth of each animal or plant. For example, 1 Operating conditions for the post-emergence stage 2 Operating conditions for the
[0027] The animal / plant growing device 110 acquires the acquired physical quantities related to the growth of the animals and plants and transmits them to the server device 105, whereby the server device 105 acquires the results of the growth of the animals and plants based on the physical quantities. The server device 105 can determine whether the operating conditions provided to the animal / plant growing device 110 were appropriate based on the growth results. Furthermore, the server device 105 can accumulate the determination results as to whether the operating conditions were appropriate and use machine learning to determine the operating conditions appropriate for the animal / plant growing device 110.
[0028] For example, when the animal / plant growing device 110 is growing a plant, the server device 105 determines whether or not tip burn has occurred on the plant based on an image captured by the imaging device. Tip burn is a known phenomenon in which the ends of a plant (e.g., the tips of leaves) turn brown. Tip burn occurs due to factors such as a lack of evaporation of water from the ends of a plant. Thus, in response to determining the occurrence of tip burn, the server device 105 can determine that the operating conditions are not appropriate.
[0029] The animal / plant growing device 110 can avoid chip burn by increasing the air volume. For example, the animal / plant growing device 110 can prevent chip burn by increasing the air volume and blowing stronger air to the plants, thereby promoting evaporation of water from the leaves of the plants.
[0030] [Server device configuration] In FIG. 1, the server device 105 is disposed outside the animal / plant growing device 110, but may be disposed inside the animal / plant growing device 110. FIG. 2 shows a configuration of the server device 105 of the present disclosure. The server device 105 is a computer and includes at least a processor 205, a storage device 210, and a communication device 215. The processor 205, the storage device 210, and the communication device 215 are connected to each other via a bus 250. The storage device 210 includes a RAM (Random Access Memory), a ROM (Read Only Memory), and the like. A large-scale database may be implemented using the storage device 210. The processor 205 controls the server device 105 based on a computer program stored in the storage device 210. The server device 105 can communicate with other devices via a communication line such as the Internet using the communication device 215. The devices 205 to 215 shown for the server device 105 are merely examples, and the server device 105 may include other devices, and some of the devices 205 to 215 may be omitted. 2, server device 105 is shown as being one physical device, but may be realized by two or more computers. In other words, server device 105 may be realized physically by one computer, or may be realized logically by one computer.
[0031] [Configuration of the animal and plant growing device] The animal / plant growing device 110 includes a physical quantity control device including at least one of a lighting device, a window, a cooler / heater, a dehumidifier, a humidifier, an odor output device, a vibration generator, and a wind generator. The animal / plant growing device 110 includes a physical quantity measuring device that measures a physical quantity controlled by the physical quantity control device. The physical quantity measuring device may include at least one of an imaging device and a mass measuring device.
[0032] Examples of the lighting device, which is a physical quantity control device, include lighting fixtures such as LEDs (Light Emitting Diodes), fluorescent lamps, etc. The light amount of these fixtures may be configured to be controlled.
[0033] An example of a window that is a physical quantity control device is a window whose transmittance can be controlled. The window may be configured to block part or all of the light coming through the window with a curtain installed near the window.
[0034] The animal / plant growing device 110 may include an optical measuring device having any one of the functions of a device for measuring luminance, illuminance, chromaticity, etc. related to light, a light quantity measuring device, a power measuring device for measuring the driving power of a lighting device, etc., as a physical quantity measuring device. For example, a photon meter may be used as the optical measuring device. The animal / plant growing device 110 can control the light from the lighting device and / or the window based on the measured value of the optical measuring device. This allows the animal / plant growing device 110 to provide appropriate light to the animal or plant to be grown. For example, when the object to be grown is a plant, light suitable for photosynthesis is provided, and when the object to be grown is an animal or plant, light for simulating daytime and light for simulating nighttime can be provided.
[0035] When the physical quantity control device is a lighting device, the driving power amount and the photosynthetic photon flux density (PPFD) of each lighting device have a predetermined correlation. The amount of light generated based on the driving power amount, i.e., the amount of light generated by the lighting device, may be defined by the PPFD. It is also known that the appropriate PPFD changes depending on the type of plant to be grown. The PPFD value may be the average value of values measured at multiple positions, or may be a value measured at a predetermined position. For example, the PPFD value may be a value measured at a position 50 mm away from the light source in a straight line. The PPFD value may be a value converted into the driving power amount of the lighting device. When the driving power amount is increased, the PPFD value increases, and when the driving power amount is decreased, the PPFD value decreases, so there is a correlation between the driving power amount and the PPFD value. In one embodiment, the correlation may be expressed by a mathematical formula.
[0036] The animal / plant growing device 110 may include a thermometer as a physical quantity measuring device to control a cooling / heating device, which is a physical quantity control device. The animal / plant growing device 110 can provide an appropriate temperature to the animal or plant to be grown. For example, if the object to be grown is a plant, a temperature suitable for germination or growth is provided. If the object to be grown is an animal, a temperature suitable for hatching from an egg or a comfortable temperature suitable for growth is provided.
[0037] The animal / plant growing device 110 may include a humidity measuring device as a physical quantity measuring device to control a dehumidifier and / or a humidifier, which are physical quantity control devices. The animal / plant growing device 110 can provide appropriate humidity to the animal or plant to be grown. For example, if the object to be grown is a plant, humidity suitable for germination and humidity suitable for growth is provided. If the object to be grown is an animal, humidity suitable for hatching from an egg and comfortable humidity suitable for growth is provided.
[0038] The animal / plant growing device 110 may control temperature and humidity in a related manner. For example, when the temperature is high, the humidity may be lowered to provide more comfort. Also, when the temperature is low, the humidity may be increased for the purpose of preventing the occurrence or spread of viruses.
[0039] The animal / plant growing device 110 may be equipped with an odor measuring device as a physical quantity measuring device to control an odor output device, which is a physical quantity control device. The animal / plant growing device 110 can provide an odor suited to the properties of the animal or plant to be grown. For example, when the object to be grown is an animal, an odor that reproduces the odor of the area where the animal naturally lives is provided.
[0040] The vibration generating device may be, for example, a controller for a game machine, a chair for a racing game in an arcade game machine, a massage chair, or the like. The vibration generating device may also be provided with, for example, a floor or a shelf placed under the target animal or plant, and may vibrate the floor or shelf. In growing plants, when nutrients (both liquid and solid) and water are concentrated in a certain part of the plant, applying vibrations allows the nutrients and water to be provided more uniformly. The vibration generating device may be used as a physical quantity control device. Furthermore, the physical quantity control device may have a function of tilting the floor or shelf in addition to the function of the vibration generating device.
[0041] The wind generating device may be, for example, an electric fan.
[0042] The animal / plant growing device 110 may include an airflow measuring device as a physical quantity measuring device to control a wind generating device, which is a physical quantity control device. The animal / plant growing device 110 can provide a comfortable amount of airflow to the animals and plants to be grown. For example, when the object to be grown is a plant, an effective airflow is provided to allow water to evaporate appropriately from the leaves.
[0043] The imaging device can be, for example, a camera, a video camera, etc. The imaging device can also be a device capable of measuring the surface temperature of an object, such as an infrared camera.
[0044] The animal / plant growing device 110 can acquire information on the object to be grown via an imaging device, which is a physical quantity measuring device. For example, if the object to be grown is a plant, and brown is detected in the image, it is determined that tip burn has occurred. If the object to be grown is an animal, and movement of the animal is detected in the video, it is determined that the animal is active, and if no movement is detected, it is determined that the animal is sleeping. The animal / plant growing device 110 measures or estimates the size of the object to be grown via the imaging device.
[0045] The animal / plant growing device 110 can obtain the mass of the object to be grown via a mass measuring device, which is a physical quantity measuring device. For example, the larger the mass of the object to be grown, the more smoothly it is determined that the object is growing.
[0046] With regard to other physical quantities, the animal / plant growing device 110 may be equipped with a sound collector such as a microphone and a volume meter, an audio output device such as a speaker, a carbon dioxide generator and a measurement device, an oxygen generator and a measurement device, etc. as a physical quantity control device and / or a physical quantity measurement device. Furthermore, when the subject to be raised is an animal, the animal / plant growing device 110 may be equipped with a physical quantity measurement device for measuring physical quantities related to medical information. The medical information may include, for example, brain waves, heart sounds, pulse, surface body temperature, deep body temperature, the difference between surface body temperature and deep body temperature, pupil movement, respiratory volume, exhaled breath components, sweat volume, blood pressure, etc.
[0047] The animal / plant cultivating device may be an animal / plant cultivating container installed in a laboratory or the like, a device installed in a vegetable manufacturing factory, or the factory itself. When the animal / plant cultivating device is a factory, the animal / plant cultivating device is a structure.
[0048] [Machine learning in server devices] The server device 105 stores a plurality of pre-set operating conditions. The operating conditions can be information input by a user or information downloadable from another computer. The server device 105 can modify the plurality of pre-set operating conditions. Hereinafter, the plurality of pre-set operating conditions and the plurality of modified operating conditions will be simply referred to as the operating conditions.
[0049] 3 shows a process flow for realizing machine learning in the server device 105 of the present disclosure. The server device 105 transmits operating conditions to the animal / plant growing device 110 (step 305). The transmitted operating conditions may be selected by a person or may be selected by the server device 105 using a predetermined logic or randomly.
[0050] Steps 310 to 320 relate to the operation of the animal / plant growing device 110. The animal / plant growing device 110 controls the physical quantity via a physical quantity control device based on the operating conditions (step 310). The animal / plant growing device 110 grows the animals / plants based on the controlled physical quantity (step 315). The animals / plants are grown for a predetermined period of time. The predetermined period can be expressed in various units such as seconds, hours, days, months, and years. When the growth for the predetermined period is completed, the animal / plant growing device 110 measures the physical quantities related to the grown animals / plants via a physical quantity measuring device (step 320).
[0051] The animal / plant growing device 110 transmits the measured physical quantities to the server device 105, and the server device 105 acquires the physical quantities from the animal / plant growing device 110 (step 325). Here, the animal / plant growing device 110 may also transmit information on the operating conditions used in the measured physical quantities, such as an identifier of the operating conditions, to the server device 105. The "physical quantity acquisition unit" in the claims may correspond to the configuration in step 325 in which the server device 105 acquires the physical quantities from the animal / plant growing device 110.
[0052] The server device 105 analyzes the acquired physical quantities (step 330). The analysis of the physical quantities may include a determination of the cultivation results. For example, the server device 105 determines the cultivation results based on the size, weight, and appearance of the cultivated animal or plant. For example, if the size and / or weight of the animal or plant is smaller than a predetermined threshold, the server device 105 determines the cultivation results to be poor, and if the size and / or weight is equal to or greater than the predetermined threshold, the server device 105 determines the cultivation results to be good.
[0053] The server device 105 generates a reward in response to the analysis of the physical quantity (step 335). For example, the server device 105 generates a reward (e.g., + "1") when the training result is good, and generates a reward (e.g., - "1") when the training result is bad. The training result may also have stages, and may include five stages such as better (e.g., + "2"), good (e.g., + "1"), average (e.g., "0"), bad (e.g., - "1"), and worse (e.g., - "2"). The stages of the training result shown here are merely examples, and the stages may include more than five stages or fewer than five stages.
[0054] The server device 105 associates the physical quantity measured by the animal / plant growing device 110 with the operating conditions used when the physical quantity was measured, and adds a reward (step 340). In this way, a reward value is obtained by adding the reward.
[0055] The “reward calculation unit” in the claims may correspond to the configuration in steps 335 and 340 where the server device 105 obtains the reward value.
[0056] The server device 105 and the animal / plant growing device 110 repeatedly perform steps 305 to 340 to accumulate reward information associated with each of the multiple operating conditions. The "learning unit" in the claims can correspond to a configuration in which the server device 105 accumulates the reward information.
[0057] In another embodiment, before step 305, i.e., before server device 105 transmits the operating conditions to animal / plant growing device 110, server device 105 may select a predetermined operating condition from the plurality of operating conditions. The "selecting unit" in the claims may correspond to a configuration in which server device 105 selects a predetermined operating condition from the plurality of operating conditions.
[0058] For example, the server device 105 selects a predetermined operating condition when the reward value associated with the predetermined operating condition is greater than a predetermined threshold value. The server device 105 also identifies a maximum reward value from reward values associated with a plurality of operating conditions, and selects the operating condition associated with the maximum reward value. Furthermore, the server device 105 may associate the number of times steps 305 to 340 have been repeated with the predetermined operating condition with the operating condition, store the number of times, and select only the operating condition for which steps 305 to 340 have been repeated a predetermined number of times or more. This improves the reliability of the evaluation of the reward value. In another embodiment, the server device 105 may select one or more operating conditions. The server device 105 may present one or more operating conditions to the user, and the user may select one operating condition from the one or more operating conditions.
[0059] [Machine learning based on mass and appearance of plants and animals] In another embodiment, the reward for the operating condition may be generated based on the mass and / or appearance of the cultivated animal or plant. For example, the server device 105 acquires the mass of the animal or plant and / or a part of the image and video image from the animal or plant cultivating device 110 as information on the physical quantity.
[0060] The server device 105 analyzes the physical quantity, and when the mass of the animal or plant is equal to or greater than a predetermined threshold, determines the growth result to be good, and when the mass is less than the predetermined threshold, determines the growth result to be poor.
[0061] The server device 105 acquires, measures, or estimates the size of the cultivated animals or plants based on the physical quantities. When the size of the animals or plants is equal to or larger than a predetermined threshold, the server device 105 judges the cultivation result to be good, and when the size is smaller than the predetermined threshold, the server device 105 judges the cultivation result to be bad. The size of the cultivated animals or plants may be measured and / or estimated based on images and / or video images, which are physical quantities.
[0062] The server device 105 can analyze images and / or video images, which are physical quantities, and determine that the cultivation results are good. For example, if the cultivated object is a plant, tip burn can be detected by analyzing the image. As an example, when the server device 105 detects a brown area around a green area in the image, it determines that tip burn has occurred, and when it does not detect a brown area, it determines that tip burn has not occurred. Furthermore, when tip burn has occurred, the server device 105 determines that the cultivation results are poor, and when tip burn has not occurred, it determines that the cultivation results are good.
[0063] For example, when the object to be grown is a plant such as leafy vegetables, the producer discards the tip of the vegetable where tip burn occurs. Therefore, if the occurrence of tip burn can be reduced, the producer can increase the amount of vegetables that can be sold.
[0064] It is also known that too much light increases the occurrence of chip burns. Therefore, since there are cases where good growth results cannot be obtained even if the light amount is too high, it is useful to appropriately control the increase and decrease of the light amount. Therefore, the operating conditions associated with the growth results may be the light amount, the PPFD value based on the light amount (which may be an average value), and / or the driving power amount of each lighting device that correlates with the PPFD value (hereinafter referred to as "values related to light amount"). The operating conditions associated with the growth results may be actual measured values of values related to light amount, or values related to light amount obtained by simulation calculation.
[0065] Furthermore, it is known that the greater the variation in light intensity, the more likely the occurrence of chip burn. Therefore, the variation in the value related to the light intensity may be used as the operating condition associated with the growth result.
[0066] The variation in the light intensity values may be expressed as a distribution of multiple normalized values measured at two or more different positions at a given height, where the variation is smaller as the range of the distribution is smaller, and the variation is larger as the range of the distribution is larger.
[0067] The variation in values related to the amount of light (distribution of the amount of light) can be calculated by simulation based on at least one of the type of light source (lighting), the number of light sources, the arrangement of the light sources, the amount of driving power for all the light sources and / or for each light source, the position of each light source and the amount of driving power for the light source at that position, the distance from the light source to the measurement position, and the angle from the light source to the measurement position. Therefore, the operating conditions associated with the growth results may be the distribution of the amount of light obtained by simulation calculation, rather than the distribution of the amount of light calculated from the actual measured values of the amount of light.
[0068] In another embodiment, the operating conditions associated with the cultivation results may include at least one of the following in addition to or instead of the distribution of the light amount: the type of light source, the number of light sources, the arrangement of the light sources, the driving power of the entire light sources and / or each light source, the position of each light source and the driving power of the light source at that position, the distance from the light source to the measurement position, and the angle from the light source to the measurement position. Note that at least one of the type of light source, the number of light sources, the arrangement of the light sources, the driving power of the entire light sources and / or each light source, the position of each light source and the driving power of the light source at that position, the distance from the light source to the measurement position, and the angle from the light source to the measurement position may be controlled by a means provided in the animal / plant cultivation device 110 or by a person. The distance from the light source to the measurement position and / or the angle from the light source to the measurement position may be configured to be changed by, for example, changing the physical arrangement of the animal / plant cultivation device 110. For example, if the animal / plant growing apparatus 110 includes a rack having one or more growing apparatus parts with a light source at the top (or bottom) and an animal / plant growing section for receiving animals / plants at the bottom (or top), the distance from the light source to the measurement position and / or the angle from the light source to the measurement position can be changed by changing the arrangement of the growing apparatus parts. For example, a rack that allows the height of the light source and / or the height of the growing apparatus parts to be changed may be used.
[0069] [others] It will be understood that in the above embodiments, some or all of the elements described as being implemented in hardware may be implemented in software, and some or all of the elements described as being implemented in software may be implemented in hardware.
[0070] The above-described processes can be implemented as a method executed by a computer by executing the processes in a predetermined processing order. Also, instructions for a computer to execute the method can be implemented as a computer program. The computer program can also be stored in a storage medium such as a hard disk drive (e.g., the storage device in FIG. 2), a readable / writable memory, or a magnetic medium.
[0071] In the processes or processing sequences described above, the processes or processing sequences can be freely changed as long as no inconsistencies in the processes or processing sequences arise, such as the use of data in a certain process that is not yet available to that process.
[0072] Regarding each of the embodiments described above, a part or all of each embodiment may be combined to be realized as a single embodiment.
[0073] The above-described embodiments are merely examples for explaining the present invention, and the present invention is not limited to these embodiments. The present invention can be embodied in various forms without departing from the gist of the present invention. [Explanation of symbols]
[0074] 100: System 105: Server device 110: Animal and plant growing device 205: Processor 210: Storage device 215: Communication equipment 250: Bus
Claims
1. A computer that learns operating conditions of a plurality of animal / plant growing devices, a physical quantity acquiring unit that acquires physical quantities of the animals or plants grown under the plurality of operating conditions of the animal or plant growing device; a reward calculation unit that calculates a reward value that is set in stages to represent the quality of the cultivation results of the animals and plants cultivated under the operating conditions of the plurality of animal / plant cultivation devices based on the physical quantities, and the reward value is set in stages to represent the quality of the cultivation results of the animals and plants cultivated under the operating conditions of the plurality of animal / plant cultivation devices. a learning unit that repeats operations of the physical quantity acquisition unit and the reward calculation unit, and learns appropriate operating conditions for the animal / plant growing device based on the reward value in association with the number of times the operations are repeated; a selection unit that selects at least one animal / plant growing device operating condition from the plurality of animal / plant growing device operating conditions based on the reward value and the number of times the operation is repeated; A computer equipped with
2. the plurality of operating conditions of the animal / plant growing device are related to a value related to a driving electric power amount and a value related to an amount of light generated based on the driving electric power amount and / or a distribution of values related to the amount of light; the physical quantity acquired by the physical quantity acquisition unit includes at least one of a mass and an image of an animal or plant; The computer of claim 1 , wherein the reward calculation unit calculates the reward value associated with the plurality of operating conditions of the animal or plant cultivating device based on at least one of a mass and an image of the animal or plant.
3. The computer of claim 2 , wherein the amount of light generated based on the amount of driving power is defined by a photosynthetic photon flux density (PPFD).
4. 4. An animal / plant growing device connected to a computer according to any one of claims 1 to 3.
5. The animal / plant growing device according to the fourth aspect of the present invention is a building.
6. 1. A computer implemented method for learning operating conditions for a plurality of animal or plant growing devices, comprising: acquiring physical quantities of the plants or animals grown under the plurality of operating conditions of the plant or animal growing device; calculating a reward value that is set in stages based on the physical quantity so as to represent the quality of the cultivation results of the animals and plants cultivated under the operating conditions of the plurality of animal / plant cultivation devices in three or more stages; a step of repeating operations of the physical quantity acquisition unit and the reward calculation unit to learn appropriate operating conditions for the animal / plant growing device based on the reward value in association with the number of times the operations have been repeated; selecting at least one operating condition of the animal or plant growing device from the plurality of operating conditions of the animal or plant growing device based on the reward value and the number of times the operation is repeated; The method according to claim 1,
7. A computer program for learning operating conditions of a plurality of animal / plant growing devices, the computer comprising: acquiring physical quantities of the plants or animals grown under the plurality of operating conditions of the plant or animal growing device; calculating a reward value that is set in stages based on the physical quantity so as to represent the quality of the cultivation results of the animals and plants cultivated under the operating conditions of the plurality of animal / plant cultivation devices in three or more stages; a step of repeating operations of the physical quantity acquisition unit and the reward calculation unit to learn appropriate operating conditions for the animal / plant growing device based on the reward value in association with the number of times the operations have been repeated; selecting at least one operating condition of the animal or plant growing device from the plurality of operating conditions of the animal or plant growing device based on the reward value and the number of times the operation is repeated; A computer program for executing the above.
8. A storage medium on which the computer program according to claim 7 is recorded.
Citation Information
Patent Citations
Data extraction apparatus, data extraction method, and program
JP2015000040A
Plant raising apparatus
JP2015043715A
Plant cultivating apparatus
JP2016182092A
Plant cultivation apparatus
JP2016198054A
Server device for state notification and state notification method
JP2018037975A